I wrote a dice rolling script with 3d rendering, collisions and friction in bash because why not [edit: code available now]

https://files.catbox.moe/8rz9m7.mp4

Daddy needs a new pair of RAM!

edit: the fps are way better in smaller terminal windows with lower character count but then it's hard to make out the dice. D:

edit2: code here (expires in 2 weeks)

419 points · 93 comments · view on lemmy.world

93 Comments

oce@jlai.lu · 33 pts · 21h (8 replies)

I'm curious why it seems the time slows down at the end compared to reality? Because more computation is needed to check if it's a stable state?

Pudutr0n@lemmy.world · 30 pts · 20h (7 replies)

the final resolution stage was a huge headache and as you can see, one I didn't fully solve. Basically when the dice finally settle, particularly those with more sides, lots and lots of micro collisions happen in close sequence, each one having to apply friction and bounce back. As velocities come close to be rounded to 0, the bounce back effect and force of gravity no longer provide movement, but the final angle of the bottom face may not be fully settled on the floor. So basically if i don't ignore all these tiny collisions I get the frame rate drop you see and if I do, I can arrive at not fully settled states which can become ambiguous result-wise in, say, the 20-sided die.

hirihit640@sh.itjust.works · 9 pts · 19h (4 replies)

Can you predict the resolution when it is close enough to settled? Worst case maybe have an algorithm that estimates the probability of each face ending up on top, and if one face is overwhelmingly more probably, choose it. Otherwise let it settle using your physics engine

Pudutr0n@lemmy.world · 9 pts · 19h (3 replies)

I can easily check for when motion stops, but checking for when any particular face is both parallel to the floor and at the same level can be computationally expensive, particularly in the dice with more sides. What you're suggesting is likely a good idea, but I wouldn't know how to get probabilities without doing what I just mentioned. I'm sure there's an optimization i can do with this that I can't think of rn.

hirihit640@sh.itjust.works · 8 pts · 19h (2 replies)

maybe just "if velocity < 0.01, find which side is closest to the floor, and if its within 1mm of the floor, take the opposite side (the one facing up) as the answer"

The only edge case I can think of is if it gets close to balancing on an edge, and then finally tips over to one side or the other. But in that case I don't think the face closest to the floor would be within 1mm of the floor, so it should still work.

Pudutr0n@lemmy.world · 6 pts · 19h (1 reply)

that's how it's implemented, mostly, but sometimes dice "stop" while they're still tilted and standing on a corner (and then roll back down, possibly going back more than 1 face) so there's also that.

hirihit640@sh.itjust.works · 3 pts · 15h

Ok well there's probably a dozen different ways to tackle this issue but I'll let you explore them since that's part of the fun (and I'm lazy lol)

Cethin@lemmy.zip · 2 pts · 5h (1 reply)

Since you're rolling on a plane, can't you simplify the collisions to only check the corners? I think that should be all that's needed. There shouldn't be a time where the edges or faces are below the lowest corner on a plane, so they can be skipped I believe.

Pudutr0n@lemmy.world · 1 pts · 4h

i need to check the corner collisions but when the die start settling and an entire edge comes close to the floor, gravity keeps pulling back on several of them making them bounce back just a tiny bit, not reaching the velocity that gets rounded to zero.

I fixed that on the current version doe!

glizzyguzzler@piefed.blahaj.zone · 29 pts · 19h (26 replies)

Get this on the berg I wanna render my dice rolls over SSH stat

Pudutr0n@lemmy.world · 13 pts · 19h (18 replies)

lol but it's not finished and everyone will see how stupid my code (and brain) is D:

GeirGrusom@lemmy.world · 21 pts · 16h

It's bash. Everyone is expecting nightmare code, and that's why it's fun!

CrazyLikeGollum@lemmy.world · 12 pts · 18h (1 reply)

Pudutr0n@lemmy.world · 2 pts · 10h

I forgot to tell you how much I appreciate this meme. heh. thanks!

promitheas@programming.dev · 9 pts · 18h (2 replies)

My suggestion is do it. Its a cool project, and clearly there is interest from others who could contribute to the project by suggestions, bug fixing, or general collaboration, helping you learn. Just add some kind of disclaimer in the project that its not ready yet, and that youre still learning and it should be fine.

Pudutr0n@lemmy.world · 6 pts · 16h

you make good points

Pudutr0n@lemmy.world · 1 pts · 6h

i'm just gonna pastebin it for now but yeah, thanks for the encouragement.

