Of note you can define a global snapping grid the size of your wall segment in the wall segment’s blueprint, so you can make it so you can only place wall segments in the correct, chunk-aligned positions.
I didn't know you could do this but wouldn't that limit you to only that grid? Say I put a chunk down on the map without that grid, then set the grid up as you say, would the next chunk line up with the chunk already on the map?
@@theotherbigfoot There are 2 modes: absolute and relative. Relative just lets you drag a blueprint in a line and it will tile down correctly, snapping relative to where you clicked your mouse button down. You can click down anywhere though. Absolute snaps relative to 0,0, with an adjustible offset. this is probably what you want, because you can place down a wall (or rail) section anywhere and know it's aligned with all your other sections. Yes, if you place down a section misaligned with this grid it won't be aligned, but you can adjust the offset in the blueprint. It's best to just place your first wall section aligned though. So yes, it would limit you to that grid, whcih is the entire point. You can make the grid half the size of your wall section of course, to get finer positioning, and then have converter sections that are half a wall section with a diagonal step. Or align the grid to where you placed your first section without the grid.
@@sea_kerman Great explanation, thank you, I didn't know any of this was possible. It's like being back in AutoCad!! You have definitely given me something to think about.
yeah, I was about to suggest that too :D If I'm not messing this up, you could either set the blueprint size to 50x50 and use absolute snapping, but that somewhat limits placement (what if you want to start your wall 10 tiles further down?) as an alternative that gives you more flexibility, you could set the blueprint size to 2x2 and enable absolute snapping. that way, you still have to pay attention when tiling segments together, but at least they will always be rail-aligned (never an odd offset from 0,0) I'm not a fan of relative snapping for walls and rails since you can't ensure things line up when starting in different places, and getting the initial starting point right sometimes isn't trivial. relative 50x50 is definitely an option too though, just not one I'd use here
@@jemand771 The grid snapping and the offsets, all interesting stuff. I'm definitely thinking about how I could make use of that in my builds. For the perimeter wall though, I think I'm pretty happy with the system I've come up with. The rails are included so it is foolproof. I can definitely see myself fu*king up the offsets.
On vanilla biter AI: Biters only actively target military structures (radars, turrets). Biters will attack other structures if they can't find a path around them. Biters have no other preferences. They do not target corners or water-edges specifically, nor radars or other structures specifically. However, a biter attack group targets the closest military structure. If your wall is rectangular, the corner is the most likely target. Consider a rectangular base. All biters directly north of the base will attack the north wall, since the closest path to the base is directly south. Similarly for all biter bases directly east, west, south. However, consider a biter nest northeast of the player base. Then, the closest target is the northeastern corner. If you imagine coloring a Factorio map by the target a biter would choose, the strip directly north targets the north wall, but the (infinite) rectangular area to the northeast targets the northeast corner. Therefore, a much larger area of map targets the corner, while a much smaller area of map spreads its attack to the whole wall. Similar reasoning explains biter tendencies to attack along water edges. Fun fact: instead of a continuous wall, a series of "bunkers" that draw the attention of biter attack groups works just as well to defend the base. Also, biters do not attack walls if they can path through them, hence the effectiveness of "dragon tooth" or "maze" designs.
I understand what you are saying but that has not been my general experience. Over multiple maps over multiple years my corners are always in more danger than my straight bits. It might just be that they are inherently weaker, with an exposed corner turret, like I explained in the video, and not necessarily the pathing of the biters. I believe it IS the pathing of the biters, I believe they are able to sense a weakness in the wall, but that belief is only from general gameplay experience and nothing more specific than that. Pathing aside, I experienced a significant decrease in biters being able to demolish turrets when I started using curved walls. There was a noticeable improvement. I think it is worth using curved perimeter walls just for that reason alone.
@@cewla3348 Yes, if we have an infinitely straight wall with an infinite number of angry biters then we would expect the wall to be attacked pretty evenly. But a corner in that wall represents a discontinuity in that assumption. It isn't a straight wall anymore, and a turret on that corner will always be less well protected than the corresponding one in a straight section.
@@theotherbigfootThey said corners are in more danger, they only clarified that there is no unique priority to the corners, a corner is just going to be the closest target for a large area, so more biters will attack a corner. 6 for 1, half dozen for another at the end of the day.
@@theotherbigfoot The issue is not the biters' initial target, but rather the pathfinding. The way I understand it, the pathfinder sees the corner as the easiest route to the target. Once in sight, the military structures become the target.
I use the following patterns for defense: 1. My walls always start and end at water. 2. I use landfill to continue the wall into the water, so that even on the edge, there's always overwhelming force. 3. I use these walls to close off large sections of land, preferrably so much that it includes all pollution. If I'm attacked this means my pollution grew so much it's time to conquer new land.
You can see the size of your pollution cloud on the map, so if you've already walled off that entire area the biters aren't coming because of pollution. Biters will split off in small groups to start a new nest in an area without a nest, where the biters in a group slowly turn into worms and spawners. The biters coming from outside the pollution cloud are from biter expansion only, and biter expansion decreases with evolution. Artillery is then also far more useful with high evolution because after nearby nests are cleared they only have to deal with occasional expansion nests.
Not sure if it was said in the video or not, but a heads-up for anyone trying to import the blueprints: They rely on pipes from the flow-control mod and will not work with vanilla pipes.
I randomly clicked on this because my youngest son and I are about to get back into Factorio. WOW! This is not only a fantastic set of blueprints , but a sound defensive mindset for biter defense. Thank you so much for this.
This is easily my most favourite of all the comments I have ever received. Just such a wholesome image. Father and son playing factorio. Someone should put it on a christmas card
About using belts. Particularly in the earlier game, getting them full can be prohibitively costly. What I've started doing is using ammo feed belts that eventually circle back on themselves and a circuit timer controlling the feed belt to have it hold back ammo and release it in chunks (eg. belt on for 120/1200 ticks, adjust by need). The feed belt itself is being replenished just behind, where the unused, returning ammo backs up, so it'll only be replenished when it there isn't enough ammo circulating for it to back up. Basically, it's just a way of controlling the ammo supply belt buffer size so you can do longer supply belts without the ammo costs getting out of hand.
I tend to embrace the ammo costs as stored value. The more ammo I can squeeze on the conveyor belt and the quicker I can do so, the less work needs to be done later.
@@theotherbigfoot Yeah, guess in the long run that works. I guess what I'm doing is just because I've been in situations where the ammo production has eaten resources I needed for other stuffs. But... that's probably just me not scaling up enough early enough. I suppose it could still be useful if trying to run depleted uranium ammo though? An advantage of this system is that with a circular belt ammo delivery gets spread out even when not saturated, and keeping the ammo moving in blobs kinda stops everything getting sucked up by one bit of the belt. Dunno though, never tried to do depleted uranium on any scale.
@@ashedtogether Oh dude, I wasn't telling you that you were doing it wrong, I was just saying why I do it differently. Feels like you have a well thought through setup. If anything, I'm gonna stick with what I do already because its easier!! You aren't wrong about the resources, but I've recently been massively scaling up the resource deposits so that I can build behemoth factories, so wasting resources is less of a problem. I haven't tried uranium ammo yet either, but I am about to. I'm going to divert everything from the perimeter wall to a new setup to convert to green ammo and then back again. I've started converting already and am building a reserve in a whole load of storage chests before a big switch over. We will see how it pans out. There is content planned.
Very interesting! Some thoughts: 1. Having all of your ammo on a single belt is asking for trouble. Not only does that mean you need to have a *LOT* of ammo travelling at once, but, if anything goes wrong and the belt breaks, the turrets after that point will eventually starve. Not Good. A better way is to use local loops that meter ammo into packets with a time interval between each one, served by a train stop that is only enabled when ammo is needed. 2. It is important to have your roboport be well back from the wall. Why? So that repair bots only enter the danger zone *AFTER* the wave of biters has been destroyed, so they don't get damaged/destroyed along with the wall/etc. 3. Dragon's Teeth that break up biter pathing is *vitally* important to both slow them down and create chokepoints for flamethrowers. 'Inverted Teacups' works really well for that, but make sure to have 2-wide paths so they don't start beating on the walls. 4. One alternative to using ammo belts is to have turrets surrounding/on either side of a requester chest, with stack inserters passing ammo from turret to turret as needed. (You *DON'T* want to take this too far, though; lest some get starved for ammo when things get hairy.) Cheers!
Wow that is quite a series of comments. Ima deal with it the same way I deal with my work emails!! 1. Yup, you are absolutely right, and, in the early stages of my games when I don't have robot coverage, a problem I frequently encounter. But in the later stages, once I have a final wall built (which I'll be doing from now on with these segments) it isn't an issue I have ever had. If it did happen, just for arguments sake, there would probably be plenty of ammunition available to the other turrets until it got repaired. 2. The roboport is pretty far back from the wall. I'm happy with its placement. Putting it any further back would ruin the aesthetic of the segment. I don't mind making a few sacrifices to get the elegance I am trying to achieve. 3. Someone else pointed this out, I don't do this because it opens the door for robots to be out in no mans land repairing bits of wall vulnerable to biters. It is a strategy I understand, I have seen the videos, I understand it's effectiveness, but that I choose not to do simply because I would lose too many robots. I understand if you completely disagree. 4. I like my ammo belts. IMO they work really well. I definitely do not want to rely on logistic networks. I never rely on logistic robots for anything other than fuel for my trains, maybe a bit of u235 and/or u238, and my personal logistic requirements. Basically, I don't rely on robots for anything mission critical. Perimeter wall is definitely mission critical. Literally everything in my factories is on a conveyor belt. Even my water!!
