"With me so far?" Absolutely not, but I watched until the last second anyways. Always love your deep dives Blade! Excited to see more as it comes out, and will catch you on Twitch in the meantime!
These take a bit of work to make, but I'm happy with how they turn out. It's cool to be involved with the trials! Make sure to watch future episodes (wink wink)
This is great blade! I’m not a speedrunner or anything but it is interesting to hear how everything works! It is quite niche but I don’t think anybody has ever done anything like this yet, I’m excited for more lessons!
I personally love this addition to the game especially with the weapons but it admittedly eats up too many resources as I’m comfortably allowing it to (It took me like 30 prismatic shards to get my gavel to Artful which is the most helpful one for me). Had to do several skull cavern runs to get these prismatic shards and it eats up the in-game time I would rather use to arranging my farm layout or managing my crop inventory.
The MouseyPounds predictor takes your game seed and gives you the enchantment list, so there's a way to predict enchants without relying on parsnip qualities! This is still a great video for going in-depth on these mechanics and how they work tho.
This was very informative, thanks for making this video! I didn't know that the enchantments vary with the day and are the same each day. I'm going to use that to try enchanting first thing every day and reset and try another day if I don't get the result I want.
I thought I knew a lot about stardew and had seen all the cool tips videos. I was so wrong! Lol I LOVE your Stardew content. This is so interesting. I'm subscribing
Is the enchanting RNG actually $+ 0.522425... (mod 1)$ or is it a more complicated RNG core not discussed as not important? If the RNG used is that simple, it's theoretically predictable by a human (with some error), whereas if it's using a proper RNG core it's definitely in the realm of external tools. I think there's merit in allowing external tools (that can't view internal state). Run notes are a primitive tool, a metronome for LoZ:WW superswims is a bit more involved, and the Sploosh-Kaboom tool is an RNG predictor. Depending on the specific tool use and how much it changes the route, it could be interesting to also have a "no internal knowledge tooling" category... though in RNG run-killer heavy runs, predicting/manipulating RNG is awesome as a way to make the run more consistent. I see it as on an axis where the endpoint is seeding; if I can figure out bits (pun intended) of the seed, how much am I allowed to do with that information? How much am I allowed to manipulate what seed I get? Speedrunning is about understanding a game at a fundamental level, so it's interesting to try to draw a line in the sand of what's "too much" tool assistance. I'm glad to live in this age of speedrunning where we get to blur the line between RTA/TA and TAS. (In my opinion: it becomes TAS when you can view internal state or you segment (rerecord) not at game-supported segments.) I hope I get to see some more blurring from the other direction too -- speedruns performed by a program on an unknown seed, not just as a piano roll of inputs, but making decisions on the fly to respond to RNG happenings.
Stardew uses the base C# Random class, According to the docs, "The current implementation of the Random class is based on a modified version of Donald E. Knuth's subtractive random number generator algorithm. For more information, see D. E. Knuth. The Art of Computer Programming, Volume 2: Seminumerical Algorithms. Addison-Wesley, Reading, MA, third edition, 1997." For the purposes of simulating RNG, this can be simplified to table; a list of intercepts and offsets e.g. # * 0: 0.7262432699679598, 0.5224253141891330 # * 1: 0.8173253595909687, 0.2934186175900598 # * 2: 0.7680226893946633, 0.6989879904775824 # * 3: 0.5581611914365372, 0.2134429305854453 The result of an RNG call is dependent on the index of the call on an initialised Random object. It is (seed * offset + intercept) % 1. So the first call (gold crops) uses the first row, the second call (silver crops) uses the second, and so on. Gold crops and enchanting both uses the first call of an RNG object. I agree that looking at internal state should not be allowed in speedruns. However, I believe we should be allowed to use the result of RNG in one area to predict RNG in another. As to what level we let external tools and trackers guide this prediction is up for debate.
Sorry if this feels like a weird question, but can you explain Mine crates? why do Skull Cavern ones sometimes respawn, but not always, and are there any ways the original crates respawn?
