Blazingly Fast Greedy Mesher - Voxel Engine Optimizations

Поділитися
Вставка
  • Опубліковано 21 гру 2024

КОМЕНТАРІ • 386

  • @cosmo9762
    @cosmo9762 8 місяців тому +802

    Hi! I wrote the "Binary greedy meshing" algorithm. Very cool to see this video on my youtube frontpage today, I love your video and explanations :)

    • @samuelcollier1764
      @samuelcollier1764 8 місяців тому +33

      glad to see people are finally seeing this now! It definitely deserves more attention

    • @tisaconundrum
      @tisaconundrum 8 місяців тому +68

      I was skeptical at first, but after some digging, damn you really are the guy that wrote the mesh algorithm 4 years ago. Nice!

    • @UnrealCatDev
      @UnrealCatDev 7 місяців тому

      @@tisaconundrum always has been

    • @MountainLabsYT
      @MountainLabsYT Місяць тому +3

      cosmos: casually has a 20 microsecond binary greedymesher.

  • @MrSofazocker
    @MrSofazocker 8 місяців тому +699

    You just casually made a spatially mapped datamodel lol

    • @daddy7860
      @daddy7860 8 місяців тому +37

      What part of this video was casual lol

    • @notthetruedm
      @notthetruedm 8 місяців тому +125

      @@daddy7860 The way he explained it felt like a friend explaining something to me rather than a teacher explaining.

    • @Pockeywn
      @Pockeywn 8 місяців тому +33

      yep. just to remake minecraft. this is what people do on the internet.. its awesome.

    • @yaboiminecraff
      @yaboiminecraff 8 місяців тому +9

      ​@@notthetruedm the best way to learn

    • @tomtravis858
      @tomtravis858 7 місяців тому +14

      @@Pockeywn voxel games existed before and after minecraft, not every voxel game is a minecraft clone

  • @redfatcatz
    @redfatcatz 8 місяців тому +390

    Thanks for the video, I can advise you not to make a greedy mesh for each type of block, but to make for all complete solid blocks, and then transfer to the GPU data structure with the help of which you can calculate the block type and texture by pixel position, it will simplify the mesh many times as well as the algorithm itself.

    • @tommycard4569
      @tommycard4569 8 місяців тому +9

      ah, this makes sense

    • @CaptTerrific
      @CaptTerrific 8 місяців тому +43

      I'd love to see the speed comparison on that, sounds promising!!

    • @charltonrodda
      @charltonrodda 8 місяців тому +8

      Is it really faster to do that lookup in the fragment shader than it is to store it in the vertex data or look it up in the vertex shader?

    • @raffimolero64
      @raffimolero64 8 місяців тому +27

      @@CaptTerrific Saves a hashmap entry access for every voxel. bets on 4x speed.

    • @TheSliderW
      @TheSliderW 8 місяців тому +1

      Same for lighting and ambient occlusion

  • @PikkelP
    @PikkelP 8 місяців тому +54

    this is insane! i have my own culled and greedy meshing implementations and i know they're not the fastest, but i'd never have thought it could get THIS fast. you could literally remesh every chunk every frame with this and still get good fps, which is mind-boggling. good job with the explanations, too.

  • @notthetruedm
    @notthetruedm 8 місяців тому +79

    When you explained the part in 14:40, where you explained how to find the faces looking right just by modify an interger, I was so surprise at how simple it is an yet amazingly complex

    • @110110010
      @110110010 8 місяців тому +9

      My thought right there was "oh, is this edge detection?" It was a really intuitive explanation

  • @Siphonife
    @Siphonife 8 місяців тому +54

    I now fully understand the concepts used to achieve such high performance. I also fully understand that if I were to try to write it. Every line of code would have an off-by-one error.

  • @GeorgeTsiros
    @GeorgeTsiros 7 місяців тому +2

    i love everything about this
    your awkward presentation, the handdrawn sketches, the weird pronounciation, the focus on speed, your manbun, your long hair that makes you look like a metalhead, the jokes, the effort you made, everything, everything in this video is just *right* .

  • @14corman46
    @14corman46 8 місяців тому +8

    This is incredible! I did something extremely close to this for counting strings within DNA sequences and got immense speedup. Binary manipulation is insanely speedy if you can comprehend it. Great job figuring this out and explaining it.

