Camera 1 is set to clear skybox, and draws a rotating skybox.
Camera 2 is set to clear depth, to draw the rest of the scene.
Both cameras have a blur effect (turned on occasionally) and a bloom effect added to them. I thought I could get away with having the image effects on just the depth camera, but that resulted in all sorts of madness where the skybox was supposed to be. Not sure why there was garbage, as the skybox camera I assume draws into the screen buffer first and there would actually be something there.
So I have 3 questions in regards to this.
Is this the proper way to set things up for full screen effects and multiple cameras, having the effect on all cameras, or can you somehow get away with an effect on a single camera?
It seems that at times I’ll be using 4 render textures, one after the other. Is this something to be concerned about? Will it make older cards (hi GMA 950) have a conniption?
Is there some page amongst the docs where the rendering pipeline in Unity is documented? What happens when, where and why. What to be aware of, etc? My knowledge is somewhat weak in this area so would like to learn more.
as i understand it in basic terms, the blur effect is applied to the image that’s about to be sent to the display. so it’s render everything to an image then adjust pixels per the post effects and then send to the display. if that’s correct, telling it to do two different things could cause unexpected results. post a simple project that shows the exact prob.
Attached is an example project for illustration. Click the mouse button to trigger the temporary motion blur in the last two scenes.
The very last scene, is the final configuration I settled on. It seems to work, but not sure if it’s the proper way to go about things. Like I said, I didn’t find much documentation or posts relating to overlapping image effects and multiple cameras!
yeah the way you set it up is what i’d do. believe it will require pixel shader 2 which the 950 supports iirc. you can see the results by using graphics emulation. performance depends on your project as a whole. too hard to say whether this one aspect will be the choke point. don’t think there’s documentation on rendering in this context.
Thaks pete. Yeah it appears to work, I just was puzzled. I initially thought the glow on the second camera only to be sufficient, assuming the skybox was already in the screen buffer. I certainly didn’t expect garbage to show up.
I sure would like to see some documentation on how all this fits together. I don’t feel comfortable knowing something works but now knowing why.
i don’t know enough about it to be definitive but my moto is: if it runs well/looks good, ship it. something’s are ok being black boxed but yeah docs on stuff like this would be useful. they might even exist. i just haven’t noticed them.
