is gdebugger wrong or do I really have 300gb in renderbuffers?
#1 Members - Reputation: 705
Posted 03 December 2012 - 12:53 PM
you know you program too much when you start ending sentences with semicolons;
#2 Members - Reputation: 692
Posted 04 December 2012 - 07:18 AM
I get false values too, I use a 1024x1024 RGBA8 cubemap texture. This should take:
1024 (width) x 1024 (height) x 6 (sides) x 4 (bytes per pixel) ~ 25 MB
instead it says 148 MB... So it is definitely wrong.
As far as I've seen it, it report correct values for 2D textures, but false ones for pretty much everything else.
#3 Moderators - Reputation: 14301
Posted 04 December 2012 - 07:43 AM
That sounds possible. What resolution are you using? At ~1080p with your described setup, and regular depth buffers, you would be somewhere around the ~280MiB mark...I think there may be a glitch in gdebugger where it is reporting MB as GB
Why do you need a multisampled G-Buffer and a non-multisampled G-buffer?
Edited by Hodgman, 04 December 2012 - 07:47 AM.
#4 Members - Reputation: 705
Posted 04 December 2012 - 02:33 PM
That sounds possible. What resolution are you using? At ~1080p with your described setup, and regular depth buffers, you would be somewhere around the ~280MiB mark...I think there may be a glitch in gdebugger where it is reporting MB as GB
Why do you need a multisampled G-Buffer and a non-multisampled G-buffer?
My resolution is 1280x1024 and the reason I need both multisampled and non multisampled is because in order for the multisampled texture to be displayed it needs to be blitted to a normal buffer.
you know you program too much when you start ending sentences with semicolons;
#5 Moderators - Reputation: 14301
Posted 04 December 2012 - 05:31 PM
Regarding MSAA, you can only display a single target, so there should be a need to resolve 3 MSAA targets for display.
Is your 3xrender-target a g-buffer for deferred rendering? If so, averaging MSAA samples is non-sensical -- If you've got a red pixel facing the camera at 0m, and a green pixel facing 90deg at 1000m, then their averaged appearance is not the same as the appearance of a purple surface facing 45deg at 500m.
#6 Members - Reputation: 705
Posted 04 December 2012 - 10:38 PM
At that res, I can't reconcile your description with the memory usage; must be a bug like mentioned earlier...
Regarding MSAA, you can only display a single target, so there should be a need to resolve 3 MSAA targets for display.
Is your 3xrender-target a g-buffer for deferred rendering? If so, averaging MSAA samples is non-sensical -- If you've got a red pixel facing the camera at 0m, and a green pixel facing 90deg at 1000m, then their averaged appearance is not the same as the appearance of a purple surface facing 45deg at 500m.
yep its a g-buffer but its for ssao not deferred rendering, I suppose it doesn't have to be multisampled since all it stores is normal vectors.
you know you program too much when you start ending sentences with semicolons;