  • @AlexYusiuk
    @AlexYusiuk 8 місяців тому +80

    You can speed up the data setup part by using stack arrays instead of using "Vec"s

    • @minecraftermad
      @minecraftermad 8 місяців тому +19

      oh yeah definitely since the array size is a known value, and doesn't need to be resized. and is small enough to fit into stack.

    • @0x4849
      @0x4849 8 місяців тому +16

      CHUNK_SIZE_P3 = 34*34*34, so the size of axis_cols is 3*34*34*34*64 = 7,546,368 bits. Additionally, we need twice that for col_face_masks, giving ~2.83MB. Honestly, I don't know whether this will fit on the stack or not. Maybe someone else can provide additional information?

    • @lengors7327
      @lengors7327 8 місяців тому

      ​@@0x4849 I believe max stack size can be changed when compiling, but the default is usually not very large.
      I would instead preallocate the vector once and then always use that one instance

    • @AlexYusiuk
      @AlexYusiuk 8 місяців тому

      @@0x4849 I had a program where I had an array of 5 Mb. So 2.83MB should be feasible. Also, the memory can be static. We really just need to benchmark the approaches and choose the best one

    • @AlexYusiuk
      @AlexYusiuk 8 місяців тому +8

      2.83mb should be feasible. I had a program that used 5mb for a stack array 😅. The memory can also be shared between calls whatever it's stack or heap based. Different approaches should be benched and there should be a room for improvement

  • @danmerillat
    @danmerillat 8 місяців тому +6

    One final thought, if you use 30x30 chunks you can fit the left/right neighbors into a 32bit int rather than expanding to 64. It'd halve your memory bandwidth requirements at minimum and if you use SIMD it will let you double the number of calculations performed per cycle.

  • @TimDrogin
    @TimDrogin 8 місяців тому +5

    Epic games had a wonderful talk about nanite, and the part that blowed my mind is: Gpu’s are very slow at rendering extremely small triangles. So what they did? They just wrote a SOFTWARE RASTERIZER, that is faster than the hardware one I think when the size of a triangle is less then 40*40 pixels. The difference is really impactfull, and they showed the code and implementation for everybody to use it!

    • @dancingtriangle2814
      @dancingtriangle2814 5 днів тому

      I know this is an old comment, but this reminds me of when the youtuber Gamehut (who worked at telltale games/sega in the 90's) would do a lot. He would frequently have some form of technical problem where the hardware wasn't capable of doing what he wanted, so he just wrote the functionality in software.

  • @nanda_gamedev
    @nanda_gamedev 8 місяців тому +54

    Oh my god I wish i had this video like 2 months ago when i was trying to write a greedy mesher. Thank you so much for this resource! Will definetly save it for the future!!

  • @NunTheLass
    @NunTheLass 8 місяців тому +5

    The brings back memories. I recognized some code I wrote about 15 years ago after being blown away by Minecraft. The & operation on the shifted bits specifically. The merging of the meshes was clever and much better than what I ever came up with. I put it all in a fancy octree though so I only rendered on-screen chunks. I hit a brick wall getting the lighting to work on merged meshes and it all fell apart once I had more than, say, 6 block types. Your code will do so too. But it's a great exercise and good job. A more modern way would no doubt be raycasting, there are many more triangles than pixels on the screen if you scale things up and it parallelizes better. Nice video, keep them coming!

  • @Unbreathable
    @Unbreathable 8 місяців тому +28

    This video is honestly so well explained and even though I don't know anything about voxel engines or game development I was able to understand it. This is probably one of the best resources for making a voxel engine. If I ever make one, I'll probably take a look at this again, thanks for your amazing work!

    • @theseangle
      @theseangle 6 місяців тому +1

      When you stare at engine development - graphics programming stares back. Takes like 15 passes of starting to learn the basics, to learn the basics. And this wasn't a typo

  • @sturdyfool103
    @sturdyfool103 8 місяців тому +6

    I haven’t tried greedy meshing but I’ve seen some demos of greedy meshes where it doesn’t care about block type, it constructs the triangles while remembering where the different block types are, so it’s possible to make the greedy meshes not slow down when you increase the block type count

  • @theStumblinbear
    @theStumblinbear 8 місяців тому +18

    Use an array for the data instead of a vector (since you know exactly how many entries it will contain) and it should have essentially zero allocation time since it'll be allocated on the stack instead of the heap

    • @Tantandev
      @Tantandev  7 місяців тому +22

      It doesn't fit on the stack on linux it's to large!
      But here is the funny thing... Someone noticed I'm allocating WAY more memory than was actually used.
      And now it does fit on the stack :) So I've changed it.
      The performance difference was only minimal though.