@@cewla3348 Bottled water is indeed hard to deal with, I have tried it in 6 or 7 factories now and it still doesn't work perfectly. But the challenge is what I love about this game, I'm also pretty stubborn, so I'm gonna keep going until I nail it.
@@theotherbigfoot I just watched it again and had another thought: 5. Based on my own experienced, I realized that each wall segment *SHOULDN'T* automatically link to the roboport in the adjacent segment. Why? Because, over time, the bots will tend to 'clump' and then have to travel long distances--sometimes over no-man's land, depending on the map--to repair/replace damaged structures. (#beentheredonethat) Roboports aren't expensive, so making each segment 64x64 (or w/e) with a (tileable) spot on the edge to hold linking roboports will allow easily dividing the wall into logical sections that keep all bots local so they can act quickly. (By the time you are using walls this expensive, bots are relatively cheap, so it's good enough to drop a steel chest full of them in each section with a speaker to alert you when it needs topping up.) All that being said, while I would never use your wall system--because I habitually build simple walls *FAR* beyond the pollution radius so biters never really become more than a nuisance--I am in absolute *AWE* at what you've designed here. Bravo, good sir!
Bigfoot didn't know about blueprint grid snapping at the time, but you should also make sure that your wall blueprints are grid-aligned with your rail blueprints. If they're chunk-aligned, even better. On a different note: One thing I enjoy having in my walls is a wide (~6 tiles) personal vehicle gate to make sure I don't crash into the walls as I'm running into safety, but those areas tend to be big weakpoints that suffer more than the rest of the wall, even with higher turret density. One solution I think might work for this wall system is having an invagination-shaped segment with many defenses, that tapers to a point on the inner side but curves to become more and more open on the outer side, similar to the mirrored semi-corner walls at 12:10. Rails, supply belts/pipes, and roboports/artillery/radars would have to be adjusted accordingly, but it would be too much of an issue, other than disrupting your usage of space.
*This* is what base design is about! I've experimented a lot with wall designs that exploit the biters' pathing AI, primarily for flamethrowers, since I've been playing a deathworld marathon save that makes ammunition costs way too high to justify. I haven't gotten to the stage of working tileable designs of this scale (still working my way to roboports in that save), but the tips here (particularly the one about aligning the rail and roboport before doing anything else) will be a life saver. Thanks for the inspiration, great video!
Great explanation. I've been looking for a new wall design, since my default one feels boring by now. I like to build within the standard factorio chunks. I protect my corners by making 'indents' like so ___ |__ |__ | | and then adding more dragon teeth, which 1. slows down biters --> clumping them up for flame throwers 2. often splits them into groups going for the sides
My first comment was me absent-mindedly pondering on the choice of conveyors vs logistics systems - so commenting again for comment god and just to say great video, your approach is exactly the kind I love seeing in games like this, it's enjoyable just to see other people vibing and working through it and sharing their reasoning with the world - it's what makes games like Factorio special I think =D
the corner advice is good to know i usually use only flame throwers, and the walls are further away from the flamers than what the spitters can shoot, so only the wall gets damaged by biters, and the spitters get too close for some reason as well
Wow that looks really cool, I'll have to try some similar designs once I'm done with my current "biter air filter" save where I experiment with limiting biter spawn positions and trick them into depolluting for me as long as I have ammo, oil or power.
Fair point, I use a mod called 'flow control' which makes this possible. Doesn't really change the overall geometry lesson though. If you were building your own segments then you could just leave a gap between the pipes.
A notch-out rather than curve would allow to target biters from two protected directions at the same time, which also solves the issues of bare corners. It may make some design aspects easier. (At least if the biter horde behavior settings aren't having attacking groups to be at ridiculous sizes.)
Bro...the biter AI doesn't target corners on purpose or otherwise. It's just that flat wall segments have small rectangles where spawned attack groups can originate to end up at those segments while corners have a full 90 degree spread of area where the corner happens to be the closest part of the base.
Ignoring the biter AI for just a moment, I feel like you agree with the general premise I am making in the video, which is that the corners of a perimeter wall are weaker than the straight portions. And if you do agree then the reason why this happens is irrelevant. Corners are weaker, get attacked more, whatever, and I thought curves are a good solution to that problem. I still think that. I'm going to be using these wall segments in every base I build from now on. With regards the biter AI, in the timelapse clip where I record the biters attacking the corner of my base there are moments when the biters readjust their position before attacking. For me they are re-orientating themselves to attack the corner at precisely a 45 degree angle. They don't just 'activate' and attack the first bit of wall they came across. They activate, reposition and then attack. I don't know anything about the inner workings of the game, how the biter AI works, I only have general experience of the game. A lot of general experience. But no technical understanding. What that repositioning suggests to me is that the biters look for a weak portion of the wall and then attack that portion. That is how the game 'feels' to me. But as I said, I don't have any proof of that other than from what I have learned from general gameplay.
I made a similar setup using chunk alignment. There was a laser wall, flame wall, and gun wall, and they would all layer over each other to beef up the wall and defenses if needed. Train gates, vehicle gates, roboports, rounded corners and bastion corners. I remember figuring out certain perfect distances so everything lined up felt really good. Probably spent 36 hoyrs of design for a wall that was completely overkill for anything besides hardcore deathworld maps, but they looked sexy as hell
Sounds like you were one step ahead of me. I would be really interested in seeing what you came up with. I think these segments took about 20 hours, and it isn't the first time I've tried this so I probably had a bit of a head start. It isn't a small undertaking but, as you say, the results are pretty damn sexy
I have been trying to build similar walls like this for basically the entire last year now. I though, I could try to have the blue prints chunk align, which means they are on a grid and will fit together, I even made sure to make every block align to a roboport and be aligned to the rail grid, but I ignored to make sure the turrets could be mirrored in the diagonals and my corners were not gradual but switched ratger abrupt from straight to diagonal. I made that work in the end, but it has been a chore and its not as elegant as these are. And since I have to redesign my chunk aligned railway grid again, I can redesign my perimeter and outpost walls, don't know if I will make my outposts into railway interchanges, but trains getting stuck outside the walls is a hassle for sure
Someone else said this exact same thing. The base I'm working on isn't quite ready for a full blown test yet. I made this video at the moment I needed to build the wall in that base and it is built and finished but the base hasn't been fully developed so it hasn't been tested. I guess now I need to do a follow up wall test video.
very useful information - I'm just about to install a walled perimeter in my first 1.1 base and that part about biters attacking the corners I didn't catch before, so thanks for that !
I don't do a perimeter wall. I do robot printed bastions with roboports in each. Surrounded by walls, they each have two corners, but are compact enough that all the turrets defend each other. I print these offensively to effectively cordon off an area, and expand; Then I move in with either powerlines and lasers on foot, or spidertrons later.
Corners are a natural weak point in all defensive structures in real life, but in some instances of technological advancement they provide a useful defensive posture. Creating stars, for instance, makes a fort's shape far more effective. A simpler solution to perimeter defense for most people is to create corner circles that present a circular exterior segment that connects to the straight section via another curve inward. Sort of like an Omega, but less angular at the turning points.
why does nobody ever use belts, they're a pretty effective add on to any perimeter. Just plop down a row or two facing outwards, prefferrably at choke points, even yellow ones support your turrets really well
Brilliant! But I'm sure i'm missing something. In my playthroughs I've gone to exclusively 2 lines of laser turrets/wall/3 lines of mines all supported by roboports and periodic artillery batteries. I have a couple of blueprints of different lengths and can snap them in easily and deconstruct them easily upon advance. I try to start/stop segments at water lines. If I must have a corner I simply add more mines/turrets at the corner. I keep buffer chests at intervals with all the wall components. Even if I invite the horde through artillery spamming the worst that happens is 2 layers of mines get used. Most attacks result in obliteration with at worst a little goo on the wall, quickly repaired by the bots. With this setup I don't need any supply network outside roboports. The buffer chests make expansions quick because all the supplies are local. I wonder if with mods or the new 2.0 update the biters will eventually get too strong for my system and I'll have to adapt. (I absolutely love flamethrower turrets, I just don't seem to need them.)
i'm still "new" to the game, having just begun using artillery turrets on my longest ever continuous save file. dont know why i never thought to practise my wall building in sandbox mode. i currently use a maze wall with dragon tooth corners which is for sure effective right now but my evolution is past 0.70 already since i had a slow nooby start to my game. like i still only have one line of weaponry on most of my walls. my brain hurts watching this but i'm inspired to fuck around in sandbox now
I played way too much deathworlds and I always build walls without bots cause I meed them already in early game. This led me to realization that it all literally doesn't matter. all you need is couple of flamers, few layers of wall and repairing walls until you get to bots. once you have bots DON'T give them repair packs and upgrade bot speed and they will safely replace walls without dying.
