Daz Studio Pro BETA - version 4.9.3.166! Release Candidate 5 (*UPDATED*)

1202123252629

Comments

  • The Darkside Style changes are nice.  It's much easier on the eyes in my opinion.

  • mjc1016 said:

    And I never understood the oft-repeated explanation of "the new Qt version in 4.0 doesn't let us do that any more".

    Don't worry, it will be just like that for 5...because Qt 5 makes a bunch of things that were in Qt4 very obsolete...but it does bring some fun new features...like support for modern displays/resolutions, so the various problems like text not resizing correctly and pop up windows being off screen will be reduced/eliminated.

    It's apparently simple-seeming things I meant; e.g. in the Scene Tab, when you click on a folder containing other folders, in D|S3 the containing folder snaps to the top of the tab view so you can see as much of the contents as possible, while in D|S4 it doesn't. Or scrolling with the mouse scroll wheel in a tab (the Content Library is most annoying); in D|S3 the scroll goes in discrete clunk-clunk-clunk steps. In D|S4 the scroll is smooth and goes wheeeeeeee way past the next item I wanted to see.

    Might not be much, I know, but all those UI changes landing on us at once put me off switching completely over to D|S4 for a long time.

  • mjc1016 said:

    And I never understood the oft-repeated explanation of "the new Qt version in 4.0 doesn't let us do that any more".

    Don't worry, it will be just like that for 5...because Qt 5 makes a bunch of things that were in Qt4 very obsolete...but it does bring some fun new features...like support for modern displays/resolutions, so the various problems like text not resizing correctly and pop up windows being off screen will be reduced/eliminated.

    We should probably be wary of extrapolating too much - this is already blending at least a couple of unconfirmed ideas.

  • To those who have had issues with the OS X CUDA Compute Capability version, and separately the issues with ILLEGAL_INSTRUCTION on AMD CPUs that do not support SSSE3 instructions (e.g., Phenom II, etc.), are those issues resolved for you in the 4.9.3.153 Public Build?

  • Hello everybody. Is the beta version still available? I bought the beta in the daz3d shop today but it is not showing up in my install manager. Its a bit frustrating coz i just bought me the ASUS GTX 1080 today and can't try it out yet.

  • prixatprixat Posts: 1,590

    To those who have had issues with the OS X CUDA Compute Capability version, and separately the issues with ILLEGAL_INSTRUCTION on AMD CPUs that do not support SSSE3 instructions (e.g., Phenom II, etc.), are those issues resolved for you in the 4.9.3.153 Public Build?

    Yes, rendering is working again on my Phenom II X6.

  • Richard HaseltineRichard Haseltine Posts: 102,464
    edited December 2016

    Hello everybody. Is the beta version still available? I bought the beta in the daz3d shop today but it is not showing up in my install manager. Its a bit frustrating coz i just bought me the ASUS GTX 1080 today and can't try it out yet.

    Click the Download Filters box at top-left of the Ready to Download tab and make sure that the Public Build option is checked.

    Post edited by Richard Haseltine on
  • Hello everybody. Is the beta version still available? I bought the beta in the daz3d shop today but it is not showing up in my install manager. Its a bit frustrating coz i just bought me the ASUS GTX 1080 today and can't try it out yet.

    Click the Download Filters box at top-left of the Ready to Download tab and make sure that the Public Build option is checked.

    Thank you for your reply but i already done that. There is no beta for me to download :-(

  • Hello everybody. Is the beta version still available? I bought the beta in the daz3d shop today but it is not showing up in my install manager. Its a bit frustrating coz i just bought me the ASUS GTX 1080 today and can't try it out yet.

    Click the Download Filters box at top-left of the Ready to Download tab and make sure that the Public Build option is checked.

    Sorry, my fault. I got it now. Thank you very much.