    • @LiamOfOzz
      @LiamOfOzz 6 місяців тому +5

      @@Tantandev you could also go full Kaze Emmanuar and save 0.5µs by rewriting the thing differently 5 times

  • @asherhaun
    @asherhaun Місяць тому

    I am building a voxel engine as my senior design project for school and I am about to start greedy meshing. This video helped so much, my mind is blown by that face culling technique and I can't wait to try it because right now that is a massive slowdown in my chunk generation.

  • @augustvctjuh8423
    @augustvctjuh8423 8 місяців тому +3

    You can combine different block types like grass & dirt into the same triangle. You would obviously still need to differentiate between them, which can be done in the fragment shader. Does introduce a lookup in the fragment shader but it may be very worth it if you have a lot of block types or if greedy meshing is your bottleneck

  • @p1tayaa
    @p1tayaa 8 місяців тому +4

    Man amazing video. You made it sound so hard but I feel like I grasp all of it pretty well. The visuals make it soo much easy to follow, hats off to your work.

  • @DreadKyller
    @DreadKyller 8 місяців тому +2

    Several people have mentioned looking into SIMD optimization, but a few other ideas:
    1) Using a fixed sized allocation instead of a Vec since the size is known. Not sure whether the entire arrays would fit on the stack but if so that may provide several speed improvements over a Vec on the heap.
    2) It might be possible to combine both positive and negative edge detection into a single operation by using an XOR, but would require a slightly different method of iterating over them to pass into the greedy meshing.
    3) Your structure for axis_cols has the data for each grid separated, a format similar as such: (y1, y2, y3, y4... x1, x2, x3, x4... z1, z2, z3, z4...) this means when setting the values you're writing into separate parts of the vec that might be far enough from each other to cause frequent cache misses. A layout where the three axis all are interwoven beside beside each others might be faster, such as (y1, x1, z1, y2, x2, z2, y3, x3, z3, y4, x4, z4...)
    4) It would require a bit of rework but this seems very reasonably practical for a compute shader.
    5) Would take a fair amount of work, but rethinking how you store the actual voxel data in general may make it faster to convert.
    6) Again it would be a change in direction, but there are approaches people have taken where you can greedy mesh any flat surface, regardless of different types of blocks. The way that achieve this is usually to pack the color data for the chunk into a 3D texture and use it in the material/shader for the chunk mesh, then, rather than each triangle having a color, the fragment shader can use world coordinates to query from the color data as a 3D texture at the position of the face. Allowing a single triangle to have multiple colors on it. This makes the fragment shader slightly more complex, but in most examples of people using this technique it tends to improve performance in both rendering and construction because it can result in a massive reduction of polygons, especially as you add more and more materials.

  • @Gin2761
    @Gin2761 8 місяців тому +13

    I'm making a game that has voxels and I implemented this also using Rust and something very similar to the slow approach. This video came out with such a great timing.
    Thanks for sharing this.
    Maybe it can be even faster if SIMD or parallelization are included? 😁

  • @jeanlouis5619
    @jeanlouis5619 8 місяців тому +56

    Looking at this made me realise that I clearly need to lurn bitwise manipulation

    • @DanKaschel
      @DanKaschel 8 місяців тому +11

      It's actually genuinely fun if you like puzzles. A lot of it is figuring out how to visualize it so you can figure out what's going on because the final product is always undecipherable (at least for me).

    • @DreadKyller
      @DreadKyller 8 місяців тому +8

      It's honestly amazing how many usecases there are for bitwise operations, I think at least some understanding, even if only basic, should be a core skill of any serious developer.

