Why would you want use to like your video? You showed an unpractical example of this feature by showing something that obstruct more than 75% of the screen.
@@realtimberstalker Yeah I understand what you mean, that's how I usually do it. Was talking more in terms of switching between cameras, and having certain cameras have their own specific UI elements to display things in 3d for example.
@@realtimberstalker You can take this another step and put all your UI in a separate scene with a separate camera. Then, simply open the scene as an additive. This way, you can simply reuse the UI elements without placing them in every scene or using prefabs.
Nice tutorial. Thanks. About 3 months ago, I upgraded a project to URP and had a heck of a time finding a tutorial on how to do this. I might need to revisit the project, now that I know. Thanks again
After spending a whole day trying to achieve some camera stacking in HDRP this video changed my mind in 2 minutes. I convert to URP right now =) Especially since I not only need 4 stacking layers by default, but also need additional dynamic cameras to render rotating ship parts in UI =) Thanks )
At 1:48 you can clearly see that the Post Processing (PP) of the UI (or cockpit in this case) is also affecting the Base Camera's output too since it will apply the PP on the whole merged image up to the overlay, with the PP of the overlay running last. The way I worked around this was (I am using URP): 1. I converted the UI camera from Overlay to Base, have the proper volume layer and culling set up for the UI only 2. Made a simple renderer asset for the UI camera, separate from the Main Camera 2. Output the result of the UI Camera to a Render Texture with an alpha channel 3. I wrote a custom Render Feature that takes the output of any camera and blits the render texture onto it 4. Instead of using the built in camera stacking of Unity, in the renderer of the Main Camera I added this render feature for merging the two The result is a perfect blend between the main camera's output and the output of the UI camera with two separate post processing effects. I don't know if this is overkill or if it kills performance, it does work fine for me with no significant performance drop. (Hope this helps anyone looking to solve this issue)
Thanks a bill'. Seeing the overlay camera here saved me a lot of loss time up some complicated alleys that I was going to use to accomplish what is so much better done with an overlay. Cheerios!
When I upgraded from 2020 to 2021, this broke for me. It was nowhere in the upgrade guide that I could see (I def could have missed it). This video was an amazing help! Thank you so much.
@@seanloughran6714 This is the main reason I've put off switching to URP or HDRP... it's so confusing and frustrating to figure out what's supporting what that it's more work than it's worth for me.
At 1:53 you toggle the post processing volume which affects ALL CAMERAS. Same happens in my project, it looks like there is no actual way of having an overlay camera with its own post processing volume. Tried everything and nothing seems to work, a post processing volume is actually no volume, its global. If anyone knows how to solve this please let me know
now solve the problem with lighting ... This rendering method is very good but there are some problems with lighting, objects drawn in the base camera will not draw shadows on objects that are drawn on the overlay camera
@@DenysShust No, not really. I tried camera stacking but that didn't really do anything. For my project, I wanted to use bloom so certain objects looked lit up. I decided in the end to make everything else in the scene dark (so they weren't really illuminated) and my target objects bright. You could possibly write a shader graph to create a glow, but that's beyond me.
@@fun2run4 Thanks man ! Glad to hear that not only I have this problem 😃. It seems like it is impossible to do post processing effect to only some of the objects.
I thought using camera stacking is discouraged in general due to high performance hit. Is stacking the cameras like in the video makes them somehow part of the same rendering pipeline each rendering at different stages so that there's less performance hit or it's a basic render-from-scratch-and-pass-the-buffers-to-next-camera approach?
Yes, it does a lot of optimisations behind the scene, such as overdraw reduction, but reusage of information like culling results is still in development
Camera stacking is not a problem as long as you manage your culling masks. Cameras become a problem if you don't specify culling masks, causing you to render game objects twice and doubling draw calls. There's a lot of really cool features that aren't easy (or are borderline impossible) without extra cameras, like special 3D HUD effects with lighting, or abilities that let you temporarily see objects through walls. An extra camera does have a little overhead even with no draw calls, but from what I've seen in my own compiler tests the overhead isn't much, so as long as you don't have an insane amount of cameras you won't notice a hit at all. Overall, if you're at the point where having 3 cameras on a player breaks your game then something else is seriously wrong already.
is camera stacking possible with hdrp? also seems like the player hud thing only responds to sunlight, since its in a physically different location, and not local point lights / indoor lighting. no?
