So I have a tabletop fire emblem campaign going, and sure, the role playing and big tactics brain plays are nice, but I swear to god every time a unit levels up, my players sing the little levelup music and go "ding ding ding" for each stat they gain. "Number to up feel GOOD"
@@thedark1owns it's one I've been developing with a very talented friend for almost a year. Based off of 3 houses system, leverages Google sheets for the character sheets + calculations
"Pokemon has random levels ups" Me: *starts writing an angry comment about how pokemon works* "At least to casual players" Me: *slowly deletes comment*
Compared to Fire emblem Pokemon sucks. The fight animations of FE alone are on a different Level, where as Pokemon has still GB Portraits that dont move while fighting. Every Pokemon game feels the same and the battles are horrendously boring and tidious.
@@christopherschlegel6412 your here sweety. Mission achieved. Pokemon is stagnating hard and not much further than red/blue. Its always the same. Fire emblem invents itself new and the fighting is actually enjoyable. As a concept i like Pokemon but it needs changes asap.
@@creativeideas5985 Why didn't you write this on a pokemon based video or something instead of banking on people on a fire emblem video knowing(or caring) about pokemon?
My favorite part of growths is when a character like Lorenz, with decent mag and str supposedly making them variable decide they don't want any of either for 8 or so levels. Please ignore how specific and bitter that sentence is.
Something that's interesting about growth rates is they allow for level ups to be bad, which isn't the case in almost any other RPG I can think of. If you really think about it, a character's level is nothing more than a measure of how much potential to grow they still have. Take FE6 Gonzales for example; you get him on either of the Western Isles routes, but on one route he's level 5 and on the other he's level 11. But his stats are the same in either case, meaning the level 11 one is objectively worse than the level 5 one (although he has the advantage of being able to promote right away). Thinking about levels like this, having a high level is actually a theoretically bad thing in the sense that you can't grow anymore. Meaning a no-stat level up doesn't just feel bad because you didn't get anything, it's worsened by the knowledge that you just wasted a bit of that unit's total potential that they can never get back.
@@seththeace6217 but it does make it more likely for him to compensate for missing with a crit. Also, it gives him a little bit more survivability if he kills with a crit on the first hit, and he can conserve weapons longer the same way.
I'm surprised you didn't mention how Gaiden sets stats on promotion to meet a benchmark rather than adding onto stats. You can get 0 points for 12 levels on a mage or peg knight and still have it be a competent fighter once promoted.
but on the flipside, all your units of the same class feel EXACTLY the same for the most part, aside from when they surpass the promotions base stats. I think a better thing would be #1: do this by unit, this should help with the lack of individuality. #2: It would be good for there to be a controlled random element. I can think of two ways. #1: Give an acceptable range for the target, then look at the average of that stat and set that to 0, you could either evenly divide or you could use standard deviations from the mean. In a simple problem of 35% growth rate of a level 1 unit, the average would be about 6.85 points over base at level 20 with a standard deviation of about 2.08 (it's an approximation), this would lead to a variance of -3.29 to 5.84 standard deviations, and then -3.5 to -2.5 would be a -3 modifier (minimum), -.5 to .5 would be no modifier, and 5.5 to 6.5 would be a +6 modifier. At level 20 promoted it's average would be 9.8 above base stats with a standard deviation of 2.52, this would lead to -5.07 to 6.02, allowing from a -5 to +6 modifier. A couple things to keep in mind is that due to basic probability laws, getting the extreme modifiers would be EXTREMELY UNLIKELY, and would still lead to a fairly minor difference as opposed to a minor modifier. Also while this took a while to manually plug in, it's basic formulas computers could do in a flash. Early promotions could be easily handled by just deducting points (levels left * growth rate rounded up), so promoting at level 10 in this scenario would be 3.5 deduction so -4 in that particular stat. Notably caps do exist and so do stat boosters, there would be a secret value holding their true stats ignoring stat boosters and caps, these would be the ones used for calculation, displayed/functional stats would first have a hidden stat booster modifier applied, and then compared to the caps. #2: Basically take the players current stats and multiply it by a multiplier to get the sum as close as possible to the actual sum total, you could make it weighted if you wanted too, but that's the basic gist of it. Less complicated to explain by a lot but not that much easier to program and IMO less cool and fixes the problem less (bc you can get blessed in a meaningless stat while getting screwed in another, weighting can help this but it's not perfect and can make what's a good or bad level up very weird). Like before there are deductions for early promotions, and stat boosters and modifiers would be applied seperately.
@@lordlouie3550 It would be very easy to implement. Here's a simple explanation: it's basically that the average is the promoted stands, and then basically how much you are above or below it is reduced by a factor to reduce the variance, I just chose the standard deviation because it's an important statistical measurement of variance (though notably this makes anything higher or lower than a +/- 2-3 VERY unlikely so perhaps slightly increase the likelihood of higher modifiers, but that could be complicated to implement)
I like the 3H solution of giving classes a minimum stat base. that way even if you cannot plan around the exact party composition, you at least plan according the minimum of the classes
I think the abundance of the cast is another reason why growth rates tend to work out - you only need *some* of the cast to come out strong, while the rest have a comfortable bench waiting for them. Add in the high quality characters you tend to get at the latter half of campaigns, and you should come out pretty well.
@@auraguard0212 Bases > growths every time. Jagen will shank your unpromoted sword-locked ass with his Silver Lance and Rescue your girl. Just as Kaga intended.
One thing to point out is that, grinding aside, Shadows of Valentia gives you generous base stats for promoted classes, so that ends up making your units RNG-proof to an extent, especially if its a busted class like Dread Fighter with base stats like 36 HP, 15 Atk, 18 Spd, and 7 Mov.
I'd say that Gaiden/SoV take that philosophy so far that it almost doesn't matter what your levels look like until your units are fully promoted, which can decrease the "aesthetic" value (if you're aware of it).
@@alxjones but stats still matter in the class before, since there are still benchmarks and everything that you want to meet. In fact, the better of level ups you get in a first or second tier class, the faster you can snowball and get to the next class. So I think growths can sometimes matter more in Valentia than in other games
@@Edgeperor You’re correct in that good level-ups still matter in the sense that they make the journey to the 3rd-tier class much easier. For example, if Faye gets a bunch of Def level-ups early on, she’ll become somewhat bulky enough to Nosferatu-tank, thus making it easier to gain EXP thus gaining access to Physic and Rescue much earlier.
I understand the value in a system like that but in my experience it just meant I get little to no stats on promo other than movement. I prefer flat bonuses bc I'll always get something. If I'm above class bases and promote I get nothing.
It gets sillier when you start stacking new game plus runs. Nothing like having your spell casters which are typically squishy even with shields effectively double their defense before shields and start marching through enemy lines as destroyers of worlds because most enemies simply do not hit hard enough. Once you achieve that on one playthrough you can just spend the resources to boost their weapon skills so that when they hit the right level then can just promote and then swap back. All thanks to the benefits of simply promoting into a class and the class boosting stats to hit its minimums.
If you just go war master and war monk with a 20 avo battalion, you can do way less investment, have a unit with 5 less avo (but the ability to use terrain tiles) and WAAAAY less investment.
Ok I want to say something real quick. Mozu might not be hitting any benchmarks by gaining stats and you COULD still give her stat buffs with skills, tonics, pair up, meals, etc etc etc. however, Mozu getting more strength on levelups means that instead of, say, spending 200 gold on a str tonic to get her to one round an enemy or even worse giving her a high tier backpack like Arthur or Keaton, you’re able to use less resources. These resources can then be spent elsewhere (usually Camilla, if were being honest). As a result, every stat point you get on a levelup that gets a unit even only a single point closer to a future benchmark is a stat point that you don’t need to provide to them via other means that could also be put towards different benchmarks.
I think in Fates Skill mitigate a lot of issue with low strength/magic bases. Mozu in particular just didn't care if she had bad strength. Mozu have access to Skill like Quick Draw, Bowfaire, Spendthrift and Life & Death. All of those combined and she will deal 54 damage using Crescent Bow. And this is not counting partner seal. If you give Mozu Elbow Room or Trample. You can one shot Takumi with Sorcerer Mozu.
@@riqua27 yes you CAN. The issue is that it takes a disgusting amount of resources. The named strategies were: - Quick Draw (Archer lv 10) - Bowfaire (Sniper lv 15) - Spendthrift (Merchant lv 15) - Life and Death (Master of Arms lv 15) - Crescent Bow (Requires a high bow rank and gold) - Elbow Room (Requires Mozu to marry Silas or worse, Xander) - Sorcerer (Takes S rank support bonuses for Odin or Leo, both good units, off the table) All of these things together let you shit on Takumi with even a base stat Mozu, but you’ve spent: - a minimum of 26 levels, 17 of those in a promoted class (10 to get quick draw and elbow room, promote then lvl 15 for one of bowfaire spendthrift etc and reclass into MoA or Sniper for faire or L&D, then do it again for the final one) - In the best case scenario, 1 master seal, 2 heart seals and a partner seal (The partner seal especially which you could spend on the person Mozu’s married to to instead get THEM into Sniper) - Insane amount of wexp grind or the purchase of one or more arms scrolls - Every single offensive stat booster you could possibly get her With all of these resources by themselves, you could probably get like half your army setup done for late game conquest, using heart seal on shura and Camilla, maybe a partner seal to get Xander into hero for hp+5 and sol, master seal for either Camilla or Xander’s backpacks, fixing up shura’s dagger rank with arms scrolls, the list goes on and on. Mozu getting stats on levelups means that by endgame, she could (and tbh probably would) have gained enough strength that she doesn’t need one or more of the skills given to her (quick draw or elbow room is probably the first one you could drop), and you come out with a lot of extra resources which you could instead use to make your other relevant combat units significantly stronger.
A less extreme example of this resource vs stat gain argument would be an Odin. Lets say you have an average Odin who’s been leveling defense and magic basically every other level. His speed is garbage but also who the fuck cares, he only uses nosferatu to begin with. If you had that Odin with a decent pair up with effie for defense or nyx for magic, you could easily get away with him meeting all his benchmarks up until maybe ch 19 without needing any tonics (except for the first chapter that you deploy him in to level him up since his base level and base stats are below satisfactory). Instead, if you had an Odin super magic or defense screwed, you would have to spend gold to buy him magic, HP, and defense tonics, cooking defense food perhaps, and doing this and that to get him to hit benchmarks instead. Every 2 points of magic/defense that Odin levels up is 400 gold you save a chapter which adds up insanely quickly. With just 2 defense and 2 magic levelups, you’re saving 2000 gold within 5 chapters, enough money to buy an extra master/heart seal.
@@lagspike7763 Of course. I'm being Hyperbolic in my comment. Turning Mozu into Sorcerer also extreamly unnecesary. That's just me being Hyperbolic. I think the more optimal build dor Mozu is to take Quick Draw and Bowfaire. Tht way Mozu can ne used for Dual Strike engine so other unit can get weapon rank and exp easily. Amaterasu also really good. So other unit will stay alive.
Something that I really liked about Radiant Dawn's bonus experience was that, because it would always boost 3 stats exactly, you could manipulate it to cover the weak points of a character who has capped their strengths. For example, Micaiah is notorious for her low HP, Speed, and Defense as well as limited accessibility before her second promotion (among other things), but because she is very likely to max out her Magic and Resistance early, alongside possibly her Skill and Luck, you can essentially force her to gain more HP, Strength, Speed, and Defense by saving your bonus experience for when she starts getting closer to level 20.
Had an interesting discussion about Jo_ about growths the other day, he asked me why people wouldn’t want to play fixed growth Project Anima more than random growths, I think it has to do with the story telling growths gives us, redemption archs, curb stomps and struggles all build in to our personal narratives about the story. Those are so much more powerful than the game telling you, this is the powerful character use them. No one ever says “yo look at this seth” they are way more excited by the stories of axe bros and archers tragic as some may be
You complained about games not having these super difficult final maps, but I’m kinda glad not all of them do. I think an easy final map with epic music and setting is better than something like shadow dragon where it’s not only easy, but probable to get softlocked on the final map of the higher difficulties without warp or a ridiculously strong army
@treeghettox I was playing hard 4 when I did it, and let me tell you it was hell on earth. Did they have forged brace weapons for you? The brave weapons nearly killed all my guys in one round besides sedgar and wolf, and poor Abel couldn’t even survive one round
@treeghettox fair enough, also gotta love how the caps of the manaketes lie on higher difficulties too lol. If you wanna tackle higher difficulties later I recommend reclassing wolf and sedgar to general so you can throw them in the middle of like 5 enemies to get a bunch of chip experience, best way to train them quickly. I think even on normal Medeus is ridiculous, but on h4 and h5 he has like 30 speed and 50 attack. Unless you’re a swordmaster you literally have to have capped hp and defense to survive getting doubled because all the other class speed caps are too low. All I can say is good fuckin luck man
@treeghettox not quite infinite, but 20 turns feels like a fuckin eternity lol. You do have tiki and the Aum staff so he’s always beatable, no crits required. But without warp getting there is the problem.
@treeghettox yeah, in fe3 book 2 you have to use it on the altar where Medeus is, and I think even in book 1 you have to use it in the altar where it is. Since you played fe3 beforehand I definitely understand making that assumption. But in fe11 and 12 you can use it anywhere. Makes a bit less sense but it’s better for gameplay.
@treeghettox you know, if you’re gonna use all three of them they actually have a triangle attack, pretty useful since they’re all available early on. The avatar is pretty shitty, no lie about that but he’s not as prevalent as Robin or Corrin. Good luck with the run though, fe12 is a great game.
tfw no mention of fixed mode from fe9, where every unit always has their average stats the pleasure of watching stats go up without the variance of random level-ups
@@Malhonn AFAIK the amount of growth points you gain when you fight is defined by growth rates, the enemy that you are fighting and the weapon you equip
Unfortunately, the only game that has Fixed Mode is ALSO the game where you can easily reroll levels ups using Bonus EXP on hardware. I'd love to see the feature in other games, but if I'm playing Path of Radiance I'm gonna pick Random growths and cheese bonus levels for good level-ups.
You discussed relatively low stat caps in Thracia as a means of controlling growth rate variability, but this is also true of Radiant Dawn. The devs are able to balance tower enemies around a theoretical maximum of unit power. Standouts like Ike can still carve through chunks of beorc enemies there, but they still represent a formidable challenge to more common player units. The fact that the designers then let you run loose with Laguz Royals is a separate balancing mess. Unrelated, but I think effective weaponry can be counted among the countermeasures to poorly developed stats. Just look at any archer whose usefulness goes up against certain enemy compositions (or in the most extreme case, when wielding crossbows against fliers).
Really enjoyed this one, and don't have much to add. I always imagined stat variance was a massive headache to deal with on the developers' end. Lately we've gotten stuff like Tonics or cooking with the Head Chef to boost stats temporarily, which are nice ways to save a particularly unlucky player but also (from my point of view at least) kind of a concession that IntSys understands just how unlucky a player could possibly get. It's also interesting to see how romhackers deal with stat variance with things like inventory items that boost stats when held.