    • @memes_gbc674
      @memes_gbc674 7 місяців тому +1

      if you know all these quirky things you can figure out pretty quickly that an odd number is determined by its first bit

    • @DanKaschel
      @DanKaschel 7 місяців тому

      @@memes_gbc674 assuming you understand endianness and therefore which bit is "first"

    • @memes_gbc674
      @memes_gbc674 7 місяців тому

      @@DanKaschel that too

  • @bassguitarbill
    @bassguitarbill 8 місяців тому +7

    This is some Big Brain Calculation right here, great video Tantan!

  • @Xeros08
    @Xeros08 3 місяці тому +4

    We miss you TanTan :(

  • @Iridescence
    @Iridescence 8 місяців тому +2

    Great video! I wrote a basic algorithm for doing this on culled meshes, but I'm glad to see it's possible with greedy meshes too!

  • @admexir
    @admexir 8 місяців тому +2

    I like how you mostly pronounce "Chunk" as "Shunk", always made me smile :D

  • @HoloTheDrunk
    @HoloTheDrunk 8 місяців тому +3

    Thank you Tantan for somehow releasing a video on the exact topic I was worried about for my next project, very cool.

  • @DanKaschel
    @DanKaschel 8 місяців тому +5

    Love this. I remember building a 2048 AI and going from loop-type grid transformations to bitwise operations. Bitwise stuff is hard to grok but there are sooo many orders of magnitude of improvement and it's so satisfying :)

    • @magfal
      @magfal 8 місяців тому

      I love how other SQL devs look at me when I explain my stored procs that utilize bitmap logic to be a million times faster than the naive approach to the same problem.

    • @DanKaschel
      @DanKaschel 8 місяців тому

      @@magfal umm. I think I'd do the same if a colleague said they were using bit manipulation in a stored proc

    • @magfal
      @magfal 8 місяців тому

      @@DanKaschel
      Calculating using the bitwise code and returning the final result set in postgres put less load on the postgres server than serving the data it's based on to application code, which then had to run the calculations.
      This is true quite often for OLAP style workloads.

    • @DanKaschel
      @DanKaschel 8 місяців тому

      @@magfal that is true, but it'd have to be pretty niche before performance trumped maintainability

    • @magfal
      @magfal 8 місяців тому

      @@DanKaschel a 10 line comment was enough for my colleague to understand and confidently make adjustments for a new requirment.
      Bitwise code isn't magic or that hard to do when you know the incoming data, the result, the intended behavior and you've got the code in front of you.
      And to go from a batch job ran once a month to an on demand real time task is quite important when the report directly generates revenue for it's users with more benefit being reaped the fresher the data being presented is.

  • @zy-blade
    @zy-blade 8 місяців тому +3

    What a coincidence, I currently need a good voxel algorithm for my project :D Will definitely look into it! thanks

  • @uncertawn
    @uncertawn 8 місяців тому +2

    this videos singlehandedly makes me wanna try to make a 3D game from scratch

  • @konkitoman
    @konkitoman 8 місяців тому +2

    Now i understood why this video take a while to be made!
    This was a really good video!

  • @heryu2630
    @heryu2630 7 місяців тому

    It's like a gift for me.
    It was a problem no matter how much I optimized it before
    but now I have no problem loading and rendering faster than before. thanks for your video

  • @lukejagg
    @lukejagg 7 місяців тому

    Woah, love the bit shift and negation. That's a great way to generate the culling indices instead of iterating through every single block.

  • @enrique6693
    @enrique6693 8 місяців тому +2

    3:00 as I always say "paint is the most important software for software developers"

  • @FM-kl7oc
    @FM-kl7oc 8 місяців тому +12

    If your friends CPU has significantly larger L2 or L3 cache, the performance difference could perhaps be cache misses? Aligning data for CPU cache optimization is another beast to tackle though 😅

    • @AmaroqStarwind
      @AmaroqStarwind 7 місяців тому

      I think it's possible to enable larger memory pages in some compilers.

  • @eugenech.2450
    @eugenech.2450 7 місяців тому

    I dont watch your videos (but still subscribed (I want to learn rust&bevy some day)), but every time I see your videos it feels like a new scientific experiment.

  • @LeBopperoni
    @LeBopperoni 8 місяців тому

    Lmao I wrote an algorithm yesterday for greedy meshing which does a bunch of neighbour checks for each block and then creates a bit mask from that.
    Definitely stealing the bitshifted comparison optimisation. This must be the most amazingly timed video I've ever seen.