Why does my overlay camera's post processing doubles total post processing effect? :/ Edit: Oh ok, I should disable post processing for the main camera, so the second camera's pp will affect to it's layer and other layers
Hi, cool video. I have one big struggle though. Changing post processing on one layer affects all the previous layers also. Is there a way I can separate it?
I didn't get the idea for this use case. Is this approach solves some problem I don't know about? And what problems it creates? Answering this questions would make this video much more valuable for me.
Most common use case are to render FPS viewmodel (Hand and gun) on overlay camera, this way you don't have to worry your viewModel will clip into world meshes
I followed this tutorial, but it seems if you don't turn on post processing on FPS camera, then the FPS object will not have post processing effect. However, if you turn on post processing on FPS camera. It will add post processing not only on FPS object but on all base camera objects as well. What could be wrong? 🤨🤨
The 30 dislikes are Unreal users, who can't stop their FPS weapons from clipping through the environment because their AAA engine doesn't have rendering layers.
If you set up your culling masks properly nothing should be rendered multiple times, the only disadvantage it has is that the processing of each camera regardless of what it's rendering requires some power.
@@realtimberstalker my personal POV is to avoid these not so important changes and work on essential things more, IDK, maybe to some, it's an important thing to have this system over the previous one.
It wasn’t quite clear for me if it can be clarified, the UI being rendered on top, is it a UI style render with UI Canvas element set to Camera mode (or World mode)? Or is it a custom ‘3D’ visible only through a ‘UI’ layer only camera? EDIT: or actually, just noticed, should open the project first and take a look.
Is there any way I can implement camera stacking while using a 2D renderer for my pipeline asset? I want to use camera stacking and 2d lights in my game, but have to choose one of the other.
You can do camera stacking and 2d lights in URP, you just can't have those stacked camera lights interact with other layers. You could do so if you did the render layers yourself, but it gets increasingly complicated
Have a question about additive scenes, if i want to seperate my UI scene from my player scene and load the these additively, can i still take advantage of this? or in this case, is splitting the UI scene from the game scene unnecessary?
By doing this you have total control of how each camera renders stuff. If you want to adjust certain aspects of one camera, you can do that without altering the other ones. Don't know how that would work with UI tho.
Chris -0 You do realize there are very few drawbacks, right? Cameras render in layers, and UI is done last. You’re just telling another camera to do that layer. It’s still the same amount of layers.
camera overlay doesnt work with 2d lights of the URP for some reason. The 2d lights are shown in the overlay camera as well even if u change the culling mask and the layer of the camera. Am I missing something or is this an actual bug?
Just a friendly reminder that URP STILL doesn't support camera stacking on the 2D Renderer, so if your object is 2D Renderer based or you have a 2D renderer somewhere on the renderer stack, it won't work.
To answer my own question because i randomly stumbled upon my comment here: "Important note: In this version of URP, Camera Stacking is not supported when using the 2D Renderer or when using the VR Multi Pass mode. Support for these will be added in upcoming versions of URP." - source: urp documentation page
How did you get the post process to NOT affect all cameras? (I noticed you had "post-process" checked on for both the base and cockpit, but the Cockpit post-process was only affecting the cockpit. How?
I always wanted to know. Is there a way to stack cameras of multiple perspectives? Like showing one object in Orthographic and the others in perspective.
I don't understand, besides the post processing soley for the cockpit model. Couldn't you achieve the same just by attaching the model to the camera via script and a normal screenspace overlay?
really late, but then the cockpit could clip through walls and other thigns in the world, the point of camera stacking is to ensure the things on seperate camera layers are always rendered on top. also others have mentioned it, but you also can control the camera layer's parameters (like fov), seperate from the base camera's
Is it possible to make unity render the first pixel skip the second render the third and so on Having a game run at the GPU requirements of roughly half the screens native resolution while looking better than just outright upscaled would be cool By all means it looks off but its way nicer to play something like this than it is to play with lower frame rates
i did this but i have 1 issue I am using this to render the guns in my game, the issue i am having is that the player is able to see the guns through the wall when the guns are not being held by the player. anyone pls help
Smash that *thumbs up* button for more URP tutorials! =)
i see you like titanfall
And HDRP. I don't understand the light and shadow something doesn't fit and looks greyish.