BasedMaquisEddington@lemmy.dbzer0.com · 8 pts · 17h (4 replies)

I hope you change your mind.

Most people care about features, not the implementation.

Those who dislike your code are probably going to generally dislike others’ code anyways, so in a sense they don’t even need to see your code to already disapprove of it 🤷‍♀️

Pudutr0n@lemmy.world · 6 pts · 16h

Well, you guys are making some p good points. I'll probably end up sharing it, but'll give it a clean up, try to fix some bugs and improve my comments before posting it publicly. For now whoever really wants it (including you) can just ask and I'll dm it, understanding it will be a hard read and still has bugs.

Pudutr0n@lemmy.world · 3 pts · 6h (2 replies)
BasedMaquisEddington@lemmy.dbzer0.com · 2 pts · 2h (1 reply)

Thanks fam. Can’t wait to use this for dramatic and comedic effect during pair programming 🤣

Pudutr0n@lemmy.world · 1 pts · 2h

Go nuts! haha. All yours.

prole@lemmy.blahaj.zone · 7 pts · 12h (1 reply)

Cut yourself some slack. These days, most people would have just pasted LLM outputs

Pudutr0n@lemmy.world · 3 pts · 11h

lazy

KeenFlame@feddit.nu · 6 pts · 13h (3 replies)

The only way to do it is to do it - me, 2026

Nah, but seriously, that insecurity is a disease that will actually make you produce embarrassing code. As soon as you feel confidence feedback take off, you will paint the functions brilliant once you de-neuter your inner codebeast. It wants to fuck. Let it.

KeenFlame@feddit.nu · 4 pts · 13h (1 reply)

I fucked that sentence up, but you understand what I mean! ∆<3

Pudutr0n@lemmy.world · 1 pts · 6h

I did and I appreciate it. here's the code if you're interested.

Pudutr0n@lemmy.world · 2 pts · 10h

Thank you for your inspiration <3

glizzyguzzler@piefed.blahaj.zone · 2 pts · 9h

Haha there is no shame in shell code, it’s already monstrous to start!

I’ve written things in shell that would look sooo nice and concise in any other language, but it became a Cthulu-ass demigorgon so quick in shell! Have no fear, fear is the mind killer, the little death!

Pudutr0n@lemmy.world · 5 pts · 6h (6 replies)

psst... here ya go

glizzyguzzler@piefed.blahaj.zone · 2 pts · 3h (5 replies)

When I opened the code I immediately scrolled to the bottom to see how many lines, and hell yeah. I read through it and I'm horrified, but I'm even more horrified it works. This is truly an chievement

Here it is running over SSH https://files.catbox.moe/qkwd4q.mp4 what more could I ask for, thank you for making this insane thing and sharing it!

Pudutr0n@lemmy.world · 2 pts · 3h (4 replies)

holy shit, I'm so flattered! Thank you for your kind support and encouragement!

It's crazy that it looks okaish right? quaternions, dude. I have implemented them in 3 projects and still have no fucking idea how they work.

Also, you can run it in --quality low if you're struggling with framerate. :)

glizzyguzzler@piefed.blahaj.zone · 2 pts · 1h (3 replies)

It's insane how many features this has for being insane in of itself!

Edit: that was running on anemic "1 core" (1 very shared core) VPS, bumping it up to 4, hooboy do those dice fly

Pudutr0n@lemmy.world · 2 pts · 44m (2 replies)

nice! I'm still working on and have kept adding features. the encouragement was really motivating, so now I'm finishing multi-dice rolls that later add the result of all rolls.

If you want an updated version later on, would be glad to share w you. :)

glizzyguzzler@piefed.blahaj.zone · 2 pts · 36m (1 reply)

Sign me up!

Pudutr0n@lemmy.world · 1 pts · 14m

'ere ya go. hope you like.

AnarchistArtificer@lemmy.world · 28 pts · 10h (1 reply)

One of my favourite things in the entire world is what I call "high effort shitposts". Running DOOM on absurd devices like pregnancy tests is the archetypical example I use when explaining this.

Well, congrats, because this is absurd enough that this makes the list too. You are ridiculous and delightful, and I am glad that you exist to create impressive horrors like this.

Pudutr0n@lemmy.world · 11 pts · 10h

Thank you so much! I love high effort shitposts too!

And if you liked that, you're going to love this: A while ago I also wrote a 3d renderer from scratch inside the picotron, an emulated fantasy console that never existed, to display a 3d model of a pudu.

I have a few more of these you might appreciate, non 3d related.

farmgineer@nord.pub · 20 pts · 19h (1 reply)

Are you often described as a masochist?