Great job! Such a pitty that outpost ones don't have flamethrowers though - what can be more fun than setting up an outpost-supply-train with liquid wagon or barrels of oil?
No you can't, you're lying!! I checked this a few years ago, there is no way the developers would add something to the game after I had already tried it, found there was no way, and ruled it out!! That has never happened before!! Damn this game.
@@theotherbigfoot i checked it myself now, but normally you can flip a blueprint with F and G. Only some minor blueprints can't flip, but then there will be given a particular reason why it cant.
@@ronphil6666 Oh dude, I was being sarcastic. This is an excellent tip you have given me here, if only I had known a week ago before I started all this.
I'm guessing this is a bigger issue with things like the Rampant mod, or Death World preset? Then again, I'm also on Rail World preset, so biters haven't been a major concern. Even so, great demonstration (still watching). My setup started with a single-layer wall with flamethrower turrets spaced kind of on feel, but ended up about 7 tiles apart I think. Between them is a single gun turret. Belt for ammo, inserter to load it, and power poles to keep everything running. Separately, with the new fluid mechanics, my wall now also needs a pumping station every 320 tiles. I made a nice little insert that pipes out 10 tiles to a tank, then a pump, then pipes back into the line of flamethrower turrets. This has two effects. One, it extends my pipeline without complicating logistics with rails. And two, it provides a small fuel reserve that will never flow back into my production facilities. That way, even if there's a run on production, there will still be at least 25k fuel in reserve per line of flamethrower turrets. In my next expansion of the wall, I plan to add laser turrets (left them out for the achievement this run), rocket turrets and artillery. Not that I need it, because even behemoth biters/spitters can't breach the single-layer wall before getting burnt to a crisp.
Is there somewhere I can get the entire set of blueprints in one string? I see the link has them all in sections but it'd be nice if they were all in one book. Nevermind, I didn't look hard enough :)
thx for blueprints , just dint figured it out to redirect the belts ... is feed them from other site . got the blueprint to or is it manualy to adjust ?
Before having the blueprints: "Holy shit, what the hell is this kind of rocket science jargon" After having the blueprints: "This is as easy as using Lego blocks"
Awesome video, although looking at the walls I usually build I must be much easier on the biter settings than you... Apart from the comments on grid alignment and mirroring blueprints, would also add that biters I believe focus on some modules over others (specifically radars) and they're 'preference' for corners is likely more to do with that - worth testing, but I tend to just go with 90 degree corners and keep radars away from being in the extreme corner.
I put the radars there specifically to provoke the biters because I was recording for the video. That isn't my normal practice. That was just the corner in my current base that was getting the most action. The radars definitely sped things up but this corner has been problematic the whole build. I have been constantly repairing it. It has been the same with the other corners, although just not as heavy, which is why I didn't use them. The straight sections have been more or less impervious the whole time. As long as I've had a continuous line of turrets and available ammo the straight sections of wall have been able to resist the biters. This is all on the standard settings. Normally yes, I do play with higher settings on the biters, and the effect is definitely magnified, but the principle is still the same. Walls with curves, in my experience, have been inherently stronger. I'm not saying you are wrong, I guess I'm just saying that this is why I go to the trouble of doing curves in the first place.
A beatufull pixel art, but I can't see the point. I've played with mods like rampant and natural evolution, I hit 99.9999% evolution and never used such thick walls and so much turrets. Your builds look so overpowered, that I don't understand what mods do you use if you need this kind of firepower? I only see the point if you had no roboports to repair, but they are the core of the build.
I originally built my walls like this, trying to cram as much turrets in as possible with lots of walls, but after reaching late game, it's just overkill. 95% of my turrets are unnecessary and just waste power and resources. I've switched designs to better exploit biter behaviour and reduce maintainence costs. Might switch back to something like this if I ever install Rampant.
Wow, I didn't even know that mod existed. If I didn't hate the biters with all my being and tolerated their existence simply because it is boring without them, then maybe I'd give this a try. I think my way of making the game difficult for myself is overhaul mods. Or silly race tracks.
A train station for mining? I was concentrating on just perimeter wall stuff. Point of entry for fuel and ammo - not 100% sure what you mean, there's a point of entry for trains. Do you run your ammo trains on a different train network on a separate set of lines?
Lol.. my first round and i got most of it right… for repair i need a robo port.. Also i build 3 pices straight, inside and outside corner.. made them 45° wall from corner to corner of the robo port since they seem weak… But i did not go for Overkill
Someone was just telling me about how the railways broke everyone's blueprints and I was like 'I'm fine, I don't really use blueprints'. Then you popped up and reminded me. I'm gonna get a video out of it so I can't tell if it's a good thing or a bad thing!!
@@theotherbigfoot hopefully not too bad to fix. Only the segment with the big loop is going to have to get one rail width larger circle. And with the extra diagonals you might be able to make a diagonal entrance. Good luck (:
Pfft. Correct answer is laser turrets and artillery and construction drones supplied with everything needed. Around lvl10 laser damage 3-4 rows of laser turrets, literally impenetrable.
This is a great tutorial on game mechanics, but *IMHO* by the time you have the resources to build defenses like this, you should have the ability to use artillery or MK2 armor with PLD to remove all biter presence beyond the reach of your pollution. At that point you only need to defend your artillery placements. This statement applies to vanilla and lightly modded play... deathworld or rampant mod might make this kind of defense more necessary.
o - been awhile since I last played took me off guard with 'artillery' lol - never understood the value of using conveyors to keep turrets full of ammo - seems needlessly troublesome when building small distribution hubs (which may or may not also be a good spot for a passenger stop as well) and using logistics distribution makes considerably more sense in terms of flow for rapid expansion after clearing an area, without needing to connect everything together by anything other than rail - which serves to transport the player and electricity anyways so would always be present. This is mostly me just being a bit puzzled tho - lots of other excellent problem solving and approach on display here. 👍
I have a 'maintenance' set of trains which whizz around all the outposts delivering ammo as well as replacement turrets, electricity poles, conveyor belts etc. Once the ammunition loop is full it only needs to be topped with what what gets used.
@@theotherbigfoot hmmm I see - oh wait I think I see what's up - the conveyor serves as a natural buffer so that you don't need to use circuits to ensure smooth supply? I've been too in love with having resupply trains on standby until an outpost requests more ammo/fuel/items/bots/etc. ever since they added the ability to turn train stations on/off >.< so I've just gotten used to always running the circuit lines along with the electricity and trains, so think that's why I'd not really considered it. Watching the network whirr to life as it's needed is just too satisfying 😁 edit: though I've still not found the most elegant solution for requesting multiple resupply trains across a vast network - always seem to get distracted by something shiny before it gets monstrous these days
@@NeoRipshaft Personally, i use "Set train limit" with circuits. Set limit to 1 if anything is required, otherwise 0. Trains will wait at loading area unless there is demand, then go to that particular place with no risk of multiple trains going to the same station or getting lost on the way. So far, works well even with dozens of destinations. Circuits are useful for controlling loading of resupply trains as well, so i do not have to set filters in wagons (just have "desired load" set by constant combinators) and i can use the same wagon for carrying junk from the wall (either from terrain or deconstruction after a new wall further out was made).
Why you don't abuse the biter pathing with a even a simple maze? A good funnel can negate most, if not all, biter damage. Also - why 2 kinds of flamethrower fuel? Either you are rich, so you use the better one, or you are poor so you use the cheaper one, no?
Ok, my theories as follows. I don't use simple mazes because when those mazes get damaged my robots have to fly out into no mans land to repair them, where they will invariably get destroyed by biters. So I just leave it to my turrets to get the job done, which are at least designed to be engaged with the enemy. The oil thing, I think the first time I used the flamethrowers I used just the one fuel and it ruined the balance of the oil processing in the factory. Entirely possible my newer factories are better designed and would cope better, but I think this is where the design decision came from. I'm not saying the way I do it is the right way, I'm just saying this is how I do it.
So I think theres something important here, each flamethrower fuel mechanically functions almost identically, Light Oil is objectively the best fuel as it gets 10% bonus damage and is the most economically available. You want to feed your turrets with light oil hands down. Heavy gets a 5% damage bonus but you can instead crack it into light oil for more total damage. I would still run two supply lines to the turrets purely for throughput but include some pumps that push the fuel in opposing loops of the base. Determine a 'seam' in that loop so the lines will fully saturate, this can be the same local as where the fuel enters the wall. Additionally, though it is a preference matter, I generally set my Flame turrets back further so they are more likely to leave fire along the wall where biters are likely to stand within the flames instead of momentarily passing though them. You mention having trouble balancing your oil production, this is resolved with a basic circuit, read the contents of your petrol, Light, and Heavy tanks, compare one to the other for each, and when you have say, more Heavy oil than light, the circuit enables a pump to crack heavy into light, the same for light to petrol. It doesnt work in reverse of course so you can use combinators to adjust the balance of this circuit as you prefer. Hope thats beneficial to you
Ok, just for you, well, and because I'm interested to see if my current factory can handle it, I'm gonna try out light fuel only in my flamethrowers. I'm pretty confident in my oil setup (thanks for the help but I already have a circuit setup I'm happy and confident with. I was so confident I made a video about it (oil and SPM)) so let's put it to the test and see if it can do light oil only. Convenient timing as well, I'm only a few hours away from hooking up the flamethrowers. It will work. I have no doubts. There are some design decisions I am still tied to from my early days playing the game and I just haven't tried anything different because habits. You say 5% bonus (I didn't know that), ima go grab it.
