Timeline for What are the common rendering optimization techniques for the geometry pass in a deferred shading renderer?
Current License: CC BY-SA 3.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 9, 2013 at 5:29 | history | edited | concept3d | CC BY-SA 3.0 | added 2 characters in body |
| Nov 26, 2013 at 11:49 | comment | added | concept3d | @MickLH edited the answer I hope it's more accurate now. | |
| Nov 26, 2013 at 11:47 | history | edited | concept3d | CC BY-SA 3.0 | added more details regarding deferred shading |
| Nov 26, 2013 at 10:31 | comment | added | MickLH | I'll retract my downvote if you make it clear that these optimizations are actually the least effective for a deferred renderer, as it stands you haven't shown him/her + googlers that the performance issue has nothing to do with deferred shading. | |
| Nov 26, 2013 at 10:26 | comment | added | concept3d | though I agree on the part that some optimization should occur on the lighting calculations, and if light calculations was not the dominant then deferred is the wrong way to go. but again that was not his problem. | |
| Nov 26, 2013 at 10:23 | comment | added | concept3d | it's really stupid to build a deferred engine without frustum culling for example, so the engine will process for example (100 million vertex) just to be able to submit the result to the g-buffer. The differed shading solves a different problem, which was not his problem, his problem was submitting all the geometry to the pipeline. | |
| Nov 26, 2013 at 9:59 | comment | added | concept3d | @MickLH Unfortunately, apparently you didn't read the question, his problem was mainly that he iterates over the whole scene everytime, and he didn't mention any bottleneck regarding the deffered shading. I mentioned at first that deffered shading solves the passes explosion problem when having a lot of lights/materials. But I then added that those are a must optimizations for any engine regardless of the shading technique forward or deffered. Regarding that these are the exact issues that the technique migitates I strongly disagree, I can't tackle the whole points here (follows) | |
| Nov 26, 2013 at 9:38 | comment | added | MickLH | Sorry, these have nothing to do with deferred shading, they are actually the exact performance issues that the technique mitigates effectively and therefore the least useful optimizations to make, the focus should be on the lighting pass(es) because if the lighting cost is not the dominant time-taker, deferred shading is probably the wrong choice. | |
| Nov 26, 2013 at 9:24 | history | edited | concept3d | CC BY-SA 3.0 | fixed some spelling and structure errors |
| Nov 24, 2013 at 21:56 | history | edited | concept3d | CC BY-SA 3.0 | added 2 characters in body |
| Nov 24, 2013 at 15:37 | history | edited | concept3d | CC BY-SA 3.0 | deleted 3 characters in body |
| Nov 24, 2013 at 11:07 | vote | accept | deniz | ||
| Nov 24, 2013 at 10:59 | history | answered | concept3d | CC BY-SA 3.0 |