🙌🙌🙌🙌👏👏👏👏😁
Why would you want use to like your video? You showed an unpractical example of this feature by showing something that obstruct more than 75% of the screen.
@@stuwustudio because they make maybe new tutorials with deeper informations, so its to basic, i do the same since days ago, without a tutorial...
That's an awesome feature, I like the way we can display UI in a separate camera, which makes it easy to toggle them :)
Hey unity look at what you are teaching your users... Great job there!
You don’t need individual cameras. Just put each of the ui’s under a game object and toggle the game objects.
@@realtimberstalker Yeah I understand what you mean, that's how I usually do it. Was talking more in terms of switching between cameras, and having certain cameras have their own specific UI elements to display things in 3d for example.
HadiLePanda Ok. Thanks for clarifying.
@@realtimberstalker You can take this another step and put all your UI in a separate scene with a separate camera.
Then, simply open the scene as an additive.
This way, you can simply reuse the UI elements without placing them in every scene or using prefabs.
Nice tutorial. Thanks. About 3 months ago, I upgraded a project to URP and had a heck of a time finding a tutorial on how to do this. I might need to revisit the project, now that I know. Thanks again
It looks like it was a feature that was just released in 2020.1
@@joshuagoldshteyn8651 I am still using Unity 2019 and the URP Camera stacking works here too :)
@@MQNGameDev Oh dang I wish I knew about this earlier too then so my weapons would stop clipping into the walls
@@joshuagoldshteyn8651 I ended up ditching the mini-map in a couple projects. Glad we know how to do it now :)
After spending a whole day trying to achieve some camera stacking in HDRP this video changed my mind in 2 minutes. I convert to URP right now =) Especially since I not only need 4 stacking layers by default, but also need additional dynamic cameras to render rotating ship parts in UI =) Thanks )
once i see the effects, titanfall2 jumped into my head. its amazing!
Thanks, definitely gonna try this out! Would be nice to see how to create the distortion effects as well
At 1:48 you can clearly see that the Post Processing (PP) of the UI (or cockpit in this case) is also affecting the Base Camera's output too since it will apply the PP on the whole merged image up to the overlay, with the PP of the overlay running last. The way I worked around this was (I am using URP):
1. I converted the UI camera from Overlay to Base, have the proper volume layer and culling set up for the UI only
2. Made a simple renderer asset for the UI camera, separate from the Main Camera
2. Output the result of the UI Camera to a Render Texture with an alpha channel
3. I wrote a custom Render Feature that takes the output of any camera and blits the render texture onto it
4. Instead of using the built in camera stacking of Unity, in the renderer of the Main Camera I added this render feature for merging the two
The result is a perfect blend between the main camera's output and the output of the UI camera with two separate post processing effects.
I don't know if this is overkill or if it kills performance, it does work fine for me with no significant performance drop.
(Hope this helps anyone looking to solve this issue)
Thank you very much, I'm from Brazil and I searched a lot for this tutorial and I finally found it. Thank you very much.
Thanks a bill'. Seeing the overlay camera here saved me a lot of loss time up some complicated alleys that I was going to use to accomplish what is so much better done with an overlay. Cheerios!
When I upgraded from 2020 to 2021, this broke for me. It was nowhere in the upgrade guide that I could see (I def could have missed it). This video was an amazing help! Thank you so much.
I think we've been doing that for years with the old renderer.
exactly lol , not sure if the per-cam postProcessing was possible before though
Yap, but the features been missing since the release of LWRP/URP until now and the setting seem to be clearer compare to built-in pipeline
wasnt supposed to be advertised as a new feature, but just showing a new way of doing it... congrats on completely missing the ball though
that inner display in the cockpit .... wow!