@@theotherbigfoot I'll admit this was the first of your videos I've yet come across, so I will have to check out your oil system video when I get a chance. I know people can do magic with circuits but I've only just started to understand them about a month ago despite having started playing before it hit steam, balancing oil with out circuits was basically just hoping that luck would do the job lol
It's no worries at all buddy, I'm just small fry on youtube really. I am no magician with circuit networks, I can do some cool stuff, but not a lot. I've made a video about that too!! I don't like plugging myself but if you are interested in understanding the mathematics of oil and spm then that tutorial is decent. I don't think anyone else on youtube has figured out what I explain in that video. When I set out on the SPM design route in my factories I certainly couldn't find a video that explained it. If mathes is gibberish, there is a spreadsheet. That is in the video as well.
Your grid is impressive, especially the railway inclusion. But why is your perimeter wall so expensive? On the length of one roboport I support a wall with only 8 gun turrets in pairs. Each pair has one requester chest requesting ammunition and some burnables the turrets get feet by burner inserters. So no electricity is needed besides the roboport. If you place buffer chests strategically this will support even long walls. Artillery will pull huge groups. I strengthen those segments with flamer turrets. Just make shure the turrest get ammunition before the artillery.
I do generally play with the biters at quite high levels, and I do generally feed the wall with just red ammo, so I have found a big thick wall to be prudent, but I am certain that if I made a video about a perimeter wall that wasn't completely over the top then nobody would watch it!! I do build my walls like this, and I have been doing for quite some time so I don't feel like I'm being misleading, but also I think to be successful on youtube these things need to be taken to their absolute limits. Besides, I generally build big factories so an expensive perimeter wall isn't really a problem. (There is nothing stopping you from taking my blueprints and removing the turrets you don't want. If you need to upgrade then you can just overlay my original blueprint on top)
@@theotherbigfoot as i said your Segments are impressive. And the video is a real eyecather. My biggest factory is a 1.2 k SpM Railworld. Biters are at normal level though so. Therefore i never experienced a bitter attack that would call for such a strong defense. Lately i found that with artillery on 5 th range extension the need for real perimitter walls shrinks to a minimum. So i build a heavily protected artillery outpost which clears the area before building other outposts when the dust settels. Your wall is certainly built to last.
@@ralfw7463 This is why I love the game, there are so many different ways to play it. Everyone has their own strategy. Very rarely will I use artillery turrets. I guess this is a consequence of having the biters at higher levels, but when you attack a huge biter nest with artillery the last ditch surge attack from that nest are really difficult to resist. They just overwhelm the perimeter wall. My strategy has been to build really tight and efficiently spaced factories with compact robust perimeter walls, and leave artillery mostly off. If everything starts sprawling then these perimeter walls are expensive to build and use a lot of power. As I am finding in my current lets play. Thank you for the compliments.
@@computercoma I had a factory, must be 5 years ago, two pipes joined, fluids mixed and I didn't notice for an hour or so. Autosaves were all nixed. The mixed fluids had got everywhere. Knowing what I know now I could probably have fixed it, but at the time, my skills were pretty limited, factorio was still early access and google was no help, and I was probably too angry to think straight. I downloaded flow control and never looked back.
@@theotherbigfoot I'm pretty sure fluid mixing is largely impossible nowadays. The amount of effort they went into to make it so systems with different fluids can't mix is intense, lol.
@@sevinolastname hmmm, maybe, I just don't trust it though. Flow control allows me to put two pipes directly next to each other. If for no other reason I would still have flow control for that feature
@@theotherbigfoot Updates per second. Basically how much CPU does it take to run this wall. Seems that it could be quite a bit if the ammo belt isn't full.
So I just googled 'marching squares' and it's a computer programming thing. I'm not a programmer so it's all gibberish to me. Is the thing I do in this video a particular technical thing that I'm not aware of?
How many signals you want, how close together they are etc is kinda up to you. Bigger trains (vs smaller trains) will need less signals spaced further apart. I didn't want to impose my rail method on what I wanted to be a general blueprint.
There is a lot of stuff I gotta re learn after using mods to make defence basically a non issue *cough* waterfill *cough*, but with engineers such as your self, the pain of learning is lessened.
Whew.. glad my current save is past all the logistics of defense. I have just reached lvl 20 lasers so I just have a couple rows of them around my entire base. About 11k at the moment, and about 8k roboports to maintain it all.
Or just supply corner roboports with resources to fix all the stuff.. It's about growth, not perfection! Or if you like rampant, put a couple of those capsule turrets with slowdown and poison capsules yk.. Vanilla biters aren't all that strong tbh
It is on my todo list. I have to finish the factory first so that it is creating the right amount of noise to attract the biters. This base has been 4 months in the making. A full perimeter wall test is coming.
...Or you can just double the amount of turrets on corners, and not worry about curves. Great video, but i'm questioning the practicality of curved sections. Having only straight, dioganal, and corner sections should work fine for most situations.
That is a fair comment. Maybe the language in the video wasn't precise enough. Replace 'curves' for 'graduated corners' as in 'the more graduated the better' (a curve is just a very graduated corner, particularly in factorio) and we would probably have an agreeable middle ground. I think the main thing I was trying to get across was that 90 degree corners are not great, or at least not in my experience. That was the main thing. If you choose to solve the problem differently, awesome, that is what makes this game great. 100 different ways to solve the same problem.
Diagonal walls, this guys achieved perfection.
the man, the myth, the legend himself.
To quote trupen, "players are obsessed with walls. you don't need walls."
i see no furnances as walls im not satisfied when my smelting area isnt my wall defence
Why did he hand build them all when he could have just made a blue print of each of the firsts? and flipped it right?
@@AthanDrakharr it's just his presentation style for the video, he hand destroys everyything as well
Of note you can define a global snapping grid the size of your wall segment in the wall segment’s blueprint, so you can make it so you can only place wall segments in the correct, chunk-aligned positions.
I didn't know you could do this but wouldn't that limit you to only that grid? Say I put a chunk down on the map without that grid, then set the grid up as you say, would the next chunk line up with the chunk already on the map?
@@theotherbigfoot There are 2 modes: absolute and relative.
Relative just lets you drag a blueprint in a line and it will tile down correctly, snapping relative to where you clicked your mouse button down. You can click down anywhere though.
Absolute snaps relative to 0,0, with an adjustible offset. this is probably what you want, because you can place down a wall (or rail) section anywhere and know it's aligned with all your other sections. Yes, if you place down a section misaligned with this grid it won't be aligned, but you can adjust the offset in the blueprint. It's best to just place your first wall section aligned though.
So yes, it would limit you to that grid, whcih is the entire point. You can make the grid half the size of your wall section of course, to get finer positioning, and then have converter sections that are half a wall section with a diagonal step. Or align the grid to where you placed your first section without the grid.
@@sea_kerman Great explanation, thank you, I didn't know any of this was possible. It's like being back in AutoCad!! You have definitely given me something to think about.
yeah, I was about to suggest that too :D
If I'm not messing this up, you could either set the blueprint size to 50x50 and use absolute snapping, but that somewhat limits placement (what if you want to start your wall 10 tiles further down?)
as an alternative that gives you more flexibility, you could set the blueprint size to 2x2 and enable absolute snapping. that way, you still have to pay attention when tiling segments together, but at least they will always be rail-aligned (never an odd offset from 0,0)
I'm not a fan of relative snapping for walls and rails since you can't ensure things line up when starting in different places, and getting the initial starting point right sometimes isn't trivial. relative 50x50 is definitely an option too though, just not one I'd use here
@@jemand771 The grid snapping and the offsets, all interesting stuff. I'm definitely thinking about how I could make use of that in my builds.
For the perimeter wall though, I think I'm pretty happy with the system I've come up with. The rails are included so it is foolproof. I can definitely see myself fu*king up the offsets.
the first half of diplomacy, before sending the eviction notices.
On vanilla biter AI:
Biters only actively target military structures (radars, turrets). Biters will attack other structures if they can't find a path around them.
Biters have no other preferences. They do not target corners or water-edges specifically, nor radars or other structures specifically.
However, a biter attack group targets the closest military structure. If your wall is rectangular, the corner is the most likely target.
Consider a rectangular base. All biters directly north of the base will attack the north wall, since the closest path to the base is directly south. Similarly for all biter bases directly east, west, south.
However, consider a biter nest northeast of the player base. Then, the closest target is the northeastern corner.