A FE that can 'measure' how powerful your army is and generate an appropriate challenge would be kinda cool. The downside is that it could probably be heavily exploited, but only the most hardcore would be doing that.
But doesn’t that in a way make the RNG-based lvls up pointless? Like it means that no matter how good or bad your units level ups are they’ll still be evenly matched with the foe. If you get 1 point of Spd and all the enemies get 1 point of extra Spd then that point of Spd is essentially pointless, the only effect being it means your other units have to also reach that new Spd benchmark.
@@CosmicToad5000 Very good point. Maybe in addition to raw stats it would measure how reliant you are on certain units and make counters for them. If you've been stomping the game with Seth the final map would throw lots of horseslayer enemies for example.
@@CosmicToad5000 Maybe you could apply that ''evenly matched'' bonus to some bosses but not abuse it. But yeah, it defeats the whole purpose of random levelups.
It's essentially the hidden % value that determines a character's level up. The higher the percent, the more likely they will level up. Think about Rolling a die with 100- If the numbers you roll are less than growth rate percentage, your character will gain a level up in said stat. That's how critical hits works too. The numbers rolled would have to be equal or less than the number value to work.
Great video. Pretty much agree on all fronts. Designing for Fire Emblem is hard af, and growths only make it harder. But they add so much to that unique Fire Emblem flavor that it's difficult to imagine the series without them. Keep up the awesome work!
@@AlexanderMartinez-kd7cz Depends on who you talk to. I like the idea that one of my units can become stat screwed; I think it makes the game more interesting strategically. It adds another layer of decision making to the game: do I keep using this unit who I know is supposed to be good in the hopes that they get better, or do I bench them for a prepromote with a good baseline but low potential?
@@AlexanderMartinez-kd7cz its not strategic decisions itself, but I like being forced to use units I wouldn't have otherwise. Especially in games where there are a lot of units to pick from, it just adds more variety
The way level ups are doled out is also what makes the level up screen in Mario and Luigi games so exciting. Especially the slot machine. The stats that the slot is giving is largely negligible, but big 4 or 5 gotten from a random chance feels so exciting.
The thing that makes it even better is that you CAN learn it, but it's pretty hard and requires you to take some risks for timing high numbers next to low numbers.
Promoted minimum stats in Gaiden, Echoes, and 3H are all ways the developers can ensure the player has a benchmark of stats to finish the game provided they take the time to actually level their units appropriately. Berwick Saga uses stat bracketing, which provides a minimum and maximum that stats can deviate from the average for each character. I think this is probably the most genius system, as it allows the devs to fine tune the endgame to expected stat thresholds and prevents players from having stat screwed characters, all while providing the satisfaction of the level up screen.
Bracketing system was in Shining Force since the first game thi, albeit in 1 it was a variance of like 10 points. Wich enabled absurdities like Zylo getting 10 HP and 5 attack in a single lvlup.
I've ranted so much about Project Ember's final chapter because of stuff like this. They just made Idunn much stronger without helping you in any way, it almost felt like a kneejerk reaction to her being considered a pathetically easy final boss. Idunn one shotted nearly half of the units I brought with me, but I could still one-round her with Roy, but that's not guaranteed. It's even more upsetting because this could have been prevented with Eliwood and Guinivere joining, but instead they do it two chapters earlier, technically giving you enough room to get them killed. Not to mention Guinivere would have probably been killed in one hit too. So many people just never think about the worst case scenarios. Of course, there's the fixed growths version of it too, which might be why the devs did what they did, but it's not really promoted to be the ultimate version of the hack, and no warning for the normal growths version is given or anything. I'm sure a lot of people really wanted to try the fixed growths version out, but I wouldn't be surprised if most went with the standard growths version anyway.
@@sleepedge8257 making the final boss one-shot half your characters isn't hard. It's just bad game design. Do you consider Medues from fe11 hardest's difficulty well designed?
I think the vast, VAST majority of people forget how statistics work. Yes, a unit CAN be stat-screwed, but if you level them up they're very likely to hit around average, even with bad levels to start. Random level-ups work because of the ability to work around them, and because of the scaling difficulty given to the player. Games without grinding, you just have to scale enemies to the expected averages and things will work out. Weaker teams will play more strategically and be fine (assuming you didn't get so unlucky as to have EVERY SINGLE UNIT stat-screwed), and stronger teams will win easier in the moment but aren't guaranteed to continue winning, and even better it rewards those good units' levels with being stronger than the enemies. That's the whole POINT of random stats! And beyond that, with a shit ton of units at your disposal, you're likely to get good ones you can stick to, and bad ones you can leave behind. I think giving things like infinite grinding is a bit odd as it leads the player to "craft their own difficulty", which typically means "grind until they stomp everything, or don't grind at all", but I still think that that's fine as that's how the player decides to play the game (though there are many players who grind and then complain about difficulty, or vice versa). In any case, I'd say growth rates are NOT bad, they're at their most difficult to predict in the earliest parts of the game, but since you're also given units in most of those chapters (who are themselves capable of doing well in the map), it's okay to base those off of averages as well. Level design doesn't need to be fundamentally impacted by the standard variation; and I'd argue that it isn't influenced by how many units to choose from either, as most units have archetypes and maps that benefit those archetypes don't REQUIRE those archetypes to function in any case.
"if you level them up they're very likely to hit around average" Have you done the math? Can you calculate the exact probability you mean when you say "very likely?"
@@WildMatsu Depends on the level, depends on the exact rate and depends on how you treat "around average". For, say, a 50% growth to hit 20 times in 39 levels (20/20 from level 1), it's about 12.5%. To hit anywhere from 17 to 22 times, it's a whopping 65%. For rates like 25%, 7-12 growths in 39 levels is ~75% likely. Variance is to be expected, but for multiple units to have key stats borked, is unlikely. With higher growth rates and more levels (like awakening) the odds only slim further. In around 50 levels, with a 75% growth rate and a cap about 30 levels higher than base, there's about a 2% chance it won't hit its cap. All in all, odds of a unit having at least one non-average stat (including too high) is decent, but if you say odds a character has at least one screwed stat specifically, you can approximately halve the loss odds. And if you train more than one unit, the bounds are even greater; and considering most casual players won't be speedrunning with a single unit, the odds of being screwed on all of them are very low. I can do calcs for multi-units on key stats if you'd like.
I absolutely loved my playthrough of FE9 with Fixed Growth Rates. Guaranteeing units to be good, so long as they are leveled properly, feels so damn good that I wish other FE's would allow a Fixed Growth feature.
I mostly play GBA Fire Emblems and have found the experience much more enjoyable when I patch the games to have fixed growths/average stats. I get to focus more on strategy and planning knowing what I can expect from each unit. The games already offer a lot of variability for me just by choosing to use different teams and even subpar units. And when I want a crazier experience, I'll straight-up use a randomizer.
@@pascalandraos Definitely, I do the same. I just wish we could eventually get an Awakening/Fates/Echoes Fixed Growth Rates patch. That would be a freaking romp
One very minor but pretty cool thing I’m reminded of it’s is in FE10 when Sanaki gets the Rudol Gem which gives her +10 def when held and is locked to her inventory right before the final map. I think this was to ensure Sanaki could survive Ashera’s physical AOE attack which she may not otherwise due to poor Hp & Def bases & shaky growths. Now they don’t get the same thing to Micaiah who’s forced but hey they did it for one unit.
Nice.This was a really interesting topic. Thought I have to admit I kind of like the chapters when you face a lot of damage sponges. Its fun (to me) to face a wall of generals with pavise and hit them with a forged hammer with Cherche and see how they still died despise all the extra defenses they put in place to avoid just that. I don't know I just like to see this ridiculous 70+ hp bar struggling to get to 0 before the battle animation it's over, because your characters can hit with the wrath of God.
I think FE should actually give up actually making the scenarios and strategies revolve around "stat checking", as it were. Not because of the idea of the level ups being random suddendly becoming a problem, because let's face it, is not. Players get stuck on maps because sometimes they made bad resource management a thing, such as not promoting actually good characters(characters that are useful promoted, like they gain access to staves or can actually use multiple weapons or they gain great movement options OR like most modern FE do, actually give them skills that can change how you approach maps) or not understanding how to make use of money to buy things at shops that matter(like Keys, Crit Weapons, Torches and Potions), making armies be really one dimensional in strategies(let Lord Dude Phalanx the attack head on and pray he doesn't get critted halfway through is so BartreBrained) and so forth. And is not like they haven't tried to do this sort of thing, where you can be a base stat chump be a hero because the map is designed to be beatable with strategy. Things like Bosses being weak or strong against certain damage types(like a Boss stands on a Throne with a Melee Weapon and a Ranged one) and gets cooked by Magic or has to break his Melee and then fight him with a much worse Ranged weapon(like he can't Range 1 attack so he just gets bulldozed at Melee). Escape maps where the idea is not to fight every single enemy, but to fight the enemies that matter and, when the player takes too long, force him to either leave or die, Defense maps where the idea is to protect a point and enemies can be "holded" or are just chump as fucc, like they actively ignore you for the load you have to protect and you can just chomp on these enemy chumps no issue even at level 1, so it becomes a map where the point now is to actually protect the load and so forth. And of course, actually making enemies use the same resources as you. It doesn't matter if a Enemy is a Bishop if all Bishops do all game is use D Tier Light magic. And in the same can be said for combat only units. Paladins having access to all weapon triangle is dumb if all they carry is just Steel Lances and maybe a few Killer Lances.
Your comments on escape and defense maps is exactly why I find a lot of conquest maps so fun. The eternal stairway being one of the few escape maps that actively push the player to escape lest they be overrun by lots of no-exp enemies that stack Chip damage.
Maybe I missed something, but how do I even know what my units' growth rates are without going to am external website to look up the tables. I can't know ahead of time what a new character is likely to be good at. And even after a playthrough, I can't be sure character X has good strength growths, or if he just got lucky with the rng last time and benefitted from being in a class that gave a bonus to strength growths. Engage does stuff like give you Anna, with the highest magic growth in the game but starting in a physical class. It's a fun idea to let players figure out that it's best to send her to a magic class, but I don't see where I would get the information needed to make that analysis within the game.
Scared stones allowed you to see the growth rates via select on the stat screen for some reason they removed this feature even though it would be just as easy to implement
I'm surprised that you didn't approach this from the angle that due to growth rates each player's run of a Fire Emblem game will be that much more unique, to tie it back into the permadeath video. Like person x's playthrough could have a terrible Tana but person y could have Tana as their MVP. Person z got her killed in chapter 13 or something idk.
it makes it highly replayable. similar to hit and avoid your firing your characters like a viewer of a sports team on to do good but your not sure of a win. Its way simpler than he approuches in my not humble but very humble opinion
Some kind of smart rate rigging system would be viable. Have a baseline stat curve that a character should follow (either as a combined total or per individual stat) and then characters below that have their growth rates tweaked to improve chances of hitting that baseline. It wouldn't have to make a unit viable, if anything the baseline should be a few points behind the expected played stats. That expectation could be made up for by skills, weapon ranks, stat boosters, etc. For harder difficulties, in addition to stronger enemies, reduce the baseline system further or remove it entirely. In addition, if some units are stat screwed, the system could accelerate other units' growths, meaning no player should be soft locked by too many unviable characters.
Joke's on you, i'm not juggernauting Awakening with a small army. I'm Juggernauting Awakening with a massive army cuz i've got nothing better to do than grind up like 40 different characters
6:14 Three Houses Their disappointment is MEASURABLE “I got cocky.” “I HAVE NO TIME FOR FAILURE!” “I am not setting a good example...” “Everyone has their off days!” XD
This is kind of why I really enjoy fire emblem games which allow you to recruit several units long into the end game since it kind of gets rid of the problem with growth maybe your Soren had really terrible growth but Hey you'll get another mage in the late game that probably help out with that so you know.
I'm surprised you didn't go over the concept of Berwick Saga's bracketing, a mechanic which prevents units from straying too far from their stat averages while still providing that aesthetic pleasure of seeing your units grow along with a bit of randomness. From what I've found they seem to stay around 3 more or less for most of their stats and this doesn't seem to be effected by statboosters. I know kaga saga isn't technically fire emblem but it's similar enough and.. has growth rates? Would you like to see this in a new fe and do you think it would make it better? And I completely disagree with the idea that every stat that isn't reaching the benchmark for something is pointless. They are important to reaching that threshold in the first place and can still have value in specific scenarios. Sure, being able to double all the enemies on a map is more important than being able to double a single enemy, but being able to double or kill that one enemy can have much more merit depending on the situation. Enemy stats can differ in very small increments and being one point short of doubling can make the be the difference between life and death is some battles. I feel like I have to mention the small stat boosts you get like avoid from speed or crit from skill but those are so little I feel like they barely even count. While the benefit from a single stat might seem minor and the likeliness if having that matter much in the grand scheme of things, having the mindset that it CAN matter and that every stat is useful has given me a lot more enjoyment from these games. ps you cant just put that funky earthbound music its tooo goood
Benchmarks are weird in FE because most of the games have very sharp benchmarks. 27 speed in RD takes a character from doubling basically nothing to nearly everything. It’s weird how truncated many of the stats are. I guess that “pointless” was the wrong word, but rather, “not immediately useful for the majority of enemies.” I think a Berwick Saga approach like what you point out would be a good idea in a game like Three Houses where basically every character is a growth unit and is meant to be carried through the whole game. All EB music is too good.
I've never understood why games like FE have stuck with these growth rate systems for so long when there are such simple changes that can basically fix all the problems they have. For the simplest option, we can look to games like Chrono Cross where characters have an "expected value" for each stat at each level, and the further the actual values are from the expected value, the more odds are tweaked to get things back on track. As in, if you miss your 30% stat up five times in a row, it'll have increased to 35, 40 or more to help make sure you never fall too far behind. Conversely, if you high-roll that stat gain five times in a row, you might only have a 20% or 15% chance to hit it again, that way you never get too strong to trivialize things purely off RNG. A system like that allows there to still be plenty of RNG and variation to keep things interesting yet also allows devs to have a reasonable expectation of how strong characters will be at any given level which they can use while designing levels.
@@user-ed4nb4ef7w If you ever play the game Chrono Cross you'll see that despite having a system like I described, there's still a great deal of variance in character stat growths from run to run. It works for two reasons. One, you can choose how much you want to weight the bias in question. Two, the impact of it is greatest at the highest levels. Even if characters wound up with the exact same stats every run at 20/20 (and I'm not suggesting going that far), that's still ~40 levels with variance in them. And how many of your units will even reach 20/20?
@@NightRaven5568 I like RNG in my hit/miss chances for much the same reason, but that doesn't mean I dislike the fact the game doesn't use true RNG for them. The same would be true here. Adding some bias toward the middle/average values wouldn't render RNG irrelevant. It would just tone the RNG down a bit.
