#💥┃post-processing
1 messages · Page 9 of 1
You could do this with a fullscreen UI image.
Or a dedicated UI shader if you want to dynamically change the shape
Or even a custom post process
I see, alright, thank you!
If you place it like this, with a fixed horizon height, woudn't it be enough ?
okay i'm not totally sure i understand anymore how it would help me render only what's on that slice?
i'm really sorry i never really went into anything that is rendering related
My solution was just to fill the gaps.
Looking at your screenshot I'm even trying to figure out how you did render the cubes in the first place.
Would help to see the camera projection when it is selected in the scene view.
I don't know eighter how your world is made. Is it an optimized generated chuck mesh, or just a bunch of brute forced cubes ?
If you really want a clear intersection, with proper holes filing, you gonna do it yourself, by generating the intersecting meshes.
Or, now that I'm saying it, it could be done with a dedicated shader, that flatens out the cube faces to match the intersection ... if the near/far planes are very close, it could work. Still tricky to do.
well i'm not the one that made this like i said it's a project made by someone about minecraft in 4D he just used this as an exemple on how 4D could work https://youtu.be/u8LMyWcKL_c?t=96
Play the Demo: https://store.steampowered.com/app/1941640/4D_Miner/
Official 4D Miner Discord: https://discord.gg/pMtZa6AVy2
Official Subreddit: https://reddit.com/r/4D_Miner
Official Wiki: https://4d-miner.fandom.com
Twitter: https://twitter.com/Mashpoe
Patreon: https://www.patreon.com/Mashpoe
The Kickstarter campaign has ended and unfortun...
i put the video with the time code showing the exemple
i know it's possible to do i just don't know were to start
From what I see, simply having a big enough far plane to cover the diagonal of a cube (the shortest required distance to not have holes in the section), and rendering the backfaces (inner faces of the cubes) seems enough.
Ah, hum, no, in some cases it would render "to much" geometry. By combining with a stencil mask, to not display back faces if front faces are visible it should do the trick.
The problem is the wall behind get rendered, i only want a slice of the world to be rendered
That's because the far plane is to big
i know but if it's too small it won't fill the clipping part
i'm using the universal render pipeline in case that's usefull
That's why I asked about the method. If your world was made of a lot of tiny cubes of the same size, it would work.
The other option is stencil masking like I mentioned.
Each object needs to be rendered twice :
- front faces first in the stencil buffer to store a mask
- then back faces, but only where the stencil was not written
But it's something not currently achievable out of the box with URP or without using custom hand writen shaders.
oh okay
well i'm gonna try to research more about stencil masking
thank you
Wait i have an idea on how i could do that
i mean without the stencil masking
I could make a really thin box collider that is trigger that mimic what should be a 2D world and that only allow to render what's in the trigger collider
i know it's really sketchy but it could work
i'm gonna try both and i will see
For some reason the Overlay camera isnt showing up in the stack list
Nevermind, think I figured it out
I've been trying to bake my scene since yesterday with 0 success...
Whenever i Attempt a bake, it get's stuck on "Baking... 0%".
It uses no GPU or CPU.
And yes, the objects i'm trying to bake are marked as Static and Contribute to GI.
I've tried to clear GI Cache, Clear Baking data, restart unity, restart my PC, but nothing i do fixes this...
I'm running an Intel i5 13600K, 32GB RAM, RTX 3060TI, and i'm on an SSD.
Do you have a directional light source?
Or any light sources set as baked or mixed?
Also, this is really unrelated to post processing. Should be in #1390346776804069396
How can I make the sword stop glowing so much from the candle? Is there a way to do this without using grey color?
Grey color for what, the light or the sword or the bloom?
Increasing Threshold, lowering Clamp and Intensity may help
But it it depends on how much bloom you need
I also would probably not use an orange tint for the bloom but a neutral white and lower the Intensity in turn instead
As with a tinted bloom lights that match the tint and material colors that respond to it will get way more bloom than other color bloom sources
If it's a 2D light you also have control over the light blend modes which affect how materials of various colors respond to it
okey, thanks 👍
I don't know where to ask, so I just ask here. I got this elevator which exists across multiple scenes. I want the outside to be affected by the Adaptive Probe Volume, But the inside should stay consistent over multiple scenes, but should still be affected if the sun is shining through the door or a red light appears. How can i do that? (It doesn't move and is not ticked as static)
#1390346776804069396
You can bake it as contributing to GI so it casts a shadow and isolates the interior from baked outdoor lights, but use it as dynamic at runtime
Still for the sun and other lights to only show through when the door moves they need to have realtime shadowcasting
Thanks I'll try that. The project works without dynamic lights, currently I found a working solution, a baked light in the elevator and then the same mesh with fliped normals as an object which can get affect by light probs. It works but its not the best way I guess
How do flipped normals help here?
I just duplicated the Faces in blender and then fliped the normals. Then I dublicated my mesh in unity and called them out-/inside elevator
do you get it?
Right, I guess that's necessary if your walls have no thickness at all
Having realistic thickness is really useful for allowing the inside and outside to more easily be affected by separate APV probes
Yeah, I should change that, don't know why I just flipped the faced, could have make them thicker
heres my night scene rn. I dont want to just add spot lights to each window to give it some volume. Whats a good alternative thats cheaper on the hardware?
Some meshes with emissive materials to fake the inner lights glowing outside ?
not post processing i dont think but how do i fix this? is a png from the pixil website thing for drawing
Seems to me like the texture is just not seamless.
If you know it's not post processing, why ask here??
#🔀┃art-asset-workflow is more appropriate.
Hi, so I have this small lake in my world and I want it to have this cute volumetric fog haze over this lake, but I dont really understand how to do it
If it's hdrp and you want it to be really volumetric, you can use the local volumetric fog component.
There are a few sample 3d textures if you import volumetric samples from the package manager
If you don't need it truly volumetric, you can just use the particle system or vfx graph and stack some smoke flipbooks (lit if you want)
UDR
yeap urp , misspell
but I got over that issue, I am occurying another problem at this time, based on wind and plants
Can I create a wind effect over grass drawn with a brush? @cedar forum
By drawn with brush, do you mean it's just a texture cutout, rather than a 3d mesh?
either way, yes
it is a 3d mesh
ok so drawn with brush has no relevance here? Either way, yes, you typically just manipulate the vertices by adding displacement such that it displaces more at the top, less in the middle and no displacement at the bottom
there are yt tutorials for that, go search it up if you're not sure how to do it
I tried to follow 1 from youtube, i did step by step and nothing changed
thats the thing XD
yeah
well I suggest you make thread in the rendering channel (that's the right one for shaders) and post your shader graph setup (assuming it's in shader graph), post your mesh with material setup pics
maybe it's as simple as you not applying the shader or material, cause you'd probably see SOMETHING happen otherwise, even if you did it wrong
maybe you can share a premade shader? maybe I did mine wrong?
Are there any universal tips on making a fullscreen shader more efficient? I just implemented one and it's dropping my fps by like 100
By 100 from what number? If it's 600->500, that a difference of just .33ms which is totally acceptable for a full screen shader.
It’s 400 > 300 on 1440p
I’m pretty new to full screen shaders so that just seemed like a lot
That's 0.8ms. Still. Could be acceptable depending on the effect.
Tip: You should stop measuring performance in fps. Compare the ms difference.
Beyond that, it's impossible to answer your question without more context.
I’m using this but slightly edited to work as a fullscreen shader if that helps at all: https://ameye.dev/notes/halftone-shader/
Seems pretty complex, so such numbers are not surprising IMHO. Anyways, to optimize it further, you'll need to understand the bottlenecks. For shaders that implies using tools like PIX to analyze what instructions are taking most of the time.
One common way to optimize post processing is to render them at lower res and then upscale to the full res, but the overhead needs to be worth it.
And there is a potential for quality loss.
I’ll definitely check out PIX and I think rendering this at lower res is pretty realistic considering it’s meant to be low quality anyways
Thanks for the tips ❤️
this laptop buttons not working please i need help
make sure to post in thr proper places
this channel is for post processing
@toxic jackal Ok I am sorry
I wanna ask, is there a way to configure post-process volume component in animation? I want to gradually increase bloom for cutscene purposes, or do I have to use script with key event
you can use animation events for that
on your animation timeline add events then do whatever you want from code
oh okay then, animation event it is
thanks
I think I put my thread in the wrong place but I'd appreciate it if anyone could take a look: https://discord.com/channels/489222168727519232/1464743512149983324
Hi all. Why am i getting this invalid error? Is this harmless?
On unity 6.3 URP
This package doesn't work with URP, you can remove it. But the error itself likely related to the current bug in Package Manager, it should be patcked up soon.
But i want post processing tho
URP has it built in
Or are u saying the post processing for URP is not this one?
So it 100% won't work for URP? Or still works?
I'll considering using this one if it's more lightweight for example. I'm targetting webgl
This package doesn't work with URP
Ok reading that thx
Note: URP does not support Post-processing on OpenGL ES 2.0.
Is webgl opengl 2.0? Or 3.0?
Does this mean it can work for webgl opengl 3.0?
Presumably that would've been mentioned if it didn't? You can look it up.
Hi Folks, I'm a mere coder, I'm looking to make a showcase for my race timing Asset that I'm building. I would like to make something that is visually appealing so that people can see the possibilities without getting distracted by programmer art :). Can anyone recommend some tutorials or a course to help me make the scene pretty - performance isn't really too much of a worry - I just need to record a few demos with some decent hardware. What would be the first thing you would do?
Here is how it looks atm
Hi. I'm looking to get the bokeh to have a crisp shape (the complete opposite of rounded), without necessarily making the far stuff so blurry or the depth of field so narrow. Is there a way to achieve this? Maybe faking it to force the bokeh have crisp shape?
There's a bokeh spot on the right side of the Journal button in the screenshot
Hey guys, can i ask why is my viginette only appearing on my scene view not my game view
because it's not setup for the game camera?
the camera that is rendering must have post processing on
and if you later don't want the effects coming in the scene view you can disable them using the effects button in the scene toolbar and disabling post processing
Make sure you are using correct Post Processing for the render pipeline you are using.
Is that "game render rectangle" a UI image ???
Hi. I'm now making a new DOF type for URP, just to get a stylized bokeh
Never tried customizing postprocessing stuff before. Any tips, guidelines, samples, etc?
Unity 6.3
Lol that's a screenshot from pubg so maybe that's intended
There's not much to it other than setting the focal distance that you want
There are no universal settings, it depends on your camera distance to the world and what kind of dof you want to achieve
Even for mockup, what I'm trying to say here is that if it is a UI image on a screen space overlay canvas, it's expected that it is rendered OVER the post process vignette.
Ah yes, you're right
No i'm not talking about using the existing DOF post processing. I'm talking about makingi a custom post process
Anyways the default DOF is designed to be photorealistic. Not what i need
I'm just reading things and will eventually figure things out. But maybe there's an interesting article or something about URP, and render graph etc
https://www.youtube.com/watch?v=oLyv3NSpPeg
This is such a good starting point
If you want to support my Patreon: https://patreon.com/user?u=92850367
Ben here, and I've been writing shaders and full systems of shaders in Unity from scratch for 6-7 years now. Shaders, especially HLSL or CG shaders seem to be less known in Unity, so I just wanted to do my part, share my experiences and really show others how to write shader...
Oh geez even with the tut and AI, everyone's struggling with URP API changes...
Hi guys i need help with this custom post processing with rendergraph..
https://docs.unity3d.com/6000.3/Documentation/Manual/urp/render-graph-write-render-pass.html
I'm using this code and there are still errors
And pretty sure have read properly
Blitter.BlitTexture(context.cmd, data.copySourceTexture,
new Vector4(1, 1, 0, 0), 0, false);
Apparently on this line, there are 3 error spams:
Assertion failed
Render Graph Execution error
ArgumentNullException: Value cannot be null.