     

  • linvanchenelinvanchene Posts: 1,382
    edited December 2016

    3rd Update/Edit:

    - removed some sentences to focus on only two issues in this post

    - added more detailed information and further examples of things tried

    - all tests done with two GTX 1080 activated, CPU deactivated in the render settings.

     

    @ any plans to improve Iray scene preparing and VRAM management?

    From other render engines like Octane I am used to work with a smooth interactive realtime preview viewport and a smart and efficient VRAM management.

    There are several challenges while working with the DAZ Studio Nvidia Iray viewport drawstyle.

    1) "Prepare Scene" time takes too long

    In a test scene the process to load geometry and textures into VRAM and display it in the viewport took

    ~ 37 seconds with the DAZ Studio Nvidia Iray viewport drawstyle

    ~ 19 seconds with the OctaneRender for DAZ Studio plugin

    -> Roughly estimated it takes the native Iray render engine twice as long to load scenes into VRAM than a 3rd party plugin.

     

    - - -

    2) Switching activities completly wipes the VRAM

    To make sure that the user interface is not clustered with tabs I made separate "activities" like Scene Setup, Rendering with Iray, Animations etc.

    Example:

    For certain activities like switching parameters, or object placement the normal OpenGL "texture shaded" drawstyle seems sufficient.

    Nevertheless for any changes that affect materials or lighting the Iray viewport would be prefered to view "realtime" changes in render quality where possible.

     

     

    Whenever the user switches from one activity to another the Iray VRAM is completly wiped and the next time the users switches back to an activity with the Nvidia Iray viewport drawstyle the whole "Prepare Scene" process restarts from scratch.

    Even if you have both activities set to Nvidia Iray drawstyle switching activities triggers a complete reload of the scene.

    On the other hand the 3rd party plugin OctaneRender for DAZ Studio is setup in a way that

    - the whole OR scene remains in VRAM even when switching activities

     

     

    - - -

    Known issues? - any changes planned?

    I) Are there any plans to reduce the time it takes to load the Iray scene into VRAM?

    II) Will users get any manual control to choose that  geometries and textures remain in VRAM when switching activites?

    III) Will there be any option to manually clear all Iray VRAM information if needed?

    - - -

    -> Do I need to submit a bug report, a feature request or has this allready been discussed and developers are working on solutions?

     

     

    Preparing Scene takes 35+ seconds.jpg
    1920 x 1080 - 303K
    switching Activities clears VRAM.jpg
    1920 x 1080 - 347K
    Post edited by linvanchene on
  • 3dOutlaw3dOutlaw Posts: 2,481

    Hi, any chance Request #204529 is being considered to be fixed?  I just updated to the Beta in hopes it was, but its not.  I just ran across it opening up one of my older scenes to use it as a base.

    (...regarding per pixel shading affecting the vertical tiling in the viewport versus 3delight/iray renders)

  • I'm not sure if this is a beta problem or not since I'm only running the beta for the Pascal GPU support and this is the first time I've used DS (was previously using C4D and InterPoser about half a decade ago), but I'm noticing that if I load a scene, render, make some changes and re-render a few times I'll get a couple of hundred iterations into a new render and my display will blank out for a second and DS will crash to the desktop. This happens with small scenes as well as more complicated ones.The log reports

    2016-12-15 07:28:58.012 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00258 iterations after 89.660s.2016-12-15 07:28:59.995 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00259 iterations after 91.644s.2016-12-15 07:29:03.595 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00260 iterations after 95.253s.2016-12-15 07:29:04.079 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.23  IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while launching CUDA renderer in core_renderer_wf.cpp:816)2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.23  IRAY   rend error: CUDA device 0 (GeForce GTX 1070): Failed to launch renderer2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): Device failed while rendering2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while initializing memory buffer)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.220 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:06.619 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00261 iterations after 98.271s.2016-12-15 07:29:09.463 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00262 iterations after 101.115s.

     

  • Cake OneCake One Posts: 381

    Hi everyone

    i'm on Mac OS X, 10.11, cuda 8.