I think something that's kinda important is that the more levels you go through the more the levels should even out, this means that the early few levels can be a wild west where one character can have a string of 3-4 really good or really bad level ups now you aren't getting these stats back, if you should have got 5 speed from your first 10 levels and you got zero it's very very unlikely you are going to get 10 speed in your next 10 but that 5 speed as a percentage of your total speed is huge. Something that can also offset the rng a bit is if you have 2 very similar characters and the game expects you to have one of that type at least that is good, I think Alan and Lance in fe6 are good examples, if Alan gets speed screwed you might have to dump him but at least you still have lance (and the cavs that come shortly afterwards) and if Lance gets strength screwed you might have to dump him but at least you have Alan, sure both can happen but the more times you roll the dice the closer you are going to get to expected outcome
One thing I think would help mellow out this issue is if Fire Emblem did what a lot of traditional RPGs do, where the level is actually put into the battle calculations. I.E. things such as hit, atk, avo, and crit would increase as level does (Obviously not like by 1 for each level. But depending on the stat.) Also hi, I'm new here, I just came from your PvP with Mekkah. Good content. :)
Another solution is the point system, where it just adds points equal to growth rate and when points equal 100 then you get a stat point, but considering that it's optional you still can't base the levels around it. Also it lack ANY sort of variance, so perhaps you can get up to +/- a few points randomly. Alternatively, you can add or subtract the chances levels up will occur dependent on how blessed or curse you are.
I can think of ways to fix level ups, Permadeath on the other hand, means that any map has to be beatable even if you only have half the number of units your even allowed
Having a garbage unit with bad bases and bad growths randomly be blessed and carry the game is what make growth rates so fun. I always remember that specific run whenever it happens.
This is the conclusion I came up with when I thought how I might go about making a fire emblem game: 1st: The only stats that consistently grow when you level are HP and Skill. The reasoning behind HP is simple in that your characters will scale with enemy damage and enemy stat increases. The reasoning behind why skill grows is simple, as you gain real life experience you become more skilled. I would actually put the baseline of skill across every single playable unit at 50% per level. That means that hit rates and crits are changed. I would make crits skill locked behind an adequate threshold and in addition to that lock it to specific classes like Priests and Swordmasters. Priests because yadda yadda divine intervention and Swordmasters because Swordmasters are actually supposed to be the most canonically skilled in the series' lore but in the games they are consistently some of the worst units almost always. The real modifiers I would make as it pertains to stat ups is gear. The thing I love most about FE is the fact that the earlier entries in the series kept numbers low and managable. It actually matters when numbers only go to something very easy to identify like 20. I would make it so that the best way to alter characters stats are to wear armour that takes up an active inventory slot and weapons that actually give bonuses to stats in addition to making S ranks on weapons change depending on what weapon a unit was carrying. For Example: A Sword unit with S rank weapon level would have +1 to every stat with an Iron Sword. +2 defence with a Steel Sword. +3 Strength with a Silver Sword. +3 Skill with a Killing Edge., or +2 Resistance with a Rune Sword. It is important to be skilled and mindful of your money as your weapons do have durability but not in the traditional way FE usually does it. The best way to make an in series comparison would be to akin it to FE 5's fatigue system. If you take a single weapon into 4 or so chapters that weapon will break after the 4th. So making careful calculations with your funds is very important as wasting it means that you won't be able to buy valuable armours and weapons. In doing so each individual weapon would cost a lot more but in exchange even something as simple as an Iron Sword is very powerful with S weapon ranks so you want to keep good track of even iron weapons. If I were to make the game like that I could manage difficulty very easily as the only thing I would have to worry about is hit rates and HP. I can scale very well by scaling to individual character stats as it pertains to possible items the characters may have purchased from the armories and blacksmiths. I fell in love with FE because of growths but I now find it consistently lacking when playing the end of these games because the early chapters are almost always so much more tight and tactical. I would make an FE were every single in game stat benchmark was determined by gear you could actively influence with decision making as opposed to uncontrollable variables. One of the defining reasonings I came to when I thought about it a while back was how much I liked the Fire Emblem 4 supports. When Lex finds the brave axe by the sand is cool. When characters have a support and an actual stats raise I find those things far more intriguing than the level up screen. The player has full control over those and by designing around them you could make levels far more tight. TLDR: Stat variables in my FE would be determined by items, weapons, and supports. Only HP or Skill could ever level up when a unit levels. It would also teach the player that promoting is valuable far more than any FE ever has period.
I prefer the randomness of growth rates because it allows for individuality among each playthrough. However FE needs to allow for growth modifiers like bands from FE9 but better.
I honestly don't get the whole thing about the _wide range of stat variance_ you talk about when designing for late game. Fire Emblem doesn't use arcane methods for determining stats, it's a basic probability. The way these probabilities function means that, over millions of players, there is going to be an average spread of stats that the designers are able to assume a character will be at. I think what would fuck with that are the more robust class systems the newer games tend to have, allowing for more variance in stat growths that the designers might not be able to 100% account for. There are still ways they can mitigate that, though, e.g. weighting growth rates to be less class-dependent and more character-dependent a la Three Houses, or having only a small number of classes to switch between per character/class a la Awakening.
@@Edgeperor Yes, I know how a normal distribution works. I also know how a normal distribution is, well, distributed. 50% of players will be below those averages, but the majority of players will still be around those averages.
Thank you! Someone else recognizes why the gorgon map is the best SS map (at least in theory) I would love to see a Fire Emblem game with totally predetermined growths - i.e. at level 5, the lord will always get +1 str, spd, luck, res. At level 6, they will always get hp, skl, spd, luck. This gives the excitement of seemingly random growths to players playing blind, and gives IS the knowledge of every units' potential stats to plan around for late game. It would also mean that the levels 11-20 for unpromoted units (in a standard FE promo system) can be weighted to either encourage or discourage early promotion. Just an idea that I'd like to see implemented, or at least explored in an optional setting or even fan hack.
One thing that could be done with fixed levels to encourage early/late promotions even for blind players is to have stat caps per character instead of classes (maybe stat boosting items can go over them in that system, maybe they can't, both have advantages). And that way you can physically modify how early some character peaks some key stats. Like, if you want one myrmidon to be trained to level 20 and the other early promote, have the first one cap strength at level 20 and make the other cap it at level 12. (this would work with class specific caps too, just easier to balance with character specific). That way you know the first one got a lot of room to grow much stronger if you take the time while the second myrmidon should be promoted if you see them struggling to deal enough damage, and you don't need to learn about how the devs calculated the stats, it's just very easy visually. That's one of my major problem with stat growth that isn't talked about in this video, it's invisible. Like, it's important enough that it can influence your decision of what character to train or not, but you have no way to know without looking it up. And then once you start looking it up it becomes hard to not be caught up in optimizing your characters, exp, etc, rather than simply using the right character for the job you have ahead of you, which makes the game less enjoyable to play. Weell that's a broader issue caused by the way level ups work in fire emblem, but hidden growth rate really emphasize the issue for me, as the game isn't designed around you knowing that but looking it up is extremely easy and can change how you approach things a lot.
Imagine a (relatively short) hack where every character's growths are 100% across the board. No luck. No randomness. No uncertainty. Boring? Maybe, maybe not. But in theory, it would make maps a lot easier to design, so they might be better.
I think you're overestimating the effect of growth rates here, it's a relatively easy activity to figure out a confidence interval for where a character's stats will be at level x
Ooh, I'm reminded of this game devil survivor's growth rates. In that game there's 4 stats (strength, magic power, defense and avo..... essentially) and at each level up you gain one of these. All the other characters choose that point they gain each level, but your player character gains whatever you want them to gain. So basically instead of the game screwing you over, you now have the capability of screwing YOURSELF over! Anyways, really interesting video!!! I definitely have nightmares of my favorite growth units becoming useless and dying due to screwed stats. I was lucky(?) in my fe8 run because my Franz became one of the godlike ones, but in return my Eirika gained 0 strength for something like 10 levels. Whose decision was it to make lords with bad growths lmao As annoying as growths can be, I do love how runs become so personalized due to them... I guess the question becomes how to make endgame doable when it's possible for you to reach it with just the lord and maybe 3 other guys 😂 maybe IS should check how many units you have and send you a couple friends if your units' levels or the number alive are below a certain threshold haha
They tried giving you extra units if your army was too small in the Shadow Dragon remake and apparently it pretty much trivializes the highest difficulty. (I don’t know myself, but I’ve been watching Excelblem’s Shadow Dragon Ironman recap clips. From what I’ve seen, the system is very abusable.)
@@nanowasabi4421 interesting!! I've played mostly newer fes (8-10, 3ds, 3h) so I didn't know how those panned out. But it sounds better than getting softlocked at least 😂
Growths are like a drug for anyone who is cursed with being a fire emblem fan, even when you recognize their flaws watching your favorite little skrunkly level strength, speed, and defense just makes you love em
Also wanna mention how I think growths are usually a one or none thing is probably because they wanted the player to be able to still calculate and plan ahead with a good enough idea of how things should go without requiring the use of a calculator if the stats were much higher
Three houses has some mitigating factors in it as far as helping you get over bad growths. Passing certification exams will automatically elevate a unit's stats to a baseline if they're not already met. For example, certifying Mercedes as a Fortress Knight will give her a massive bonus to defense. These bonuses are *permanent*, so even if you swap Mercedes back to a Bishop afterwards, she keeps all those bonus points. In this way, you can take any unit lagging in a certain stat and guarantee they can at least achieve an expected minimum. Additionally, Three Houses also does not allow for blank level ups, and will guarantee you at least 1 stat up.
I prefer the gains from games like steamworld heist, where you get a fixed amounth of extra perks, it allows for better level design, and requires smarter play, since every unit is guaranteed to have a niche and role.... But damn, turning a blessed gaius into a myrmidon on a forest tile to see him go ham on on enemies without being hit once *is* fun too I also hope the clip of gilliam being crited into the arena becomes a staple. I wish to see this clip in every odd video of yours now
i love athos. He contradicts everything what you would see usually in japan culture based games. An westernized bearded wizard, comparable to gandalf. Who can solo the last map....lol. and carries around artifacts for your heroues next to the mastership of all magic classes allowing him to use all of the great books recieved.
Growth rates should be PSEUDO-random, as in if your average growth rate is significantly below or above what it should be the odds should be tilted in order to fix that.
What I was thinking might work is have units have three or so stats that are almost garunteed to level up (80% +) and have there other growths be abysmal (30% -) this way units are predictable in what they can perform with some level of variance and making really good level ups more exiting. I think a good example of this is aran from FE10. he has a 70% growth in str, skl, def, so you know and devs know that aran is likely going to be a hard hitting resilient Frontline with little variance.
As usual you dropped another banger video. I disagree with alot of your takes but you generate good discussions and question things no other fe tubers do imo. Also the background music is absolute fire.
Once I started using emulators, understood what savestates were, and randomizing games, I realized how much I cared about making sure I got good levels. My Fir actually turned out better than Rutger, through MANY savestates to get more than skill and luck. But I really appreciate them in normal playthroughs, just because it adds that element of chaos that these games desperately need at times
I think a lot was overlooked here. The entire video is pretty much bashing the existing system, exaggerates the problem, and doesn't take a serious look at the benefits, instead opting for "growths are exciting!" It also does not compare the current system to alternatives. The most obvious reason I can see for the growth rate system is to keep the game fresh. On any given playthrough, I cannot rely on any one character being good, so I'd be likely to change up my team based on turnout. It forces the player to improvise based on whatever stats happen to come their way. Additionally, the issue of being stat screwed, and of the developers being unable to design around how powerful a character's team will be is greatly exaggerated. You can certainly have bad turnouts for a few characters, but this simply won't be the case for your army as a whole. Given the size of the team you use and the number of level ups throughout the game, the total stats gained across teams with similar growth rates will be very similar as a whole. The difficulty in balancing the game is caused more by the double strike threshold, and the combat system being addition based. High defense values can make characters basically unkillable by low power units, and as stats grow, there's a wider range in values, which leads to very high and very low damage spreads, which is problematic. The difficulty in designing around double-strikes should be straightforward. Pokémon uses multiplication based combat, which accommodates high and low stats much better, but it also makes things difficult to calculate. Fire Emblem needs battle outcome calculations to be easy to assess for the player as the consequences for making mistakes can be severe. Finally, we look at the alternative systems. Fixed growth rates function pretty similarly. It prevents stat screwing, but ensures a reliable path forward for the player so they won't need to improvise. A no stat growth system would make it easy to balance battle difficulty, but removes the player's strategy of balancing good growth rates vs. good initial stats. The player simply selects the strongest/most situationally appropriate units. It would be much like pre-deployed maps in Advance Wars. This isn't really sufficient to base an entire game around, so Advance Wars adds resource management. Lastly, we can reduce the impact of stat variability by working with larger numbers, like most traditional RPGs. Each level up gets you say, 3+/-1 points of strength, so a low and high gain wouldn't be too different. This is kind of a middle ground between fixed and random growths. It's not completely reliable, and requires some improvising and replayability, but not drastically. The main disadvantage of this system is that it leads to large numbers overall, which can become difficult to calculate. Figuring out "these 3 enemies have a total attack power of 73, will attack 3 times, and my character has a defense of 15, so the maximum damage my character can take is 73-3*15=28" is doable, and the type of calculation a good player will be doing regularly. "There 3 enemies have a combined attack power of 281 and will attack 3 times, my character has a defense of 36, so the maximum damage possible is 281-3*36=173" is substantially more cumbersome, so the player would then be more likely to make ballpark estimates, which leads to more deaths. So in any case, criticism of the existing system is fair. But it's not without merit, and I can't find any alternatives that are strictly better (I personally advocate for random growths, but seeded at the game start to prevent save-scumming. Also, generally keep defenses low due to the addition based combat.)
This video is fantastic and very well thought out, accurate and thoroughly developed. And I say tha completely disagreeing with your view. Yes, growths can be annoying and rewarding. Sometimes both. But this is exactly what gives the player the freedom to beat the game how they want. There is no set path. Nothing says you have to use or level certain characters to reach the final chapter. This is why I find the "lord defeats antagonist and only the lord can do it" a terrible mechanic, even with a solid lord. Because it removes that freedom and insists you use a specific character with a wide variable of stats after 20+ levels to kill a boss with fixed stats. The beauty of Fire Emblem is being able to steamroll the game with a diligently raised and lucky Florina if I want to. I can raise a balanced army, or I can train 1-2 decent characters, get lucky and annihilate the opposition. OR if I get unlucky with a supposedly good growth character, I can change plans and train someone else. It adds so much replay-ability and characterization to a game that is highly character driven. It also subsequently allows character that would normally turn out average-bad, like Mekkahs Speed-Bartre, to become exceptional given luck.