Parameter name: value
Oh ok i changed this
renderPassEvent = RenderPassEvent.BeforeRenderingPostProcessing;
//renderPassEvent = RenderPassEvent.AfterRendering;
Now there's no error. But i hope this is correct
@wispy tapir here
what i need to read?
so look in your hierarchy on the left, there is Global volume
yes
I've linked you the manual already, but you can also find it pinned here (pin button is top right). Select your pipeline there.
not really
Not sure what you are asking, but Pst Processing is a camera space effect, so if you want to have objects not affected or affected with different effects you need two separate cameras not scenes.
so are you able to see the effects now
Make sure to follow the manual. You need to set the space affected (can set global for simplicity) and other things
enable on camera component which might not be by default as well
ok you only have a few ui elements, maybe it is not obvioius enough but the pp should be working fine
not working(
and i can't understand why
i am sure that i am doing something wrong
maybe from the start
but i can't understand what did i wrong
change the Canvas to screen space
in your hierrachy you see Canvas
yeas
then click on it and change the overlay to screen space
make sure to enable individual properties on effects as well
Set properties like that to test
Bloom works only when under threshold
open a new urp project and test it there
new unity project?
you messed up a few things previously. Yes new unity urp project.
and only use the built in global volume dont download any packages as you did before
I am actually seeing it as well. Or rather not seeing on 6.3.61f
It doesn't affect UI even in screen space
for some reason
hmm , could it be a bug
wait why reinstall unity.
I think i know why it doesnt work'
when unity was installing i removed that from installing
but it installed
anyway thank you guys
Make sure to set the camera in the canvas
UI needs to know what camera renders it to show post processing.
@wispy tapir ^^
If you are on the latest version you don't need to reinstall. Make sure you've read my messages how to make it work.
@wispy tapir ^^
Hi, im struggling to use post processing in URP. Could i have some help?
You could, yes.
How do you set post processing in urp
Whenever I try to add an override, the camera effects do not change
Check the pinned message for the link to the manual with step-by-step guide.
Thanks
i added this to my shader #pragma multi_compile _ _ADDITIONAL_LIGHT_SHADOWS
shadows are working, but when I toggle my light to soft or hard shadow type it doesn't seem to change for my shader, am I missing something to get that softer appearance? docs seem rather lacking in much information https://docs.unity3d.com/6000.5/Documentation/Manual/urp/use-built-in-shader-methods-shadows.html
These keywords control that #pragma multi_compile_fragment _ _SHADOWS_SOFT _SHADOWS_SOFT_LOW _SHADOWS_SOFT_MEDIUM _SHADOWS_SOFT_HIGH
Oh I need all of those keywords?
Assuming you want to support all shadow qualities, yes.
Or you could just force one of them by #define _SHADOWS_SOFT_MEDIUM for example.
Or just do the shadow sampling yourself instead of relying on Unity's builtin keywords etc. (But probably not recommended unless you have a good reason)
I have created a Global Volume and custom post processing file (1st photo). The overrides I've added, like Bloom, Tonemapping are working fine. But I also noticed in URP asset settings that there's already some sample post processing profile which also has Bloom and other similar stuff (photo 2). Question: What's the use of that sample profile in URP asset?
It's the default volume profile for that URP Asset
Iirc there's another global default to be specified in Project Settings > Graphics
Volume profiles work on a priority system
The render pipeline asset's profile overrides the global default, and the volume components in scene override both of those, and each other by priority
On a per-override basis
is there a way to apply pp to the whole stack in a camera stack?
When adding the PP effect to the last camera in the stack, the PP applies to everything that has been rendered before in the stack. It is very unlikely you want to add PP to multiple cameras (let alone all of them), since that would render the PP effects multiple times on top of each other (for example adding bloom to an image that has already bloom added to it by the earlier camera in the stack).
Hi. I'm learning render graph, trying to make custom DoF so i have more stylized control of the bokeh effect
Having no knowledge about PP and the technicals, i got some questions
Are DoF highlight hotspots (that's used as the bokeh spots) affected by bloom?
So if i make a custom DoF, it should be AfterPostProcessing, or Before?
If i in the future make a custom bloom (is it even likely? For stylized, would u end up having a custom bloom? What're some examples?), should this also be Before or After PP?
What's the relations between DoF and Bloom with each other, and with other PPs ?
I'm just seeing this OnAfterRendering and PostProcessPassRenderGraph.RenderPostProcessingRenderGraph and even if i can see the order of things internally, among others i dont wanna end up playing with the numbers of the params (across several different PPs) without any formal understanding of their relations
Hi just confirming, (since i can't find where it's mentioned) so shadergraph shaders are compatible for shaders for URP rendergraph postprocessing, right?
Is there still any use to use handwritten shaders for postprocessing purpose?
But i think i saw a limitation that shadergraph shader can't do something in the context of custom postprocessing, forgot what
But even then, can shadergraph be used to achieve anything without touching any written shader?
The shader
you use with the Blitter API must be a hand-coded shader. Shader Graph shaders aren’t compatible with the Blitter API.
But ya my question is still that. Can everything possibly be implemented in SG instead of written shaders?
Basically yes. If something is not available via nodes, you can make a custom function node.
And blitting is basically just copying a texture, so even if you need to write a shader for it, it should be pretty simple.
I'm just really not sure which approach/APIs to use, for my use case. Ok say i'm open to either SG or written shader
But what's the point of using Blitter API over SG shader + not using Blitter? I guess AddBlitPass is useable throughout?
I'm targeting Webgl, URP 17.3, unity 6.3, rendergraph
Like my question is like, what's the difference between AddCopyPass, AddBlitPass, AddRasterRenderPass?
I guess which one i use also depends on that i'm making webgl?
https://docs.unity3d.com/Manual//urp/render-graph-blit.html
I'm actually not sure if this page is the most up to date. The page looks kinda broken for me
Remove the extra /, https://docs.unity3d.com/Manual/urp/render-graph-blit.html
I believe AddCopyPass & AddBlitPass are using the Blitter API (Blitter.CopyTexture & Blitter.BlitTexture) behind the scenes, they are just a pass that's handled for you rather than needing to write out your own RasterRenderPass
Lol omg
All this time
AddCopy/Blit works with shadergraph shader? I see a sample that does that
So what's the limitation about really?
I'm taking notes on these 3 methods' description side by side hold on.. kinda lengthy
I'm pretty sure I've used "Fullscreen" graph types with Blitter functions and it worked fine, so this docs statement confuses me a bit. It might be referring to that some functions (like the BlitOctahedral/Cube ones wouldn't have graph support?) but IDK for sure if that's what they mean
The way I understand it, a copy pass is exact 1 to 1 copy of the texture or render target into another one. You don't have control over how it's done. It likely uses a graphics api or unity internal shader for that. Blit is a render pass set up for you automatically that takes one input texture and renders to one output texture. You provide the shader here, so you can decide how it's rendered. Raster render pass is the lower level way that provides you the most freedom, but also requires more manual setup. Blit is using it under the hood if I get it right.
Yea seems like it after reading some
What kinda freedom does RasterRender have over Blit? They both uses a shader to process it?
This question is kinda style related. But i'm doing 2.5D, environment uses pixel art texture. I just assume that i won't need MSAA whatsoever
Apparently AddBlitPass does something about MSAA handling?
If i want no MSAA...
Uhh i'm not sure what's best at this point but i think i'll use RasterRenderPass for now
Modifying the gpu state, setting render targets and binding resources, the ability to execute several draw calls too I'd assume.
To be honest, I'm not sure you should be messing with it in URP or HDRP. This is a method from the Core RP library and is largely for writing your own render pipeline I think. Might want to look at what the URP is providing you with instead.
Im wanting to make a custom post processing and that led me to render graph samples, which uses RasterRenderPass (in renderer feature and render pass)
Are these urp samples?
Yes. There are several samples and this one is titled rendergraph
This is the wrong channel, you can do #🏆┃daily-win for one-off showcases or #1180170818983051344 for ongoing projects
Oh okay man
thnaks!
I have a question you folks might be able to answer about video, but it's somewhat out of pocket.
I don't know anything about video stuff in particular.
If you wanted to track motion in a video by various factors, say bands of color + size, and track it as a single axis of data over time, how would you do that?
so for example a video of a surfer with a camera on a stick. The water tube surrounds them, and the visual goes from open sunny sky to being inside a water tube. How would I sample the appearance of that tube in terms of its position over time?
Here's a simpler example. Imagine a video game scene of a 1x1 square. the left side is x: 0, the right side is x: 1. As the character moves from x: 0 to x: 1, the character has a position value changing over time. You can infer an acceleration, moveSpeed, etc. from the visual over time by watching it. I want to essentially do this kind of "measuring" but with an algorithm.
I want to use this data normalized on one axis though, which is why I thought you VFX dudes might have an insight.
Hi, im basically trying to setup post processing on my 3D unity project but i cant find why it doesnt work
I added Post Processing in package manager, created an empty object with a Post-process volume component and a post-process layer component on my camera
i also added a Post-processing layer on both
and still i cant see any movement when i add an effect
If you are on any "modern" unity version, you are probably using URP or HDRP, that don't work with the old post processing package but ship with their own PP
For example in URP, here is the doc : https://docs.unity3d.com/6000.3/Documentation/Manual/urp/integration-with-post-processing.html
Probably dumb question but how can i find which im using (URP, HDRP) ?
Im currently in the last unity version
If you didn't pay attention to pick a specific project template when you create a project, you are using URP
I picked universal 3D template
Yep, Universal (Render Pipeline) 3D template.
So URP.
Else you ca be sure by checking the default render pipeline in the graphic settings : https://docs.unity3d.com/Manual/class-GraphicsSettings.html#SRLoop
Alright ty a lot for your help, i found a video tutorial on Post Processing in URP i guess it will work 😉
hi
Do you have a relevant question?
Hi, im trying to add a glow effect with URP on my project, but actually i only found how to make a glowing material in engine, but i wanted to know if its possible to make an existing material glow, like using the existing colors of the material to make them glow (actually i can only use a mono color material for a glow)
you can use emission color on most materials, but afaik it wont affect lighting if you want that in URP (only baked)
I have basically this kind of material that i want to glow, but because it is composed with two colors i am only able to make it glow in one color ?
Im using emission like this but i cant find any parameter no parameters that are added to the already existing material so as not to take over
Does anyone know how to fix this volumetric fog glitch? It keeps disappearing evertime my player enters this white box made of quads and planes and I have to turn my spot light crazy high to be able to light an object
Sorry if this is not in the rendering channel cuz I need this fixed asap for a game jam
Answered here : #1391720450752516147 message
why did this pixelation get rid of my material for eyes? They are planes parented by my mesh's head bone with a material on it
also whats with that weird artifact between the beak and head and also on the body? Its not visible during gameplay or at a distance
Whatever you're using for the pixelation likely uses the color texture, which is rendered after opaques but before transparents (which is so that transparent materials can utilize it)
The artifacting is SSAO, configurable in renderer asset
so how would I use materials with alpha?
You don't really use the scene color texture for that
I would rather suggest using a different method for pixelating the screen
thank you
oh i just realized that it worked when i changed render queue. Is this a good fix?
Not likely
Transparents are in their own render queue for a reason
what issues might it pose?
Performance issues and sorting artifacts
https://docs.unity3d.com/6000.3/Documentation/ScriptReference/Rendering.RenderQueue.html
Opaque render queue is better for performance, while transparent queue is required for correct semitransparent depth sorting
Scene color texture is limited to opaques for the reason that transparent materials don't show up on it and get rendered recursively
Full screen post processing effects shouldn't use it unless you want to exclude transparents for some purpose
And for pixelation specifically it may be better to decrease the source rendering resolution, like URP's render scale does
!collab @zinc oar. This server is English only.
:loudspeaker: Collaborating and Job Posting
We do not accept job or collab posts on Discord.