I can, and this would make a good topic for a future video. Crates that spawn around the side of the map are based on dailyLuck. You have a 50% to have 1-4 of them spawn, adjusted (-2 to +2) by daily luck. You can break as many of these as you want and they will still spawn. Crates on container platforms work differently. The very first time you load a floor that has a container platform on it, each tile on the container platform has a 40% chance to spawn a container. Once the floor is loaded, that number of containers that spawned on the platforms are saved to the save file. This number is the maximum number of containers that can spawn on the container platform on this level. On subsequent loads of the same floor, each tile still has a 40% chance to spawn a container, but once the maximum number of containers has been reached, no more containers will spawn on the platforms. (priority is left to right, top to bottom). When you break a container on a container platform, this maximum number is decreased. Once this number his 0, no more containers will spawn ever on that mines floor. These numbers work exactly the same in skull caverns - the maximum number of containers is saved against the floor number. However, since the floor layout is randomised you can roll a layout that has no containers. The maximum number is set the first time you roll a floor layout that has containers. Some floor layouts have more valid container platform spots than others. If you roll a small floor layout the first time the container platforms are to be set.... too bad!
Hey Blade! Love this and am planning a run with my sister using the grid for day 5 (even though Haboo used another one in his most recent perfection run for day 18) I was wondering if the Dark blue tiles meant it starts with Crusader or a different enchant or if they are colored different for a different reason
I like the way CA did this rng. It's nice to know that when I turn my game back on I can expect certain things to stay the same and I don't loose a really awesome rng roll because my son turns Stardew off and I realize I didn't save by sleeping lol
I JUST made a post on the reddit that seems to have gotten 2 downvotes before getting a few upvotes on basically this very same thing only using a different method that isn't nearly as technical of RNG/Code savvy but should work for like 99% of people playing the game. People don't believe me. If you've like I can link it.
"With me so far?" Absolutely not, but I watched until the last second anyways. Always love your deep dives Blade! Excited to see more as it comes out, and will catch you on Twitch in the meantime!
I only now realised that the channel name is pronounced as “blade” and not “bla - dee”
Yes I know I mixed up Axe and Pickaxe...
"Let's say your name is Haboo..."
Great video, this man is gonna be the pannenkoek of Stardew
yes
this is a great idea for a series. cant wait for the next Stardew University. Also loved seeing you in the Stardew Trials today!
These take a bit of work to make, but I'm happy with how they turn out.
It's cool to be involved with the trials! Make sure to watch future episodes (wink wink)
@@BlaDeSDV heck yeah. Will do
👀👀👀
I didn't understand a word of this but I really found it fascinating. I'm glad I watched it!
please PLEASE do more of these
I've watched this 5 times and I'm still lost. Love it though. I'll watch it another 5 times and see what sticks, lol.
So glad Z gave us your channel. Love this stuff. Please do more of these.
This is great blade! I’m not a speedrunner or anything but it is interesting to hear how everything works! It is quite niche but I don’t think anybody has ever done anything like this yet, I’m excited for more lessons!
I personally love this addition to the game especially with the weapons but it admittedly eats up too many resources as I’m comfortably allowing it to (It took me like 30 prismatic shards to get my gavel to Artful which is the most helpful one for me). Had to do several skull cavern runs to get these prismatic shards and it eats up the in-game time I would rather use to arranging my farm layout or managing my crop inventory.
Really enjoyed this video.. even though I didn’t understand anything.. very soothing Temuera Morrison like voice
Yes! This is amazing and everything I could possibly ask for. Thank you so much, and I can't wait for the rest of the series!!
I understood 0.9% of this but I am entranced. Speedrunners are on another level, I salute you.
Wow, Blade, you’re a rock star. And enchant and all the other stuff. Thanks!
The MouseyPounds predictor takes your game seed and gives you the enchantment list, so there's a way to predict enchants without relying on parsnip qualities! This is still a great video for going in-depth on these mechanics and how they work tho.
I like you funny words magic man
for real though, great video!
This was very informative, thanks for making this video! I didn't know that the enchantments vary with the day and are the same each day. I'm going to use that to try enchanting first thing every day and reset and try another day if I don't get the result I want.
I thought I knew a lot about stardew and had seen all the cool tips videos. I was so wrong! Lol
I LOVE your Stardew content. This is so interesting. I'm subscribing
I love Stardew University
Great video!
Z just increased your sub count by like 200
It was amazing. Shoutout to @UnsurpassableZ!
Is the enchanting RNG actually $+ 0.522425... (mod 1)$ or is it a more complicated RNG core not discussed as not important? If the RNG used is that simple, it's theoretically predictable by a human (with some error), whereas if it's using a proper RNG core it's definitely in the realm of external tools.
I think there's merit in allowing external tools (that can't view internal state). Run notes are a primitive tool, a metronome for LoZ:WW superswims is a bit more involved, and the Sploosh-Kaboom tool is an RNG predictor. Depending on the specific tool use and how much it changes the route, it could be interesting to also have a "no internal knowledge tooling" category... though in RNG run-killer heavy runs, predicting/manipulating RNG is awesome as a way to make the run more consistent. I see it as on an axis where the endpoint is seeding; if I can figure out bits (pun intended) of the seed, how much am I allowed to do with that information? How much am I allowed to manipulate what seed I get? Speedrunning is about understanding a game at a fundamental level, so it's interesting to try to draw a line in the sand of what's "too much" tool assistance.