This is awesome!!! And so easy to do!
Would be also cool to see more tutorials on hdrp.
thank you soo much, i was about to lose it because i couldnt get this to work
Be aware that each extra camera adds overhead in the form of culling time and more
I love this feature! It just saved my day :D Thank you Unity! ❤
this lady helped me a lot, thank you ^^
This is great. Thank you for the info, explanation and the demo file to look at
Good that this feature is now in URP
Okey we need HDRP version of this tutorial now
Just, wow! You did my week!
Welcome aboard, Pilot
Cool, now bring it to HDRP.
Just tried this in HDRP and you're right, it's not available. Why is everything so split and mismatched???
@@seanloughran6714 This is the main reason I've put off switching to URP or HDRP... it's so confusing and frustrating to figure out what's supporting what that it's more work than it's worth for me.
This is the reason I subscribed :)
Very cool thank you for sharing!!! Please more URP Tutorials :D
At 1:53 you toggle the post processing volume which affects ALL CAMERAS. Same happens in my project, it looks like there is no actual way of having an overlay camera with its own post processing volume. Tried everything and nothing seems to work, a post processing volume is actually no volume, its global. If anyone knows how to solve this please let me know
Exactly this. Official tutorial giving misinformation. There is likely a reason why this was not presented but only verbalized.
Dealing with the same over here!!
This solve my problem in my fps game thanks!
NIIIIIIIIIICCCCEEEEE. You are my Advil. Thank You so much, I subbed.
Exactly what i needed, thanks
Greyed out in URP for Overlay
Thanks boss, good video.
Thanks for this short video.
Sorry, I can't get it. What is the difference of this approach compared to the previous one of setting Clear flags and Depth?
actually this feature been missing since the release of LWRP/URP, but now they finally added the camera stacking to URP
@@walaoRIKi Oh, I'm so lost of new changes :D
OMG thanks for this!!!!!!!
now solve the problem with lighting ... This rendering method is very good but there are some problems with lighting, objects drawn in the base camera will not draw shadows on objects that are drawn on the overlay camera
How do I have the PP only affect a single layer? I've created a bloom PP, but it affects everything else in the scene.
s. same here. couldn't find a solution. someone help on this pls...
@@macaronia007 Did you find the solution ?
Did you find the solution ?
@@DenysShust No, not really. I tried camera stacking but that didn't really do anything. For my project, I wanted to use bloom so certain objects looked lit up. I decided in the end to make everything else in the scene dark (so they weren't really illuminated) and my target objects bright. You could possibly write a shader graph to create a glow, but that's beyond me.
@@fun2run4 Thanks man ! Glad to hear that not only I have this problem 😃. It seems like it is impossible to do post processing effect to only some of the objects.
Thanks a lot! it helped with my issue
I thought using camera stacking is discouraged in general due to high performance hit.
Is stacking the cameras like in the video makes them somehow part of the same rendering pipeline each rendering at different stages so that there's less performance hit or it's a basic render-from-scratch-and-pass-the-buffers-to-next-camera approach?
It's a sad day when you can't even enjoy watching Unity release videos anymore, its just fluff, you go to touch it and it turns to mush...
Yes, it does a lot of optimisations behind the scene, such as overdraw reduction, but reusage of information like culling results is still in development
@@neonage6294 Sounds good. I'll take it that for now I should avoid stacking in production projects.
@@Chris_t0 There's nothing fluff about rendering layers. They're useful, if not necessary, in some cases. You can't really make an FPS without them.
Camera stacking is not a problem as long as you manage your culling masks. Cameras become a problem if you don't specify culling masks, causing you to render game objects twice and doubling draw calls. There's a lot of really cool features that aren't easy (or are borderline impossible) without extra cameras, like special 3D HUD effects with lighting, or abilities that let you temporarily see objects through walls.
An extra camera does have a little overhead even with no draw calls, but from what I've seen in my own compiler tests the overhead isn't much, so as long as you don't have an insane amount of cameras you won't notice a hit at all. Overall, if you're at the point where having 3 cameras on a player breaks your game then something else is seriously wrong already.
thanks this was helpful!
is camera stacking possible with hdrp?
also seems like the player hud thing only responds to sunlight, since its in a physically different location, and not local point lights / indoor lighting. no?