Please, use Discussions to promote yourself as job-seeking, advertise commercial job offers, or look for non-commercial projects to participate in:
• ** Collaboration & Jobs**
Hi all. What's this Post Processing checklist for? In my scene view my post processing still shows
To clarify, this is a custom PP that still shows in the scene view even when the PP checkmark is off
The default PP like bloom is properly disabled in scene view
Is the custom PP implemented through the volume framework?
Yes
That's the only guess I had
Why?
I mean the default bloom is also in volume, yet it can get disabled in scene view just fine
If it had not been, that could've been a reason
So what can i do then?
This custom PP i tried on Before and After PostProcessing, but either one still persists showing in the scene view
How to make my game look better
it looks really cool already, what look are you trying to get?
Kinda like this
@mighty gazelle
are you in URP?
the hell am i doing wrong lol?
Likely a mismatch between the render pipeline your project is configured to use and the render pipeline your shader is targetting
Hi all. I'm making custom DoF postprocessing. Unity 6.3, URP 17.3. Targeting webgl, Forward+ path
Has separate controls for the near and far blurs
The blur has 2 parts, the interior shape's blurring, and the exterior, expanding outwards blur
Yes, i'm using AI. It's really close, but can't seem to exactly get it right
I'm having issues about when the pixel being blurred (if in this case is the near object's blur) is next to pixels that comes from the sharp objects, or from the far objects.
On some version the exterior doesn't show if the nearby pixels come from the sharp
Then the exterior of the far objects don't show
Or the exterior of the far objects can blend/expands to the near object's interior
Anyways, my general question is, why is this so hard?
In my head this seems simple to do. Just draw the far objects first, with its interior/exterior blur as needed
Then draw the objects in the sharp focus distance
Then draw the near objects
Is this not possible? Is this where this forward/deferred gets into play? So does that mean i should change to Deferred? But then does it work for webgl?
Is there a way to make this work for Forward, or a fake version so it's cheap/fast for webgl but still works?
Attached is the HalfBlur shader. The whole renderer feature is 2 pass, the blur pass and the recombine pass (which now according to the AI is not needed and all the logic's are in the blur pass)
at some reason both cameras (UI and base camera) rendering postprocessing which doesn't exist at scene or prefab
I have this scene, i want the neon to have volumetric lighting, i watched hundreds of videos and i still don't understand why it does not work
is it just me or is the lighting all wack
Directional light is not shadowed by the ceiling
Realistically there wouldn't be a sun at all if it's an interior scene
Environment ambient light and reflections don't match the interior either, they are lit by the sky that's occluded as well
There are no visible light sources for illuminating the interior instead of the sun and the sky
But how you want to deal with that is up to your artistic vision, not just realism
#1390346776804069396 is for lighting questions
Volumetric fog is a camera effect
First get it working in your scene, then use the volumetric toggle on lights to exclude them from interacting with the fog
Then you can turn up the volumetric multiplier on your light and turn down the density of fog in the override to make just that light "volumetric"
why is it when i scale up my sprite it gets foggy? how do i fix it
why post that randomly in a channel that has nothing to do with it
use a higher resoloution, use something like 256x256
kk thanks, what channel is correct if i have more questions?
i´d say #🖼️┃2d-tools or #💻┃unity-talk
thanks
Set filtering to none/point, rather than increasing the resolution
Consider also using high/none compression with pixel art textures
with higher reso i ment when importing into unity
Yes
Using higher resolutions kinda works by "bruteforcing" the bilinear filtering
But that requires the texture to be scaled much, much bigger to hide the bilinear smoothing, which is pretty wasteful
It also makes the sprite incompatible with typical pixel graphics workflows
Kind of a crude alternative to shader based texture antialiasing for a style of smooth-edged pixel art
But it is very simple, although not as simple as simply switching to point filtering
Hello! I am making a game in Unity, but i just cant seem to figure out how to make one singular object glow. I have seen a lot of tutorials where i just crank some stuff up in bloom settings in a global volume, but that makes the intire project glow.
Any tips?
And i am new to Unity, so i dont realy know how to find my way into shadergraphs.
There is no simple way to do that because bloom is a full screen effect
The simplest is to enable bloom, but increase both bloom threshold and the material's emission intensity
That way only the particular material is likely to glow
Thank you! I tried it out and it worked! That probably saved me tonnes of time!

Hey guys! A question about the 2d urp renderer. Is it possible to have a depth of field post processing effect? I have changed the camera to be a perspective one and did offset my objects in z-space, but cant seem to make it work, did anyone had success with that ?
DoF is a post processing effect so it has to rely on data that has been stored while rendering, the only useful data for that is the depth texture. The issue with that is how transparent objects usually don't write to the depth buffer and sprites are usually regarded as transparent whether they have semi-trasnparent parts or not. I'm sure there are many ways to workaround that but it would require knowing more of what exactly you are trying to achieve visually to give any viable alternatives.
So the effect I want to a achieve is a depth of field where objects in far z are more blurred, on z pos 0.0 where the player is as sharp as it can be and the closer to the camera the blurrier it gets. Objects are sprites and it’s 2d urp renderer
Do you need it to be fully interpolated and not like 1 blur for foreground, 1 for background (same amount for each in these layers)? Are the sprites semi transparent or always either 0 or 1 alpha (alpha clipping possible)?
I would appreciate a fully interpolated way. If it’s too hacky/complicated/demanding I would go with 1 blue level for bg/fg , don’t have semi transparent sprites for now but I guess it might become a thing ? Just wanted to check my options and couldnt find much in documentation or on YouTube about that topic specific with 2d rendere.
I guess the 2d lights won’t work with a 3d urp renderer right ?
And I thought it’s as easy as turning dof as postprocess on in the camera, but I guess it’s not possible or there is more to it
That is right. There isn't any clean way to do these kinds of post processing effects on semi transparent geometry since those can't render to the depth buffer. Making the sprites opaque with alpha clipping (might require custom shader) is one way to make it work but there's no semi-transparent then. One possible solution would be to make the blur happen in the shader of each sprite but that could be bit bad for performance since those nice post processing tricks (like doing vertical and horizontal pass separately) won't work
Of course using preblurred sprite images would be a solution too but that could be quite bit of work and still not great (changing between blur levels would be hard)
Thank you, will try to find a proper solution
Hello, I have a problem. I’m making a 2D game with a moving camera, but when the camera moves, the pixels at the edges of my sprites are flickering.
Not related to post processing but in either case post only to one channel that seems relevant, please
terrible suggestion
whar are you saying, he was using a really small resoloution
if you import the sprite with a proper resoloution there is no need for changing the filters
thats desirable when working with pixel art, like the other guy said,
- no compression
- nearest neighbour/point filtering
- change the pixel unit scale
exporting at a higher resolution is a waste, especially if you have multiple frames of animation
and its easy to export the files back out to make edits too
did you even read what i said? i said when importing into unity...
"Proper" scale with pixel art would be no upscaling (or other scaling) of any kind
Upscaling is wasteful, but it kind of does work for addressing the specific issue
And admittably easier than doing everything the right way with pixel perfect rendering, or texture antialiasing
#1390346776804069396 a better place for this
Hi all. I'm making custom DoF
URP Unity 6. So, I have an oil paint texure I want to apply as a post processing effect. It looks how I want when I just give it UV0 (picture 1). The problem is I would like it to keep the texture in the same spot relative to world objects when the camera moves. I figure that just means changing my UVs to read from position world (Picture 2). The problem is when I do that I can't get the texture to appear at all no matter how I change parameters (Picture 3). this feels basic but I just can't get it to work.
Well, you need to think thoroughly as to what you want to happen. As it is now, you're using x, y of world position as the uvs. Is that really what you want? Not x, z?
With some really high tiling values I can get some striping with xz and yz splits. But still not the intended result.
Otherwise it just looks the same
What does it look like with actual xz?
XZ just looks like
I am trying a different methods subtracting the camera starting position from its current position to compensate the texture as it moves but I can't figure out how to convert from how much the camera position changes to how much the UV coordinates need to shift.
I figured it out. My methods only works because this is orographic projection so there is an easy relationships between world units/pixels/uv coordinates. But here it is (warning messy):
Hey, I tried some post processing methods and values but the scene looks out of touch. What would you recommend I change in the overrides?
Depends on what you aim for.
Like realistic vibe or cartoony.
Realistic
Realistic rendering is not solely about post processing, but mostly made with well authored materials, texture and lighting.
Do you have a basic approach or an article on that by any chance?
Since you're aiming for lego bricks, I'd consider searching how Lego Builder's Journey was done.
IMHO it has the best realistic rendering of these plastic bricks.
Here's a GDC video : https://www.nvidia.com/en-us/on-demand/session/gtcspring21-e32773/
From what I remember :
- Physically accurate materials
- Physically accurate light intensities
- Realistic reflections with RTX raytracing (but can be approximated with screen space reflection and reflection probes)
- And add imperfections : random fingerprints on the materials, random unaligments on bricks ....
Oh sweet!
Let me have a watch
Hi! I would love some insight on how to accomplish something in my scene - I CANNOT figure this out for the life of me.
I have one post process scene with fog for my underwater zone - which I like. I think i did a good job and it fits. But I transition into a darker zone (you can see darker walls, etc.) and I would preferably like that zone (seperated by trigger box colliders) to get darker / maybe fog darkens or switches to a different color. I'm using URP so there isn't any advanced options like HDRP has. Any solution to make this transition work where I can change the fog? Yes I can use RenderSettings.Fog to change this stuff with a script and maybe the solution is halfway through the first zone start transitioning into the other fog settings so it doesn't instantly cut or look odd if you look back up?
either way im not sure, any insight would greatly be appriciated
You could write yourself a simple VolumeComponent that you can integrate in volumes in the scene to control the fog.
Then you need to write a small monobehaviour or URP renderer feature that fetches the interpolated values and applies it to the scene fog.
Anyone have any ideas on how to draw a full screen shader behind UI elements? Can it target different layers? I wanted to achieve a sort of blurred vignette effect (which I am confident I can do in shader graph), but I obviously wouldn't want the UI blurred, it would need to be rendered above the effect.
I guess this would be called a tilt shift effect. Using URP btw.
You can't separate by layer without quite some custom render stuff. That isn't often a problem since the overlay canvas (and UIToolkit by default afaik) is rendered on top after post processing is applied. If you use some other canvas mode, you can consider using camera stacking to render the UI on top after PP
How not to blur the UI is typically trivial
Blurring it would be hard
It sounds like camera stacking may be the route I have to take. UI is not rendered using Overlay canvases.
Can you clarify whether you want to render the UI after the blur, without the blur
Or to include UI in the blurring somehow?
Sorry if that wasn't clear in my original question. UI would need to be clear, don't want players unable to read text for example. The blur effect would just be for some visual flare.
But for other reasons regarding how my UI is currently set up, I wouldn't want to switch to overlay canvases. I do have a blur effect which I can render to a panel for certain UI elements, and had considered using that, but I don't want to be rendering a whole new camera every frame just for that.
I think you can control the rendering of screen space camera and world space canvases with Render Objects renderer feature
So they only render after post processing
Thanks for the tip. Going to look into that now.
Yeah I think that is probably what I want. Thanks for the tip @mighty stump !
Good info.
It is!
Also, the render objects feature renders objects additionally, doesn't change how they're rendered otherwise
So if you don't want them rendered twice you'd remove them from being rendered the first time
Iirc from the layer masks in the renderer asset
Why does panini projection make my camera blurry the higher its fov is?
The package that fixes the issue is paid. SURELY there must be some other way
Because it streches things.
I guess that makes sense, but how do I resolve that issue?
The kind of only way to solve this is by overscanning : rend to a higher resolution target to compensate the way panini project warps things.
does anyone know how to make a overlay in unity 2d urp to make it look glitchy, like the five nights at freddys 1 main menu static effect
Anyone have some quality guide for getting custom luts to work in unity?
Hey folks. I'm trying to make some small effects in my game, namely the game become blander with a vignette effect when close to gameover. I'm using a global volume post process effects for that, however, I have some issues. The color adjustment works but the vignette do not. In fact even in the editor, when I change the intensity, the vignette doesn't appear anymore.
i read in a random post that you are not supposed to dynamically change those values ? What should I do ?