  • @Cluxiu
    @Cluxiu 8 місяців тому +1

    I came from Dani's video, and I'm glad I did :) Great video!

  • @gregbigwood4532
    @gregbigwood4532 8 місяців тому +2

    phenomenal video explaining this. you are very good at explaning these topics

  • @jerobarraco
    @jerobarraco 3 місяці тому +1

    nice explantion and walk through code. i loved that. thanks

  • @lukidoescode
    @lukidoescode 2 місяці тому

    Fantastic video! I very much appreciate all the work you put into these. You have a gift for explaining things. Can’t wait for you to publish a game. Make sure to introduce it in a video, pls 😊

  • @downey2294
    @downey2294 7 місяців тому +2

    1:03 missed opportunity for the vsauce intro ost

  • @katech6020
    @katech6020 8 місяців тому +2

    I would love to see a full bevy tutorial on your channel

  • @sentinelav
    @sentinelav 8 місяців тому

    You're using bitwise operations to calculate binary derivatives. That's dope :')

  • @mek101whatif7
    @mek101whatif7 8 місяців тому +24

    Now do it with SIMD

    • @angeldude101
      @angeldude101 7 місяців тому +3

      At least on x86, bitwise operations like count trailing/leading zeros are only available on the more recent AVX-512 processors, so adding SIMD might make it faster of their friend's CPU, but could actually make it slower in parts on their own. There are probably some places where it'd be beneficial anyways though.

    • @GeorgeTsiros
      @GeorgeTsiros 7 місяців тому

      @@angeldude101 you sure about that? Let me check... FELIIIIIX, WE NEED YOUR SITE AGAIN

  • @VulpeculaJoy
    @VulpeculaJoy 6 місяців тому +1

    Someone on the internet makes a fast greedy voxel mesher.
    *distant horizons 2.0 has entered chat

  • @godmode8687
    @godmode8687 4 місяці тому +1

    I have an exam in relational databanks in 3 days. And this video and the bitwise manipulation really, really didnt help me at all. But its very cool stuff and im looking forward to try something like this as soon as i finished my exam

  • @gnif
    @gnif 3 місяці тому

    You could further speed this up by making use of SIMD as you can do all these bit-wise operations on several values at once reducing your need to loop as many times. The XMM registers are 128bit wide allowing you to stuff 4 32bit values into a register would allow you to calculate 4 values in one operation.

    • @tomtravis858
      @tomtravis858 3 місяці тому

      True, but I typically try to avoid SIMD because of portability.
      I'm not sure what Tantan is expecting to support.

  • @bosine9431
    @bosine9431 8 місяців тому +8

    Just want you to kmow that this video was so good that at 11:27 there was a solid 5 seconds where I actually scrambled to rewind the video to try to desperately see the code

    • @Syerjchep
      @Syerjchep 3 місяці тому

      Code's in the description

  • @thaddaeusmarkle1665
    @thaddaeusmarkle1665 8 місяців тому

    Wow man, mad props. That was some heavy stuff and yiu actually explained it extremely well. Thanks, and keep up the good work!

  • @macawls
    @macawls 8 місяців тому +10

    BLAZINGLY FAST

  • @AmitBen
    @AmitBen 8 місяців тому +1

    Amazing work, You can make this dramatically faster using SIMD now that its a bitwise op game

  • @meetem7374
    @meetem7374 6 місяців тому

    Oh! Great catch. Initially in my rendering I've used 64 bitmasks, because my chunks (not rendering chunks) were always 4x4x4 voxels. Tho I haven't implemented a greedy meshing, because I need to support much more than a solid block, so different shapes etc. End up with custom rasterizer.

  • @Teflora
    @Teflora 8 місяців тому

    You succeeded very well in explaining something complex in a simple manner! Well done!

  • @fugoogle_was_already_taken
    @fugoogle_was_already_taken 8 місяців тому +1

    The reason, why the code was equally fast is, that bit operations are single instructions. The reason, why new CPUs are usually faster, is other than clock speeds, which are generally about the same, because chip engineering hit a limit on that one (any faster and CPUs melt basically).
    However, I am not sure, why the standard mesher is faster. But I am suspecting CPU cache size, since improvement in pipelining or superscalar execution would also lead to improvements in the first comparison and the clock speed is about the same, as I mentioned before

  • @OctagonalSquare
    @OctagonalSquare 8 місяців тому

    15:04 this was the point I verbally said “this guys psychotic” but in a good way. This is a crazy way to think about this data but it makes so much sense! Good work man!