HDRP will have Compositor tool as more efficient replacement
Neonage Hello, could you explain how to use it?
@@yuier2883 I don't use HDRP, my 7 year old laptop would not allow that :P
Why does my overlay camera's post processing doubles total post processing effect? :/ Edit: Oh ok, I should disable post processing for the main camera, so the second camera's pp will affect to it's layer and other layers
I disabled the pp effects for the main camera, and checked the UI camera's pp but it didn't work.
this is amazing
Hi, cool video. I have one big struggle though. Changing post processing on one layer affects all the previous layers also. Is there a way I can separate it?
so helpful! Liked
Thanks, Christine!
Excellent. Thank you!!!!
I didn't get the idea for this use case. Is this approach solves some problem I don't know about? And what problems it creates? Answering this questions would make this video much more valuable for me.
Most common use case are to render FPS viewmodel (Hand and gun) on overlay camera, this way you don't have to worry your viewModel will clip into world meshes
@@awesomeniac well, not that there is other ways to solve it,but at least a good use case.
@@dadlord689 True, there's a lot of way to render viewmodel and avoid clipping. But that is one of the easiest example i can find at the moment :p
@@awesomeniac Thanks
@@dadlord689 It's the proper way to do it.
I followed this tutorial, but it seems if you don't turn on post processing on FPS camera, then the FPS object will not have post processing effect. However, if you turn on post processing on FPS camera. It will add post processing not only on FPS object but on all base camera objects as well.
What could be wrong? 🤨🤨
I had the same thing. Let me know if you find an answer.
make a separated object and store the volume in that object, and then set the object's layer to the second camera layer.
The 30 dislikes are Unreal users, who can't stop their FPS weapons from clipping through the environment because their AAA engine doesn't have rendering layers.
what are performance like compared to built in forward?
Like always it depends
@@AngryApple What about under the same circumstances?
@@Mehrdad995GTa than it depends on the circumstances :D
Finally Thank you! but does this work for 2019.3.f1? or does this only work for versions 2019.4 upwards?
I think it requires unity 2020.1
Finally! Thank you!
Is the scene render twice? Like VR rendering the scene for LR camaras. Drags down performances?
From my experience I do get noticeable performance hits.
If you set up your culling masks properly nothing should be rendered multiple times, the only disadvantage it has is that the processing of each camera regardless of what it's rendering requires some power.
mehrdad But isn’t that the point? To let you process them differently before they are added?
@@realtimberstalker my personal POV is to avoid these not so important changes and work on essential things more,
IDK, maybe to some, it's an important thing to have this system over the previous one.
Yeah. Second thought. Less performance impact than volume scattering anyway.
now i can make my own titanfall 2 game
It wasn’t quite clear for me if it can be clarified, the UI being rendered on top, is it a UI style render with UI Canvas element set to Camera mode (or World mode)? Or is it a custom ‘3D’ visible only through a ‘UI’ layer only camera?
EDIT: or actually, just noticed, should open the project first and take a look.
Is there any way I can implement camera stacking while using a 2D renderer for my pipeline asset? I want to use camera stacking and 2d lights in my game, but have to choose one of the other.
You can do camera stacking and 2d lights in URP, you just can't have those stacked camera lights interact with other layers. You could do so if you did the render layers yourself, but it gets increasingly complicated
ah so it was the second renderer I was missing... used to work with just marking layers pre-urp
Rarest things on earth
1 finding diamond
2 *Unity hearts your comment*
Have a question about additive scenes, if i want to seperate my UI scene from my player scene and load the these additively, can i still take advantage of this? or in this case, is splitting the UI scene from the game scene unnecessary?
What's the advantage to just making these overlays ui elements?
Thanks for any kind of help :D
By doing this you have total control of how each camera renders stuff. If you want to adjust certain aspects of one camera, you can do that without altering the other ones. Don't know how that would work with UI tho.
The advantage is that your room will be hotter from all that extra wasted power going to rendering your UI on a seperate camera.