If you imagine coloring a Factorio map by the target a biter would choose, the strip directly north targets the north wall, but the (infinite) rectangular area to the northeast targets the northeast corner.
Therefore, a much larger area of map targets the corner, while a much smaller area of map spreads its attack to the whole wall.
Similar reasoning explains biter tendencies to attack along water edges.
Fun fact: instead of a continuous wall, a series of "bunkers" that draw the attention of biter attack groups works just as well to defend the base. Also, biters do not attack walls if they can path through them, hence the effectiveness of "dragon tooth" or "maze" designs.
I understand what you are saying but that has not been my general experience. Over multiple maps over multiple years my corners are always in more danger than my straight bits. It might just be that they are inherently weaker, with an exposed corner turret, like I explained in the video, and not necessarily the pathing of the biters. I believe it IS the pathing of the biters, I believe they are able to sense a weakness in the wall, but that belief is only from general gameplay experience and nothing more specific than that.
Pathing aside, I experienced a significant decrease in biters being able to demolish turrets when I started using curved walls. There was a noticeable improvement. I think it is worth using curved perimeter walls just for that reason alone.
Well, technically, if you generate an infinite amount of biters who are all angry, all walls will be attacked the exact same amount - infinite
@@cewla3348 Yes, if we have an infinitely straight wall with an infinite number of angry biters then we would expect the wall to be attacked pretty evenly. But a corner in that wall represents a discontinuity in that assumption. It isn't a straight wall anymore, and a turret on that corner will always be less well protected than the corresponding one in a straight section.
@@theotherbigfootThey said corners are in more danger, they only clarified that there is no unique priority to the corners, a corner is just going to be the closest target for a large area, so more biters will attack a corner.
6 for 1, half dozen for another at the end of the day.
@@theotherbigfoot The issue is not the biters' initial target, but rather the pathfinding. The way I understand it, the pathfinder sees the corner as the easiest route to the target. Once in sight, the military structures become the target.
I use the following patterns for defense:
1. My walls always start and end at water.
2. I use landfill to continue the wall into the water, so that even on the edge, there's always overwhelming force.
3. I use these walls to close off large sections of land, preferrably so much that it includes all pollution.
If I'm attacked this means my pollution grew so much it's time to conquer new land.
You can see the size of your pollution cloud on the map, so if you've already walled off that entire area the biters aren't coming because of pollution. Biters will split off in small groups to start a new nest in an area without a nest, where the biters in a group slowly turn into worms and spawners. The biters coming from outside the pollution cloud are from biter expansion only, and biter expansion decreases with evolution. Artillery is then also far more useful with high evolution because after nearby nests are cleared they only have to deal with occasional expansion nests.
im just the mad one who builds my walls through the water, maybe in 10 years water biters will exist and im already set 😎
Not sure if it was said in the video or not, but a heads-up for anyone trying to import the blueprints:
They rely on pipes from the flow-control mod and will not work with vanilla pipes.
I randomly clicked on this because my youngest son and I are about to get back into Factorio. WOW! This is not only a fantastic set of blueprints , but a sound defensive mindset for biter defense. Thank you so much for this.
This is easily my most favourite of all the comments I have ever received. Just such a wholesome image. Father and son playing factorio. Someone should put it on a christmas card
An absolute Godsend for less skilled and knowledgeable Engineers such as myself
Truly, thank you for taking the time to go through the motions c:
those walls inspired me so much they fundamentally changed the way i aproach building in this game holy shit
About using belts. Particularly in the earlier game, getting them full can be prohibitively costly. What I've started doing is using ammo feed belts that eventually circle back on themselves and a circuit timer controlling the feed belt to have it hold back ammo and release it in chunks (eg. belt on for 120/1200 ticks, adjust by need). The feed belt itself is being replenished just behind, where the unused, returning ammo backs up, so it'll only be replenished when it there isn't enough ammo circulating for it to back up.
Basically, it's just a way of controlling the ammo supply belt buffer size so you can do longer supply belts without the ammo costs getting out of hand.
I tend to embrace the ammo costs as stored value. The more ammo I can squeeze on the conveyor belt and the quicker I can do so, the less work needs to be done later.
@@theotherbigfoot Yeah, guess in the long run that works. I guess what I'm doing is just because I've been in situations where the ammo production has eaten resources I needed for other stuffs. But... that's probably just me not scaling up enough early enough.
I suppose it could still be useful if trying to run depleted uranium ammo though? An advantage of this system is that with a circular belt ammo delivery gets spread out even when not saturated, and keeping the ammo moving in blobs kinda stops everything getting sucked up by one bit of the belt. Dunno though, never tried to do depleted uranium on any scale.
@@ashedtogether Oh dude, I wasn't telling you that you were doing it wrong, I was just saying why I do it differently. Feels like you have a well thought through setup. If anything, I'm gonna stick with what I do already because its easier!! You aren't wrong about the resources, but I've recently been massively scaling up the resource deposits so that I can build behemoth factories, so wasting resources is less of a problem.
I haven't tried uranium ammo yet either, but I am about to. I'm going to divert everything from the perimeter wall to a new setup to convert to green ammo and then back again. I've started converting already and am building a reserve in a whole load of storage chests before a big switch over. We will see how it pans out. There is content planned.
@@theotherbigfoot 🙂I'm interested to see that!
I just use a solid line of flamethrowers using crude oil and call it a day
Very interesting! Some thoughts:
1. Having all of your ammo on a single belt is asking for trouble. Not only does that mean you need to have a *LOT* of ammo travelling at once, but, if anything goes wrong and the belt breaks, the turrets after that point will eventually starve. Not Good. A better way is to use local loops that meter ammo into packets with a time interval between each one, served by a train stop that is only enabled when ammo is needed.
2. It is important to have your roboport be well back from the wall. Why? So that repair bots only enter the danger zone *AFTER* the wave of biters has been destroyed, so they don't get damaged/destroyed along with the wall/etc.
3. Dragon's Teeth that break up biter pathing is *vitally* important to both slow them down and create chokepoints for flamethrowers. 'Inverted Teacups' works really well for that, but make sure to have 2-wide paths so they don't start beating on the walls.
4. One alternative to using ammo belts is to have turrets surrounding/on either side of a requester chest, with stack inserters passing ammo from turret to turret as needed. (You *DON'T* want to take this too far, though; lest some get starved for ammo when things get hairy.)
Cheers!
Wow that is quite a series of comments. Ima deal with it the same way I deal with my work emails!!
1. Yup, you are absolutely right, and, in the early stages of my games when I don't have robot coverage, a problem I frequently encounter. But in the later stages, once I have a final wall built (which I'll be doing from now on with these segments) it isn't an issue I have ever had. If it did happen, just for arguments sake, there would probably be plenty of ammunition available to the other turrets until it got repaired.
2. The roboport is pretty far back from the wall. I'm happy with its placement. Putting it any further back would ruin the aesthetic of the segment. I don't mind making a few sacrifices to get the elegance I am trying to achieve.
3. Someone else pointed this out, I don't do this because it opens the door for robots to be out in no mans land repairing bits of wall vulnerable to biters. It is a strategy I understand, I have seen the videos, I understand it's effectiveness, but that I choose not to do simply because I would lose too many robots. I understand if you completely disagree.
4. I like my ammo belts. IMO they work really well. I definitely do not want to rely on logistic networks. I never rely on logistic robots for anything other than fuel for my trains, maybe a bit of u235 and/or u238, and my personal logistic requirements. Basically, I don't rely on robots for anything mission critical. Perimeter wall is definitely mission critical. Literally everything in my factories is on a conveyor belt. Even my water!!
@@theotherbigfoot Fair enough. All good points.
@@theotherbigfoot 4 just seems a bit strange, bottled water is so hard to deal with!
@@cewla3348 Bottled water is indeed hard to deal with, I have tried it in 6 or 7 factories now and it still doesn't work perfectly. But the challenge is what I love about this game, I'm also pretty stubborn, so I'm gonna keep going until I nail it.
@@theotherbigfoot I just watched it again and had another thought:
5. Based on my own experienced, I realized that each wall segment *SHOULDN'T* automatically link to the roboport in the adjacent segment. Why? Because, over time, the bots will tend to 'clump' and then have to travel long distances--sometimes over no-man's land, depending on the map--to repair/replace damaged structures. (#beentheredonethat) Roboports aren't expensive, so making each segment 64x64 (or w/e) with a (tileable) spot on the edge to hold linking roboports will allow easily dividing the wall into logical sections that keep all bots local so they can act quickly. (By the time you are using walls this expensive, bots are relatively cheap, so it's good enough to drop a steel chest full of them in each section with a speaker to alert you when it needs topping up.)
All that being said, while I would never use your wall system--because I habitually build simple walls *FAR* beyond the pollution radius so biters never really become more than a nuisance--I am in absolute *AWE* at what you've designed here. Bravo, good sir!
Bigfoot didn't know about blueprint grid snapping at the time, but you should also make sure that your wall blueprints are grid-aligned with your rail blueprints. If they're chunk-aligned, even better.