I haven't had issues before changing vignette or other effects' intensity, so I wonder if you still have that random post for providing some clues
It may be more convenient to have a volume profile with just the vignette override and modify that volume component's weight rather than modifying the volume profile at all
nevermind I found my issue. I had two global volume in the scene for some reason and the game didn't like it much
You can have two or more global volumes, but they should have a different priority values so their overrides can be applied in logical order
what can i do to make my game look like this in terms of post processing? ive done research and tried bloom pp and it hasnt worked
ive tried:
- setting the quality in my urp renderer to HDR
- enabling pp on the camera
- post process layers etc
- ive made sure that its not just selecting to avoid viewing post processing in scene/game view
post process layers etc
Post process layers are from the old post process stack V2 and should no more be used with URP, in favor of the new volume system.
To get a look like the image you've posted, indeed bloom and some tweaked color post processing should be enough.
If you could post screenshots of you current setup and rendering output that would greatly help us, to help you.
I dont have screenshots currently, on my phone.
However, my setup looks like this:
Camera Parent G.O.
- Main Camera (pp layer, with the pp layer set)
- Gun clip prevention (same as main cam but child of main cam)
PP gameobject with Global PP Volume, GameObject layer is set to the PP Layer
does that sort of help?
@fluid fern
ok so ive followed your directions, by utilising the new post processing system and its working
i thought there was only one system, i do apologise. thanks so much!!
how can i make the scene l;ook more realistic lighting wise? i only have 10 lux sun and a rectangle area light where the emissive material on the roof is. Im really bad at scene composition
any help would be greatly apprietiated
To make the scene look more realistic lighting-wise, the main thing you need to improve is how light behaves in the environment rather than just increasing brightness. Right now, the lighting looks a bit flat because there isn’t enough global illumination, so enabling baked or mixed global illumination in Unity and baking your lighting will allow light to bounce naturally around the room, which instantly adds realism. You should also tweak your area light by slightly lowering its intensity, increasing its range, and enabling soft shadows so it feels more like a real ceiling light instead of a harsh source. Additionally, make sure your ceiling panel material uses emission with a slightly warm tone and contributes to global illumination, as this helps simulate how real light panels glow and affect nearby surfaces. Small adjustments like these -combined with softer shadows and subtle color variation - will make a big difference in how natural and believable your scene looks.
It still doesn’t look right( sorry for the picture). Is there any chance you could go into a bit more details in dms maybe?
It does look way better than before.
Please use screenshot instead of photos of the screen.
Now, to make it better to your taste, what "looks right" to you needs to be defined. A lot of the answer can be influenced by art style and direction.
Yeah I’m sorry, I didn’t have discord on my laptop at the time.
How do i use bakery ? I just got it but i kinda dont know how to use it
as usual, you would read the documentation or check their website and asset store for more infos. also in unity you would look for new content that has something to do with bakery for example context menu, toolbar, asset folders etc
I have a question regarding post processing in URP.
I think it likely is not possible but can custom effects be injected inbetween post processing effects?
Looking at the source code it seems like they are all in one recordrendergraph function or a similar function so i feel like it's likely not possible.
why isnt the scene complete darkness. There are no light sources.
Likely either ambient lighting or reflections. Don't know how either work in HDRP
Hello, I need some help. About 11 years ago, I was looking for a way to make video game graphics smoother, specifically to remove jagged edges (aliasing) from walls and textures, and also to fix mouse movement. After a lot of searching and struggling — it took about a year and a half — I think I managed to do it using things like CUDA, cuDNN, CNTK, Python, and similar tools. Could anyone help me figure out where to start learning and how to do this again?"
I don't think anyone would be able to teach you about some secret technique you came up with years ago.
From the tools and apis you mentioned it sounds like some custom machine learning approach.
As for aliasing, there is a huge number of different antialiasing techniques, so I'm not sure why you'd need a custom solution at all.
Also, aliasing doesn't really come from textures, it comes from mesh edges.
Alpha cutout textures could be a problem too (primarily to MSAA) though any PP AA should take care of that similar to mesh edges. ”Fixing mouse movement” also sounds very odd, would need more context on that too to even know what that means
Can you have a bloom effect on specific emissive materials, rather than just anything bright in the scene? This is because I need to fake some realtime area lights or other types of light and then add bloom but I do not want every bright part of the scene to have bloom! I guess would require stencil buffer and custom bloom
With the default bloom, you could use camera stacking and apply the bloom to a specific camera only but that could mess up some other things like shadows
yeah camera stacking is useful for lots of simple rendering tricks, but I am not sure URP is best optimised for it
Yeah, I just don't think the Bloom effect provides any other kind of filtering than the brightness threshold. If the threshold isn't enough, a custom effect probably is the best bet. Unfortunately efficient bloom effect isn't particularly easy to implement (I think the unity effect recursively downscales to calculate the effect and then upscales back to full resolution)
Raising the bloom threshold and emission intensity in tandem works fine in like 90% of use cases I'd say
Also for faking lighting you can make use of "flare" sprites as old games did
There's even some pretty advanced techniques in that category:
https://simonschreibt.de/gat/doom-3-volumetric-glow/
If you draw the bloom on the base camera, non-blooming objects from the overlay camera will be over the bloom itself too
If you draw the bloom on the overlay camera, bloom would be applied to everything
An exception could be the alpha processing feature that according to description could do something much like this, but I haven't seen any example how it's used
I am working on a project using Unity HDRP and volumetric fog. However, when I use TextMeshPro UI elements in World Space (Canvas), they are not being affected by the volumetric fog in the scene. The fog does not visually influence these UI elements like it does with other 3D objects. How can I resolve this issue or achieve a similar fog effect on World Space TextMeshPro UI?
Nice. Yes I use simple divided planes procedurally for lighting effects, at least until URP has realtime area effects. It looks ok but that doom 3 glare is interesting
True, totally forgot how the no-bloom objects would be rendered on top of the bloom. I just thought one could render the objects in base camera, apply bloom and then render everything else and call it a day. That is nonsense of course now that I think about it. I should play around with camera stacking more to get better intuition to them.
If you can find out what it takes to get alpha processing to work I'd be very curious
Oh and I think you're right that the included bloom may be a bit complex to imitate
But if the chosen solution is implementing bloom from scratch, it seems like the existing bloom code could be copied and modified
Hi guys im trying to acheieve a look like the reference picture to the left - but at most I am doing whats to the right. I have NO idea what to look at or what im doing wrong for this. any tips or help would be appriciated! ❤️
at least - in terms of colors and lighting
there .. thats my best attempt : D
If I want my color space to be linear in my project, where would I set that? I need to get rid of whatever is shifting my colors away from what they should be
Like in this case, 0.5,0.5,0 in photoshop should be 128,128,0, but it's giving me something closer to 188,188,0
Somewhere in the project settings - player. Gamma <-> linear I think. Though, there might be other color correcting steps in the full pipeline.
It was already set to linear, so there must be some other setting somewhere
Not seeing anything else in the pipeline asset, unless I'm looking at the wrong spot
But I'm no expert on rendering. Just trying to figure out color lookup tables, but no matter how I make the 3d texture it doesn't seem to map correctly, and multiple input colors end up mapping to the same UV
I don't think there's anything beyond that to configure.
The gamma correction should be happening at later stage though(I think) , so perhaps the color space is unrelated to your issue.
I'd debug the relevant shaders in PIX to see what's going on.
So I have a funky one for ya'll that I've never seen in all my years.
We're using a PSX shader combined with Unity's URP.
Thus far we've been experimenting with some solid results, but my colleagues captured PNGs are behaving extremely weird with dithering.
When displaying these at full size on a phone, computer screen, etc the dithering looks fantastic, however if you look at the photo in portrait mode the scaling algorithm breaks the dithering.
This happens on my android phone gallery and when trying to upload to Instagram.
The post processing stack may need to be adjusted or else we're gonna go through hell trying to promote the project with any screenshots. I'm wondering if there's a way to "flatten" the post processed display for the user's display so game captures and screenshots don't look horrendous.
By far one of the strangest things I've ever seen developing before.
Moiré patterns are quite expected when a dithered image is downsampled smoothly for preview or otherwise
The simple solution is to disable dithering for the screenshots
With such a fine pixel pattern they won't be visible unless the images are viewed at least at 100% zoom which is unlikely, and the dither looks to be largely consumed by image compression anyway
Alternatively the dither pattern scale could be increased for screenshots/video, but it just makes the problem gradually less likely to appear rather than fixing it
I know this is super vague, but how do i make my shadows look better? My game seems very dark, i tried toning down shadow intensity but it didnt look good either
My game isnt supposed to be hyper realistic, but right now the shadowy areas just looks off
Would appreciate any advice
Adjust the ambient/env light color.
Also lights have shadow intensity settings.
Ok ill give it a try
This is using baked lighting, right?
You should show your lighting settings and scene window's baked lightmap view from the same angle
Its realtime
Certain? These black smudges look like baked AO, but that would mean they're drawn in the assets' AO maps and/or base color
I did enable ao
Im using urp
From where? There are many kinds
But im pretty sure i didnt do any baking
You do have to at least generate lighting to get correct ambient light from the sky so you probably did that
And by default it might easily trick you into enabling baked lightmapping
Showing the lighting settings would help determine that
Ok ill show it later, im not home rn
Tyy
Of course
These I mean, to be exact
Including the box area at the bottom of lighting settings which reports lightmaps if they exist
Thanks mate ill get back to u soon
Hi guys im working on my enviroment and - I like it - but it doesn't feel polished for me. I'm not sure whats quite of. I am going for a stylized feel / look which I have somewhat down but something is off. I dont have the eye for it - can someone give me a sense of direction
"Stylized" could mean anything, and if it feels off it is because it doesn't match the direction you have in your mind even if you cannot quite define it with precise terms
Artistic direction is pretty personal and something you determine that informs your choices, that would be references and guidelines like what you want the result to feel like and what kind of shapes and elements to use to achieve that
There are techniques and technical tips that may help, but mostly this can be solved only by getting experience with trying to replicate certain styles and getting familiar with techniques
A practical tip I have is to watch your assets' color contrast
If you have dark or saturated colors like deep greens, you cannot affect the object's appearance much with lighting or post processing
yeah, good point. i mean - do you think I should try to see what the scene would look like maybe with brighter more vibrant materials - maybe at least even so slightly?
like - I think it looks pretty okay right now but its not like idk
Suprisingly I was originally going to go for something like this - but clearly my lighting and art direction gives more of a realism vibe. Not really sure how I did that it just kind of happened. Wouldn't even know how to make it look closer to think - but maybe brighter materials / more vibrant ?
I had to re do the fences too, and removed shadows from grass, i think this looks even better
ok the issue is when i look at this screenshot the scene looks more "crisp" thats the best way I can describe it
does tihs mean my shadows are baked?
Based on this you've enabled baked lightmapping but have not generated anything, I believe
Assuming it is the baked lightmap view and none other
And that this box in the bottom of lighting settings doesn't report any baked lighting
And assuming you aren't using adaptive probe volumes
how can i achieve this look?
is this mainly post processing or lighting or textures
I mean, how much of it is really 3d? It could very much be a drawing. There's even an author name at the bottom left.
its all 3d its a 3d rendering
Even if it's a render, some of it could just be sprites in 3d scene while actual 3d objects limited to maybe the robot character.
It's a common method of creating such stylistic renders.