I used to love growth rates until I came across Awakening Lunatic. When your units get stat screwed, you might as well let Robin do all the work because going against strong and speedy units like Heroes are just a nightmare to deal with.
In Fire Emblem, and Tactical RPGs generally, there are two forces in tension: dramatic uncertainty and planning. They begin as something close to a puzzle game where many (but still not all) factors are definite and known, and gradually diverge toward a large field of uncertain possibilities, mostly due to growths. That means the early game can be a precisely tuned experience (though sometimes it's Awakening Lunatic+ instead), but has limited replay value. For late game, if the game isn't going to be dynamically procedurally balanced (which would be possible, but exploitable), and isn't going to give us granular difficulty options, designers have to target the difficulty of a one-size-fits-all solution for each difficulty level against some set of “reasonable” expectations. They guess. I assume they aim well below theoretical maximums most of the time, and that is why Fire Emblem's hardest difficulties tend to become easy to outscale in the late game (and are not like Apotheosis). The interplay of experience gain, over-leveling, stat caps, stat growths, RNG abuse, and boosters (including equipment, rallies, etc.) allows you to put units significantly above the difficulty curve. And you usually only need one such unit to dominate the game. But ultimately, Fire Emblem is not designed to beat you. That's the illusion. It is designed to try to put up a competent fight. The player is supposed to win, and should be able to win in a variety of ways. That (partially) excuses the imbalance of players becoming OP. And on the flip side, Fire Emblem's gradual trend toward anti stat-screwed mechanics over time is promising. I do wish IS gave us more options to tailor the experience beyond self-limitation, like an average stats or 0% growths / no experience mode, for players more interested in the challenge than the victory.
I was wondering why my Mercedes has a low attack stat since I was trying to make her a physical hitting unit but then I remembered growth rates are a thing and then figured out she had a 25% str rate 😐
I wish they just had a mechanic where you should have minimum stats for level intervals. Like if you're not at 12 strength by level 14 the game keeps pushing you to get that.
Fe on lunatic i think fixes a lot of the problems i have with level ups. A lot of my frustration is when you get a really good level up and then die and have to restart. On lunatic your characters get a fixed level up seed. Levels are still randomized, but your characters levels are determined the moment you get them. So you can't lose a good level up, but also cant save scum when you get a bad one
There needs to be a way to control the environment more. To me, this is the hall mark every good strategy based rpg, the ability to execute crowd control through AoE spells. This should be the real utility of mages, instead of just being archers who target a different stat. Why can’t characters who control the elements use those abilities to control the environment? Obviously you would have to figure out how to balance that, but you could easily tie it with weapon usage.
There’s this cool unit called every other Tellius prepromote who has good bases and growths thus destroying the need for Ulki and making me use palette swapped Hugh
@@Edgeperor I’ve taken longer than I,I’d care to admit trying to figure out who palette swapped Hugh is. Is it Bastian? Soren? I love the Tellius cast but I legitimately don’t know who you’re talking about lol.
Ngl when team building I always get stuck on growths. This is usually because I design for endgame, though funnily enough I don't consider whether I'll get screwed or blessed. Maybe the common denominator would be fixed growths + a small % to add to the other stats.
You know I’ve always thought, what if we could devise a system that scales enemy stats based on how RNG blessed or screwed the player’s units are. Like if you keep getting good levels the stats fo future enemies slowly increase. I suppose that in a way is the same as just removing RNG together as it means getting RnG blessed or screwed doesn’t change anything. A tricky problem to overcome.
yeah, what's the point of leveling at all if it doesn't change anything to your relation to ennemy units? at this point you might aswell remove levels entirely (which honestly might not be a bad idea).
Yeah I definitely resonate w this video. My most recent play through of DS Shadow Dragon saw my fully leveled Marth ending the game with a whopping 10 strength. Only was able to beat the final boss on a lucky Excalibur crit
On another note, Clash isn't actually just a red vs. blue killing field where red is just a ton of mooks. I mean, it is, but it isn't. Particularly on high difficulties. Clash introduced a second game over condition while also oriented on this flat, open prairie such that you had to approach your objective diagonally. This diagonal funnel sees you lodged against two flanks, and if your units aren't sufficiently powerful enough or you leave gaps, you will be pushed into, and your second, fragile game over condition will net you a loss. On lower difficulties you just send a strong unit in each direction and the map pressure tends to not exist. On the Japanese Maniac mode, there are additional pressures that prevent this. Chiefly a large influx of siege tome users. Also, though, more reinforcements and more promoted reinforcements than usual. Due to their enhanced numbers and stats, you feel the actual map pressures, and you have to find a way to redirect enemies away from your new game over condition, while pressing out as hard as you can. But even with an army of gods, you have one Achilles heel that you have to account for. I think it's one of the best designed straight up open maps in the series for that alone.
The Gaiden system fixes rng screwed players for the most part but on the flipside, all your units of the same class feel EXACTLY the same for the most part, aside from when they surpass the promotions base stats. I think a better thing would be #1: do this by unit, this should help with the lack of individuality. #2: It would be good for there to be a controlled random element. I can think of two ways. #1: Give an acceptable range for the target, then look at the average of that stat and set that to 0, you could either evenly divide or you could use standard deviations from the mean. In a simple problem of 35% growth rate of a level 1 unit, the average would be about 6.85 points over base at level 20 with a standard deviation of about 2.08 (it's an approximation), this would lead to a variance of -3.29 to 5.84 standard deviations, and then -3.5 to -2.5 would be a -3 modifier (minimum), -.5 to .5 would be no modifier, and 5.5 to 6.5 would be a +6 modifier. At level 20 promoted it's average would be 9.8 above base stats with a standard deviation of 2.52, this would lead to -5.07 to 6.02, allowing from a -5 to +6 modifier. A couple things to keep in mind is that due to basic probability laws, getting the extreme modifiers would be EXTREMELY UNLIKELY, and would still lead to a fairly minor difference as opposed to a minor modifier. Also while this took a while to manually plug in, it's basic formulas computers could do in a flash. Early promotions could be easily handled by just deducting points (levels left * growth rate rounded up), so promoting at level 10 in this scenario would be 3.5 deduction so -4 in that particular stat. Notably caps do exist and so do stat boosters, there would be a secret value holding their true stats ignoring stat boosters and caps, these would be the ones used for calculation, displayed/functional stats would first have a hidden stat booster modifier applied, and then compared to the caps. #2: Basically take the players current stats and multiply it by a multiplier to get the sum as close as possible to the actual sum total, you could make it weighted if you wanted too, but that's the basic gist of it. Less complicated to explain by a lot but not that much easier to program and IMO less cool and fixes the problem less (bc you can get blessed in a meaningless stat while getting screwed in another, weighting can help this but it's not perfect and can make what's a good or bad level up very weird). Like before there are deductions for early promotions, and stat boosters and modifiers would be applied seperately.
I'm increasingly thinking about why fixed growths would mean missing anything long-term. You could create a little divergence with fixed growths even(or clever conditions that affect it). Playing blind I think many of us assumed fixed growth the first time until replaying after having to reset a map a few times, or very offbeat growths. Still, even with fixed growths the shiny levelup experience doesn't go away. It might mean losing a little of what is a unique FE thing, but it's got a lot of potentional to design other parts better, or tweak the effects stats have knowing the player can grow it consistently(I'm looking at you dexterity). The video discusses failsafes, alternative sources of stats, and different map design goals, but I'm still not convinced it's a nessecary hassle. I mean you could swap it around and make random growths unlockable as ng+ with the risk the game is poorly playtested for it, just to have random growths. But I feel the wealth of gameplay with more fixed growths it worth it. It's mostly about something the game teases, but never quite follows up on: needing to decide to train growth units, even when it's harder in order to prepare for later. In practice and with less efficient play this happens a fair bit anyway, but still, making it outright optimal might make for a interesting game/tier list discussion.
I know it's late, but the more consistent things are the less challenge there is and the worse replayability it has. As far as I'm aware of most of the games vanilla versions do not actually tell you what the growth rates are, so specifically the intended way to play most of the time is to simply react to how well or not units are growing. You basically lose both replayability and variety if you make everything fixed. Which is why I pretty much the easy way to get through any game is the over invest into only a small amount of characters, especially with things like stat boosters that are consistent. It basically would just make progressive decision-making more shallow, and creates even less incentive to use characters that come in later on. Because the way fire emblem is actually built to be played is to continuously play almost like your iron manning, and not restart every time something goes wrong. If you were intended to do that it would actually just give you a game over like when a lord dies, but you can't stop people from metagaming the systems.
I never really get excited by the level-ups in my limited FE experience. It is either just "okay, I guess this makes up for that last bad level up." Or "Well that was pointless." It never excites me and I would just rather a more consistent levelling experience. It would also really suck to be invested in a specific character for their personality/story and then they get screwed out of viability due to RNG. And I care about the social sim aspects way more than the gameplay so that would be a deal-breaker on a playthrough for me.
In Hypothetical Mozu gained 0 strength. She would still deal more damage than any unit because She gain access to Spendthrift, Life & Death, Bowfaire, Quick Draw. So even with 0 strength Mozu will deal 29 damage. And this is only if you use heart seal. If partner seal is added as variable she can deal 34 damage with trample. Mozu is good Bopper. Speedy as the name suggest is too hasty.
@@Ramsey276one Yes. The only unit that can deal more damage than Mozu is Xander from Revelation and Corrin in any route except Birthright. Xander setup Heart seal= Trample = +5 Friendship seal (Ryoma) = Swordfaire +Life & Death = +15 Heart Seal (F-Corrin) = spendthrift = +10 Personal Skill = Chivalry + Elbow Room= +5 Corrin pair up = + 2 Total = 37 damage even with 0 strength Corrin can reclass to any class she want. Best build is Mozu build but you swap bowfaire with Swordfaire. The reason why Corrin damage pool is higher is because she can pair up with Gunther. And Gunther give +3 which mean Corrin can deal the same damage as Xander thanks to Gunter Personal Skill.
As a true casual the best way to look at growth rates is to ignore them and like the highest difficulty where the strategies required are so batshit that you dont even feel like you are playing a tactical war game and more of a "overpowered badass kills everything" just ignore that as well.
Yo so not to talk about *Heroes* but... I like the fixed cap for characters in heroes. Personally i'd have a fixed statline at max level unpromoted, that you reach randomly. For example, you're destined to get 10 speed levels before you hit unpromoted 20, but you've only got 2 so far and you're already at level 10. You hold off on promoting that unit, because they've now become an est in so far as most levels they get from here to 20 will be great. On the other hand, if your guy is destined to only get 3 strength levels, and by level 10 he's got 'em all, you can decide to promote right away since you got blessed. It keeps the randomness (and short term challenge from bad levels) but doesnt let a character you want get screwed all the way to the bench. Plus you can design around it more easily.
Another way to mitigate bad level ups is to do a compromise between fixed level ups and random variability, by clamping the amount of upticks that can happen each level up. Meaning, instead of there being a chance that you don't get an uptick at all, the lowest threshold is always 2 or 3 upticks spread out accordingly to each stats' rate, and to balance it out on the other end maybe the max upticks would be 6 or 7 stats each level. That way you get the best of both worlds, you don't eliminate the probem per se, but you might mitigate the worst variability that would lead to polarising gameplay. So basically, the worst level up would be 3 upticks in your worst stats, whereas the best level up would be 7 upticks in your best stats.
Great video! Really got the level up screen feel on the nose. One small thing -- the music was really loud for me and made it hard to hear you clearly.
I thought growth rates were random only in terms of when they happen. Meaning a maxed out character will always have the same stats. Or is it just how it works in Fire Emblem Heroes?
Low power curve Fire Emblem is the best Fire Emblem. There are 2 other ways of dealing with random stats FE2, 15, 16: Minimum stats on promotion FF7: Stat Ranges (Unaugmented, the Dexterity of every FF7 char will vary by 7 points by level 99, depending on RNG)
@@JohnDoe-kn7ex i still prefer it over the random chance to get a stat that determines it, and you usually use one weapon type with a unit most of the time, it really isn't that bad
![Blox Fruits Dragon Rework Update [Full Stream]](http://i.ytimg.com/vi/EqDAp8udhm0/mqdefault.jpg)
So I have a tabletop fire emblem campaign going, and sure, the role playing and big tactics brain plays are nice, but I swear to god every time a unit levels up, my players sing the little levelup music and go "ding ding ding" for each stat they gain. "Number to up feel GOOD"
Oh? Is there a ruleset for this?
@@thedark1owns it's one I've been developing with a very talented friend for almost a year. Based off of 3 houses system, leverages Google sheets for the character sheets + calculations
@@Leelee-Brown Oh my god, if you have some way of sharing it, please do.
@@Leelee-Brown that sounds cool. i would love to try it someday
@@Leelee-Brown hot damn. I'll have to find this.
"Pokemon has random levels ups"
Me: *starts writing an angry comment about how pokemon works*
"At least to casual players"
Me: *slowly deletes comment*
Compared to Fire emblem Pokemon sucks. The fight animations of FE alone are on a different Level, where as Pokemon has still GB Portraits that dont move while fighting. Every Pokemon game feels the same and the battles are horrendously boring and tidious.
@@creativeideas5985 here's everyone that asked or cares:
@@christopherschlegel6412 your here sweety. Mission achieved. Pokemon is stagnating hard and not much further than red/blue. Its always the same. Fire emblem invents itself new and the fighting is actually enjoyable. As a concept i like Pokemon but it needs changes asap.
@@creativeideas5985 Why didn't you write this on a pokemon based video or something instead of banking on people on a fire emblem video knowing(or caring) about pokemon?
@@krenkrombo9616 okay boomer
My favorite part of growths is when a character like Lorenz, with decent mag and str supposedly making them variable decide they don't want any of either for 8 or so levels. Please ignore how specific and bitter that sentence is.
I had characters gaining only HP and ONE STAT
“I AM BUYING SHIELDS FOR *EVERYONE* “
Something that's interesting about growth rates is they allow for level ups to be bad, which isn't the case in almost any other RPG I can think of. If you really think about it, a character's level is nothing more than a measure of how much potential to grow they still have. Take FE6 Gonzales for example; you get him on either of the Western Isles routes, but on one route he's level 5 and on the other he's level 11. But his stats are the same in either case, meaning the level 11 one is objectively worse than the level 5 one (although he has the advantage of being able to promote right away). Thinking about levels like this, having a high level is actually a theoretically bad thing in the sense that you can't grow anymore. Meaning a no-stat level up doesn't just feel bad because you didn't get anything, it's worsened by the knowledge that you just wasted a bit of that unit's total potential that they can never get back.
Unless you're playing Awakening that is (or later)
He isn’t really objectively worse on the level 11 route, with his low skill he could desperately use the extra reliability that +30 crit brings
@@Edgeperor yeah, but +30 crit doesn't give any extra hit...
@@seththeace6217 but it does make it more likely for him to compensate for missing with a crit. Also, it gives him a little bit more survivability if he kills with a crit on the first hit, and he can conserve weapons longer the same way.