On a different note: One thing I enjoy having in my walls is a wide (~6 tiles) personal vehicle gate to make sure I don't crash into the walls as I'm running into safety, but those areas tend to be big weakpoints that suffer more than the rest of the wall, even with higher turret density.
One solution I think might work for this wall system is having an invagination-shaped segment with many defenses, that tapers to a point on the inner side but curves to become more and more open on the outer side, similar to the mirrored semi-corner walls at 12:10.
Rails, supply belts/pipes, and roboports/artillery/radars would have to be adjusted accordingly, but it would be too much of an issue, other than disrupting your usage of space.
*This* is what base design is about! I've experimented a lot with wall designs that exploit the biters' pathing AI, primarily for flamethrowers, since I've been playing a deathworld marathon save that makes ammunition costs way too high to justify. I haven't gotten to the stage of working tileable designs of this scale (still working my way to roboports in that save), but the tips here (particularly the one about aligning the rail and roboport before doing anything else) will be a life saver. Thanks for the inspiration, great video!
Ammo is useless, don't see a reason to use gun turrets unless you are playing on easier settings.
For me, using just laser turrets has pretty severe power implications.
@@reuven2010 Its the only realistic option for defence early game.
Great explanation. I've been looking for a new wall design, since my default one feels boring by now. I like to build within the standard factorio chunks. I protect my corners by making 'indents' like so
___
|__
|__
|
|
and then adding more dragon teeth, which
1. slows down biters --> clumping them up for flame throwers
2. often splits them into groups going for the sides
My first comment was me absent-mindedly pondering on the choice of conveyors vs logistics systems - so commenting again for comment god and just to say great video, your approach is exactly the kind I love seeing in games like this, it's enjoyable just to see other people vibing and working through it and sharing their reasoning with the world - it's what makes games like Factorio special I think =D
Amazing, I love the symmetry of your thinking, now i must design Blueprints with your teachings in mind.
the corner advice is good to know
i usually use only flame throwers, and the walls are further away from the flamers than what the spitters can shoot, so only the wall gets damaged by biters, and the spitters get too close for some reason as well
Holy moly. That video was packed. Impressive engineering. Totally different concept as mine, but really interesting.
Wow that looks really cool, I'll have to try some similar designs once I'm done with my current "biter air filter" save where I experiment with limiting biter spawn positions and trick them into depolluting for me as long as I have ammo, oil or power.
This is nice and all, but it is impossible to put pipes like that next to each other.
Fair point, I use a mod called 'flow control' which makes this possible. Doesn't really change the overall geometry lesson though. If you were building your own segments then you could just leave a gap between the pipes.
A notch-out rather than curve would allow to target biters from two protected directions at the same time, which also solves the issues of bare corners. It may make some design aspects easier. (At least if the biter horde behavior settings aren't having attacking groups to be at ridiculous sizes.)
Bro...the biter AI doesn't target corners on purpose or otherwise. It's just that flat wall segments have small rectangles where spawned attack groups can originate to end up at those segments while corners have a full 90 degree spread of area where the corner happens to be the closest part of the base.
Ignoring the biter AI for just a moment, I feel like you agree with the general premise I am making in the video, which is that the corners of a perimeter wall are weaker than the straight portions. And if you do agree then the reason why this happens is irrelevant. Corners are weaker, get attacked more, whatever, and I thought curves are a good solution to that problem. I still think that. I'm going to be using these wall segments in every base I build from now on.
With regards the biter AI, in the timelapse clip where I record the biters attacking the corner of my base there are moments when the biters readjust their position before attacking. For me they are re-orientating themselves to attack the corner at precisely a 45 degree angle. They don't just 'activate' and attack the first bit of wall they came across. They activate, reposition and then attack. I don't know anything about the inner workings of the game, how the biter AI works, I only have general experience of the game. A lot of general experience. But no technical understanding. What that repositioning suggests to me is that the biters look for a weak portion of the wall and then attack that portion. That is how the game 'feels' to me. But as I said, I don't have any proof of that other than from what I have learned from general gameplay.
I made a similar setup using chunk alignment. There was a laser wall, flame wall, and gun wall, and they would all layer over each other to beef up the wall and defenses if needed. Train gates, vehicle gates, roboports, rounded corners and bastion corners. I remember figuring out certain perfect distances so everything lined up felt really good.
Probably spent 36 hoyrs of design for a wall that was completely overkill for anything besides hardcore deathworld maps, but they looked sexy as hell
Sounds like you were one step ahead of me. I would be really interested in seeing what you came up with.
I think these segments took about 20 hours, and it isn't the first time I've tried this so I probably had a bit of a head start. It isn't a small undertaking but, as you say, the results are pretty damn sexy
These are really well thought out! I'll try to make my own walls like this in the future.
I have been trying to build similar walls like this for basically the entire last year now. I though, I could try to have the blue prints chunk align, which means they are on a grid and will fit together, I even made sure to make every block align to a roboport and be aligned to the rail grid, but I ignored to make sure the turrets could be mirrored in the diagonals and my corners were not gradual but switched ratger abrupt from straight to diagonal. I made that work in the end, but it has been a chore and its not as elegant as these are. And since I have to redesign my chunk aligned railway grid again, I can redesign my perimeter and outpost walls, don't know if I will make my outposts into railway interchanges, but trains getting stuck outside the walls is a hassle for sure
so happy you distilled it down to a set of principles, super useful
Chunk aligned walls... Why not like that?
Would that work on the 45 degree sections?
Great concept that I'll certainly make use of, thank you very much.
Would have loved to see your wall put to the test at the end!
Someone else said this exact same thing. The base I'm working on isn't quite ready for a full blown test yet. I made this video at the moment I needed to build the wall in that base and it is built and finished but the base hasn't been fully developed so it hasn't been tested.
I guess now I need to do a follow up wall test video.
@@theotherbigfoot Haha, I'll give you a subscribe then. Looking forward to it.
very useful information - I'm just about to install a walled perimeter in my first 1.1 base and that part about biters attacking the corners I didn't catch before, so thanks for that !
I don't do a perimeter wall. I do robot printed bastions with roboports in each. Surrounded by walls, they each have two corners, but are compact enough that all the turrets defend each other. I print these offensively to effectively cordon off an area, and expand; Then I move in with either powerlines and lasers on foot, or spidertrons later.
VERY well-thought out wall layout! Might need to steal these blueprints for myself later.
Corners are a natural weak point in all defensive structures in real life, but in some instances of technological advancement they provide a useful defensive posture. Creating stars, for instance, makes a fort's shape far more effective.
A simpler solution to perimeter defense for most people is to create corner circles that present a circular exterior segment that connects to the straight section via another curve inward. Sort of like an Omega, but less angular at the turning points.
Bro, im 8mos clean on cracktorio. thanks to you here comes my relapse.
Lol, my eyes hurt i've been playing so long.
why does nobody ever use belts, they're a pretty effective add on to any perimeter. Just plop down a row or two facing outwards, prefferrably at choke points, even yellow ones support your turrets really well
Brilliant! But I'm sure i'm missing something. In my playthroughs I've gone to exclusively 2 lines of laser turrets/wall/3 lines of mines all supported by roboports and periodic artillery batteries. I have a couple of blueprints of different lengths and can snap them in easily and deconstruct them easily upon advance. I try to start/stop segments at water lines. If I must have a corner I simply add more mines/turrets at the corner. I keep buffer chests at intervals with all the wall components.
Even if I invite the horde through artillery spamming the worst that happens is 2 layers of mines get used. Most attacks result in obliteration with at worst a little goo on the wall, quickly repaired by the bots.
With this setup I don't need any supply network outside roboports. The buffer chests make expansions quick because all the supplies are local. I wonder if with mods or the new 2.0 update the biters will eventually get too strong for my system and I'll have to adapt. (I absolutely love flamethrower turrets, I just don't seem to need them.)
That was a great watch. Here's two mods I think you might have liked to know when making this: Blueprint Flip and Turn, Blueprint Lab
This is actually so good though
You like?
@@theotherbigfoot Yeh, saw your video actually utilising the design. It's sweet man, gotta try it out myself.
i'm still "new" to the game, having just begun using artillery turrets on my longest ever continuous save file. dont know why i never thought to practise my wall building in sandbox mode.
i currently use a maze wall with dragon tooth corners which is for sure effective right now but my evolution is past 0.70 already since i had a slow nooby start to my game. like i still only have one line of weaponry on most of my walls. my brain hurts watching this but i'm inspired to fuck around in sandbox now
I use philosophy that perfection doesn't matter, just put flamers walls and don't give repair packs to bots.
I played way too much deathworlds and I always build walls without bots cause I meed them already in early game. This led me to realization that it all literally doesn't matter. all you need is couple of flamers, few layers of wall and repairing walls until you get to bots. once you have bots DON'T give them repair packs and upgrade bot speed and they will safely replace walls without dying.
I am a great fan of what I call 'the great wall of dakka' , some great Ideas in this vid , hank you Sir :)
Like the design and the presentation. Very nice!
Great job!