For the objects that are 3d probably a custom stylized shader is used. It's very unlikely to just be post processing.
this is kind of a shot in the dark, and i'm not very well versed with screen shaders at all, but does anyone know how you could recreate this sort of effect in hdrp?
the closest thing i could get was like really intense motion blur, but i couldnt make it delayed enough to get what i wanted
https://www.rxddit.com/r/scpcontainmentbreach/comments/1flno3p/
Kinda looks like they capture a frame into a texture when the effect begins then blend it into frames that follow for some time. Or something like that.
yea i figured its doing something similar to that i just don't understand how to execute it
i think the engine that game is in has it as a built in effect
i dont think unity has anything like that, so i'd have to make it from scratch and i have no clue how to do that :-(
Hi guys. I made a custom postprocessing. But it's not listed under RG_UberPost, it's not affected by the "PostProcessing" toggle in the scene view, and now i also noticed that this custom prop is in effect in inspector's asset preview
I think this is a bug/wrong/shouldn't be like this, or maybe there's a way to tag this custom prop to be considered a normal postprocessing like others?
Here's a pic
I assume it's a custom renderer feature? If so you can add checks under the AddRenderPasses method to prevent the pass being queued.
The RenderingData struct/param passed in has access to .cameraData.isPreviewCamera to prevent it affecting asset previews, and .postProcessingEnabled (untested but sounds like the scene view toggle would affect that)
Yess it works now thanks
Hi guys! Kinda dealing with the same issues i've been struggling with in terms of post processing and volumes in my project for a few weeks now. I spent alot of time learning and finally got my above ground/outside scene to look basically exactly like I wanted (Thanks @mighty stump for the direction by the way!)
Now I have another issue: I have four different "zones" in my game all attached in images in order from zone 1 to 4. Now - they're not terrible. especially the first and second zone. I think they're kind of meant to be a little simple and normal "what you'd expect if you're underwater".
Now for my third scene , I wanted a dreamy / magical kinda feel and darker colors. I kind of acheived that but to me it still just looks off and im not sure how. Although it has that darker vibe to it - it still feels just that: darker than the others.
The fourth zone I cant get to feel how I want it at all, period. I just feel very limited to my options and contol with my post process volumes and stuff. (I'll get more technical in my setup below).
I also attached a short video to show the transitions between these zones. The other zones look better this is just the worst one.
Technical setup:
So each zone has their own script I made which takes a post process volume. I have on that script properties for Fog settings, and the post volume. the same game object will use a box collider Trigger = true to detect when a player enters what zone. (I also have one global underwater post process).
Between each zone, I have a transition box collider that has two properties "From Zone, To Zone", which will transition between each post process volume into the next zone.
TL;DR questions:
How can I get a better look for my underwater zones?
How can I make the transitions better?
You know you can set PP volumes to local and it will automatically handle the transitions and triggers for you
As for underwater you can add a wobbly screen effect
For Looks (zones):
Dont just make areas darker—use color grading and fog together. For the dreamy zone, shift colors (like blue/purple), lower contrast, add slight bloom, and tint the fog so it feels magical instead of just dark. For the last zone, push one strong idea (heavy fog, strong color, low visibility) rather than trying to balance everything.
For Transitions:
You can use Unitys volume blending, but dont rely only on the automatic distance-based blend. Instead, when entering a zone, manually lerp the volume weight over time (0 to 1 in 1–2 seconds) for smoother, more controlled transitions.
Setup tip:
Keep a global volume and use local volumes with priority, then control their weights via script instead of hard switching.
It's a valid option to either use local volume blending or some kind of lerp or nonlinear transformation to interpolate between volumes
But you may want your fog to use the same blending, whether implementing it as a custom volume component or replicating the interpolation
Or if you use a different blend for fog, it should be intentional
Blending fog in or out can easily look weird because it has an infinite distance, so the foggy area will look fogless before you enter it and vice versa
If you don't want to implement some more complex local fog shaders or post processes, another option is to bend the path or obscure it with objects to break lines of sight, or use foglike particle effects
Particle effects in general can improve the atmosphere
Watery places tend to have floating particulate, and some gentle Noise can make the particles flow as if along currents
The light blue area seems to have global light and shadows from a sun, which is very unrealistic
But it can very well work to make the place more surreal if that is your intent
no kidding... I had no idea. Ill experiment with that
thats actually exactly how I handle it currently! I couldn't explain muhc in detail in my original question because of limited text space
yup that scenario you described is exactly what im dealing with when it comes to the fog. Maybe bending the path down with like a transition layer might be the move. This way the transition happens during that bend as suggested.
sorry, i was using GLOBAL volumes with priority which is where you suggested different, yeah I think definitely changing them to local volumes will make my setup much more clean and easier to work with
The main difference is whether you want the transition to occur over distance rather than over time
I suppose for my project it would be over distance that would make the most sense... the closer I get to the next zone it starts transitioning into that PP
which means global.. might still be the best?
Local volumes do exactly that
it looked ok for the other zones I created. Although after i posted this I ended up disabling all the scripts because i am going to completly re do it , or else id show you right now how it looked ;p
oooo, okay okay
i overcomplicated the s*** out of the way I did it then
With distance blending the player can practically control the blend with their movement
Stopping and reversing it if they start going backwards halfway
Or making the blend occur very quickly if they have a way to move very quickly
If that's an issue, blending over time is one solution
But the drawback is that if the player can reverse course and head back to the first zone, the blend to the second zone would continue
More of a design decision than a technical one
Yeah - which means blending over distance really is the only option
its just a matter of then making that look good
I had no idea local PP volumes blended by distance automatically lol
i also cant believe i didn't check before doing it the way i did it lmao
You can also blend any property using the local volumes as I mentioned
So fog, even if it's not a post processing override
Just not sure what's an up to date guide for making custom volume components in current year
thats fine. I'll do some research. Key word - custom volume components
thanks , it'll point me in a good direction I think 😄
update: yeah , local PP volumes so much easier. lol
Great to hear!
!ask
:thinking: Asking Questions
:mag: Search the internet for your question!
:book: Use the API Scripting Reference and User Manual and this troubleshooting site for commonly posted issues.
:wrench: Attempt to debug your issue.
:thought_balloon: Find an appropriate channel by reading the name and description in #🔎┃find-a-channel
:grey_question: And don't ask to ask, ask a full question illustrating with screenshots if needed.
-# For more posting guidelines, go to #🌱┃start-here
@mighty stump I ended up deleting the entire underwater enviroment of the game - I just could not be satisfied with it. Side note- I was thinking maybe a better solution would be one global underwater PP volume, and then add a decension based change to it.
So the deeper I go maybe the darker it gets, fog color or ambient lighting changes, etc. (idk how ill do it but just that as an example).
I think - yes it would make each zone less "unique" but it will give the overal enviroment / game a much more polished and better feel to it. what are your thoughts on that?
this way i dont have to deal with transition layers and things like that
Hello, I want to have a different post processing on selected objects, that belong to an extra layer we can call "Outlined". (I need to keep my outline effect through post processing, no shader or geometry methods).
Is it possible to have different post processing volumes by layer through Renderer Features? I can do this with objects by layer, but I can't manage to do this with Global Volumes, even if they're set to "Outlined". Camera includes "Outlined" in the Volume Mask and Culling mask.
(I'm exploring possibilities with Renderer Features since I discovered it's impossible with the multiple cameras trick, been browsing on internet for days now I think).
No, you can't have different post processing on selected objects.
You'd need to find an outline post process than handles out of the box to specify outline objects.
Renderer Features is indeed probably the way to go.
Sorry I don't understand, you say I "can't" with Renderer Features, then you say Renderer Features is the way to go..
So what should I do?
You can't have different post processing on selected objects. Post processing is global per camera.
Ok, so, since I was already saying that I'm using Renderer Features, and asking if it's possible with that,
what should I do?
I don't know where to add a script if I were to do it by code somehow, as clicking "Add Renderer Feature" doesn't seem to prompt for any script option.
Here is the documentation page about how to create a new custom renderer feature.
To make an outline around specific objects in the scene, the renderer feature would need to work in at least two passes :
- render only the wanted objects in a separate color/depth target (not the main camera)
- draw the outline on the main camera by using 1. as source for outline
Thanks but what is illustrated in the documentation is exactly what I did already for objects, and that is what I'm referring about in my original message: that I can't do the same with Global Volumes, it seems, but only with other objects.
(I don't know how to add scripted Renderer Features still, if I have to)
Now to your added comment, how do I do point 1? In a separate color/depth target like? You mean excluding them with a Layer from the main camera?
Edit: I found a workaround that works: in the Renderer asset I disabled the Default layer from the Prepass Layer Mask. This way it actually does what I need: outline only on elements that are in the Outline layer. 👍 Hoping it doesn't cause cons.
You can simply change a variable based on your Y position. The lower the y, the darker the blue (or whatever your preferred effect is). Easy. I'd still put it inside a local volume if it's only supposed to be underwater.
If this is hdrp, there is also built in water that has a bunch of detailed underwater effects like that, but you might want to change how the surface water looks to match your stylized look
I don't quite see what makes that a better solution
Local volumes seem like a useful solution to the problem, rather than something you "have to deal with"
Global volumes are also an option, but I'd prefer to control the post processing by changing the weight of overlapping global volumes in that case
Modifying just one volume to me seems clunky
The result is the same but you'd be avoiding the methods that already exist for doing what you need to do
Hi, why cant i disable post processing? I have tried disabling it for the editor, for cameras, changing values, making a new profile with nothing in there but it wont be disabled. I can disable post processing overall but i want to change values actually not disable but that isnt working
A volume component with a volume profile would contain overrides
So an empty profile or a disabled component simply will not change anything from the default volume profile which likely exists in the Settings folder
A volume can disable post processing effects by overriding an effect with a deactivated state or 0 intensity
how do I apply post processing to ONLY ui in URP?
may i ask a Lighting question here?
This is for post processing effects specifically. Lighting in general would better fit at #1390346776804069396 or #1391720450752516147
Perfect thank you!
someone have a page for Post Processing Shaders? like Vignette, Bloom, etc (page like Github or other)
By "page" do you mean documentation for the post processing overrides or something else
My English is bad sorry, but I mean like the GitHub repository or reddit
I mean to ask what is the purpose?
Trying to use Unity's post processing
Or make your own perhaps?
Make my own perhaps, I mean i want edit shaders
For my game
If you use URP here's documentation how to make custom post processing effects which you can now make in Shader Graph:
https://docs.unity3d.com/6000.3/Documentation/Manual/urp/post-processing/custom-post-processing.html
The source code of the effects themselves is also available, though not documented
https://github.com/Unity-Technologies/Graphics/tree/internal/6000.3/staging/Packages/com.unity.render-pipelines.universal/Shaders/PostProcessing
Thank you so much!
Effects like bloom require temporary textures and typically compute shaders which shaders alone can't do, so for those you'd use the Render Graph in modern unity versions
https://docs.unity3d.com/6000.3/Documentation/Manual/urp/render-graph-introduction.html
Bloom relies on series of subsequent downsampling and then upsampling steps which makes it a good example of something that is not possible in single post processing shader (well of course possible but would be wayyyy too slow). Well this multi pass setup maybe is technically a single shader idk: https://github.com/Unity-Technologies/Graphics/blob/master/com.unity.postprocessing/PostProcessing/Shaders/Builtins/Bloom.shader.
hey guys, not sure where to ask this and probably not a post-processing question, but where can i set my max shadow render distance in urp? i cant find the URP asset t-t
It's a #1390346776804069396 question
You can find a field referencing the URP asset in the currently active quality level in Project Settings>Quality
ok, found it, thanks!
Hello guys!
To summarise my problem in a few words: I want two different post-processing blooms on the red and blue. Red is the background (single image), blue is the foreground (tilemap, spriterenderers, particle systems etc.). This is in isometric 2d space, ortographic camera. What I want to do is have a post-process bloom in one color on the red background, and a different post-process bloom on the blue foreground. No matter how I set it up the blue foreground bloom always propagates additively on the red background. Does anybody have any experience tackling similar issues?
I can provide a more detailed image and more detailed explanation if this is not enough (sorry if that is the case)
Not very practically doable
How do you need the bloom to be "different"?