  • @Kevroa1
    @Kevroa1 8 місяців тому

    WOW you did a really phenomenal job at explaining your algorithm

  • @zennii
    @zennii 8 місяців тому +2

    Coincidentally I just implemented nearly the same thing a week ago, though I support octree blocks so it's a bit more involved, but cool to compare implementations. I made use of xor to detect my faces, never thought of just flipping the neighbor... My meshing ended up about 50% faster somehow after implementing it, even though it feels like more work is being done

    • @infinitasium
      @infinitasium 8 місяців тому +1

      The expression he did for his mesher is actually one half of an xor (A xor B = A*(!B) + (!A)*B), and since CPUs have built-in support for all binary operations, your algorithm does the work at once instead of going through it twice by choosing the two paths at once. The only caveat here being two bits are on instead of one, but that difference is irrelevant as they are guaranteed to be next to each other.

    • @zennii
      @zennii 8 місяців тому

      Interestingly I tried switching my system to just flipping bits instead of xor, I was already flipping the bits for another part so surely it should be free gains. Weirdly, it ended up very slightly slower which is perplexing. I don't think it's worth diving into it enough to find out why or what changes the compiler has here, but thought I'd note what I found...

  • @skilz8098
    @skilz8098 6 місяців тому

    Pretty cool project and interesting technique and considering that it is within the context of a voxel engine, it is very fitting to say the least.
    Outside of that, if one isn't referring to a pure voxel engine; there are many other interesting techniques that are also very interesting. These range from:
    Quad and Octrees, The use of Quaternions, Octonions, Sparse Matrices, FFTs (Fast Fourier Transforms) and their inverses, Instancing through referencing, other types of procedural generation techniques, Perlin Noise Sampling, Batch Processing through a priority queue, Fractal Generation, and so much more. Don't forget, you can also do a lot of interesting things with a plethora of available techniques not just within the engine itself, but also within the context of shaders. The Graphics Pipeline in general is a very interesting field of study.

  • @blinkblade6962
    @blinkblade6962 8 місяців тому +1

    Loved the Flight of the Conchords reference

  • @VegetableJuiceFTW
    @VegetableJuiceFTW 8 місяців тому

    Fun fact, you can further reduce polygon count by allowing polygons cross (each other in) the same block type or culled space.
    z-fighting is not an issue as it is the same texture or not visible. I have demo and thesis on this.
    The following "donut" example requires only a single polygon :D
    01110
    01x10
    01110

    • @DreadKyller
      @DreadKyller 8 місяців тому +1

      While Z-fighting isn't an issue, you still then may have to deal with overdraw.

    • @VegetableJuiceFTW
      @VegetableJuiceFTW 8 місяців тому

      @@DreadKyller yup, it's a tradeoff.

  • @meanmole3212
    @meanmole3212 8 місяців тому +1

    That is cool revelation and use of bits.

  • @MeepMu
    @MeepMu 2 місяці тому

    Now SIMD it!
    Granted LLVM might auto-vectorize it already, but you might be able to optimize even more by doing it manually :)

  • @sotojared22
    @sotojared22 8 місяців тому

    Thanks to practicing image manipulation in JS, this was surprisingly easy to understand and clicked right away for me. 1D data models and traversal is not simple, so I understand your pain.

  • @purplepixeleater
    @purplepixeleater 8 місяців тому

    Thanks from a godot developer (csharp) this is very useful there as well since bitwise operations work very similarly and especially with multimesh instancing! Cheers :)

    • @DreadKyller
      @DreadKyller 8 місяців тому +1

      bitwise operators are basically universal, they aren't language specific, you can do them in every language I know of. So very useful and easily transferable skill to know.

  • @DigitHallMusic
    @DigitHallMusic 6 місяців тому

    This is cool, I remember playing with greedy mesher but end up going back to a traditional one because I didn't find a good way to get rid of T-junction artifacts

  • @mattrommel9521
    @mattrommel9521 8 місяців тому +13

    Why can't the mesher be happy with what it has

    • @cubee4108
      @cubee4108 6 місяців тому

      because it runs out

  • @diontryban5645
    @diontryban5645 8 місяців тому

    This is awesome! I'm looking forward to attempting to implement this myself. I'd love it if you would cover ambient occlusion in the future or at least provide some resources for where you learned about it

  • @a1r592
    @a1r592 8 місяців тому +1

    You explained this really well! Thanks!