It looks neat. I can't imagine wanting to write it in bash, though, heh.

Pudutr0n@lemmy.world · 16 pts · 19h

XD I am! And it was a complete nightmare. hahaha

MonkderVierte@lemmy.zip · 16 pts · 18h (2 replies)

(hopefully) Constructive critique:

  • it looks like you're handling the dice in physics as a ball? Takes forever to settle at the end.
  • the throw has a visual glitch.
Pudutr0n@lemmy.world · 15 pts · 18h (1 reply)

Thanks! Yeah, there's definitely A LOT of room for improvement. collisions are handled per vertex, but torque is applied to angular impulse to the dice as a whole (like a ball I guess). The settling is a huge issue, yes. the visual glitches I've been trying to get rid of but man... fix one and then another one pops up in a different case. Thanks for your critique. :)

Atherel@lemmy.dbzer0.com · 6 pts · 16h

I first read it as "there's a lot of DOOM for improvement" but as the rest of your comment confirmed this, I'll keep it that way in my head!

RheumatoidArthritis@mander.xyz · 16 pts · 21h (11 replies)

Haha nice! How long did it take you? Integer-only math?

SpaceNoodle@lemmy.world · 23 pts · 21h (6 replies)

All computer math is integer math if you go deep enough

RheumatoidArthritis@mander.xyz · 9 pts · 21h (5 replies)

I'm sure implementing floating point directly in bash would work great

RheumatoidArthritis@mander.xyz · 17 pts · 20h (2 replies)

I didn't have the patience to do it myself bit wanted to see just how complex it would get:

fp32_mul() {
    local a=$1 b=$2
    local sa=$(( (a >> 31) & 1 ))
    local sb=$(( (b >> 31) & 1 ))
    local sign=$((sa ^ sb))

    local ea=$(( (a >> 23) & 0xff ))
    local eb=$(( (b >> 23) & 0xff ))
    local fa=$(( a & 0x7fffff ))
    local fb=$(( b & 0x7fffff ))

    # NaN / infinity / zero handling
    if (( ea == 255 )); then
        if (( fa != 0 )); then
            printf '%08x\n' $((0x7fc00000))
            return
        fi
        if (( eb == 0 && fb == 0 )); then
            printf '%08x\n' $((0x7fc00000))   # inf * 0 = NaN
            return
        fi
        printf '%08x\n' $(((sign << 31) | 0x7f800000))
        return
    fi

    if (( eb == 255 )); then
        if (( fb != 0 )); then
            printf '%08x\n' $((0x7fc00000))
            return
        fi
        if (( ea == 0 && fa == 0 )); then
            printf '%08x\n' $((0x7fc00000))
            return
        fi
        printf '%08x\n' $(((sign << 31) | 0x7f800000))
        return
    fi

    if (( ea == 0 && fa == 0 || eb == 0 && fb == 0 )); then
        printf '%08x\n' $((sign << 31))
        return
    fi

    # Convert subnormals to a normalized significand/exponent.
    # m is a 24-bit significand for normals.
    local ma mb
    if (( ea == 0 )); then
        ma=$fa
        ea=1
        while (( (ma & 0x800000) == 0 )); do
            ma=$((ma << 1))
            ((ea--))
        done
    else
        ma=$((fa | 0x800000))
    fi

    if (( eb == 0 )); then
        mb=$fb
        eb=1
        while (( (mb & 0x800000) == 0 )); do
            mb=$((mb << 1))
            ((eb--))
        done
    else
        mb=$((fb | 0x800000))
    fi

    # Multiply the two 24-bit significands.
    # Product is up to 48 bits.
    local p=$((ma * mb))
    local e=$((ea + eb - 127))

    # Normalize product.
    #
    # ma*mb has binary point after bit 46.  If bit 47 is set,
    # product is [2,4), otherwise [1,2).
    local shift
    if (( p & 0x800000000000 )); then
        shift=24
        ((e++))
    else
        shift=23
    fi

    # Extract 23 fraction bits plus guard/round/sticky information.
    local frac=$(( (p >> shift) & 0x7fffff ))
    local guard=$(( (p >> (shift - 1)) & 1 ))
    local round=$(( (p >> (shift - 2)) & 1 ))
    local sticky=0

    if (( shift >= 3 )); then
        local mask=$(( (1 << (shift - 2)) - 1 ))
        (( (p & mask) != 0 )) && sticky=1
    fi