Well, in short:
Terrain sprites (red background) use a single color palette (think orange, grey, brown, yellow)
Foreground sprites use a multitude of different color palettes (yellow, blue, red, cyan)
The end result im going for is to simulate emission on the brighter parts of the terrain by blooming the lighter colors (yellow) and doing the same on the foreground but on the different colors (red, blue etc)
However if I apply yellow bloom to the background and a blue bloom to the foreground, I get green bloom
ive tried using URP and the built-in pipeline, ive tried dual camera setups, overlays, base, tried the post-process layer and post-process volume components but no matter how I spin it i run into the same issue
Sure you can't use just one color neutral bloom volume?
Given a sufficiently intense color, each sprite will glow its own color
The building in the center shows the desired bloom amount, however the terrain arround gets way too much bloom. They share a volume.
The first image is with no bloom, the second is with bloom applied
If the parts you want to glow are limited to specific materials or textures, a workaround is to increase the bloom threshold until background no longer glows, then increase material emission intensity to compensate
If you specifically want to use bloom for twinkling lights, additive billboard sprites would likely be a better option
The builing and the background are both sprites on sprite renderer components. Could you tell me more about limiting the bloom to materials, textures in this situation?
Higher threshold means more colors on screen will be excluded from bloom
Higher color value on a material means that material is more likely to surpass the higher bloom threshold
Sprite materials don't expose Emission or other types of intensity so it takes a couple of extra steps to make a sprite brighter
This workaround requires HDR to be enabled, so brightness values above 1 aren't excluded from rendering
This is for a mobile project, wouldnt HDR be too heavy/overkill?
Five years ago there were warnings against using HDR (or linear color space at all) on mobile devices, but now I haven't heard such cautions
I don't develop for mobile so I don't have any first hand knowledge
I'd test the performance / power use impact before committing to this method
On mobile bloom in general is also in that category to avoid if you can help it
Because your objects are sprites, that means any bloom or glow can easily be achieved with local glow sprites
Or even glow baked directly into the sprite graphics themselves
I see. Thank you for the insight, if I have any further questions i will circle back on that
I wouldn't even call it a "workaround"; you're just using sensible colors!
bright stuff blooms because it's bright
If there's punctual lights with the default inverse square falloff, you may have to set the bloom threshold and emission senselessly high to try to filter the bloom without guarantee of it working in every case
As it really isn't a per-object bloom and lacks the advantages that feature would have, it is a workaround to achieving that effect specifically
baffled that URP lacks a "light radius" option
(softening shadows is hard, but just limiting the brightness isn't)
Indeed
I appreciate the idea of PBR accurate lighting model as the default, but with infinitely small point of origin that kind of falloff doesn't really make sense in practice
At least not without size or limit or an easy way to customize it
Specular aliasing also combined with bloom is a similar kind of constant rock in the shoe
Hi yall, I am not at all good at post processing, but i would definitively apprechiate some help getting color grading to work and change this (1st image) to this (2nd image) this here is the post processing i have right now(3rd image)
Is the first image with your post processing, without post processing or with default post processing?
If so, what overrides
1st is no post processing, the 3rd is with these
Oh no, that is a volumetric sandstorm asset
Not actual bloom, I can see the confusion for sure though
This glow looks very much like bloom
I dont think there should be any bloom huh
You have at least two "global volume" gameobjects in the scene as seen in hierarchy
And URP projects by default have a few post processing overrides on in the default volume profile
Oh I see, okay, I removed the 2nd one but he still kinda glows
Hi, does anyone have a nice transparant overlay for the old photograph creases, Im making the slow time ability have like a sepia effect and it would rlly improve the feel
hi guys, idk if its the right channel, but how the heck can i make sun rays using hdrp with a light source??
What have you tried and where has it failed? HDRP has volumetric fog feature which is probably what you want though it is not very clear what kind of sun rays you are looking for
Also #1390346776804069396 is probably a better channel
to answer, im trying to make a god ray from light and fog, using volumetric fog. i tried many settings to do it with a combination of those two things but so far i failed, im a beginner, in this, i will transfer the chat in rendering anyway
ok i found the solution
I'm tryna get lighting to darken areas that are closed off but for some reason it is remaining lit up and my sun object that has a direction light in it doesn't hit off objects and goes through
See there's visible shadows but under the terrain object theres no shadow or darkening to the map.
The underside looks to be occluded from the directional light, but not from environmental light from the scene sky, which is everywhere
If you set environment light and reflection source to none, all shadows would get extremely dark
Realtime lighting doesn't simulate environment light occlusion at all, because it cannot
Baked lightmaps and light probes would be required for that
Lighting is a topic for #1390346776804069396
What you should do depends on what kind of light occlusion you want, and what kind of compromises you'd be willing to make
Turning down environmental light and reflections would be the simple option, but like I said that makes all shadows very dark
Baked GI is totally static, so you can't "update" them at runtime, if that's what you refer to
"Realtime GI" on the other hand is a system that can update lightmaps and light probes, but only if they are totally static (though probes do light dynamic objects)
I want it so that place like mines are dark from inside and when looking into them.
Or caves or facilities
All three of these methods do that, with different drawbacks
But I so need it to update lighting if lights are toggles or characters are walking in front of lights to give it shadows
And I also have doors that move too
Baked GI workflow may not support any dynamic lighting itself, but that's not necessarily a problem
Lightmaps + light/reflection probes could be baked just for the environment lighting occlusion, so interiors aren't illuminated by the sky through walls
Any dynamic light sources placed in those interiors work just as you'd expect with realtime lighting and shadowcasting
With baked GI there's also options for situations like those, like Mixed type lights, or including objects like dynamically moving doors in the static bake to better control the occlusion
But Baked GI is a big topic so expect you have quite a bit to learn and practice
Turning down the environment light and reflections globally is the simpler alternative you could try before that, but it's likely not satisfactory
If you deal with GI, note that Realtime GI like Baked GI also depends on static geometry, and generates its own lightmaps using different techniques, so not fully "realtime", and the two systems shouldn't be used together before you understand them both
How would I force the game's particular area to be darker than the rest?
I tried changing the light setting to baked and it's not generating shadows
Easiest way probably via local volume where you would change exposure or whatever else you want
My main issue is I want to be able to look into the cave or facility and see the shadows
So you don't get any shadows at all from your directional light? Check it and see if the shadowmap is disabled or if the shadow dimmer is turned up.
Or maybe your cave mesh has shadows disabled. Hard to asses blind like this, you would need to show some settings and inspector
Like I explained previously you need to determine specifically what type of light occlusion you want
And try the different methods
If you cannot use them successfully when testing, you cannot know whether the method is a good fit or not
If you want help with light baking make a thread in #1390346776804069396 detailing your setup and explaining the issue
Or help with modifying lighting in general
The default post processing setup is a mystery for me in URP, ignoring the sample scene with its own profile in there, the URP asset has an embedded scene profile which adds Neutral Tone Mapping (very grey overtone to scene), bloom and vignette. Then there is AO with some of the worst default settings imho as very noisy. Shouldn't it be setup for the best default appearance?
What the default setup should be is subjective
What a "best appearance" is is subjective
So you likely can't come up with a default setup that'd make everyone happy
If HDR is enabled it will not appear quite correctly unless a tonemapping profile is enabled, and also unless there's bloom to allow hdr color values to glow instead of just becoming saturated
Neutral tonemapping alone is there to be technically neutral as the starting point for you to apply color correction that fits the visual style and exposure level of your scene
If the default look was not neutral, it would risk the average user erroneously compensating for it with material properties and lighting
I believe in 90% of cases the default SSAO settings are fine, but sometimes perhaps due to clipping planes, camera fov, scene scale or other factors the noise gets overblown
SSAO as default could be justified as giving some clarity of depth when working with meshes in the editor, but mostly it as well as vignette are there just to look good
Vignette in particular is a bit of a trend since many 3D programs have it, despite only developers ever seeing it
super basic question i'm aware, but i've followed line for line by too many tutorials at this point: i have a maincamera with a post-processing layer, but no options to set a volume. any idea what's going wrong? the maincamera is the child of another object in the scene, but i don't see why that should affect anything
here are the camera settings, if that helps
this is unity 2022.3.18
wait, lol nvm i'm stupid and misunderstood all those tutorials. obviously the volume doesn't go in the layer container, you make a new layer for the whole scene and apply that there
default noise and shadows (over dark SSAO, over noisy)
The neutral tone mapping means white is grey at the start so this is not a good starting point to me
SSAO strength seems to mainly depend on distance/scale, which doesn't make sense for light or shadow
But it explains why some have a bad experience with it and others don't
That I think is a misunderstanding of HDR color space
With or without tonemapping what's "white" is arbitrary
You might assume that white is grey with neutral tonemapping, because in the default scene nothing is bright enough to truly be white
If you add a bright light or an intensely emissive material, you'll see it will appear white
Not having any tonemapping causes brightness values to cut into solid white early, so they appear brighter, but it also makes it impossible to see the full brightness range or to utilize it with your own post processing
No tonemapping vs neutral vs ACES
The point of tonemapping is to expand the usable dynamic range, without it you are not getting much if any benefits of HDR rendering
"If you add a bright light or an intensely emissive material, you'll see it will appear white". OK this is really about HDR - which I am not using on web. It nearly reaches 255 white with HDR and full emissive white, but disable HDR and it is 209 white, so maybe they should have a separate template for non-HDR with no tonemapping or something.
So you are right for default setup which I failed to check as turned of HDR initially :/
The default web setup is actually terrible as has different settings by quality and a stupid 4 light limit per model
But where are webgl and mobile on HDR now?
WebGL 1.0 used to have many limitations related to HDR and other features, but it is so old I believe WebGL 2.0 has replaced it entirely:
https://docs.unity3d.com/6000.3/Documentation/Manual/WebGL2.html
2.0 has none of the previously listed limitations, I think it didn't use to support more modern rendering features like VFX Graphs, Forward+ rendering, or Resident Drawer
Aside from the last one there's no mention of having that limitation now, so might support them too now
WebGPU is in process of replacing WebGL, which'll bring it completely up to speed with desktop graphics APIs, but right now even if a browser supports WebGPU it might not enable it by default
I don't develop for mobile so I don't know what the state of things is, but I hear you have to check whether the devices that you are targetting do support graphics API like Vulkan, and the graphical features that you plan to use, mainly Compute Shader and Shader Storage Buffer Objects support for the likes of VFX Graph
Probably was more of an issue a couple of years ago
So the other reason though I do not use HDR anyway is I need MSAA (not doing fast moving stuff).
So basically I can use PC setup and kill mobile web default and any default volume with tone mapping and HDR toggle
Light loop in forward rendering path is relatively expensive regardless of platform, so 4 is pretty reasonable default
But now Deferred, Forward+ and even Deferred+ are pretty well supported across platforms, which eliminate the light loop cost
The web default has 0.8 render scale FFS so why set as web default I do not get
yes I use Forward+ which is default - though again looks like mobile web default is not +
The first linked article says that is a WebGL 1.0 limitation, which likely isn't relevant to you
Still if you are not doing fast moving stuff you should consider TAA, as it's cheaper and better than other AA methods, but fails with fast motion
Also, check early what the limitations of WebGL are on a mobile platform
Even if the mobile platform supports a modern API like Vulkan, it might not be able to utilize it if it has to use WebGL instead of WebGPU
I may have misunderstood TAA - if can do sub pixel it could work. That is what I need and FXAA, SMAA are not
Afaik TAA like FXAA and SMAA is post processing only AA method. MSAA is the only anti-aliasing that I know which is sub pixel in a sense that it actually renders at sub pixel resolution and averages it to the native resolution. Is that what you mean by sub pixel?
And why is "sub pixel" a requirement?
sub-pixel means pixel thin lines and other very thin things render very well. Without MSAA you can not achieve the same. I have specific needs for drawing fine detail.