    So far, the JIT compilation used in some shaders used to crash CUDA totally (so renders were done using CPU only).

     

    On the last BETA relase : no more JIT error 

    So thank you for that

    BUT

    (and that is a big BUT)

    Shader don't crash Cuda anymore but don't look as they are supposed to :

    Here is OOT Cathy hair with Hairblending shader (known to crash cuda in MAC) rendered in Daz Studio 4.9 :

    https://gyazo.com/6e3e90ea99ba62659b5fb7decae37e44

    It looks like it shoud, based on the highlighted DUF file on the left. It was rendered by CPU, obviously, as CUDA crashed.

     

    Here is the same scene, rendered in the last beta :

    https://gyazo.com/26beb0107b46cbb81d417afcb5e18ed3

     

    Cuda doesn't crash, the GPU renders but the hair looks nothing like it.

    Can any other mac user confirm this difference in JIT shaders in both versions?

    Thanks

    C.

     

     

     

  • terahurts said:

    I'm not sure if this is a beta problem or not since I'm only running the beta for the Pascal GPU support and this is the first time I've used DS (was previously using C4D and InterPoser about half a decade ago), but I'm noticing that if I load a scene, render, make some changes and re-render a few times I'll get a couple of hundred iterations into a new render and my display will blank out for a second and DS will crash to the desktop. This happens with small scenes as well as more complicated ones.The log reports

    2016-12-15 07:28:58.012 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00258 iterations after 89.660s.2016-12-15 07:28:59.995 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00259 iterations after 91.644s.2016-12-15 07:29:03.595 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00260 iterations after 95.253s.2016-12-15 07:29:04.079 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.23  IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while launching CUDA renderer in core_renderer_wf.cpp:816)2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.23  IRAY   rend error: CUDA device 0 (GeForce GTX 1070): Failed to launch renderer2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): Device failed while rendering2016-12-15 07:29:04.095 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while initializing memory buffer)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.111 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.142 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.189 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:04.220 WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.5   IRAY   rend error: CUDA device 0 (GeForce GTX 1070): unknown error (while de-allocating memory)2016-12-15 07:29:06.619 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00261 iterations after 98.271s.2016-12-15 07:29:09.463 Iray INFO - module:category(IRAY:RENDER):   1.0   IRAY   rend info : Received update to 00262 iterations after 101.115s.

     

    How much RAM does your system have? It could be that it is running out of RAM if you are restarting renders in the same session. Make sure your drivers are up to date.

  • GaryHGaryH Posts: 66

    Are you closing your previous render windows?

  • Cake One said:

    Hi everyone

    i'm on Mac OS X, 10.11, cuda 8.

    So far, the JIT compilation used in some shaders used to crash CUDA totally (so renders were done using CPU only).

     

    On the last BETA relase : no more JIT error 

    So thank you for that

    BUT

    (and that is a big BUT)

    Shader don't crash Cuda anymore but don't look as they are supposed to :

    Here is OOT Cathy hair with Hairblending shader (known to crash cuda in MAC) rendered in Daz Studio 4.9 :

    https://gyazo.com/6e3e90ea99ba62659b5fb7decae37e44

    It looks like it shoud, based on the highlighted DUF file on the left. It was rendered by CPU, obviously, as CUDA crashed.

     

    Here is the same scene, rendered in the last beta :

    https://gyazo.com/26beb0107b46cbb81d417afcb5e18ed3

     

    Cuda doesn't crash, the GPU renders but the hair looks nothing like it.

    Can any other mac user confirm this difference in JIT shaders in both versions?

    Thanks

    C.

     

     

     

    Could you please attach a simple scene which reproduces the issue, listing any needed content.