@@seththeace6217 his promotion does give a ton of skill though which does up his hit rate by a notable margin.
I'm surprised you didn't mention how Gaiden sets stats on promotion to meet a benchmark rather than adding onto stats. You can get 0 points for 12 levels on a mage or peg knight and still have it be a competent fighter once promoted.
This is true of Shadow Dragon and the Blade of Light, and SoV as well.
Yeah that's something I love
but on the flipside, all your units of the same class feel EXACTLY the same for the most part, aside from when they surpass the promotions base stats. I think a better thing would be #1: do this by unit, this should help with the lack of individuality. #2: It would be good for there to be a controlled random element. I can think of two ways.
#1: Give an acceptable range for the target, then look at the average of that stat and set that to 0, you could either evenly divide or you could use standard deviations from the mean. In a simple problem of 35% growth rate of a level 1 unit, the average would be about 6.85 points over base at level 20 with a standard deviation of about 2.08 (it's an approximation), this would lead to a variance of -3.29 to 5.84 standard deviations, and then -3.5 to -2.5 would be a -3 modifier (minimum), -.5 to .5 would be no modifier, and 5.5 to 6.5 would be a +6 modifier. At level 20 promoted it's average would be 9.8 above base stats with a standard deviation of 2.52, this would lead to -5.07 to 6.02, allowing from a -5 to +6 modifier. A couple things to keep in mind is that due to basic probability laws, getting the extreme modifiers would be EXTREMELY UNLIKELY, and would still lead to a fairly minor difference as opposed to a minor modifier. Also while this took a while to manually plug in, it's basic formulas computers could do in a flash. Early promotions could be easily handled by just deducting points (levels left * growth rate rounded up), so promoting at level 10 in this scenario would be 3.5 deduction so -4 in that particular stat. Notably caps do exist and so do stat boosters, there would be a secret value holding their true stats ignoring stat boosters and caps, these would be the ones used for calculation, displayed/functional stats would first have a hidden stat booster modifier applied, and then compared to the caps.
#2: Basically take the players current stats and multiply it by a multiplier to get the sum as close as possible to the actual sum total, you could make it weighted if you wanted too, but that's the basic gist of it. Less complicated to explain by a lot but not that much easier to program and IMO less cool and fixes the problem less (bc you can get blessed in a meaningless stat while getting screwed in another, weighting can help this but it's not perfect and can make what's a good or bad level up very weird). Like before there are deductions for early promotions, and stat boosters and modifiers would be applied seperately.
@@megarotom1590 That sounds absurdly complicated. You’re not wrong about the samey-ness issue though.
@@lordlouie3550 It would be very easy to implement. Here's a simple explanation:
it's basically that the average is the promoted stands, and then basically how much you are above or below it is reduced by a factor to reduce the variance, I just chose the standard deviation because it's an important statistical measurement of variance
(though notably this makes anything higher or lower than a +/- 2-3 VERY unlikely so perhaps slightly increase the likelihood of higher modifiers, but that could be complicated to implement)
I like the 3H solution of giving classes a minimum stat base. that way even if you cannot plan around the exact party composition, you at least plan according the minimum of the classes
You mean shadow dragon. But I know that one is different enough to sometimes be a detriment since it could also lower your stats.
It also have the side effect of giving units like dedue free 23 res or Lysithea 25 defense without committing to the class
Agreed
I think the abundance of the cast is another reason why growth rates tend to work out - you only need *some* of the cast to come out strong, while the rest have a comfortable bench waiting for them. Add in the high quality characters you tend to get at the latter half of campaigns, and you should come out pretty well.
Growth Units really are terrible if they're not broken... and that's just depressing to think about.
@@auraguard0212 Bases > growths every time. Jagen will shank your unpromoted sword-locked ass with his Silver Lance and Rescue your girl. Just as Kaga intended.
One thing to point out is that, grinding aside, Shadows of Valentia gives you generous base stats for promoted classes, so that ends up making your units RNG-proof to an extent, especially if its a busted class like Dread Fighter with base stats like 36 HP, 15 Atk, 18 Spd, and 7 Mov.
I'd say that Gaiden/SoV take that philosophy so far that it almost doesn't matter what your levels look like until your units are fully promoted, which can decrease the "aesthetic" value (if you're aware of it).
@@alxjones but stats still matter in the class before, since there are still benchmarks and everything that you want to meet. In fact, the better of level ups you get in a first or second tier class, the faster you can snowball and get to the next class. So I think growths can sometimes matter more in Valentia than in other games
@@Edgeperor You’re correct in that good level-ups still matter in the sense that they make the journey to the 3rd-tier class much easier. For example, if Faye gets a bunch of Def level-ups early on, she’ll become somewhat bulky enough to Nosferatu-tank, thus making it easier to gain EXP thus gaining access to Physic and Rescue much earlier.
You say this but then SOV also has boey a unit which has a noteworthy stat benchmark on 16 out of 22 chapters.
I understand the value in a system like that but in my experience it just meant I get little to no stats on promo other than movement. I prefer flat bonuses bc I'll always get something. If I'm above class bases and promote I get nothing.
Confidence+Sword avoid+Alert Stance plus+10 avoid from wyvern lord is the way three houses makes Ferdinand Von Aegir always have good unit feel.
Going dancer just for sword avoid is pretty cringe but sure 👍
dancer ferdinand is based fuck you
Don't forget the avoid ring, avoid from battalion, and swordfaire and sword crit+10 from swordmaster
It gets sillier when you start stacking new game plus runs. Nothing like having your spell casters which are typically squishy even with shields effectively double their defense before shields and start marching through enemy lines as destroyers of worlds because most enemies simply do not hit hard enough. Once you achieve that on one playthrough you can just spend the resources to boost their weapon skills so that when they hit the right level then can just promote and then swap back. All thanks to the benefits of simply promoting into a class and the class boosting stats to hit its minimums.
If you just go war master and war monk with a 20 avo battalion, you can do way less investment, have a unit with 5 less avo (but the ability to use terrain tiles) and WAAAAY less investment.
Ok I want to say something real quick. Mozu might not be hitting any benchmarks by gaining stats and you COULD still give her stat buffs with skills, tonics, pair up, meals, etc etc etc. however, Mozu getting more strength on levelups means that instead of, say, spending 200 gold on a str tonic to get her to one round an enemy or even worse giving her a high tier backpack like Arthur or Keaton, you’re able to use less resources. These resources can then be spent elsewhere (usually Camilla, if were being honest). As a result, every stat point you get on a levelup that gets a unit even only a single point closer to a future benchmark is a stat point that you don’t need to provide to them via other means that could also be put towards different benchmarks.
I think in Fates Skill mitigate a lot of issue with low strength/magic bases. Mozu in particular just didn't care if she had bad strength. Mozu have access to Skill like Quick Draw, Bowfaire, Spendthrift and Life & Death. All of those combined and she will deal 54 damage using Crescent Bow. And this is not counting partner seal. If you give Mozu Elbow Room or Trample. You can one shot Takumi with Sorcerer Mozu.
@@riqua27 yes you CAN. The issue is that it takes a disgusting amount of resources. The named strategies were:
- Quick Draw (Archer lv 10)
- Bowfaire (Sniper lv 15)
- Spendthrift (Merchant lv 15)
- Life and Death (Master of Arms lv 15)
- Crescent Bow (Requires a high bow rank and gold)
- Elbow Room (Requires Mozu to marry Silas or worse, Xander)
- Sorcerer (Takes S rank support bonuses for Odin or Leo, both good units, off the table)
All of these things together let you shit on Takumi with even a base stat Mozu, but you’ve spent:
- a minimum of 26 levels, 17 of those in a promoted class (10 to get quick draw and elbow room, promote then lvl 15 for one of bowfaire spendthrift etc and reclass into MoA or Sniper for faire or L&D, then do it again for the final one)
- In the best case scenario, 1 master seal, 2 heart seals and a partner seal (The partner seal especially which you could spend on the person Mozu’s married to to instead get THEM into Sniper)
- Insane amount of wexp grind or the purchase of one or more arms scrolls
- Every single offensive stat booster you could possibly get her
With all of these resources by themselves, you could probably get like half your army setup done for late game conquest, using heart seal on shura and Camilla, maybe a partner seal to get Xander into hero for hp+5 and sol, master seal for either Camilla or Xander’s backpacks, fixing up shura’s dagger rank with arms scrolls, the list goes on and on. Mozu getting stats on levelups means that by endgame, she could (and tbh probably would) have gained enough strength that she doesn’t need one or more of the skills given to her (quick draw or elbow room is probably the first one you could drop), and you come out with a lot of extra resources which you could instead use to make your other relevant combat units significantly stronger.
A less extreme example of this resource vs stat gain argument would be an Odin. Lets say you have an average Odin who’s been leveling defense and magic basically every other level. His speed is garbage but also who the fuck cares, he only uses nosferatu to begin with. If you had that Odin with a decent pair up with effie for defense or nyx for magic, you could easily get away with him meeting all his benchmarks up until maybe ch 19 without needing any tonics (except for the first chapter that you deploy him in to level him up since his base level and base stats are below satisfactory). Instead, if you had an Odin super magic or defense screwed, you would have to spend gold to buy him magic, HP, and defense tonics, cooking defense food perhaps, and doing this and that to get him to hit benchmarks instead. Every 2 points of magic/defense that Odin levels up is 400 gold you save a chapter which adds up insanely quickly. With just 2 defense and 2 magic levelups, you’re saving 2000 gold within 5 chapters, enough money to buy an extra master/heart seal.
@@lagspike7763 Of course. I'm being Hyperbolic in my comment. Turning Mozu into Sorcerer also extreamly unnecesary. That's just me being Hyperbolic. I think the more optimal build dor Mozu is to take Quick Draw and Bowfaire. Tht way Mozu can ne used for Dual Strike engine so other unit can get weapon rank and exp easily. Amaterasu also really good. So other unit will stay alive.
Something that I really liked about Radiant Dawn's bonus experience was that, because it would always boost 3 stats exactly, you could manipulate it to cover the weak points of a character who has capped their strengths. For example, Micaiah is notorious for her low HP, Speed, and Defense as well as limited accessibility before her second promotion (among other things), but because she is very likely to max out her Magic and Resistance early, alongside possibly her Skill and Luck, you can essentially force her to gain more HP, Strength, Speed, and Defense by saving your bonus experience for when she starts getting closer to level 20.
Had an interesting discussion about Jo_ about growths the other day, he asked me why people wouldn’t want to play fixed growth Project Anima more than random growths, I think it has to do with the story telling growths gives us, redemption archs, curb stomps and struggles all build in to our personal narratives about the story. Those are so much more powerful than the game telling you, this is the powerful character use them. No one ever says “yo look at this seth” they are way more excited by the stories of axe bros and archers tragic as some may be
My shadow dragon Jagen got an HP/STR/SKL/SPD/DEF level up and I popped off lmao
ohoho, scripted bopper makes my day
You complained about games not having these super difficult final maps, but I’m kinda glad not all of them do. I think an easy final map with epic music and setting is better than something like shadow dragon where it’s not only easy, but probable to get softlocked on the final map of the higher difficulties without warp or a ridiculously strong army
@treeghettox I was playing hard 4 when I did it, and let me tell you it was hell on earth. Did they have forged brace weapons for you? The brave weapons nearly killed all my guys in one round besides sedgar and wolf, and poor Abel couldn’t even survive one round
@treeghettox fair enough, also gotta love how the caps of the manaketes lie on higher difficulties too lol. If you wanna tackle higher difficulties later I recommend reclassing wolf and sedgar to general so you can throw them in the middle of like 5 enemies to get a bunch of chip experience, best way to train them quickly. I think even on normal Medeus is ridiculous, but on h4 and h5 he has like 30 speed and 50 attack. Unless you’re a swordmaster you literally have to have capped hp and defense to survive getting doubled because all the other class speed caps are too low. All I can say is good fuckin luck man
@treeghettox not quite infinite, but 20 turns feels like a fuckin eternity lol. You do have tiki and the Aum staff so he’s always beatable, no crits required. But without warp getting there is the problem.
@treeghettox yeah, in fe3 book 2 you have to use it on the altar where Medeus is, and I think even in book 1 you have to use it in the altar where it is. Since you played fe3 beforehand I definitely understand making that assumption. But in fe11 and 12 you can use it anywhere. Makes a bit less sense but it’s better for gameplay.
@treeghettox you know, if you’re gonna use all three of them they actually have a triangle attack, pretty useful since they’re all available early on. The avatar is pretty shitty, no lie about that but he’s not as prevalent as Robin or Corrin. Good luck with the run though, fe12 is a great game.
tfw no mention of fixed mode from fe9, where every unit always has their average stats
the pleasure of watching stats go up without the variance of random level-ups
@@Malhonn AFAIK the amount of growth points you gain when you fight is defined by growth rates, the enemy that you are fighting and the weapon you equip
smh alex
Unfortunately, the only game that has Fixed Mode is ALSO the game where you can easily reroll levels ups using Bonus EXP on hardware. I'd love to see the feature in other games, but if I'm playing Path of Radiance I'm gonna pick Random growths and cheese bonus levels for good level-ups.
You discussed relatively low stat caps in Thracia as a means of controlling growth rate variability, but this is also true of Radiant Dawn. The devs are able to balance tower enemies around a theoretical maximum of unit power. Standouts like Ike can still carve through chunks of beorc enemies there, but they still represent a formidable challenge to more common player units.
The fact that the designers then let you run loose with Laguz Royals is a separate balancing mess.
Unrelated, but I think effective weaponry can be counted among the countermeasures to poorly developed stats. Just look at any archer whose usefulness goes up against certain enemy compositions (or in the most extreme case, when wielding crossbows against fliers).
Really enjoyed this one, and don't have much to add. I always imagined stat variance was a massive headache to deal with on the developers' end. Lately we've gotten stuff like Tonics or cooking with the Head Chef to boost stats temporarily, which are nice ways to save a particularly unlucky player but also (from my point of view at least) kind of a concession that IntSys understands just how unlucky a player could possibly get. It's also interesting to see how romhackers deal with stat variance with things like inventory items that boost stats when held.
A FE that can 'measure' how powerful your army is and generate an appropriate challenge would be kinda cool. The downside is that it could probably be heavily exploited, but only the most hardcore would be doing that.
But doesn’t that in a way make the RNG-based lvls up pointless? Like it means that no matter how good or bad your units level ups are they’ll still be evenly matched with the foe. If you get 1 point of Spd and all the enemies get 1 point of extra Spd then that point of Spd is essentially pointless, the only effect being it means your other units have to also reach that new Spd benchmark.
@@CosmicToad5000
Very good point. Maybe in addition to raw stats it would measure how reliant you are on certain units and make counters for them. If you've been stomping the game with Seth the final map would throw lots of horseslayer enemies for example.