You could elaborate on the specific needs
Different AA methods have different specific advantages and disadvantages
In some cases transparent mipmapping is way more effective for smoothing fine detail than AA on geometry
TAA is "sub-pixel" AA but only temporally, so the effect breaks down in motion and is a bit wiggly
"Render Scale" or supersample AA is the most effective, but least efficient by far
Others depend more on the use case
im having this issue with URP, where i want to record a clip using the Recorder window, and my game uses .3 scaling resolution for a low res pixelated look. when i record it that way, it creates weird artifacting in the image and scales it wrongly
this is with scale 1
and this is with .333333
i want to have the pixelation effect in the recording, its part of all of my shaders
Nearest neighbor upscaling can be achieved in many ways, even if Render Scale were not supported by unity recorder
An option is to reduce the recording resolution, and then upscale the output externally if it's an image sequence or a lossless video
I'm not familiar with the recorder so it's possible it has its own output upscale setting as well
Another option still is to do the "old fashioned" pixel upscaling by rendering to a low resolution RT texture and displaying that in front of your main camera as a UI raw image
I think a bit of vignette would work
Hi guys does anybody know how I should make a shader to get graphics like this ? https://store.steampowered.com/app/606160/ROUTINE/
█ ROUTINE is a First Person Sci-Fi Horror title set on an abandoned lunar base designed around an 80s vision of the future.
Curious exploration turns into a need for survival when a lunar base goes completely quiet. Searching for answers puts you face to face with an enemy who is certain the main threat is you. Discoveries lead to deeper u…
$24.99
6009
79
We have the 3d models for this level of graphics we just need a shader
Also I’m wanting to bake everything because the game is going to have a vr version
There's really no "shader" that can do all that
The look is composed of high quality PBR materials, carefully designed lighting and heavy color adjustment post processing with some film-like effects on top of those
I want to know how to make that beautiful shader effect
Which one?
The over all fpv annolog camera look
Do you think it’s possible?
Most of the time there's just color correction and film grain
Color correction looks a bit more complex that what the built-in overrides can do so it's likely Color LUT used to import adjustments from software dedicated for it
The analog distortion effect is pretty complex, as it swaps between many styles on the fly
Generally that kind of effect has an animated overlay, animated distortion, pixelation / coordinate rounding of sorts, sharpening, chromatic aberration, image trailing / blending with previous frames, and a variety of color correction profiles
If you research vhs / analog camera post processing effects, you can find all these techniques
Though most implementations have only some of these features so you will need to compare a few
Thanks
Ok I will see what I can find
Hello, could someone help me with baked lighting? I’m making a vr game and I’m struggling with the baked lighting
!ask
:thinking: Asking Questions
:mag: Search the internet for your question!
:book: Use the API Scripting Reference and User Manual and this troubleshooting site for commonly posted issues.
:wrench: Attempt to debug your issue.
:thought_balloon: Find an appropriate channel by reading the name and description in #🔎┃find-a-channel
:grey_question: And don't ask to ask, ask a full question illustrating with screenshots if needed.
-# For more posting guidelines, go to #🌱┃start-here
looks like fog with a little desaturation and low contrast
Can I ask a lightning problem here?
Well seeing other messages i'll do ot
Why do the borders of this dark room look bright?
Sorry for bad quality
could have something to do with ambient occlusion
Hello... this is my 2D game, i want to keep the textures minimalist
But i also want to add realism, i have already added:
Bloom
shadows
slight vignette
do you people have some more suggestions?
You should use #1390346776804069396 instead
And explain what kind of lighting you are using
And take a screenshot
or at least a photo that's in focus 
Yeah sorry didn't have time
I will
Hello
Basically, I want to apply a bloom to this text (and to the rest of the canvas). The end goal is to make a world-space screen, and it will have lots of UI elements.
I am beginner, so apologies if this question is dumb
By the way I am using URP
So is the problem about adding bloom in general or adding bloom only to the canvas?
Sorry for the late reply, but yes, bloom only to this canvas.
The end goal is to make a bright computer monitor/TV
Where do I start?
The default bloom component doesn't allow applying for only certain objects, it will always apply to whole screen. The common approach is to enable HDR and increase the bloom threshold which makes anything below that threshold in brightness not affected. Increase the brightness of the UI to make it affected by bloom. I don't think the default UI elements have HDR color input so to add brightness, you could make a custom UI shader that takes in HDR colors
Again, apologies if this is a dumb question but how do I enable HDR?
Nevermind, thanks
I adjust the threshold via this, right?
By the way how do I actually apply the shader to the text
Do you know why this happens?
Is an scene of a mod map of a game, there is no camera or player
Hey guyes I am new to post processing I want to know is there any way to do some post processing but not effect on all object on the scene (Based on different layer) ???? :
for example make some change for tower 1 and tower 2 but not effect third tower ????
isk if this will help, but whenever you bake your lighting, if you don’t set static to those two, it may work
Hey everyone, hope y’all well,
Question regarding depth of field post processing effect and opaque vs transparent materials
How do people get around the limitations with the camera seemingly unable to focus on transparent textures? Maybe I’m missing something obvious but is something I’ve had difficult with for years and thought maybe someone knows a solution!
hey guys I am so frustrated about how my textures look in unity while they look in good substance. it uses omni shader currently even tho tho the colors look so washed up. how can I fix this any recommendations?
The largest contributor to a material's appearance is lighting
Substance likely has a well defined HDRi for its environment lighting
The second most important thing is post processing
By default unity has neutral tonemapping with no additional tweaks to color
That's meant to give you a blank canvas for color adjustments, and to squash the color range as little as possible so you can see all variation in lighting as clearly as possible
I came here to ask this exact same question and the fact that someone else has asked the exact same thing just a few days ago with no response has killed me
SURELY someone has faced and solved this issue before? It's practically a fundamental. Unity 6.3LTS, URP 1.7.3 for context. Trying to have DoF apply to large transparent masses of fluid.
There are some pseudo solutions I've found that might help you tdharding. You can try these and see if they stick, I didn't have any luck with them, except for one, that didn't work out because it just compromised too much for my use case.
-
This is the one that actually worked kind of, but it's not good enough for my case: Add the transparent materials you need to a render feature (Render Objects) you can direct specific materials or layers to just skip post processing entirely. The elephant in the room though is that it skips post processing entirely. Meaning water wont be bloomed and color graded to match the rest of your scene. It also wont blur into the distance, this can probably all be offset a little by writing all this straight into the material itself, but... that's pretty complex and kind of defeats the point of post processing.
-
You could try writing a custom depth texture that injects/overwrites the depth texture right before post processing begins. This one I just couldn't get to work at all. Maybe someone better with shaders can solve it.
-
You could try writing a custom LOD screen effect that occurs later, again, way beyond the scope of my skills, I tried it and it just didn't work, I think that probably still requires option 2, because without the complete depth texture, I don't know how it could be done.
If you find something that sticks, or anyone else has faced and solved this, tag me please! Also, if anyone wants to DM and work together to solve this problem, you can DM me.
There is also a youtuber called XR Masiso who has a video called "Fix UI Post-Processing in Unity! (Finally!) - Render Graph URP Camera Stacking Tutorial (2025)" This could potantially be a solution, I haven't tried yet. He uses a custom render feature called "Push pop"
Hi guys, I've been trying to make a certain place of my map to be total darkness (a dungeon). I've tried various ways it doesn't seem to make it totally dark.
- I have a global volume and directional light which is used for the exterior.
- I have another volume for the dungeon itself but due to the directional light on the exterior is making it not to become total darkness.
- Im using URP
Does anyone know about the method to achieve this?
You have a local volume in the dungeon but the light still bleeds in?
You could try placing a quad or some kind of mesh to block the entrance and set it to "shadow only" in the mesh settings
Running at 130fps without ray tracing in hdrp on a 5060
Does anyone know how i can control the sun glare without tampering with the lighting controls?
I'm not sure whether this is a light bleeding in issue or something else. I'm using a Directional Light to light my exteriors. The dungeon is located below of the exterior map (Teleported into the dungeon) with a panes shaped into a cube blocking the dungeon's light source. I tried to set it to "shadow only" it still doesn't work.
The image is what I'm encountering, as you can see the floor is still very bright instead of "pitch black", as for the far right, there's some light bleeding in caused by the far clip (ignore this).
In this case, this should be due to to ambient lighting, where that's the minimum amount of light even without light sources (does it look the same if you disable the directional light).
You can decrease/remove ambient lighting, either locally or globally
iit's look weird ??
You need to ask a specific question, as well as what you are trying to achieve with post processing.
So I don't do anything wrong? It is just how it is supposed to be?
Probably
I don't know what the difference is because you didn't show
But for the reasons I listed any objects in the purely default scene will likely be at least somewhat duller in appearance compared to other rendering software
And the default scene is there to be changed
Thanks for the info, haven't had the time to try this yet.
why does my model look dirty when i import it?
SSAO renderer feature
You can turn it off, or tweak its settings like decrease radius to make it not overcompensate on small meshes
Oh additionally SSAO can mess up like that if your mesh has inverted normals with both sided rendering on
thankns
Hi, I tried every tutorial on youtube but my post processing is not working
I'm using URP
Neither the post-processing volume or the normal volume works
I fixed it. It was because inside UniversalRP.asset volume field was empty from default
You shouldn't have the "post-processing volume" component at all
To avoid confusion it's best to uninstall the PPsv2 package entirely
Thanks, there is no recent tutorial about it, so I thought it was the same as 2022 version
The Post-Processing package was deprecated in 2019, and unitl that early versions of URP used to support it
So confusion is understandable, especially with generic feature names like these
But when in doubt it's best to check the documentation's instructions for your specific editor and render pipeline versions and follow them to the letter
Hey thankyou for your thorough response dude! The struggle is real! Thankyou for your pointers, i have been playing around with the renderer features too… the effect i have now isnt exactly what i intended but looks ok. I will look into your suggestions and certainly let you know if i find anything conclusive… maybe unitys ears are burning and this will be solved sooner rather than later
With URP, is it possible to have a camera with post processing on, and then an overlay camera (e.g. for UI) that uses its own post processing, without affecting the underlying base camera?
I don't think that is possible with the camera stacking alone. The stack is always built one camera at a time and post processing on a camera always applies to the whole stack below it (everything rendered so far). From resource usage standpoint that is very reasonable, it only requires one target texture where everything can be rendered one camera at a time. Would rendering one of them to a render texture and then overlying that on top work for you?
facing two issues with volumes
-
i have a global volume on a scene and no volume on another. the volume's effect persists on the other scene. is this intended behavior? how do i stop it?
-
local volume isn't local at all. i set a collision box, made it a trigger, switched to local, and even set blend distance to 0, but the volume's effect still persists everywhere in the scene. i am guessing this's related to the first issue somehow
- Are you using a multi-scene setup, or does the effect persist when you fully switch the scenes ?
- And if you have no volume at all in the scene, no effect ?
- it persists even when i simply switch scenes through the editor. i haven't tried switching through code during runtime yet
- yes, i forgot to mention this. having no volume or even removing the existing one doesn't do anything. even after restarting the editor. like it's being cached somewhere
I think you've edited the default volume profile ^^
If you go in Project Settings > Graphics, scroll to "Pipeline Specific Settings", in the URP tab you'll find the "Default Profile". It's the volume used as default values for all renderings. If by accident you've assigned the same volume profile asset to one of the scene volumes, and edited it, it is saved instantly and still used (because it is the default one).
When working with volumes, you need to have different profile assets, and only override the needed settings.
just checked that out and unfortunately it's unedited. i tried setting my scene's volume asset to the default one and it looks normal (second pic). the default profile in the URP tab is the same default asset as well.
the actual default behavior is that of "ShopVolume" though for some reason (first pic)
Hum, I doubt that it could be it, but you can also set a base volume profile per renderer asset ... maybe this ?
oh you were right that was it
i have no idea when i changed that
thanks for the help!!
Hi, first post, Noice! Is there a way to limit the number of variants in the PP Uber Shader? I don't need 5k variants for Color Correct and Bloom :P
Just open the shader and remove all the shaderfeatures you don't want.
