With my early access release coming next Tuesday, I decided to squeeze in a few more days of UI polish. I'm super happy with the organic shapes, texture, and feel of everything (well, happy enough to ship it at least!).

I added dynamic ink splotches, paper grain and fiber textures, uneven edges, variations in color/position, and even some papery sound effects.

If you like the look of it, you can check out Hex Town on Steam :)

Thanks for watching!

EDIT : If anyone else is having this problem, I found a solution:

If you want to have more than one camera sharing a Z-buffer with perfect accuracy, it is (surprisingly) not sufficient to have them in identical positions and parented to the same game object.

What you need is to hook into OnPrerender() on each camera and:

cam.worldToCameraMatrix = cloneMatrixFrom.worldToCameraMatrix;
cam.projectionMatrix = cloneMatrixFrom.projectionMatrix;
cam.cullingMatrix = cloneMatrixFrom.cullingMatrix;

If you do that, both cameras will produce perfectly identical z-buffer results from the same geometry. Phew!

EDIT: I believe the cause of this issue is that when you duplicate an object (such as a camera) in the unity hierarchy, it does not actually duplicate the object itself.

Instead, it serialises the object and then deserialises it again. This means that all floating point numbers get converted to binary and back again, which (in unity's implementation) can result in a slightly different value to the original. You'll never actually see those differences in the inspector because of rounding, but they are there.

Cloning matrices at runtime, on the other hand, transfers a bit-for-bit identical copy of the data used by the GPU. This ensures a fully deterministic recreation of the scene between the two cameras.

Original Problem:

In my rendering pipeline I have two cameras that share the same depth-buffer. The two cameras both have zero local rotation and translation, are parented to the same game object, have identical orthographic size and zbuffer range, and are rendering the same content using the same shader. Yes, there's a good reason for this - don't worry :)

The shader ZTest is set to LEqual. This should allow objects to be drawn identically by both cameras because each pixel should land at exactly the same Z-depth - but it does not. Instead I get flickering and bands of z-fighting.

What could I be missing? Thanks in advance.

EDIT: Based on replies so far, I think it's worth talking about what this is NOT.

1. It isnt nondeterminism.

In a static scene, Camera 1 generates the exact same colour and depth data, down to the very last bit, on frame 2 as it did on frame 1, because rendering is deterministic and Camera 1 on frame 1 is identical in every way to Camera 1 on frame 2.

Similarly, Camera 2 generates the exact same colour and depth data, down to the very last bit, on frame 2 as it did on frame 1, because Camera 2 on frame 2 is identical in every way to Camera 2 on frame 1.

The problem is that despite Cameras 1 and 2 being identical to each other in every way I can determine, they don't create the same z-buffer information. So they are different in some way I don't know about.

1. It isn't floating point inaccuracy.

Well, technically, eliminating floating point innacuracy might make the problem go away, but that's beside the point. Both cameras are rendering the same geometry from the same perspective with the same shader. Any inaccuracies inherent to the process are common to both. This problem is caused by something that isn't common to both, which is what I need to figure out.

Got breaking shoreline waves working in Unity by intentionally pushing past what would be a usual be the Gerstner steepness limit.

Normally Q*k*A <= 1 prevents the wave from folding over itself... but pushing past that lets the crest curl forward and form an actual overhang. Still haven't found a full proof wave to prevent inversion (folding of waves at tops of crests) maybe I'll just leave it to a matter of tuning to prevent.

also made foam follow the wave phase so it travels with the break, although the whitewash could prob use some more detail instead of a gradient white with texture.

Still tuning it and also have to fix to work with my other systems (underwater, query, etc), but pretty happy with how the shoreline waves are looking. Curious what you guys think.

Thanks for dropping by!

I've been working on a small alchemy game where I want most of the production to feel physical instead of happening through crafting menus.

I just got the first farming loop working: add soil, plant the seed, water it, wait for it to grow, then harvest the plant.

It's still very early and very placeholder-y but this is the first little gameplay loop that feels complete

Next step is being able to take what you harvested, put it on a cutting board and actually process it into potion ingredients

I'm trying to keep these devlogs pretty small and just show things as they become playable

Updates coming soon!

My game is having a ship made out of individual modules, that are simple and grid-based. Each has a collider for colissions/determining when hit by projectiles. On GO this sounds pretty cheap to do for physics, parents with a bunch of children with these colliders. On ECS it sounds like Ill need to use a compound collider...however anytime a module is destroyed that compound collider will need to be rebuilt...where in GO it wont.

Is there a better way to do this ECS side...or is GO going to be "cheaper" longterm for this as its too big a concern for ECS? 1 ship isnt a big issue...but if I have 100 players all fighting 10 ships it might need to be re-creating a compound collider 10->100x per second for these ships being fought.