Such a pitty that outpost ones don't have flamethrowers though - what can be more fun than setting up an outpost-supply-train with liquid wagon or barrels of oil?
you can mirror blueprints with F and G
No you can't, you're lying!! I checked this a few years ago, there is no way the developers would add something to the game after I had already tried it, found there was no way, and ruled it out!! That has never happened before!! Damn this game.
@@theotherbigfoot i checked it myself now, but normally you can flip a blueprint with F and G. Only some minor blueprints can't flip, but then there will be given a particular reason why it cant.
@@ronphil6666 Oh dude, I was being sarcastic. This is an excellent tip you have given me here, if only I had known a week ago before I started all this.
@@theotherbigfoot yes, i thought so already, but it was also meant for the ones who didnt notcied your sarcasm :)
@@ronphil6666 Fair
Thanks for this video, you went over a ton of pitfalls that would have made me give up and go back to an ugly and simple square wall
4:40, you might say that the roboports will be QPU aligned.
I'm guessing this is a bigger issue with things like the Rampant mod, or Death World preset? Then again, I'm also on Rail World preset, so biters haven't been a major concern.
Even so, great demonstration (still watching). My setup started with a single-layer wall with flamethrower turrets spaced kind of on feel, but ended up about 7 tiles apart I think. Between them is a single gun turret. Belt for ammo, inserter to load it, and power poles to keep everything running.
Separately, with the new fluid mechanics, my wall now also needs a pumping station every 320 tiles. I made a nice little insert that pipes out 10 tiles to a tank, then a pump, then pipes back into the line of flamethrower turrets. This has two effects. One, it extends my pipeline without complicating logistics with rails. And two, it provides a small fuel reserve that will never flow back into my production facilities. That way, even if there's a run on production, there will still be at least 25k fuel in reserve per line of flamethrower turrets.
In my next expansion of the wall, I plan to add laser turrets (left them out for the achievement this run), rocket turrets and artillery. Not that I need it, because even behemoth biters/spitters can't breach the single-layer wall before getting burnt to a crisp.
Is there somewhere I can get the entire set of blueprints in one string? I see the link has them all in sections but it'd be nice if they were all in one book.
Nevermind, I didn't look hard enough :)
thx for blueprints , just dint figured it out to redirect the belts ... is feed them from other site . got the blueprint to or is it manualy to adjust ?
now have an optimal dragons teeth addon
And here i am, still planting nothing but laser turrets from early to late game, wondering why people would try anything else in vanilla.
solar is too ineffective, nuclear is too lategame, coal/solid fuel is too ineffective.
Before having the blueprints: "Holy shit, what the hell is this kind of rocket science jargon"
After having the blueprints: "This is as easy as using Lego blocks"
Why not just allign the defence grid with the rail grid? Are you afraid of placing roboports two tiles closer to each other?
Awesome video, although looking at the walls I usually build I must be much easier on the biter settings than you... Apart from the comments on grid alignment and mirroring blueprints, would also add that biters I believe focus on some modules over others (specifically radars) and they're 'preference' for corners is likely more to do with that - worth testing, but I tend to just go with 90 degree corners and keep radars away from being in the extreme corner.
I put the radars there specifically to provoke the biters because I was recording for the video. That isn't my normal practice. That was just the corner in my current base that was getting the most action. The radars definitely sped things up but this corner has been problematic the whole build. I have been constantly repairing it. It has been the same with the other corners, although just not as heavy, which is why I didn't use them.
The straight sections have been more or less impervious the whole time. As long as I've had a continuous line of turrets and available ammo the straight sections of wall have been able to resist the biters.
This is all on the standard settings. Normally yes, I do play with higher settings on the biters, and the effect is definitely magnified, but the principle is still the same. Walls with curves, in my experience, have been inherently stronger.
I'm not saying you are wrong, I guess I'm just saying that this is why I go to the trouble of doing curves in the first place.
Impressive work! What a level of detail in design .... what amount of time you put in this ...
Will this design work with the new rails since the release of Space Age?
No it won't. An update is coming. At some point.
why are you using 2 different fuels? is it in case one runs out? why would one run out?
Erm, because I think it's cool.
You say corners are vulnerable so we should avoid them.
I say that's bait and exploitable.
and you didn't mention anti-aliasing once! Great video
its much more work than just aligning by chunks tbh
A beatufull pixel art, but I can't see the point. I've played with mods like rampant and natural evolution, I hit 99.9999% evolution and never used such thick walls and so much turrets. Your builds look so overpowered, that I don't understand what mods do you use if you need this kind of firepower? I only see the point if you had no roboports to repair, but they are the core of the build.
it's like a membrane of a living organism 😮
I originally built my walls like this, trying to cram as much turrets in as possible with lots of walls, but after reaching late game, it's just overkill. 95% of my turrets are unnecessary and just waste power and resources. I've switched designs to better exploit biter behaviour and reduce maintainence costs. Might switch back to something like this if I ever install Rampant.
Rampant Mod: "our battle will be legendary"
Wow, I didn't even know that mod existed. If I didn't hate the biters with all my being and tolerated their existence simply because it is boring without them, then maybe I'd give this a try.
I think my way of making the game difficult for myself is overhaul mods. Or silly race tracks.
@@theotherbigfoot i really should try making racetracks tbh lol
Getting the perimeter railways correct is the most painful part...
will you be adding a train station and a point of entry for fuel and ammo segments ?
A train station for mining? I was concentrating on just perimeter wall stuff.
Point of entry for fuel and ammo - not 100% sure what you mean, there's a point of entry for trains. Do you run your ammo trains on a different train network on a separate set of lines?
Great tutorial! Also it's all made simpler now with second floor rails and blueprint flipping. Do you plan to update your wall?
Yes I do
Lol.. my first round and i got most of it right… for repair i need a robo port..
Also i build 3 pices straight, inside and outside corner.. made them 45° wall from corner to corner of the robo port since they seem weak…
But i did not go for Overkill
just any fyi; the supply limit is 50x50, the construction is 110x110. You dont need your roboports so close. but i dont think it matters anyways. lol
Unless you want a continuous logistic network on your wall.
RIP all thes beautiful walls with the 2.0 rails
Someone was just telling me about how the railways broke everyone's blueprints and I was like 'I'm fine, I don't really use blueprints'. Then you popped up and reminded me. I'm gonna get a video out of it so I can't tell if it's a good thing or a bad thing!!
@@theotherbigfoot hopefully not too bad to fix. Only the segment with the big loop is going to have to get one rail width larger circle. And with the extra diagonals you might be able to make a diagonal entrance. Good luck (:
Wow! This is amazing!
Pfft. Correct answer is laser turrets and artillery and construction drones supplied with everything needed. Around lvl10 laser damage 3-4 rows of laser turrets, literally impenetrable.
This is a great tutorial on game mechanics, but *IMHO* by the time you have the resources to build defenses like this, you should have the ability to use artillery or MK2 armor with PLD to remove all biter presence beyond the reach of your pollution. At that point you only need to defend your artillery placements. This statement applies to vanilla and lightly modded play... deathworld or rampant mod might make this kind of defense more necessary.
o - been awhile since I last played took me off guard with 'artillery' lol - never understood the value of using conveyors to keep turrets full of ammo - seems needlessly troublesome when building small distribution hubs (which may or may not also be a good spot for a passenger stop as well) and using logistics distribution makes considerably more sense in terms of flow for rapid expansion after clearing an area, without needing to connect everything together by anything other than rail - which serves to transport the player and electricity anyways so would always be present.
This is mostly me just being a bit puzzled tho - lots of other excellent problem solving and approach on display here. 👍
I have a 'maintenance' set of trains which whizz around all the outposts delivering ammo as well as replacement turrets, electricity poles, conveyor belts etc. Once the ammunition loop is full it only needs to be topped with what what gets used.
@@theotherbigfoot hmmm I see - oh wait I think I see what's up - the conveyor serves as a natural buffer so that you don't need to use circuits to ensure smooth supply? I've been too in love with having resupply trains on standby until an outpost requests more ammo/fuel/items/bots/etc. ever since they added the ability to turn train stations on/off >.< so I've just gotten used to always running the circuit lines along with the electricity and trains, so think that's why I'd not really considered it. Watching the network whirr to life as it's needed is just too satisfying 😁
edit: though I've still not found the most elegant solution for requesting multiple resupply trains across a vast network - always seem to get distracted by something shiny before it gets monstrous these days
@@NeoRipshaft Personally, i use "Set train limit" with circuits. Set limit to 1 if anything is required, otherwise 0. Trains will wait at loading area unless there is demand, then go to that particular place with no risk of multiple trains going to the same station or getting lost on the way. So far, works well even with dozens of destinations.
Circuits are useful for controlling loading of resupply trains as well, so i do not have to set filters in wagons (just have "desired load" set by constant combinators) and i can use the same wagon for carrying junk from the wall (either from terrain or deconstruction after a new wall further out was made).
Why you don't abuse the biter pathing with a even a simple maze? A good funnel can negate most, if not all, biter damage. Also - why 2 kinds of flamethrower fuel? Either you are rich, so you use the better one, or you are poor so you use the cheaper one, no?
Ok, my theories as follows.