Unity won't compile those variants at the expense of those effects not being available, of course @mint sonnet
No automatic way, of course
A script that does this automatically is pretty easily achieved though
There's some stripping you can do for builds https://blogs.unity3d.com/2018/05/14/stripping-scriptable-shader-variants/
^I'm not sure if this was the blog I was looking for - I think there's been a few
Massively reduce Player build time and data size by allowing developers to control which Shader variants are handled by the Unity Shader compiler and inclu...
@long surge I don't really like solutions that involve changing code that tends to update frequently. It's a hard lesson I learned long ago 😛 Even an automated solution with Unity's package manager isn't great cos you either have to negate the package manager by bringing the package into the project directly, or you have to change the shader for all projects. I'm looking for a built in solution.
Nah, ain't possible then
@marsh frigate THAT's interesting, this may do what I need. Thanks!
@long surge "just", "easily" 🤣
I assure you there's nothing complicated. The guys themselves structured everything in a specific way so that all of the things can be easily unplugged or extended
maybe I misunderstand - do you mean to just remove #pragma multi_compile .... lines? Because then I agree it is not overwhelming... it does leave you with hard-to-upgrade-shaders though :´(
(and stuff in the UI will of course stop having an effect unless you modify that too...)
hmmmm
I have a bit odd question about unreleased tech again 😄
on HDRP PPv3 , can you at the moment tweak TAA parameters somewhere?
PP AA has moved to camera there
but I don't see any place where I can tweak TAA or FXAA settings there
wondering here if I need to toggle some value from code for these to pop up somewhere or if the settings for these simply haven't been done yet
@cobalt wing I think that may be in the High defintion Render Pipeline Asset settings
I cant seem to get PP to work in 2018.3 any ideas?
I created a profile, attached a volume to the Camera, nothing happening
@clear gazelle it's not in HD Asset or the Scene Asset, but this is still very WIP thing I'm testing, it's literally from latest SRP HDRP master (not released yet)
also to your issue
o ok sorry bud 😦
have you used PPv2 before?
uh i think this is the first time for me
i used the beta stuff before, when there was a diff way of doing it
I recommend you to put a new layer for postprocessing unless your project already has one
ok! was wondering what that was about
then put ALL three, camera, your postprocessing layer script and volume to have that postprocessing layer
Ok so what is the volume for?
and then from that postprocessing layer script you have on camera, make it target that "postprocessing" from layer list
this basically tells your PP only affect objects that are on that layer
you'd also want to set the volume to unbounded or global, whatever the setting is if you only ever have one
if you want to to transition between multiple volumes, you keep it unchecked
got ya
it's there so you can for example do transitions from one set of PP settings to another directly on your level
totally get it like UE
you don't need additional code for that
just put extra volume and tweak the location and settings
thanks a lot man! so helpful
right on i think this will do it
the initial setup of the PPv2 stack can be confusing but it actually makes a lot sense when you start to using it in practice
v cool
yea i set up the volume and was like....why isnt this working.
but yeah, if you have a global volume, you don't have to position it in any way, it'll apply to everywhere regardless
np
Is there a sharpening option for post processing, im missing this and a LUT input in the post processing stack
it should be there
@clear gazelle did you find it ?
I mean for the sharpening
look at the Color Grading effect
it has Tone setting which also includes Contrast
for LUT itself, it's bit more tricky
you have to change the whole Color Grading Mode to External
and I think they only accept 3d texture for it which you can only author on some specific packages, I this PS can do it, also some video editors
you will also lose all manual controls to color grading on Unity side if you do this
o damn!
thanks again, ok so the sharpening is the Tone trackball?
Im familiar with LUTs so I can workaround that
@cobalt wing Wasnt able to find sharpening, did see the contrast etc,
oh snap
I mixed up things
there might not be sharpening there, I think TAA itself has if you use it
no probs
but I dunno if it's really doing what you want
yeah, sorry for the side trip 😄
@clear gazelle for what it's worth, this does both the sharpening and LUT as extension to PPv2 but it's paid package: https://assetstore.unity.com/packages/vfx/shaders/fullscreen-camera-effects/sc-post-effects-pack-108753
(but sharpness seems to be on that TAA still, just checked)
I bought that pack on last sale myself, just tested it
sharpen on it goes past to what TAA sharpen goes (in strength)
but I dunno if you'd ever use such high values
awesome thanks!
Quick note that if you change the Color Grading mode to LDR you can then use a 2D LUT like in previous packages. This means however that all Color Grading effects will be applied in LDR, which changes some things, including no tonemapping support.
So if your camera is rendering in HDR, it will simply be clamped to LDR when Color Grading starts
We dont recommend this for HDR rendering basically, its mostly a path for mobile games.
One message removed from a suspended account.
heh, I always forget that man has about everything on his github 😄
I wanted to visit him when we went to Japan! But we didn't.. next time hehe
I honestly wonder how he managed to get so many repos there
most of them even seem like they are all genuinely his own work, has to be some machine instead of a man
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
❤ Keijiro. I used to work with him on PostFX team. Can confirm he is a machine.
PaRappa the Rapper was weird and fun too
as another topic
I recently found out that HDPR was never supposed to support SMAA even (which was one of the reasons why PPv3 doesn't have it)
I'm still amazed it worked through all the HDRP releases without official support 😃
Hi there, does anyone has a branch of the post processing stack which would support dynamic resolution properly by any chance ? Please : D Before I do it myself
@forest summit what you mean by that?
All the Render Textures used by the post processing stack don't use the "useDynamicScale" setting, so it doesn't scale properly when using it. Which breaks most effects (AO is probably the most visible)
oh I mean, the details are super scarce here
what you use dynamic resolution with, LWRP? some specific platform
I know that they've only lately added dyn res support on SRPs
are these even in any official releases yet(?)
We are not using SRP, we are on 2017.4, our platforms where we use dynamic resolution are all the main console (PS4, Xbox One, Switch)
ah
you probably know that tho 😄
Yeah : ( I was just hoping someone had fixed it on their own version and maybe had shared it
I'll do it myself if not : )
thanks though
it breaks for you even on the older PP versions?
I don't think it's a similar issue
also considering that PPv2 is supposed to be stable, I think you can file bug report on it breaking
We are using 2.0 I think though
that totally sounds like a thing that should work
Yeah, there may be reasons why not though. Also they tend to use the RenderTextureDescriptor API to get the RT, but that struct doesn't include a parameter for dynamic scale. I'm going to rewrite that tomorrow to allow it, see if it works. I may do a PR if I can spend enough time to test
Someone was creating post processing shaders with amplify shader editor? I was trying for an hour to get depth post processing effect but without success, for start I wanted at least something like https://catlikecoding.com/unity/tutorials/advanced-rendering/depth-of-field/circle-of-confusion/coc.png
Hello!
I am facing a problem with 2D sprites in a 3D space. I want to have 2D characters in my 3D game and I want to ignore Post-FX on the 2D sprites so I added them to a 2nd camera and set the 'Clear Flags' to 'Depth Only'. Everything works fine and I am able to ignore post-fx for 2D sprites.
But now, the problem is, the 2D characters overlap my 3D models. How do I fix this? I want to have some kind of layering between my 2D and 3D assets. Is this possible?
made a forum post about the PPv3 Bloom now that 5.3.0 is "almost out": https://forum.unity.com/threads/hdrp-5-3-0-ppv3-bloom-destroys-overall-image-quality.620272/
@broken birch I almost changed the title as it's bit aggressive but it's not technically wrong 😃
So agressive. Such hatred.
ikr
I don't really want to sound too hostile, people tend to get more defensive when they think you are attacking their work
and that's definitely not the intention (therefore I changed the title)
new bloom only has two 0-1 sliders, tint color + lens dirt effect texture and slider for that
aaaaand, they have new "advanced" button on the top that I missed
even still, it doesn't change the end result, but I think it does explain why it gets blurred
they only process the effect at half resolution
on advanced, you can toggle betwen half and quarter, quarter blurs it way more
here's one test at 0.1 intensity but with 300x emissive on the paint can
I can't get similar glow with more modest emissive values unless I crank up the bloom intensity and it trashes the whole image at 0.2 already
but yeah, I think I can work with this, it just seems tad hacky as I feel like I'm working around the blur issue by boosting the emissive (and probably need to do same with lighting)
it feels even as if bloom is now dictating your whole scene lighting, instead of being true postprocessing effect which you tweak afterwards
yeah all of this makes more sense if you've been following the probes/fps demo discussion
yeah, but didn't they do the exact opposite there?
to reduce the specular aliasing
anyway, really curious to see the next hd template
I hope the spend some time putting nice defaults to it
they have to redo the whole scene volume thing for it because old PP volume doesn't do anything anymore
it would make no sense for them to just copy/paste the PPv2 values to it as it clearly doesn't work right that way
current hd template on staging is from early december
same version on regular registry too
yeah they said repeatedly we have to redo all our settings and i've been preparing for this
So, I'm loosing on average 60 FPS w/ post processing engaged.
I'm using HDRP, a lot of the heavier stuff has been disabled (like SSS samples and volumetric resolution increase).
IMO @Olento a lot and he will quickly know why anything sucks for you
@cobalt wing is the master..
I haven't actually done extensive perf analysis for HDRP as it keeps evolving at such high speed. but right now few big perf hits are at least the volumetrics (which you've disabled already) and SSR if you use it
SSR is ridiculously expensive if you've upped the sample count from default
oh wait
you are using volumetrics? just not the high res?
that stuff definitely has it's toll
@honest loom
It seems to be post processing mainly..
and according to Unity, stack lit shader shouldn't even be used in realtime games (so I'd expect that to be super slow)
are you on 2018.3?
because 2019.1 just got new PP which has perf optimizations
new PPv3 is on 5.3.0 and 5.3.1 but those are still at staging as far as I know
expect to see 5.3.1 on package manager soon though unless Unity QA finds some issues on it
also, Unity 2019.1 betas are starting any moment now
so you may want to wait for getting the beta before jumping to 2019.1
I know the first beta is out internally already
Wait till 2019.1 goes beta then the packages will be guraranteed for PM
this should be this week so i've heard or next i guess
AFAIK they wanted something in time for the show
got lots of time to get 2019.1 stable
(or maybe not lol)
my educated guess would be end of this April
when I tried 19.1 alpha it was just crash after crash
definately recommend waiting for beta anyway
it was the first alphas that got me whenever i entered play mode - boom, due to a problem in HDRP with unlit
(with SRP batcher enabled)
anyway post shouldn't be reducing something from 130 to 60
seems a bit harsh just for post
mind you it was an empty ish scene so thats the true cost maybe
I'd disable all the post filters and then enable them one by one
mac/metal not optimised for it
to see where the costly bits are
the workbench one?
many post effects also have variations where some settings for the same effect are way cheaper
oh
and 5.x.x SRPs also run PP in async afaik
so that should help a bit as well
yeah keep those wavefronts occupied with little surfer bros
they run alot of other things async now too
only people who don't know his work aren't ;D
I'm really happy Unity managed to hire him
but their HDRP team is full of talented people
if he asks for a 2nd office they simply should do it
Seb needs one office to orgnise the staff of his main office
I still wonder about his working hours 😄
yeah, kept pushing commits last night through
if Unity let him go they will be truly pimple-brained
it's not like the guy hasn't defined pbr optimisations in most of the AAA games played today
Thing is HDRP is not able to render at maximum efficiency is it? I learned a while back there's a number of legacy bottlenecks involved
I'm mainly worried about the low end scalability, not really that much on what it can push at higher settings
when you start enabling fancier feats, those are not things you'll be able to do on built-in renderer anyway so it's not really fair comparison
but what kind of bottlenecks you mean?
Yeah, it's the workbench scene.
No idea, just saw staff mention that HDRP is still held back in a few ways with the unity engine.
It's just scattered hints there and here.
I'd not be surprised though, it's weaving all kinds of workarounds like piggybacking on existing components, working around existing limitations etc