I'm glad to live in this age of speedrunning where we get to blur the line between RTA/TA and TAS. (In my opinion: it becomes TAS when you can view internal state or you segment (rerecord) not at game-supported segments.) I hope I get to see some more blurring from the other direction too -- speedruns performed by a program on an unknown seed, not just as a piano roll of inputs, but making decisions on the fly to respond to RNG happenings.
Stardew uses the base C# Random class, According to the docs, "The current implementation of the Random class is based on a modified version of Donald E. Knuth's subtractive random number generator algorithm. For more information, see D. E. Knuth. The Art of Computer Programming, Volume 2: Seminumerical Algorithms. Addison-Wesley, Reading, MA, third edition, 1997."
For the purposes of simulating RNG, this can be simplified to table; a list of intercepts and offsets e.g.
# * 0: 0.7262432699679598, 0.5224253141891330
# * 1: 0.8173253595909687, 0.2934186175900598
# * 2: 0.7680226893946633, 0.6989879904775824
# * 3: 0.5581611914365372, 0.2134429305854453
The result of an RNG call is dependent on the index of the call on an initialised Random object. It is (seed * offset + intercept) % 1. So the first call (gold crops) uses the first row, the second call (silver crops) uses the second, and so on.
Gold crops and enchanting both uses the first call of an RNG object.
I agree that looking at internal state should not be allowed in speedruns. However, I believe we should be allowed to use the result of RNG in one area to predict RNG in another. As to what level we let external tools and trackers guide this prediction is up for debate.
Why is blade so underrated ;-;
Only university I trust.
Thank you for this video. Losing Prismatic Shards to RNG has not been fun.
Sorry if this feels like a weird question, but can you explain Mine crates? why do Skull Cavern ones sometimes respawn, but not always, and are there any ways the original crates respawn?
I can, and this would make a good topic for a future video.
Crates that spawn around the side of the map are based on dailyLuck. You have a 50% to have 1-4 of them spawn, adjusted (-2 to +2) by daily luck. You can break as many of these as you want and they will still spawn.
Crates on container platforms work differently.
The very first time you load a floor that has a container platform on it, each tile on the container platform has a 40% chance to spawn a container. Once the floor is loaded, that number of containers that spawned on the platforms are saved to the save file. This number is the maximum number of containers that can spawn on the container platform on this level.
On subsequent loads of the same floor, each tile still has a 40% chance to spawn a container, but once the maximum number of containers has been reached, no more containers will spawn on the platforms. (priority is left to right, top to bottom).
When you break a container on a container platform, this maximum number is decreased. Once this number his 0, no more containers will spawn ever on that mines floor.
These numbers work exactly the same in skull caverns - the maximum number of containers is saved against the floor number. However, since the floor layout is randomised you can roll a layout that has no containers. The maximum number is set the first time you roll a floor layout that has containers.
Some floor layouts have more valid container platform spots than others. If you roll a small floor layout the first time the container platforms are to be set.... too bad!
Hey Blade! Love this and am planning a run with my sister using the grid for day 5 (even though Haboo used another one in his most recent perfection run for day 18) I was wondering if the Dark blue tiles meant it starts with Crusader or a different enchant or if they are colored different for a different reason
I got kinda lucky: I got artful on my hammer first try and then swift on my pick first try.
I wonder why they did it like this instead of making it seed completely unpredictably
The Random is seeded so that if you restart the day you get the same results. There is a predictable pattern because that is the nature of the C# RNG
@@BlaDeSDV why not make a new seed at the start of each day instead of for the whole file?
@@als_pals There are ways to make it more random yes. We just have what CA decided to go with
I like the way CA did this rng.
It's nice to know that when I turn my game back on I can expect certain things to stay the same and I don't loose a really awesome rng roll because my son turns Stardew off and I realize I didn't save by sleeping lol
I'm your 1000 sub
shamans be like:
I JUST made a post on the reddit that seems to have gotten 2 downvotes before getting a few upvotes on basically this very same thing only using a different method that isn't nearly as technical of RNG/Code savvy but should work for like 99% of people playing the game. People don't believe me. If you've like I can link it.
Blade, could you do something on how to manipulate the casino RNG?
:)
Nz accent?
Yep