  • @simonhartley9158
    @simonhartley9158 8 місяців тому +1

    Really cool. I wonder how this would relate to the optimizations that Vercidium uses to get voxel rendering running at a claimed 12000 fps.

  • @liz2k
    @liz2k 2 місяці тому

    Very nice work, and mathematical ideas!
    Good inspiration...

  • @scotthooper6460
    @scotthooper6460 7 місяців тому

    You can go farther. In my greedy mesher I store both block and ambient-occlusion lighting in textures, as bytes, using a bin-packing algorithm. One 2kx2k texture has always been enough, but I also added the ability to track which is needed by each chunk in case I needed many.
    This is particularly useful in city-like terrain where the geometry has a lot of flat faces made up of different types of blocks.

  • @realdlps
    @realdlps 8 місяців тому +2

    God damn bitwise wizards, I really have to learn how to use that stuff, because in theory I understand it, but I don't know how to use it

  • @gunnerstone120
    @gunnerstone120 3 місяці тому

    My osrs hydra ptsd came back when your music started playing around 2:25

  • @Skeffles
    @Skeffles 8 місяців тому

    Amazing to see the level of performance you can get out of using the binary representation and this has me wondering if I can use any in my own projects. I suspect I will need something similar to create an AI navmesh in the near future.
    Fantastic video once again TanTan!

  • @Gwilo
    @Gwilo 8 місяців тому +1

    please would you turn the in-game render distance up as far as you can and show us how fast it runs?!

  • @btarg1
    @btarg1 8 місяців тому

    This voxel engine looks incredibly advanced and would make a brilliant base for games! For future videos I'd love to see you implement a scripting language into an existing Rust project like Lua or Angelscript.

  • @devpenguin0
    @devpenguin0 8 місяців тому

    I tried out the mesher in my own project with a different chunk storage scheme. I ran benchmarks on my project and got around 500 µs (microseconds) per chunk. When I ran your benchmark I was getting around 32 µs. Initially I thought it was just inefficiencies in retrieving voxel data since I'm using palette based compression. After some more testing I found that if I use your chunk generation code the benchmark result was around 50 µs. Turns out there's only a few solid voxels in the benchmark chunk which is why it runs much faster. The first chunk I tested/benchmarked had solid voxels in a sphere shape. Still my voxel retrievel from the chunk is significantly slower then simply indexing into a Vec. Mainly because I use bitpacking for the indices.

  • @thygrrr
    @thygrrr 6 місяців тому +2

    I think your memory layout can probably benefit a little from ordering by which stage of execution needs the direction.
    The newer CPU has a better cache and branch prediction, and that's why it outperformed on the culled renderer. The greedy mesher caused it to stall more, so these advantages were likely nullified.
    Your greedy mesher seems to tax the cache a lot - your blocks are 32x32x32 bits (right?), and interleaved in triples; but it could be better to somehow make them fit into 64x8=8x8x8 bits (64 bytes, a cache line) for each of their core operations.
    So, making the block smaller by a factor of 2 to 4 in every dimension AND ordering the memory according to execution order / ordering execution to be doing operations on directly adjacent memory blocks directly after one another could probably give you another big boost.

  • @thestealthtarget
    @thestealthtarget 3 місяці тому

    This is awesome! i love to see the Rust code and see how you are using Rust. As a new to low level programmer, this helps a ton! keep up the content!

  • @bramhuis3571
    @bramhuis3571 2 місяці тому

    This is a beautiful piece of engineering😍😍

  • @NabekenProG87
    @NabekenProG87 8 місяців тому +3

    I wonder if the data layout could be improved as it looked like you use sequences of array indices that are far apart from another. Depending on how large the data is, this could theoretically lead to cache misses as not the entire array is loaded into the cache at the same time. But it's only 0.8% of runtime and the Compiler probably already optimizes this. But if there was a slowdown caused by cache misses, improving the data layout could speed up the code a lot

  • @JayAether
    @JayAether 3 місяці тому

    i barely understood this but great video! gonna return to this video one day

  • @charetjc
    @charetjc 8 місяців тому

    Excellent video. The animations are easy to follow along with. Thanks for sharing.
    I'm curious about the method you used to profile your code to determine the execution time of various sections. I didn't see any particular video in your catalog that seemed to cover this, so perhaps a "How Tantan profiles his Rust code" could be an idea for another video.