@@CosmicToad5000 right thats what happened to fallout 3 imo
@@CosmicToad5000 Maybe you could apply that ''evenly matched'' bonus to some bosses but not abuse it. But yeah, it defeats the whole purpose of random levelups.
I have no idea what you are talking about, but I love watching fe units gaining level ups while drinking coffee xD
same
It's essentially the hidden % value that determines a character's level up. The higher the percent, the more likely they will level up. Think about Rolling a die with 100- If the numbers you roll are less than growth rate percentage, your character will gain a level up in said stat. That's how critical hits works too. The numbers rolled would have to be equal or less than the number value to work.
Great video. Pretty much agree on all fronts. Designing for Fire Emblem is hard af, and growths only make it harder. But they add so much to that unique Fire Emblem flavor that it's difficult to imagine the series without them. Keep up the awesome work!
I thought FE was a tactical rpg.
Wouldn't it taste better without the garbage?
@@AlexanderMartinez-kd7cz
Depends on who you talk to. I like the idea that one of my units can become stat screwed; I think it makes the game more interesting strategically. It adds another layer of decision making to the game: do I keep using this unit who I know is supposed to be good in the hopes that they get better, or do I bench them for a prepromote with a good baseline but low potential?
@@nanowasabi4421 why do you need strategic decisions based on characters randomly being trash?
Are FE battles that bad?
@@AlexanderMartinez-kd7cz its not strategic decisions itself, but I like being forced to use units I wouldn't have otherwise. Especially in games where there are a lot of units to pick from, it just adds more variety
@@sonjaa.martin6269 sounds like FE needs some new game + locked character mode not broken leveling
The way level ups are doled out is also what makes the level up screen in Mario and Luigi games so exciting. Especially the slot machine. The stats that the slot is giving is largely negligible, but big 4 or 5 gotten from a random chance feels so exciting.
The thing that makes it even better is that you CAN learn it, but it's pretty hard and requires you to take some risks for timing high numbers next to low numbers.
Rolf > Good bases > Good Growths
Sorry I don't make the rules.
Sorry man but Rolf is a poor man's Wolt
@@carlosrod100 How dare you speak ill of the Son of a Shepherd!
Promoted minimum stats in Gaiden, Echoes, and 3H are all ways the developers can ensure the player has a benchmark of stats to finish the game provided they take the time to actually level their units appropriately. Berwick Saga uses stat bracketing, which provides a minimum and maximum that stats can deviate from the average for each character. I think this is probably the most genius system, as it allows the devs to fine tune the endgame to expected stat thresholds and prevents players from having stat screwed characters, all while providing the satisfaction of the level up screen.
Bracketing system was in Shining Force since the first game thi, albeit in 1 it was a variance of like 10 points. Wich enabled absurdities like Zylo getting 10 HP and 5 attack in a single lvlup.
I've ranted so much about Project Ember's final chapter because of stuff like this.
They just made Idunn much stronger without helping you in any way, it almost felt like a kneejerk reaction to her being considered a pathetically easy final boss.
Idunn one shotted nearly half of the units I brought with me, but I could still one-round her with Roy, but that's not guaranteed.
It's even more upsetting because this could have been prevented with Eliwood and Guinivere joining, but instead they do it two chapters earlier, technically giving you enough room to get them killed.
Not to mention Guinivere would have probably been killed in one hit too.
So many people just never think about the worst case scenarios.
Of course, there's the fixed growths version of it too, which might be why the devs did what they did, but it's not really promoted to be the ultimate version of the hack, and no warning for the normal growths version is given or anything.
I'm sure a lot of people really wanted to try the fixed growths version out, but I wouldn't be surprised if most went with the standard growths version anyway.
So its bad because Idunn is no longer a joke and is an almost challenging final boss now? That makes total sense 🙄
@@sleepedge8257 It sounds more like it's "harder", except applying the Binding Blade to Idunn's face is still the obvious solution, so not really.
@@sleepedge8257 making the final boss one-shot half your characters isn't hard. It's just bad game design. Do you consider Medues from fe11 hardest's difficulty well designed?
12:26 I guess this is why most games don't have weapon ranks as a stat anymore.
THANK SOTHIS
XD
I think the vast, VAST majority of people forget how statistics work. Yes, a unit CAN be stat-screwed, but if you level them up they're very likely to hit around average, even with bad levels to start.
Random level-ups work because of the ability to work around them, and because of the scaling difficulty given to the player. Games without grinding, you just have to scale enemies to the expected averages and things will work out. Weaker teams will play more strategically and be fine (assuming you didn't get so unlucky as to have EVERY SINGLE UNIT stat-screwed), and stronger teams will win easier in the moment but aren't guaranteed to continue winning, and even better it rewards those good units' levels with being stronger than the enemies. That's the whole POINT of random stats! And beyond that, with a shit ton of units at your disposal, you're likely to get good ones you can stick to, and bad ones you can leave behind.
I think giving things like infinite grinding is a bit odd as it leads the player to "craft their own difficulty", which typically means "grind until they stomp everything, or don't grind at all", but I still think that that's fine as that's how the player decides to play the game (though there are many players who grind and then complain about difficulty, or vice versa).
In any case, I'd say growth rates are NOT bad, they're at their most difficult to predict in the earliest parts of the game, but since you're also given units in most of those chapters (who are themselves capable of doing well in the map), it's okay to base those off of averages as well. Level design doesn't need to be fundamentally impacted by the standard variation; and I'd argue that it isn't influenced by how many units to choose from either, as most units have archetypes and maps that benefit those archetypes don't REQUIRE those archetypes to function in any case.
"if you level them up they're very likely to hit around average"
Have you done the math? Can you calculate the exact probability you mean when you say "very likely?"
@@WildMatsu Depends on the level, depends on the exact rate and depends on how you treat "around average". For, say, a 50% growth to hit 20 times in 39 levels (20/20 from level 1), it's about 12.5%. To hit anywhere from 17 to 22 times, it's a whopping 65%. For rates like 25%, 7-12 growths in 39 levels is ~75% likely. Variance is to be expected, but for multiple units to have key stats borked, is unlikely. With higher growth rates and more levels (like awakening) the odds only slim further. In around 50 levels, with a 75% growth rate and a cap about 30 levels higher than base, there's about a 2% chance it won't hit its cap.
All in all, odds of a unit having at least one non-average stat (including too high) is decent, but if you say odds a character has at least one screwed stat specifically, you can approximately halve the loss odds. And if you train more than one unit, the bounds are even greater; and considering most casual players won't be speedrunning with a single unit, the odds of being screwed on all of them are very low. I can do calcs for multi-units on key stats if you'd like.
I absolutely loved my playthrough of FE9 with Fixed Growth Rates. Guaranteeing units to be good, so long as they are leveled properly, feels so damn good that I wish other FE's would allow a Fixed Growth feature.
I mostly play GBA Fire Emblems and have found the experience much more enjoyable when I patch the games to have fixed growths/average stats.
I get to focus more on strategy and planning knowing what I can expect from each unit.
The games already offer a lot of variability for me just by choosing to use different teams and even subpar units.
And when I want a crazier experience, I'll straight-up use a randomizer.
@@pascalandraos Definitely, I do the same. I just wish we could eventually get an Awakening/Fates/Echoes Fixed Growth Rates patch. That would be a freaking romp
@@ElectricLegend14 I agree that it should at least be an option.
One very minor but pretty cool thing I’m reminded of it’s is in FE10 when Sanaki gets the Rudol Gem which gives her +10 def when held and is locked to her inventory right before the final map. I think this was to ensure Sanaki could survive Ashera’s physical AOE attack which she may not otherwise due to poor Hp & Def bases & shaky growths. Now they don’t get the same thing to Micaiah who’s forced but hey they did it for one unit.
Hades Music :)
Nice.This was a really interesting topic. Thought I have to admit I kind of like the chapters when you face a lot of damage sponges. Its fun (to me) to face a wall of generals with pavise and hit them with a forged hammer with Cherche and see how they still died despise all the extra defenses they put in place to avoid just that. I don't know I just like to see this ridiculous 70+ hp bar struggling to get to 0 before the battle animation it's over, because your characters can hit with the wrath of God.
I think FE should actually give up actually making the scenarios and strategies revolve around "stat checking", as it were. Not because of the idea of the level ups being random suddendly becoming a problem, because let's face it, is not. Players get stuck on maps because sometimes they made bad resource management a thing, such as not promoting actually good characters(characters that are useful promoted, like they gain access to staves or can actually use multiple weapons or they gain great movement options OR like most modern FE do, actually give them skills that can change how you approach maps) or not understanding how to make use of money to buy things at shops that matter(like Keys, Crit Weapons, Torches and Potions), making armies be really one dimensional in strategies(let Lord Dude Phalanx the attack head on and pray he doesn't get critted halfway through is so BartreBrained) and so forth.
And is not like they haven't tried to do this sort of thing, where you can be a base stat chump be a hero because the map is designed to be beatable with strategy. Things like Bosses being weak or strong against certain damage types(like a Boss stands on a Throne with a Melee Weapon and a Ranged one) and gets cooked by Magic or has to break his Melee and then fight him with a much worse Ranged weapon(like he can't Range 1 attack so he just gets bulldozed at Melee). Escape maps where the idea is not to fight every single enemy, but to fight the enemies that matter and, when the player takes too long, force him to either leave or die, Defense maps where the idea is to protect a point and enemies can be "holded" or are just chump as fucc, like they actively ignore you for the load you have to protect and you can just chomp on these enemy chumps no issue even at level 1, so it becomes a map where the point now is to actually protect the load and so forth.
And of course, actually making enemies use the same resources as you. It doesn't matter if a Enemy is a Bishop if all Bishops do all game is use D Tier Light magic. And in the same can be said for combat only units. Paladins having access to all weapon triangle is dumb if all they carry is just Steel Lances and maybe a few Killer Lances.
Your comments on escape and defense maps is exactly why I find a lot of conquest maps so fun. The eternal stairway being one of the few escape maps that actively push the player to escape lest they be overrun by lots of no-exp enemies that stack Chip damage.
Maybe I missed something, but how do I even know what my units' growth rates are without going to am external website to look up the tables. I can't know ahead of time what a new character is likely to be good at.
And even after a playthrough, I can't be sure character X has good strength growths, or if he just got lucky with the rng last time and benefitted from being in a class that gave a bonus to strength growths. Engage does stuff like give you Anna, with the highest magic growth in the game but starting in a physical class. It's a fun idea to let players figure out that it's best to send her to a magic class, but I don't see where I would get the information needed to make that analysis within the game.
Scared stones allowed you to see the growth rates via select on the stat screen for some reason they removed this feature even though it would be just as easy to implement
no it didn't
I'm surprised that you didn't approach this from the angle that due to growth rates each player's run of a Fire Emblem game will be that much more unique, to tie it back into the permadeath video. Like person x's playthrough could have a terrible Tana but person y could have Tana as their MVP. Person z got her killed in chapter 13 or something idk.
it makes it highly replayable. similar to hit and avoid your firing your characters like a viewer of a sports team on to do good but your not sure of a win. Its way simpler than he approuches in my not humble but very humble opinion
I miss these types of videos
Some kind of smart rate rigging system would be viable. Have a baseline stat curve that a character should follow (either as a combined total or per individual stat) and then characters below that have their growth rates tweaked to improve chances of hitting that baseline. It wouldn't have to make a unit viable, if anything the baseline should be a few points behind the expected played stats. That expectation could be made up for by skills, weapon ranks, stat boosters, etc. For harder difficulties, in addition to stronger enemies, reduce the baseline system further or remove it entirely.
In addition, if some units are stat screwed, the system could accelerate other units' growths, meaning no player should be soft locked by too many unviable characters.
Joke's on you, i'm not juggernauting Awakening with a small army.
I'm Juggernauting Awakening with a massive army cuz i've got nothing better to do than grind up like 40 different characters
The worst is when you level up and even the character states their disappointment
6:14 Three Houses
Their disappointment is MEASURABLE
“I got cocky.”
“I HAVE NO TIME FOR FAILURE!”
“I am not setting a good example...”
“Everyone has their off days!”
XD
This is kind of why I really enjoy fire emblem games which allow you to recruit several units long into the end game since it kind of gets rid of the problem with growth maybe your Soren had really terrible growth but Hey you'll get another mage in the late game that probably help out with that so you know.
I'm surprised you didn't go over the concept of Berwick Saga's bracketing, a mechanic which prevents units from straying too far from their stat averages while still providing that aesthetic pleasure of seeing your units grow along with a bit of randomness. From what I've found they seem to stay around 3 more or less for most of their stats and this doesn't seem to be effected by statboosters. I know kaga saga isn't technically fire emblem but it's similar enough and.. has growth rates?
Would you like to see this in a new fe and do you think it would make it better?
And I completely disagree with the idea that every stat that isn't reaching the benchmark for something is pointless. They are important to reaching that threshold in the first place and can still have value in specific scenarios. Sure, being able to double all the enemies on a map is more important than being able to double a single enemy, but being able to double or kill that one enemy can have much more merit depending on the situation. Enemy stats can differ in very small increments and being one point short of doubling can make the be the difference between life and death is some battles. I feel like I have to mention the small stat boosts you get like avoid from speed or crit from skill but those are so little I feel like they barely even count. While the benefit from a single stat might seem minor and the likeliness if having that matter much in the grand scheme of things, having the mindset that it CAN matter and that every stat is useful has given me a lot more enjoyment from these games.
ps you cant just put that funky earthbound music its tooo goood
Benchmarks are weird in FE because most of the games have very sharp benchmarks. 27 speed in RD takes a character from doubling basically nothing to nearly everything. It’s weird how truncated many of the stats are. I guess that “pointless” was the wrong word, but rather, “not immediately useful for the majority of enemies.”
I think a Berwick Saga approach like what you point out would be a good idea in a game like Three Houses where basically every character is a growth unit and is meant to be carried through the whole game.
All EB music is too good.
I've never understood why games like FE have stuck with these growth rate systems for so long when there are such simple changes that can basically fix all the problems they have. For the simplest option, we can look to games like Chrono Cross where characters have an "expected value" for each stat at each level, and the further the actual values are from the expected value, the more odds are tweaked to get things back on track. As in, if you miss your 30% stat up five times in a row, it'll have increased to 35, 40 or more to help make sure you never fall too far behind. Conversely, if you high-roll that stat gain five times in a row, you might only have a 20% or 15% chance to hit it again, that way you never get too strong to trivialize things purely off RNG.
A system like that allows there to still be plenty of RNG and variation to keep things interesting yet also allows devs to have a reasonable expectation of how strong characters will be at any given level which they can use while designing levels.
@@user-ed4nb4ef7w If you ever play the game Chrono Cross you'll see that despite having a system like I described, there's still a great deal of variance in character stat growths from run to run. It works for two reasons. One, you can choose how much you want to weight the bias in question. Two, the impact of it is greatest at the highest levels.