  • mjc1016mjc1016 Posts: 15,001
    Cake One said:
     

    Shader don't crash Cuda anymore but don't look as they are supposed to :

    Here is OOT Cathy hair with Hairblending shader (known to crash cuda in MAC) rendered in Daz Studio 4.9 :

    https://gyazo.com/6e3e90ea99ba62659b5fb7decae37e44

    It looks like it shoud, based on the highlighted DUF file on the left. It was rendered by CPU, obviously, as CUDA crashed.

     

    That looks like hair with a high Translucency weight and too light a color/no map.

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    I have a slight difficulty here with Iray.

    WARNING: dzneuraymgr.cpp(307): Iray ERROR - module:category(IRAY:RENDER):   1.0   IRAY   rend error: Cannot render: found no usable devices.

    Is the GTX960 not an Iray capable device ???

    txt
    txt
    493beta_NoGo_001.txt
    93K
    Post edited by ZarconDeeGrissom on
  • It should be, as I have a GTX 980. You have 2 different cards installed on the same system?

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    It should be, as I have a GTX 980. You have 2 different cards installed on the same system?

    Yes, and I just tried it and the CPU. It (DS 4.9.3.153) will only render on the CPU, not the GT730 (driving the monitors) nor the GTX960 compute-only card.

    It worked fine on all of the above in DS4.9.2.70 and older.

    CaseAndPoint_001.png
    1024 x 700 - 373K
    Post edited by ZarconDeeGrissom on
  • You have the latest NVIDIA drivers installed?

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    for win7 that would be 347.88? There was one newer one (368.39) that had horrendous video payback and desktop stuttering issues.

    I see a 376.33, I'll give it a try and chime back after the reboot (and saving FAH from it's spasms).

    DriverVer_001.png
    562 x 509 - 10K
    Post edited by ZarconDeeGrissom on
  • for win7 that would be 347.88? There was one newer one (368.39) that had horrendous video payback and desktop stuttering issues.

    I see a 376.33, I'll give it a try and chime back after the reboot (and saving FAH from it's spasms).

    It needs at least 370.xx as it uses CUDA 8.

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    I don't know about this, the 376.33 driver keeps crashing and resetting a lot of apps including the browser.

    (EDIT) that's an epic fail, it disabled the GTX960 card.

    Post edited by ZarconDeeGrissom on
  • See if you can find 376.19 somewhere. I've been using that without issues.

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    stutters just as bad as 368.39, and it also adds a second plus delay to mouse clicks on the desktop. It appears to at least recover from it's occasional crashing unlike 376.33.

    Well, as long as the driver dose not crash in the middle of rendering

    it appears to be functional, in the most minimal that can be considered 'functional', it now renders on the cards.

    I'm use to a desktop that has done it before I have released the mouse button, "Click and wait" is not part of my vocabulary nvidia, lol.

    Ah_Ideas_001.png
    420 x 108 - 4K
    Success_NowAboutTheStuttering_001.png
    1040 x 660 - 340K
    Post edited by ZarconDeeGrissom on
  • mjc1016mjc1016 Posts: 15,001

    You'll need to go the full 'clean it all out' and install 376.19 'clean'...

    There seems to be too big a gap to bridge going from 345 to 37x drivers...I had to do that on the HTPC machine.

  • I always do a "clean install" when updating the drivers.

  • ZarconDeeGrissomZarconDeeGrissom Posts: 5,412
    edited December 2016

    If only it was as easy and painless as it sounds. After 98se, windows started doing counterproductive stuff. The process if uninstalling a driver involves a reboot, and once windows starts to get to the desktop after the reboot it automatically starts putting back in the driver you were trying to remove. The only way to completely rid yourself of an old driver, is a wipe of the boot drive and install windows from scratch, that will have to wait for a day or two that I don't need a computer (not today).

    Post edited by ZarconDeeGrissom on
  • mjc1016mjc1016 Posts: 15,001

    No...you just have to boot, manually, into Safe Mode, before Windows automatically reboots.

    Plus this helps...https://www.bleepingcomputer.com/download/display-driver-uninstaller/

Sign In or Register to comment.