  • @Siphonife
    @Siphonife 8 місяців тому +3

    Now write it in SIMD using WIDE bit registers. imagine what you could do with 4x256 bits :P

    • @DreadKyller
      @DreadKyller 8 місяців тому

      My goodness I don't think the world is prepared for that much power...

  • @l4zycod3r
    @l4zycod3r 8 місяців тому

    Buffers can be prealocated and reused, that should speed up a bit more

  • @redpug5042
    @redpug5042 4 місяці тому

    Idk if this would be any better, but I'd like to see an algorithm that doesn't mesh it at all, but uses the GPU to figure it out.
    Basically, you have a 32x32x32 chunk, with 32 quads from each of 3 directions, all facing the camera. When it comes time to render, you send those quads to the GPU along with an array with the 32768 voxel types, and the GPU would be able to draw each quad with the respective colors.
    To draw it in the correct layer position, draw all of the chunks that the camera isn't in order from furthest to closest, and do the same with the quads.

  • @oglothenerd
    @oglothenerd 8 місяців тому +1

    Yes! He's back! Let's goooo!

  • @ToonedMinecraft
    @ToonedMinecraft 8 місяців тому +1

    This was so clearly explained! You finally made me understand a usecase for bit-shifting!

  • @diegoaugusto1561
    @diegoaugusto1561 8 місяців тому

    a tip go further decrease the data creation time: You're always creating and releasing memory with those Vec's. You should find a way to allocate memory once and reuse it instead. Also, don't use the stack because it can heavily limit you.

  • @HanProgramer
    @HanProgramer 8 місяців тому +1

    Babe wake up another tantan video has dropped

  • @superblaubeere27
    @superblaubeere27 8 місяців тому

    I think this can be made even faster using SIMD-instructions. Most of these problems are similar to problems in parsing where I know that those instructions can make a big difference. Especially in that data preparation step.

  • @rinoturtle738
    @rinoturtle738 8 місяців тому +1

    Thak You! That video and research are so usefull!
    After watching your video, my greedy mesher looks so sloooooow :c

  • @jerobarraco
    @jerobarraco 3 місяці тому +1

    2:49 but acerola... have is your culled mesher optimized as much yet?

  • @jackjohnes2623
    @jackjohnes2623 7 місяців тому

    I imagine, there should be a way to optimise the procedure at ~3:35 so that the whole meshing step is complete in the smallest number of expansions possible (for the general case).

  • @jamesalewis
    @jamesalewis 7 місяців тому

    This is the same algorithm as bitmap edge detection. Shift-not-and-ing is really common in other applications 🤙

  • @eboatwright_
    @eboatwright_ 8 місяців тому

    You love to see it! I've also been optimizing the Rust code of my Chess engine, although this seems exponentially more complicated 😅

  • @user-dm7sk5wf5w
    @user-dm7sk5wf5w 7 місяців тому

    Very cool. I bet you can double the performance with some tweaks to how you manage memory. I see a lot allocations happening in loops when you could make 1 allocation outside the loop and reuse the variable for each cycle of the loop.

  • @tommycard4569
    @tommycard4569 8 місяців тому +2

    Brilliant! how does splitting data by block type affect the memory footprint as more types are added tho? Is there an optimal sized chunk to limit the unique block types that can occur within each versus the number of iterations to cover every chunk?

  • @Hellbending
    @Hellbending 8 місяців тому +1

    Would it be possible to bitmask the x, y and z coords somewhere? So as you can calculate the face-5ets in one pass, without having to triple loop and storing the bite set for it somewhere external to the loop? Whether it's worth it or not is down to testing performance but is it even possible?
    Could it be possible to use the remaining space in the byte to store block type data too? (IE: air/dirt/grass etc.) it would grow as you add more block types, but if you're taking up 3, for xyz, perhaps possible to find a way for the remaining 5 to be useful for each parse?