    # Round-to-nearest, ties-to-even.
    if (( guard && (round || sticky || (frac & 1)) )); then
        ((frac++))
        if (( frac == 0x800000 )); then
            frac=0
            ((e++))
        fi
    fi

    # Overflow -> infinity.
    if (( e >= 255 )); then
        printf '%08x\n' $(((sign << 31) | 0x7f800000))
        return
    fi

    # Normal result.
    if (( e > 0 )); then
        printf '%08x\n' $(((sign << 31) | (e << 23) | frac))
        return
    fi

    # Underflow into the subnormal range.
    #
    # At this point the normalized significand represented by
    # (1.frac) must be shifted right by 1-e positions.
    local mant=$((0x800000 | frac))
    local rshift=$((1 - e))
    local lost=0
    local halfway=0
    local low=0

    if (( rshift >= 25 )); then
        # Everything rounds to zero (unless the exact value is
        # sufficiently close, which it cannot be here).
        mant=0
    else
        low=$((mant & ((1 << rshift) - 1)))
        mant=$((mant >> rshift))

        halfway=$((1 << (rshift - 1)))

        if (( low > halfway || (low == halfway && (mant & 1)) )); then
            ((mant++))
        fi
    fi

    # Rounding a subnormal can produce the smallest normal.
    if (( mant >= 0x800000 )); then
        printf '%08x\n' $(((sign << 31) | (1 << 23)))
    else
        printf '%08x\n' $(((sign << 31) | mant))
    fi
}
RheumatoidArthritis@mander.xyz · 15 pts · 20h (1 reply)

1440 multiplications per second on my computer!

Womble@piefed.world · 18 pts · 20h

Wow, over a kiloflop!

Pudutr0n@lemmy.world · 7 pts · 19h

hahaha you seem to be familiar with the issues I ran across

tabular@lemmy.world · 4 pts · 16h

Invert the gravity value and then all the dice will be floating point 🫣

Pudutr0n@lemmy.world · 14 pts · 19h (3 replies)

Jebus I don't even know how long I've been working on this. It started out as a way to teach myself bash, combined with my long obsession with rendering platonic solids. I previously coded a huge galaxy of hundreds of thousands of polyhedra you could fly across in python with opengl shaders and got to reuse/translate a lot of that code. The quaternion stuff I was grateful to not have to rethink that much again. haha. And yes, integer-only! Some params are "floating point" but i just parse them and add a bunch of zeroes so i can later do all the operations with equally blown up values of pi and trig stuff from "lookup tables" (case matching).

MonkderVierte@lemmy.zip · 6 pts · 18h (1 reply)

Btw, you could pipe it to bc for the odd float. Or even some complex math. Just in case you didn't know.

Pudutr0n@lemmy.world · 6 pts · 18h

yeah, I was trying to stay as pure-bash as possible, but thanks!

RheumatoidArthritis@mander.xyz · 4 pts · 18h

Strong Amiga coding vibes.

thingsiplay@lemmy.ml · 14 pts · 15h (3 replies)

I just woke up and this is the first thing I see. Did I really woke up?

dejected_warp_core@lemmy.world · 7 pts · 13h (2 replies)

Hey baby, wake up from your asleep.

https://www.youtube.com/watch?v=TQhyDYQ081U

Crackhappy@lemmy.world · 5 pts · 11h

Oh my God that's amazing.

thingsiplay@lemmy.ml · 5 pts · 12h

A banger to wake up to, as alarm clock song. :D

Aqivex@fedinsfw.app · 13 pts · 11h (1 reply)

@Pudutr0n@lemmy.world - You are certifiably insane. There are a million more effective ways to achieve this, with less resources... I LOVE IT.

Pudutr0n@lemmy.world · 8 pts · 11h

This is exactly the kind of flattery that truly touches my heart. <3

adhocfungus@midwest.social · 12 pts · 15h

Absolutely insane. Part of me wants to see the source code, but part of me is terrified of how complex it must be. Plus I've only been writing bash for 20 years, so it would probably look incomprehensible to me.

alphabethunter@lemmy.world · 12 pts · 19h (1 reply)

This is beautiful, and insane! But truly beautiful hahaha

Pudutr0n@lemmy.world · 6 pts · 19h

Thank you! haha

spacegoat@lemmy.world · 10 pts · 5h (1 reply)

Do you think God stays in heaven because he too lives in fear of what he's created?

