- Jul 10, 2022
- 54
- 1,135
I can buy myself flowers..
- Mar 22, 2018
- 227
- 7,827
- Nov 9, 2016
- 170
- 1,510
You don't have permission to view the spoiler content.
Log in or register now.
Last edited by a moderator:
- Feb 2, 2022
- 92
- 404
- Nov 10, 2022
- 183
- 5,585
Carmine
Please consider supporting !
Please consider supporting !
You must be registered to see the links
.
You must be registered to see the links
.
You must be registered to see the links
- Feb 19, 2023
- 52
- 374
- Apr 9, 2018
- 255
- 6,109
- Dec 24, 2020
- 592
- 12,849
- May 11, 2018
- 35
- 355
- Mar 11, 2023
- 13
- 266
Can anyone explain the connection in render times to me?
1280x720(HD) = 921.600 pixels
1920x1080(FHD) = 2.073.600 pixels
3840x2160 (UHD) = 8.294.400 pixels
Logically, with the same scene, same lights, same everything; you would assume that it first doubles the time from HD to FHD and then quadruples from FHD to UHD
The image filenames have a description if you hover over them. 1K took 3m42s to render, then it approximately doubles to 7m30s for the FHD render. THEN it fucking just explodes in render time and I had to stop it manually @ 85% convergence after 1h26m... how does this compute?
I do not have CPU fallback activated, so normally it just puts out a black screen, or it doesn't start up and I can see it in the error log. But today the GPU just somehow ran with only ~55% load during the whole UHD render. Is it something in between the NVIDIA drivers and DAZ software, or is there something I am missing?
Following the logical connection in pixel sizes and render times of the first two renders it should've only taken around 30 minutes for 95% convergence and not 1h26m for 85% :-/
Daz is so fucking weird sometimes; and it is infinitely frustrating both rendering normally and animating and making sense of some of the things in the programme. How old was DAZ again? xD (and yes I have been told it is a small team, but I feel like no progress is made at all).
1280x720(HD) = 921.600 pixels
1920x1080(FHD) = 2.073.600 pixels
3840x2160 (UHD) = 8.294.400 pixels
Logically, with the same scene, same lights, same everything; you would assume that it first doubles the time from HD to FHD and then quadruples from FHD to UHD
The image filenames have a description if you hover over them. 1K took 3m42s to render, then it approximately doubles to 7m30s for the FHD render. THEN it fucking just explodes in render time and I had to stop it manually @ 85% convergence after 1h26m... how does this compute?
I do not have CPU fallback activated, so normally it just puts out a black screen, or it doesn't start up and I can see it in the error log. But today the GPU just somehow ran with only ~55% load during the whole UHD render. Is it something in between the NVIDIA drivers and DAZ software, or is there something I am missing?
Following the logical connection in pixel sizes and render times of the first two renders it should've only taken around 30 minutes for 95% convergence and not 1h26m for 85% :-/
Daz is so fucking weird sometimes; and it is infinitely frustrating both rendering normally and animating and making sense of some of the things in the programme. How old was DAZ again? xD (and yes I have been told it is a small team, but I feel like no progress is made at all).
- Sep 26, 2018
- 3,040
- 3,135
- Jul 25, 2021
- 582
- 1,189
Several things effect render times, not just size of image, lighting, textures, figures, levels of detail, how much is off screen... Is it a Full Moon, half moon or an american football team moon....
In your case the body hair areas, arms, chest, back, legs is as much detail as any 5 figures, plus the figures & set, so in effect it's 7 figures plus set, lighting, detail levels(1k, 2k, 4k, 8k or 16k for the set & table)... Then there's how many morphs did you use, plus how many sets of morphs do you have loaded for whichever Genesis figures you've used, if it's 10 sets of morphs loaded for females, & 7 for males, that's 17 sets of morphs carried even if you haven't used them. The only way to only have those morphs used, is to save as a 'subset', & reload as saved parts...
Last edited:
- Mar 16, 2018
- 4,126
- 29,361