Even if characters wound up with the exact same stats every run at 20/20 (and I'm not suggesting going that far), that's still ~40 levels with variance in them. And how many of your units will even reach 20/20?
I really like how unique every character gets in FE it means no matter how many times I play one FE game my units would always be different
@@NightRaven5568 I like RNG in my hit/miss chances for much the same reason, but that doesn't mean I dislike the fact the game doesn't use true RNG for them. The same would be true here. Adding some bias toward the middle/average values wouldn't render RNG irrelevant. It would just tone the RNG down a bit.
I think something that's kinda important is that the more levels you go through the more the levels should even out, this means that the early few levels can be a wild west where one character can have a string of 3-4 really good or really bad level ups now you aren't getting these stats back, if you should have got 5 speed from your first 10 levels and you got zero it's very very unlikely you are going to get 10 speed in your next 10 but that 5 speed as a percentage of your total speed is huge.
Something that can also offset the rng a bit is if you have 2 very similar characters and the game expects you to have one of that type at least that is good, I think Alan and Lance in fe6 are good examples, if Alan gets speed screwed you might have to dump him but at least you still have lance (and the cavs that come shortly afterwards) and if Lance gets strength screwed you might have to dump him but at least you have Alan, sure both can happen but the more times you roll the dice the closer you are going to get to expected outcome
Where are the StarCraft 2 Terran music compliments?! Here's one: excellent music choice.
And Kirby Air Ride ⭐
@@DarknessChampion Etrian Oddysey 4
For like 5 seconds, but still
One thing I think would help mellow out this issue is if Fire Emblem did what a lot of traditional RPGs do, where the level is actually put into the battle calculations. I.E. things such as hit, atk, avo, and crit would increase as level does (Obviously not like by 1 for each level. But depending on the stat.)
Also hi, I'm new here, I just came from your PvP with Mekkah. Good content. :)
Another solution is the point system, where it just adds points equal to growth rate and when points equal 100 then you get a stat point, but considering that it's optional you still can't base the levels around it. Also it lack ANY sort of variance, so perhaps you can get up to +/- a few points randomly. Alternatively, you can add or subtract the chances levels up will occur dependent on how blessed or curse you are.
I can think of ways to fix level ups, Permadeath on the other hand, means that any map has to be beatable even if you only have half the number of units your even allowed
Considering some Speedrun/Challenge runs I have seen...
XD
Having a garbage unit with bad bases and bad growths randomly be blessed and carry the game is what make growth rates so fun. I always remember that specific run whenever it happens.
This is the conclusion I came up with when I thought how I might go about making a fire emblem game:
1st: The only stats that consistently grow when you level are HP and Skill.
The reasoning behind HP is simple in that your characters will scale with enemy damage and enemy stat increases.
The reasoning behind why skill grows is simple, as you gain real life experience you become more skilled. I would actually put the baseline of skill across every single playable unit at 50% per level. That means that hit rates and crits are changed.
I would make crits skill locked behind an adequate threshold and in addition to that lock it to specific classes like Priests and Swordmasters. Priests because yadda yadda divine intervention and Swordmasters because Swordmasters are actually supposed to be the most canonically skilled in the series' lore but in the games they are consistently some of the worst units almost always.
The real modifiers I would make as it pertains to stat ups is gear. The thing I love most about FE is the fact that the earlier entries in the series kept numbers low and managable. It actually matters when numbers only go to something very easy to identify like 20. I would make it so that the best way to alter characters stats are to wear armour that takes up an active inventory slot and weapons that actually give bonuses to stats in addition to making S ranks on weapons change depending on what weapon a unit was carrying.
For Example: A Sword unit with S rank weapon level would have +1 to every stat with an Iron Sword. +2 defence with a Steel Sword. +3 Strength with a Silver Sword. +3 Skill with a Killing Edge., or +2 Resistance with a Rune Sword.
It is important to be skilled and mindful of your money as your weapons do have durability but not in the traditional way FE usually does it. The best way to make an in series comparison would be to akin it to FE 5's fatigue system. If you take a single weapon into 4 or so chapters that weapon will break after the 4th. So making careful calculations with your funds is very important as wasting it means that you won't be able to buy valuable armours and weapons.
In doing so each individual weapon would cost a lot more but in exchange even something as simple as an Iron Sword is very powerful with S weapon ranks so you want to keep good track of even iron weapons.
If I were to make the game like that I could manage difficulty very easily as the only thing I would have to worry about is hit rates and HP. I can scale very well by scaling to individual character stats as it pertains to possible items the characters may have purchased from the armories and blacksmiths. I fell in love with FE because of growths but I now find it consistently lacking when playing the end of these games because the early chapters are almost always so much more tight and tactical.
I would make an FE were every single in game stat benchmark was determined by gear you could actively influence with decision making as opposed to uncontrollable variables.
One of the defining reasonings I came to when I thought about it a while back was how much I liked the Fire Emblem 4 supports. When Lex finds the brave axe by the sand is cool. When characters have a support and an actual stats raise I find those things far more intriguing than the level up screen. The player has full control over those and by designing around them you could make levels far more tight.
TLDR: Stat variables in my FE would be determined by items, weapons, and supports. Only HP or Skill could ever level up when a unit levels. It would also teach the player that promoting is valuable far more than any FE ever has period.
I prefer the randomness of growth rates because it allows for individuality among each playthrough. However FE needs to allow for growth modifiers like bands from FE9 but better.
I honestly don't get the whole thing about the _wide range of stat variance_ you talk about when designing for late game. Fire Emblem doesn't use arcane methods for determining stats, it's a basic probability. The way these probabilities function means that, over millions of players, there is going to be an average spread of stats that the designers are able to assume a character will be at.
I think what would fuck with that are the more robust class systems the newer games tend to have, allowing for more variance in stat growths that the designers might not be able to 100% account for. There are still ways they can mitigate that, though, e.g. weighting growth rates to be less class-dependent and more character-dependent a la Three Houses, or having only a small number of classes to switch between per character/class a la Awakening.
There might be averages that characters have, but remember that 50% of players will be below those averages.
@@Edgeperor Yes, I know how a normal distribution works. I also know how a normal distribution is, well, distributed. 50% of players will be below those averages, but the majority of players will still be around those averages.
Thank you! Someone else recognizes why the gorgon map is the best SS map (at least in theory)
I would love to see a Fire Emblem game with totally predetermined growths - i.e. at level 5, the lord will always get +1 str, spd, luck, res. At level 6, they will always get hp, skl, spd, luck. This gives the excitement of seemingly random growths to players playing blind, and gives IS the knowledge of every units' potential stats to plan around for late game. It would also mean that the levels 11-20 for unpromoted units (in a standard FE promo system) can be weighted to either encourage or discourage early promotion.
Just an idea that I'd like to see implemented, or at least explored in an optional setting or even fan hack.
One thing that could be done with fixed levels to encourage early/late promotions even for blind players is to have stat caps per character instead of classes (maybe stat boosting items can go over them in that system, maybe they can't, both have advantages). And that way you can physically modify how early some character peaks some key stats.
Like, if you want one myrmidon to be trained to level 20 and the other early promote, have the first one cap strength at level 20 and make the other cap it at level 12. (this would work with class specific caps too, just easier to balance with character specific). That way you know the first one got a lot of room to grow much stronger if you take the time while the second myrmidon should be promoted if you see them struggling to deal enough damage, and you don't need to learn about how the devs calculated the stats, it's just very easy visually.
That's one of my major problem with stat growth that isn't talked about in this video, it's invisible. Like, it's important enough that it can influence your decision of what character to train or not, but you have no way to know without looking it up. And then once you start looking it up it becomes hard to not be caught up in optimizing your characters, exp, etc, rather than simply using the right character for the job you have ahead of you, which makes the game less enjoyable to play. Weell that's a broader issue caused by the way level ups work in fire emblem, but hidden growth rate really emphasize the issue for me, as the game isn't designed around you knowing that but looking it up is extremely easy and can change how you approach things a lot.
Imagine a (relatively short) hack where every character's growths are 100% across the board. No luck. No randomness. No uncertainty.
Boring? Maybe, maybe not. But in theory, it would make maps a lot easier to design, so they might be better.
These essays are so helpful for the game I’m developing. Thanks Bopper
I think you're overestimating the effect of growth rates here, it's a relatively easy activity to figure out a confidence interval for where a character's stats will be at level x
The problem with that is that the confidence interval is extremely wide.
Ooh, I'm reminded of this game devil survivor's growth rates. In that game there's 4 stats (strength, magic power, defense and avo..... essentially) and at each level up you gain one of these. All the other characters choose that point they gain each level, but your player character gains whatever you want them to gain. So basically instead of the game screwing you over, you now have the capability of screwing YOURSELF over!
Anyways, really interesting video!!! I definitely have nightmares of my favorite growth units becoming useless and dying due to screwed stats. I was lucky(?) in my fe8 run because my Franz became one of the godlike ones, but in return my Eirika gained 0 strength for something like 10 levels. Whose decision was it to make lords with bad growths lmao
As annoying as growths can be, I do love how runs become so personalized due to them... I guess the question becomes how to make endgame doable when it's possible for you to reach it with just the lord and maybe 3 other guys 😂 maybe IS should check how many units you have and send you a couple friends if your units' levels or the number alive are below a certain threshold haha
They tried giving you extra units if your army was too small in the Shadow Dragon remake and apparently it pretty much trivializes the highest difficulty. (I don’t know myself, but I’ve been watching Excelblem’s Shadow Dragon Ironman recap clips. From what I’ve seen, the system is very abusable.)
@@nanowasabi4421 interesting!! I've played mostly newer fes (8-10, 3ds, 3h) so I didn't know how those panned out. But it sounds better than getting softlocked at least 😂
Growths are like a drug for anyone who is cursed with being a fire emblem fan, even when you recognize their flaws watching your favorite little skrunkly level strength, speed, and defense just makes you love em
Also wanna mention how I think growths are usually a one or none thing is probably because they wanted the player to be able to still calculate and plan ahead with a good enough idea of how things should go without requiring the use of a calculator if the stats were much higher
Three houses has some mitigating factors in it as far as helping you get over bad growths.
Passing certification exams will automatically elevate a unit's stats to a baseline if they're not already met.
For example, certifying Mercedes as a Fortress Knight will give her a massive bonus to defense. These bonuses are *permanent*, so even if you swap Mercedes back to a Bishop afterwards, she keeps all those bonus points. In this way, you can take any unit lagging in a certain stat and guarantee they can at least achieve an expected minimum.
Additionally, Three Houses also does not allow for blank level ups, and will guarantee you at least 1 stat up.
But if we don’t have growth rates, we can’t have Supertank Oscar, and that makes me sad.
You've got my like for that Etrian Odyssey music alone.
If you listened to the Super Arrange version of this song, it makes it even better.
No idea if this was on purpose, but this is pretty neat
I prefer the gains from games like steamworld heist, where you get a fixed amounth of extra perks, it allows for better level design, and requires smarter play, since every unit is guaranteed to have a niche and role....
But damn, turning a blessed gaius into a myrmidon on a forest tile to see him go ham on on enemies without being hit once *is* fun too
I also hope the clip of gilliam being crited into the arena becomes a staple. I wish to see this clip in every odd video of yours now
I want an fe that expects you to have a few good units in end game but gives you an Athos who can carry if you're willing to play meticulously
i love athos. He contradicts everything what you would see usually in japan culture based games. An westernized bearded wizard, comparable to gandalf. Who can solo the last map....lol. and carries around artifacts for your heroues next to the mastership of all magic classes allowing him to use all of the great books recieved.
12:06 Ayo was that the Forgotten Capital theme from Etrian Odyssey 4? Nice
Growth rates should be PSEUDO-random, as in if your average growth rate is significantly below or above what it should be the odds should be tilted in order to fix that.
The nice part about the growths is that it's a free casino, we're getting the thrill of the roulette every time one of our characters gains a level.
What I was thinking might work is have units have three or so stats that are almost garunteed to level up (80% +) and have there other growths be abysmal (30% -) this way units are predictable in what they can perform with some level of variance and making really good level ups more exiting.
I think a good example of this is aran from FE10. he has a 70% growth in str, skl, def, so you know and devs know that aran is likely going to be a hard hitting resilient Frontline with little variance.
I heard that Etrain Odyssey music there, don't think I didn't!
F
O
E
O_o
I remember one time when my unit gain level-up, but don't gain any stat increases... it's so disappointing.
As usual you dropped another banger video. I disagree with alot of your takes but you generate good discussions and question things no other fe tubers do imo. Also the background music is absolute fire.
Does anybody know the songs he uses in this video? I know they're Fire Emblem tracks but I don't know which ones 😅
Once I started using emulators, understood what savestates were, and randomizing games, I realized how much I cared about making sure I got good levels. My Fir actually turned out better than Rutger, through MANY savestates to get more than skill and luck. But I really appreciate them in normal playthroughs, just because it adds that element of chaos that these games desperately need at times
I think a lot was overlooked here. The entire video is pretty much bashing the existing system, exaggerates the problem, and doesn't take a serious look at the benefits, instead opting for "growths are exciting!" It also does not compare the current system to alternatives.
The most obvious reason I can see for the growth rate system is to keep the game fresh. On any given playthrough, I cannot rely on any one character being good, so I'd be likely to change up my team based on turnout. It forces the player to improvise based on whatever stats happen to come their way.
Additionally, the issue of being stat screwed, and of the developers being unable to design around how powerful a character's team will be is greatly exaggerated. You can certainly have bad turnouts for a few characters, but this simply won't be the case for your army as a whole. Given the size of the team you use and the number of level ups throughout the game, the total stats gained across teams with similar growth rates will be very similar as a whole.
The difficulty in balancing the game is caused more by the double strike threshold, and the combat system being addition based. High defense values can make characters basically unkillable by low power units, and as stats grow, there's a wider range in values, which leads to very high and very low damage spreads, which is problematic. The difficulty in designing around double-strikes should be straightforward. Pokémon uses multiplication based combat, which accommodates high and low stats much better, but it also makes things difficult to calculate. Fire Emblem needs battle outcome calculations to be easy to assess for the player as the consequences for making mistakes can be severe.
Finally, we look at the alternative systems. Fixed growth rates function pretty similarly. It prevents stat screwing, but ensures a reliable path forward for the player so they won't need to improvise.
A no stat growth system would make it easy to balance battle difficulty, but removes the player's strategy of balancing good growth rates vs. good initial stats. The player simply selects the strongest/most situationally appropriate units. It would be much like pre-deployed maps in Advance Wars. This isn't really sufficient to base an entire game around, so Advance Wars adds resource management.
Lastly, we can reduce the impact of stat variability by working with larger numbers, like most traditional RPGs. Each level up gets you say, 3+/-1 points of strength, so a low and high gain wouldn't be too different. This is kind of a middle ground between fixed and random growths. It's not completely reliable, and requires some improvising and replayability, but not drastically. The main disadvantage of this system is that it leads to large numbers overall, which can become difficult to calculate.
