driver updated 4.14 takes 50% longer to render
alan bard newcomer
Posts: 2,230
updated drivers to solve the black renders in 4.14 (they were fast though).
---
however
4.12 with new driver 2 min 05.79 secs to render
4.14 with new driver 2 min 59.87 secs to render
---
gee, I was expecting the new version to do better than the old one.
----
scene created in 414 then opened in 412 same camera same sky
---
first one is 412 second is 414
----
And as a side effect to restarting the system for the new video driver
Dim has decided there is nothing new to download and it's even ignoring the 500 updates it wanted to download.
Hopefully this is some kind of a server issue?
Daz 3D is part of 
Connect
DAZ Productions, Inc.
7533 S Center View Ct #4664
West Jordan, UT 84084
Licensing Agreement | Terms of Service | Privacy Policy | EULA
© 2025 Daz Productions Inc. All Rights Reserved.
Comments
Note that if you keep a render you've already done open, the following ones will render faster, especially if the scene is reasonably small.
Alan,
I just did a test on my machine after reading your post and I honestly got the opposite result.
Win 10 1909 x64 - Zotac AMP Extreme 980TI (x2) - Nvidia Driver 457.30
Test scene was the first image shown here - https://www.artstation.com/artwork/w8YEJ5- but rendered to 1000 samples only each time
Daz Studio 4.12.2.7 - 7m 11s
Daz Studio 4.14.0.8 - 6m 14s
now I will have to run a couple more tests ... to see what happens next time
---
first pic 4.12 second 4.14
----
I have a 980ti and a titanX
---
but these were the first two I did.,
----
6 vs 7 minutes either way is close
but the basically 2 to 3 minute dif could be a half hour more on what used to be a one hour render.
---
so off to test.
---
now I will have to run a couple more tests ... to see what happens next time
---
first pic 4.12 second 4.14
----
I have a 980ti and a titanX
---
but these were the first two I did.,
----
6 vs 7 minutes either way is close
but the basically 2 to 3 minute dif could be a half hour more on what used to be a one hour render.
---
so off to test.
---
since one was in 4.14 and the other 4.12 ... one would expect the rendering to run differentr
But I do use an old 412 that will open while 414 is open ... can they use the same scene file at the same time?
I've never consiously tried that ... with two 412s I've often rendered from one and worked in the other
did a rerun (three, actually ... used a larger more complex scene.
---
open 414 load scene render scene 54m 29 sec close 414 wait ten minutes
open 412 load scene render scene 38m 29 sec close 412 wait ten minutes
open 414 load scene render scene 50m 42 sec
----
gpu ram was up to 8g during renders dropped back to 2g when daz was released.
---
but it's still at about a 25% increase in rendering times.
---
For now even if I work in 414 I may render in 412... and of course haven't checked the render time in the beta.
---
Art Station is throwing a 404 on your render.
Perhaps your two versions have different render settings for the hardware. You said you had to different cards. I assume they are different speeds and VRAM. Which ones are being used in each version of DAZ for rendering? There's a hardware tab under the Render settings. Use only your most powerful card as they don't combine VRAM (the data is duplicated to all cards, so if one runs out, they all run out). If you run out of VRAM for a scene, it drops to CPU rendering which increases rendering time significantly (check your Task Manager for maxed CPU usage). Hopefully this helps in some way.
same file opened in both programs same settings same size..
both cards are used ..
which is a question because it increases both cars while rendering and even on a render that ran the titan up to 11888g .. it still used 5g on the 980
so i guess as long as you don't go over the data limit of the lower card it uses both and uses the larger one more for processing
--
but scene was created in 412 aways back .. then opened and saved in 414 so it's the same scene and same settings
the original render of the new building was created in 414 and same settings both cards marked for use
Sorry, my mistake - https://www.artstation.com/artwork/w8YEJ5
For future tests, set the max samples limit to 1000. That will be easier than letting a render run for an hour or more.
another set of tests
---
opened a file made in 412 in 414 rendered in 16m 20s
the saved it and open in 412 ...
interesting because this time it turned the sky dome on when it saved (was not active in the render).
Giving me the second render which I thought was running in filament? Until I discovered the sky dome change
and the third render with skydome turned off
and it was 16m 25s
---
which is right on . in that 414 and 412 took the same time.... with just scene lighting (lava, a ball light with the lave shader burning bright and a small light behind the tank hatch cover.
---
Added the captioned image .. sometimes makes my artwork make sense.
---
itereations converge out at about 800 ... 4k x 2k renders
Probably, but my point is that you have to take this into consideration when you are testing. If you do one test where a window with a previous render of the same scene is open and then another without any open render windows you may get misleading results.
No idea, never tried that.
all of these were run one at a time with one instance of scene running.......
open 414 load scene, render, close, wait , open 412, load scene, render, close, wait, open daz again. repeat as necessary.
---
this backup folder of my daz scene and render files probably still has some duplicate files but then I need to (note to self) need to add all the 2020 stuff and make sure it's all backed up to multiple places.
I have done alot of render and scene work