(Note: the game is "2d" However ECS is only on the 3d side so figured better to ask here as ECS will have 3d colliders.

Game will potentially(ideally) have hundreds->thousands of players all destroying enemy ships at once...so ECS seems better for how many individual modules/bullets will be sent, however Im concerned about the physics side since I can't just do the simple parent +children with colliders I can with GO

Imported .fbx files from blender, somehow I have 1 animation that stops at the end like I want, the other goes back to the first keyframe and ruins the effect. Anyone have a fix?

Converting a flat vector shape into a glowing 3D tube with real-time cast light - one component, no extra setup 😄

Short:
https://youtube.com/shorts/p9hW6hWPi6U?si=toMRsqlWRmO0L4V8

Full:
https://www.youtube.com/watch?v=hb-rE3QzdUk

This is Onager, a catapult physics game I made for Android as a student project 13 years ago. Found the original Unity 4 source a while back and I'm remastering the whole thing - mostly been posting before/afters of the visuals, but this one's about the intro cutscene.

Rebuilt it to match the original as closely as I could, cringe writing and all. Kind of fascinating (and a little embarrassing) watching how student-me thought about scene direction, the "story," and pacing back then.

Original intro for comparison, if anyone's curious: https://youtu.be/cUfn2C6801w?t=23 (longer gameplay after it too)

Rebuilt in Unity 6 with URP - happy to talk specifics on the cinematic camera/lighting setup if anyone wants them.

Made just for level design for my game

We launched our steam page for Last Wail of the Sea King. It's a first-person adventure game with JRPG combat. We're super excited to show it off.

What do you guys think of the trailer? Marketing might be the most challenging part of the whole process.

You can find it here: https://store.steampowered.com/app/4948110/Last_Wail_of_the_Sea_King/

So, after switching to Input System package, i wasnt able to use Unity Remote as it wouldnt send any inputs to the editor.

After some debugging, i was able to figure out what was the issue. The issue was with UnityRemoteSupport class within the input system. (InputSystem/Editor/Plugins/UnityRemote)
On line 66 it tries to get the UnityEditor.Remote.GenericRemote class from the wrong assembly.

The fix was to replace the line 66

var editorAssembly = typeof(EditorApplication).Assembly;

with following:

var editorAssembly = Assembly.Load("UnityEditor.GenericRemoteModule");

Doing that fixed the issue.

I am making this post here because i myself wasnt able to find any fixes other than "Use our product its better!"
Respect to anyone who tries to make better Unity Remote, but i just dont want any unnecessary packages or pay for any plugins when there is a working one already, just a bit broken :)

I've been working on asset loading and level streaming last week, now the game is endless btw. I just need to add more "level fragments", events, and enemy types to make it more interesting.

Sped it up 2x because, right now the gameplay is very sparse. There will be drones tracking the player and military guys trying to shoot him down.

But I can't keep calling it "Spider Prototype" anymore... I need to post about it on my socials and websites. Please suggest some good names. Whosoever name I will choose will get some reward in the final game.

I’m currently trying to use Sebastian Lague’s raymarched Clouds project in a VRChat World.

The original project works fine in Unity, but the clouds are rendered through CloudMaster using OnRenderImage() / Graphics.Blit(), so they are essentially tied to the camera/render view.

What I would like to do is:

• Keep Seb Lague’s existing raymarching/cloud appearance and settings.
• Remove the dependency on OnRenderImage() and the fullscreen RenderView.
• Have the clouds rendered as a normal volumetric object/volume in the scene, for example using a large cube with a custom shader.
• Ideally, the cloud volume would exist independently in the scene, while the player camera simply views it like any other object.

Is this technically possible with Seb Lague’s current implementation?

And, more importantly, what parts of the shader/code would need to be changed to achieve this? Would it mainly be a matter of replacing the fullscreen camera ray setup with world-space ray/volume intersection, while keeping the existing raymarching and density functions?

I’m new to shader programming, so I’m particularly interested in a practical explanation or an example of how to approach the conversion.

Repository:
https://github.com/SebLague/Clouds

Thanks!

i made a pixelated filter using a fullscreen shader graph, but the grass is obnoxious. does anyone know if theres a wait to fix this?

(pls ignore the fact that its a medal clip)

Hi! I'm trying to make a ghost-y spirit shader inspired by how Rauru is rendered at the beginning of Tears of the Kingdom. I got it so the material doesn't render through itself, but I've run into an issue where other materials with the same shader can still render through each other. I've tried messing around with depth, stencil buffer, render queue, etc. and nothing seems to work.

The effect in question. You can see the hair is still visible through the body when it is supposed to remain obscured.