Figuring out "these 3 enemies have a total attack power of 73, will attack 3 times, and my character has a defense of 15, so the maximum damage my character can take is 73-3*15=28" is doable, and the type of calculation a good player will be doing regularly.
"There 3 enemies have a combined attack power of 281 and will attack 3 times, my character has a defense of 36, so the maximum damage possible is 281-3*36=173" is substantially more cumbersome, so the player would then be more likely to make ballpark estimates, which leads to more deaths.
So in any case, criticism of the existing system is fair. But it's not without merit, and I can't find any alternatives that are strictly better (I personally advocate for random growths, but seeded at the game start to prevent save-scumming. Also, generally keep defenses low due to the addition based combat.)
This video is fantastic and very well thought out, accurate and thoroughly developed. And I say tha completely disagreeing with your view. Yes, growths can be annoying and rewarding. Sometimes both. But this is exactly what gives the player the freedom to beat the game how they want. There is no set path. Nothing says you have to use or level certain characters to reach the final chapter. This is why I find the "lord defeats antagonist and only the lord can do it" a terrible mechanic, even with a solid lord. Because it removes that freedom and insists you use a specific character with a wide variable of stats after 20+ levels to kill a boss with fixed stats. The beauty of Fire Emblem is being able to steamroll the game with a diligently raised and lucky Florina if I want to. I can raise a balanced army, or I can train 1-2 decent characters, get lucky and annihilate the opposition. OR if I get unlucky with a supposedly good growth character, I can change plans and train someone else. It adds so much replay-ability and characterization to a game that is highly character driven. It also subsequently allows character that would normally turn out average-bad, like Mekkahs Speed-Bartre, to become exceptional given luck.
I used to love growth rates until I came across Awakening Lunatic. When your units get stat screwed, you might as well let Robin do all the work because going against strong and speedy units like Heroes are just a nightmare to deal with.
In Fire Emblem, and Tactical RPGs generally, there are two forces in tension: dramatic uncertainty and planning.
They begin as something close to a puzzle game where many (but still not all) factors are definite and known, and gradually diverge toward a large field of uncertain possibilities, mostly due to growths. That means the early game can be a precisely tuned experience (though sometimes it's Awakening Lunatic+ instead), but has limited replay value. For late game, if the game isn't going to be dynamically procedurally balanced (which would be possible, but exploitable), and isn't going to give us granular difficulty options, designers have to target the difficulty of a one-size-fits-all solution for each difficulty level against some set of “reasonable” expectations. They guess. I assume they aim well below theoretical maximums most of the time, and that is why Fire Emblem's hardest difficulties tend to become easy to outscale in the late game (and are not like Apotheosis). The interplay of experience gain, over-leveling, stat caps, stat growths, RNG abuse, and boosters (including equipment, rallies, etc.) allows you to put units significantly above the difficulty curve. And you usually only need one such unit to dominate the game. But ultimately, Fire Emblem is not designed to beat you. That's the illusion. It is designed to try to put up a competent fight. The player is supposed to win, and should be able to win in a variety of ways. That (partially) excuses the imbalance of players becoming OP. And on the flip side, Fire Emblem's gradual trend toward anti stat-screwed mechanics over time is promising. I do wish IS gave us more options to tailor the experience beyond self-limitation, like an average stats or 0% growths / no experience mode, for players more interested in the challenge than the victory.
I was wondering why my Mercedes has a low attack stat since I was trying to make her a physical hitting unit but then I remembered growth rates are a thing and then figured out she had a 25% str rate 😐
I wish they just had a mechanic where you should have minimum stats for level intervals. Like if you're not at 12 strength by level 14 the game keeps pushing you to get that.
Shining force has it in every game, Kaga put it in Berwick Saga.
Fe on lunatic i think fixes a lot of the problems i have with level ups. A lot of my frustration is when you get a really good level up and then die and have to restart. On lunatic your characters get a fixed level up seed. Levels are still randomized, but your characters levels are determined the moment you get them. So you can't lose a good level up, but also cant save scum when you get a bad one
There needs to be a way to control the environment more. To me, this is the hall mark every good strategy based rpg, the ability to execute crowd control through AoE spells.
This should be the real utility of mages, instead of just being archers who target a different stat.
Why can’t characters who control the elements use those abilities to control the environment?
Obviously you would have to figure out how to balance that, but you could easily tie it with weapon usage.
There is this Cool unit called ulki who has both good bases and growths thus destroying the political identity of fire emblem and making me go 😎
Donnel is better
@@cringekid3993 O_o
There’s this cool unit called every other Tellius prepromote who has good bases and growths thus destroying the need for Ulki and making me use palette swapped Hugh
@@Edgeperor I’ve taken longer than I,I’d care to admit trying to figure out who palette swapped Hugh is. Is it Bastian? Soren? I love the Tellius cast but I legitimately don’t know who you’re talking about lol.
Great video! This is an interesting analysis coupled with cool footage and a very clear diction. Looking forward to your next "FE essays"
Man that guy really paid you to provide your own blackmail
My Tobin in echoes was godly. Growths make units good or bad with rng so it's fun but they have trends so it kinda ends up being the same run to run.
Ngl when team building I always get stuck on growths. This is usually because I design for endgame, though funnily enough I don't consider whether I'll get screwed or blessed. Maybe the common denominator would be fixed growths + a small % to add to the other stats.
You know I’ve always thought, what if we could devise a system that scales enemy stats based on how RNG blessed or screwed the player’s units are. Like if you keep getting good levels the stats fo future enemies slowly increase. I suppose that in a way is the same as just removing RNG together as it means getting RnG blessed or screwed doesn’t change anything. A tricky problem to overcome.
yeah, what's the point of leveling at all if it doesn't change anything to your relation to ennemy units? at this point you might aswell remove levels entirely (which honestly might not be a bad idea).
Yeah I definitely resonate w this video. My most recent play through of DS Shadow Dragon saw my fully leveled Marth ending the game with a whopping 10 strength. Only was able to beat the final boss on a lucky Excalibur crit
On another note, Clash isn't actually just a red vs. blue killing field where red is just a ton of mooks. I mean, it is, but it isn't. Particularly on high difficulties.
Clash introduced a second game over condition while also oriented on this flat, open prairie such that you had to approach your objective diagonally. This diagonal funnel sees you lodged against two flanks, and if your units aren't sufficiently powerful enough or you leave gaps, you will be pushed into, and your second, fragile game over condition will net you a loss.
On lower difficulties you just send a strong unit in each direction and the map pressure tends to not exist.
On the Japanese Maniac mode, there are additional pressures that prevent this. Chiefly a large influx of siege tome users. Also, though, more reinforcements and more promoted reinforcements than usual. Due to their enhanced numbers and stats, you feel the actual map pressures, and you have to find a way to redirect enemies away from your new game over condition, while pressing out as hard as you can.
But even with an army of gods, you have one Achilles heel that you have to account for. I think it's one of the best designed straight up open maps in the series for that alone.
i want a whole video on the arena real talk. such a genius risk reward system
Would creating resources that are only effective to the enemies in the final map (like special weapons) work? hmm now that I write it it seems meh
The Gaiden system fixes rng screwed players for the most part but on the flipside, all your units of the same class feel EXACTLY the same for the most part, aside from when they surpass the promotions base stats. I think a better thing would be #1: do this by unit, this should help with the lack of individuality. #2: It would be good for there to be a controlled random element. I can think of two ways.
#1: Give an acceptable range for the target, then look at the average of that stat and set that to 0, you could either evenly divide or you could use standard deviations from the mean. In a simple problem of 35% growth rate of a level 1 unit, the average would be about 6.85 points over base at level 20 with a standard deviation of about 2.08 (it's an approximation), this would lead to a variance of -3.29 to 5.84 standard deviations, and then -3.5 to -2.5 would be a -3 modifier (minimum), -.5 to .5 would be no modifier, and 5.5 to 6.5 would be a +6 modifier. At level 20 promoted it's average would be 9.8 above base stats with a standard deviation of 2.52, this would lead to -5.07 to 6.02, allowing from a -5 to +6 modifier. A couple things to keep in mind is that due to basic probability laws, getting the extreme modifiers would be EXTREMELY UNLIKELY, and would still lead to a fairly minor difference as opposed to a minor modifier. Also while this took a while to manually plug in, it's basic formulas computers could do in a flash. Early promotions could be easily handled by just deducting points (levels left * growth rate rounded up), so promoting at level 10 in this scenario would be 3.5 deduction so -4 in that particular stat. Notably caps do exist and so do stat boosters, there would be a secret value holding their true stats ignoring stat boosters and caps, these would be the ones used for calculation, displayed/functional stats would first have a hidden stat booster modifier applied, and then compared to the caps.
#2: Basically take the players current stats and multiply it by a multiplier to get the sum as close as possible to the actual sum total, you could make it weighted if you wanted too, but that's the basic gist of it. Less complicated to explain by a lot but not that much easier to program and IMO less cool and fixes the problem less (bc you can get blessed in a meaningless stat while getting screwed in another, weighting can help this but it's not perfect and can make what's a good or bad level up very weird). Like before there are deductions for early promotions, and stat boosters and modifiers would be applied seperately.
I'm increasingly thinking about why fixed growths would mean missing anything long-term. You could create a little divergence with fixed growths even(or clever conditions that affect it).
Playing blind I think many of us assumed fixed growth the first time until replaying after having to reset a map a few times, or very offbeat growths. Still, even with fixed growths the shiny levelup experience doesn't go away.
It might mean losing a little of what is a unique FE thing, but it's got a lot of potentional to design other parts better, or tweak the effects stats have knowing the player can grow it consistently(I'm looking at you dexterity).
The video discusses failsafes, alternative sources of stats, and different map design goals, but I'm still not convinced it's a nessecary hassle. I mean you could swap it around and make random growths unlockable as ng+ with the risk the game is poorly playtested for it, just to have random growths. But I feel the wealth of gameplay with more fixed growths it worth it.
It's mostly about something the game teases, but never quite follows up on: needing to decide to train growth units, even when it's harder in order to prepare for later. In practice and with less efficient play this happens a fair bit anyway, but still, making it outright optimal might make for a interesting game/tier list discussion.
I know it's late, but the more consistent things are the less challenge there is and the worse replayability it has.
As far as I'm aware of most of the games vanilla versions do not actually tell you what the growth rates are, so specifically the intended way to play most of the time is to simply react to how well or not units are growing. You basically lose both replayability and variety if you make everything fixed. Which is why I pretty much the easy way to get through any game is the over invest into only a small amount of characters, especially with things like stat boosters that are consistent. It basically would just make progressive decision-making more shallow, and creates even less incentive to use characters that come in later on. Because the way fire emblem is actually built to be played is to continuously play almost like your iron manning, and not restart every time something goes wrong. If you were intended to do that it would actually just give you a game over like when a lord dies, but you can't stop people from metagaming the systems.
I never really get excited by the level-ups in my limited FE experience. It is either just "okay, I guess this makes up for that last bad level up." Or "Well that was pointless." It never excites me and I would just rather a more consistent levelling experience.
It would also really suck to be invested in a specific character for their personality/story and then they get screwed out of viability due to RNG. And I care about the social sim aspects way more than the gameplay so that would be a deal-breaker on a playthrough for me.
Lmao I love that all the chads are in the "Gilliam" tier
In Hypothetical Mozu gained 0 strength. She would still deal more damage than any unit because She gain access to Spendthrift, Life & Death, Bowfaire, Quick Draw. So even with 0 strength Mozu will deal 29 damage. And this is only if you use heart seal. If partner seal is added as variable she can deal 34 damage with trample. Mozu is good Bopper. Speedy as the name suggest is too hasty.
HOLY CRAP
SERIOUSLY?!
O_o
@@Ramsey276one Yes. The only unit that can deal more damage than Mozu is Xander from Revelation and Corrin in any route except Birthright.
Xander setup
Heart seal= Trample = +5
Friendship seal (Ryoma) = Swordfaire +Life & Death = +15
Heart Seal (F-Corrin) = spendthrift = +10
Personal Skill = Chivalry + Elbow Room= +5
Corrin pair up = + 2
Total = 37 damage even with 0 strength
Corrin can reclass to any class she want. Best build is Mozu build but you swap bowfaire with Swordfaire. The reason why Corrin damage pool is higher is because she can pair up with Gunther. And Gunther give +3 which mean Corrin can deal the same damage as Xander thanks to Gunter Personal Skill.
@@riqua27 WOW
As a true casual the best way to look at growth rates is to ignore them and like the highest difficulty where the strategies required are so batshit that you dont even feel like you are playing a tactical war game and more of a "overpowered badass kills everything" just ignore that as well.
Yo so not to talk about *Heroes* but...
I like the fixed cap for characters in heroes. Personally i'd have a fixed statline at max level unpromoted, that you reach randomly.
For example, you're destined to get 10 speed levels before you hit unpromoted 20, but you've only got 2 so far and you're already at level 10. You hold off on promoting that unit, because they've now become an est in so far as most levels they get from here to 20 will be great.
On the other hand, if your guy is destined to only get 3 strength levels, and by level 10 he's got 'em all, you can decide to promote right away since you got blessed.
It keeps the randomness (and short term challenge from bad levels) but doesnt let a character you want get screwed all the way to the bench.
Plus you can design around it more easily.
Logically speaking, growth rates are bad and inefficient.
Psychologically speaking: stats go burr burr, give me more gambling.
Another way to mitigate bad level ups is to do a compromise between fixed level ups and random variability, by clamping the amount of upticks that can happen each level up. Meaning, instead of there being a chance that you don't get an uptick at all, the lowest threshold is always 2 or 3 upticks spread out accordingly to each stats' rate, and to balance it out on the other end maybe the max upticks would be 6 or 7 stats each level. That way you get the best of both worlds, you don't eliminate the probem per se, but you might mitigate the worst variability that would lead to polarising gameplay. So basically, the worst level up would be 3 upticks in your worst stats, whereas the best level up would be 7 upticks in your best stats.
Great video! Really got the level up screen feel on the nose.
One small thing -- the music was really loud for me and made it hard to hear you clearly.
I thought growth rates were random only in terms of when they happen. Meaning a maxed out character will always have the same stats. Or is it just how it works in Fire Emblem Heroes?
Low power curve Fire Emblem is the best Fire Emblem. There are 2 other ways of dealing with random stats
FE2, 15, 16: Minimum stats on promotion
FF7: Stat Ranges (Unaugmented, the Dexterity of every FF7 char will vary by 7 points by level 99, depending on RNG)
Finally someone praising weapon ranks and their genius!
i know a couple of people who wanted to use Wlv stat in SRPG Studio and i'm just like
why
@@dmas7749 Because it applies to every usable weapon type so you don’t have to go through a tedious weapon rank grind
@@JohnDoe-kn7ex i still prefer it over the random chance to get a stat that determines it, and you usually use one weapon type with a unit most of the time, it really isn't that bad