Pudutr0n@lemmy.world · 8 pts · 4h

God comes over
creates some bullshit
stays in heaven
refuses to elaborate

mike_wooskey@lemmy.thewooskeys.com · 8 pts · 17h (17 replies)

I'm viewing this thread in Lemmy and am unable to find any mention of a link to the source code or a demo, and I can't see the attached media (genetic image icon - is it an animated gif or a video?).

Based on all the comments, I appear to be in the [significant] minority. Would you mind posting a link to your source or demo or image/video?

Pudutr0n@lemmy.world · 18 pts · 17h (15 replies)

I haven't shared the source code yet because criticism is terrifying and I'm a horrible coder, but with the encouragement of the the community I might give it a shot. I don't even have a codeberg account yet though.

The post contains these two video links: https://files.catbox.moe/8rz9m7.mp4 https://files.catbox.moe/5d3n2k.mp4

If you still can't see them, maybe you can suggest a different hosting service, one that works for you? Or if you really really want the source code I suppose I could send it by DM on the meantime.

oce@jlai.lu · 10 pts · 17h

There are millions of professional coders who couldn't do what you did there.

vrek@programming.dev · 6 pts · 16h (5 replies)

Just letting you know the original post video didn't load and gave a 500 error in jerboa but those two links work. Looks impressive but I'm confused on how to read the die. Yes there is a giant banner telling me the value it chose but I'm struggling to figure out how the visuals map to those values, can you give some details on how to read the die?

Pudutr0n@lemmy.world · 6 pts · 16h (4 replies)

Yes, the values are rendered on top of each face with "pixel" maps (ascii art) that have the character of the digit they represent as each pixel. These are later manipulated by the perspective projection.

For example on this d6 you can see the top face has 1s written down (in the shape of a sideways one) and the front one has 4s written down in the shape of a skewed and sideways 4.

In this specific color scheme there's not a lot of contrast between the values, but even in the palettes with better contrast, good readability is something I'm kind of giving up on solving well.

edit: might be clearer here:

vrek@programming.dev · 2 pts · 7h (3 replies)

In the second picture (with the d20) I'm guessing the face with "999" is actually a 9 but what is the value of the face with "999#111"...there is no value of 91 on a d20 :)

Pudutr0n@lemmy.world · 1 pts · 6h (2 replies)

actually I was working on the code and realized i got the textures mirrored due to negative-positive mix up. so it was reflecting the mirror image of 19

vrek@programming.dev · 2 pts · 6h (1 reply)

Ah that makes sense. Overall looks impressive, was just confused on my end.

Pudutr0n@lemmy.world · 1 pts · 6h

thanks! :D

mike_wooskey@lemmy.thewooskeys.com · 3 pts · 11h (2 replies)

That's pretty darn cool! Thanks for sharing. Ik forward to cloning your repo someday.

Pudutr0n@lemmy.world · 2 pts · 6h

here ya go

Pudutr0n@lemmy.world · 1 pts · 11h

I'll do a follow up post when I upload

SystemDisc@feddit.org · 2 pts · 12h (4 replies)

Pudutr0n@lemmy.world · 1 pts · 11h (3 replies)

what video hosting service works for you?

SystemDisc@feddit.org · 2 pts · 42m (2 replies)

I have no idea. I self-host a lot

Pudutr0n@lemmy.world · 1 pts · 40m (1 reply)

well i don't know how to deliver the media to you then D: sry. you can snag the code from pastebin though. I linked on my last edit.

SystemDisc@feddit.org · 2 pts · 26m

No worries. That service is just refusing the connection or 500ing for people.

Pudutr0n@lemmy.world · 3 pts · 6h

here's the source code. :)

Gsus4@mander.xyz · 8 pts · 18h (1 reply)

3D dice rolling...ok...😐 ... in bash ...🤨 !?

PS: I remember seeing something vaguely like this recently...this you?

Pudutr0n@lemmy.world · 2 pts · 6h

no, but I LOVE that guy

cristian64@reddthat.com · 7 pts · 17h (1 reply)

I didn't know I needed this.

Pudutr0n@lemmy.world · 2 pts · 6h

and now you have it

albbi@piefed.ca · 6 pts · 18h (1 reply)

Looks like a great sl replacement! Gonna need this on every machine.

Pudutr0n@lemmy.world · 2 pts · 6h

if you want it on yours, here it is. Thanks for your support. :)

bridgeenjoyer@sh.itjust.works · 6 pts · 14h

Holy crap how cool

HeHoXa@lemmy.zip · 2 pts · 59m (1 reply)

I think I might love you

Pudutr0n@lemmy.world · 2 pts · 41m

ily2 bb <3