Chris -0 You do realize there are very few drawbacks, right? Cameras render in layers, and UI is done last. You’re just telling another camera to do that layer. It’s still the same amount of layers.
camera stacking seems to be like the older style renderpass
For some reason its not working
Thank you so much!!!
Not existing for hdrp, if u want to exclude UI from postprocessing -> use unlit shader and choose after processing and depth test disabled
camera overlay doesnt work with 2d lights of the URP for some reason. The 2d lights are shown in the overlay camera as well even if u change the culling mask and the layer of the camera. Am I missing something or is this an actual bug?
thanks a lot sis :)
Cool tuto. Now, do the same for HDRP.
Why would you do this for your UI rather than a canvas? Also, why would you not just use render targets for the cockpit?
Just a friendly reminder that URP STILL doesn't support camera stacking on the 2D Renderer, so if your object is 2D Renderer based or you have a 2D renderer somewhere on the renderer stack, it won't work.
Does this work in VR as well?
It's a feature of the Universal Render Pipeline. I believe the URP is compatible with VR, so it should be!
Not entirely sure though.
To answer my own question because i randomly stumbled upon my comment here: "Important note: In this version of URP, Camera Stacking is not supported when using the 2D Renderer or when using the VR Multi Pass mode. Support for these will be added in upcoming versions of URP." - source: urp documentation page
Это круто!
AWESOME!!!, BUT CAN YOU DO THIS WITH VUFORIA´S AR CAMERA?
How did you get the post process to NOT affect all cameras? (I noticed you had "post-process" checked on for both the base and cockpit, but the Cockpit post-process was only affecting the cockpit. How?
I dont see a render type overlap on my game
You need to install URP to Unity
Id this more performant
WIth this setup now, how can we set particles in front of UI?
Thanks u so much !!!
Where's the target display dropdown in unity 2020?
Does this work with cinemachine as well?
What about going inside the buildings, wouldn't shading be off if your weapons are hanging somewhere directly under light?
Thank you
I always wanted to know. Is there a way to stack cameras of multiple perspectives? Like showing one object in Orthographic and the others in perspective.
Absolutely
Nice!
I don't understand, besides the post processing soley for the cockpit model.
Couldn't you achieve the same just by attaching the model to the camera via script and a normal screenspace overlay?
really late, but then the cockpit could clip through walls and other thigns in the world, the point of camera stacking is to ensure the things on seperate camera layers are always rendered on top.
also others have mentioned it, but you also can control the camera layer's parameters (like fov), seperate from the base camera's
How is this on performance?
I was such a fool not doing this.. it's also a fix for the solid color which might appear
Do we have camera stacking in HDRP?
the yellow spaceship UI is a texture2d?
how i can do this with camera render URP 2d and 3d?
I DON'T HAVE a Render Type in Camera component, even tho I'm using post processing( so I gues I have URP) (v. 202.2.7f1)
Did you find a fix?
@@HugoCortell yes, but give me some time to explain, because at the moment I don’t remember how I got it to work
@@GnuSnu No need, I actually just deleted my project and spent the last 4 hours re-importing old files into URP and remaking the rest.
больше при большое спасибо, тебе добрый человек, мне глупому камушку все объяснила! 2 часа решение искал
does camera stack work with SteamVR?
What's the right way to do this with HDRP?
Is it possible to make unity render the first pixel skip the second render the third and so on
Having a game run at the GPU requirements of roughly half the screens native resolution while looking better than just outright upscaled would be cool
By all means it looks off but its way nicer to play something like this than it is to play with lower frame rates
Thank you so much!
Why is it so hard to get this information!!??
how to add camera stacking with script ,, URP guide is not enough
How is this more performant than having 1 camera? 🤔
yeah
i did this but i have 1 issue
I am using this to render the guns in my game, the issue i am having is that the player is able to see the guns through the wall when the guns are not being held by the player.
anyone pls help
Why Post Processing affects only 2nd camera? What would happen if I activated Post Processing in the 1st or 3rd camera's settings and why?
idk post processing working 4me not with 2 cameras, what do i wrong?
Titanfall 2 vibes