I don't use simple mazes because when those mazes get damaged my robots have to fly out into no mans land to repair them, where they will invariably get destroyed by biters. So I just leave it to my turrets to get the job done, which are at least designed to be engaged with the enemy.
The oil thing, I think the first time I used the flamethrowers I used just the one fuel and it ruined the balance of the oil processing in the factory. Entirely possible my newer factories are better designed and would cope better, but I think this is where the design decision came from. I'm not saying the way I do it is the right way, I'm just saying this is how I do it.
So I think theres something important here, each flamethrower fuel mechanically functions almost identically, Light Oil is objectively the best fuel as it gets 10% bonus damage and is the most economically available. You want to feed your turrets with light oil hands down. Heavy gets a 5% damage bonus but you can instead crack it into light oil for more total damage. I would still run two supply lines to the turrets purely for throughput but include some pumps that push the fuel in opposing loops of the base. Determine a 'seam' in that loop so the lines will fully saturate, this can be the same local as where the fuel enters the wall.
Additionally, though it is a preference matter, I generally set my Flame turrets back further so they are more likely to leave fire along the wall where biters are likely to stand within the flames instead of momentarily passing though them.
You mention having trouble balancing your oil production, this is resolved with a basic circuit, read the contents of your petrol, Light, and Heavy tanks, compare one to the other for each, and when you have say, more Heavy oil than light, the circuit enables a pump to crack heavy into light, the same for light to petrol. It doesnt work in reverse of course so you can use combinators to adjust the balance of this circuit as you prefer.
Hope thats beneficial to you
Ok, just for you, well, and because I'm interested to see if my current factory can handle it, I'm gonna try out light fuel only in my flamethrowers. I'm pretty confident in my oil setup (thanks for the help but I already have a circuit setup I'm happy and confident with. I was so confident I made a video about it (oil and SPM)) so let's put it to the test and see if it can do light oil only. Convenient timing as well, I'm only a few hours away from hooking up the flamethrowers.
It will work. I have no doubts. There are some design decisions I am still tied to from my early days playing the game and I just haven't tried anything different because habits. You say 5% bonus (I didn't know that), ima go grab it.
@@theotherbigfoot I'll admit this was the first of your videos I've yet come across, so I will have to check out your oil system video when I get a chance. I know people can do magic with circuits but I've only just started to understand them about a month ago despite having started playing before it hit steam, balancing oil with out circuits was basically just hoping that luck would do the job lol
It's no worries at all buddy, I'm just small fry on youtube really. I am no magician with circuit networks, I can do some cool stuff, but not a lot. I've made a video about that too!!
I don't like plugging myself but if you are interested in understanding the mathematics of oil and spm then that tutorial is decent. I don't think anyone else on youtube has figured out what I explain in that video. When I set out on the SPM design route in my factories I certainly couldn't find a video that explained it. If mathes is gibberish, there is a spreadsheet. That is in the video as well.
what mods do you use for new pipes?
Flow control. The first mod I ever downloaded!!
@@theotherbigfoot ach, thanks, you should also write that in description of video and blueprints, without it the blueprint doesn't work :D
@@warfoxltu I thought I had!! I will check
@@theotherbigfoot the blueprint books does not have that mentioned :)
That's so pretty~
Your grid is impressive, especially the railway inclusion.
But why is your perimeter wall so expensive? On the length of one roboport I support a wall with only 8 gun turrets in pairs. Each pair has one requester chest requesting ammunition and some burnables the turrets get feet by burner inserters. So no electricity is needed besides the roboport. If you place buffer chests strategically this will support even long walls.
Artillery will pull huge groups. I strengthen those segments with flamer turrets. Just make shure the turrest get ammunition before the artillery.
I do generally play with the biters at quite high levels, and I do generally feed the wall with just red ammo, so I have found a big thick wall to be prudent, but I am certain that if I made a video about a perimeter wall that wasn't completely over the top then nobody would watch it!! I do build my walls like this, and I have been doing for quite some time so I don't feel like I'm being misleading, but also I think to be successful on youtube these things need to be taken to their absolute limits.
Besides, I generally build big factories so an expensive perimeter wall isn't really a problem.
(There is nothing stopping you from taking my blueprints and removing the turrets you don't want. If you need to upgrade then you can just overlay my original blueprint on top)
@@theotherbigfoot as i said your Segments are impressive. And the video is a real eyecather. My biggest factory is a 1.2 k SpM Railworld. Biters are at normal level though so. Therefore i never experienced a bitter attack that would call for such a strong defense.
Lately i found that with artillery on 5 th range extension the need for real perimitter walls shrinks to a minimum. So i build a heavily protected artillery outpost which clears the area before building other outposts when the dust settels.
Your wall is certainly built to last.
@@ralfw7463 This is why I love the game, there are so many different ways to play it. Everyone has their own strategy.
Very rarely will I use artillery turrets. I guess this is a consequence of having the biters at higher levels, but when you attack a huge biter nest with artillery the last ditch surge attack from that nest are really difficult to resist. They just overwhelm the perimeter wall. My strategy has been to build really tight and efficiently spaced factories with compact robust perimeter walls, and leave artillery mostly off. If everything starts sprawling then these perimeter walls are expensive to build and use a lot of power. As I am finding in my current lets play.
Thank you for the compliments.
What Mod are you using for the pipes to get the pipes to lay next to eachother without interacting with eachother and mixing fluids?
Flow control. The first mod I ever installed and still my most favourite. I hate vanilla pipes so much.
@@theotherbigfoot Thanks! I will have to look into that one. I'm also not a big fan of vanilla pipes but have always just lived with it lol.
@@computercoma I had a factory, must be 5 years ago, two pipes joined, fluids mixed and I didn't notice for an hour or so. Autosaves were all nixed. The mixed fluids had got everywhere. Knowing what I know now I could probably have fixed it, but at the time, my skills were pretty limited, factorio was still early access and google was no help, and I was probably too angry to think straight.
I downloaded flow control and never looked back.
@@theotherbigfoot I'm pretty sure fluid mixing is largely impossible nowadays. The amount of effort they went into to make it so systems with different fluids can't mix is intense, lol.
@@sevinolastname hmmm, maybe, I just don't trust it though.
Flow control allows me to put two pipes directly next to each other. If for no other reason I would still have flow control for that feature
burner inserter fed gun turrets gang represent!
You’re a god
Classic Factorio, over-engineered as always.
1:28 part tuah? say that again
loved this, thankyou
This video is the dogs bollocks, thankyou
Wait a minute, how UPS friendly is this?
UPS?
@@theotherbigfoot Updates per second. Basically how much CPU does it take to run this wall. Seems that it could be quite a bit if the ammo belt isn't full.
@@niamhleeson3522 I didn't notice anything significant but to be honest I wasn't looking. Seems fine on my computer
Just expand your wall beyond the pollution cloud. You won't even need a crazy gun wall then.
Bro casually discovered marching squares
So I just googled 'marching squares' and it's a computer programming thing. I'm not a programmer so it's all gibberish to me. Is the thing I do in this video a particular technical thing that I'm not aware of?
By definition is an algorithm for creating geometry based on a grid of nodes, or in this case the perimeter wall. Great video btw
i dont see any signaling on the rails
How many signals you want, how close together they are etc is kinda up to you. Bigger trains (vs smaller trains) will need less signals spaced further apart. I didn't want to impose my rail method on what I wanted to be a general blueprint.
There is a lot of stuff I gotta re learn after using mods to make defence basically a non issue *cough* waterfill *cough*, but with engineers such as your self, the pain of learning is lessened.
Whew.. glad my current save is past all the logistics of defense.
I have just reached lvl 20 lasers so I just have a couple rows of them around my entire base.
About 11k at the moment, and about 8k roboports to maintain it all.
You can't escape it for you next factory though!!
I don't think I have even got to level 10 lasers, ever. You have taken your factory deep.
When the resident is Evil
Or just supply corner roboports with resources to fix all the stuff.. It's about growth, not perfection!
Or if you like rampant, put a couple of those capsule turrets with slowdown and poison capsules yk..
Vanilla biters aren't all that strong tbh
Holy crap u make curves in factorio :0
How could you not show it in action! :(
It is on my todo list. I have to finish the factory first so that it is creating the right amount of noise to attract the biters. This base has been 4 months in the making. A full perimeter wall test is coming.
Actively drone fed minefield FTW
...Or you can just double the amount of turrets on corners, and not worry about curves. Great video, but i'm questioning the practicality of curved sections. Having only straight, dioganal, and corner sections should work fine for most situations.
That is a fair comment. Maybe the language in the video wasn't precise enough. Replace 'curves' for 'graduated corners' as in 'the more graduated the better' (a curve is just a very graduated corner, particularly in factorio) and we would probably have an agreeable middle ground. I think the main thing I was trying to get across was that 90 degree corners are not great, or at least not in my experience. That was the main thing. If you choose to solve the problem differently, awesome, that is what makes this game great. 100 different ways to solve the same problem.
Master
time to redesign for the 2.0 rails and new curves xD
Oh no... an excuse to remake one of my favourite videos... whatever will I do?!
@@theotherbigfoot oh noooooo.........
thank you kind sir