Page 7 of 9
Posted: Mon Feb 24, 2014 7:24 pm
by v-cube
Hallo Eric,
rhino is a bit complicated when it comes to display performance.
This has been debated in several treads in the rhino discussion group. One observation is that since the GF 285 nvidia has crippled its gaming cards for open gl use so certain newer cards might work worse than your old 9800.
I would suggest to use a used Quadro 4000 card for display together with a powerful gaming card (780, titan, 590...) for octane. I run this configuration and it runs quite smoothly...
Best
Andreas
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Mon Feb 24, 2014 9:05 pm
by eric_clough
Thanks ...
I'd like to try your suggestion. I just bought the GeForce 760 GTX so I see if I can find the Quadro 4000 and see if they work together. I am not unhappy with the GeForce 760 running alone, though.
cheers,
eric
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Mon Feb 24, 2014 9:52 pm
by face_off
Regarding cards....I'm running the Intel on-board graphics (4600 on one, 4400 on the other) as the display adapters on my PC's, and they both run the Rhino render window with no lag whilst the Octane render is running on the NVidia card - works perfectly. So I think the 9800GT might be the issue as Andreas suggested.
Paul
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Mon Feb 24, 2014 11:00 pm
by eric_clough
Hi Paul ...
It sounds then like an EVGA GeForce GT 640 might work as the display only card with the 760 GTX being used for the Octane display.
If I set it up correctly it might help the overall process ???
cheers,
eric
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Tue Feb 25, 2014 4:26 am
by face_off
Not sure Eric - sorry.
Is "Use acellerated hardware modes" ON or OFF in the Rhino OpenGl settings (under Rhino Options->View->OpenGL)? From what I can tell, it needs to be OFF, otherwise it uses your Octane Rendering card for the Rhino OpenGl view.
Paul
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Tue Feb 25, 2014 8:37 am
by eric_clough
Hi Paul ...
It did help to turn off the open GL accelerator. Now I can used the 'render' display in Rhino at the same time the Octane window is open. It is not totally smooth as the Octane window lags a little behind and more or less snaps into a correct view when the mouse stops. But it is very functional ... the actual speed of sampling slowed by about 1/3 down to about 1 sample per second instead of about 1000 per minute. It is still rendering the image very rapidly though.
I am going to purchase another single cuda card for running the everyday stuff so that the more powerful dual card can be devoted entirely to Octane. I think it will improve the overall system and will be worth ... I hope I am correct.
Now if Octane only had a plant library similar to Flamingo NxT. But, if wishes were horses ..
Thanks again,
eric
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Tue Feb 25, 2014 11:42 am
by face_off
I have refreshed the installer at the top of this thread with:
1.32.0.11
- Fixed lag issue associated with block instances
- Fixed cuda errors from importing rendertargets saved with Octane 1.20
- Resolved issue where nested instances were not being loaded into the Octane scene
- Resolved issue where instances assigned to a non-visible layer were still being loaded into the Octane scene
- Fixed issue where geometry with no vertices was stopping the scene from loading into Octane
- Fixed Rhino crash when trying to load invalid background files
- Added a new color chooser. Selecting this color chooser will give a live update of the color change in the Viewport. Opening the color chooser will always restart the render, even if you do not change the color and cancel.
Paul
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Tue Feb 25, 2014 6:04 pm
by eric_clough
Thanks Paul ...
It works perfectly here.
eric
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Wed Feb 26, 2014 1:51 pm
by v-cube
Hello Paul,
thank you for the new build, works flawless so far over here, the new color picker is awesome!
I am doing a few IES light tests here, and I am quite happy with the results so far. I have a general question:
when doing interior illumination I usually arrange the lamp geometry first and then try to mimic its orientation with the IES material, which is applied to a small surface in front of the lamp, by altering the rotation values .
I would love to have the option that the IES emitting direction could be connected to the normal vector of the geometry. This way I could bind my surface to the lamp geometry and if I alter the lamp orientation the light emitting directory would be changed accordingly... (please see pic no 2)
Is this possible ?
best regards
Andreas
P.S. one thing I observed is that when either blocking an ies emitter plane or activating the "live update" option in the octane geometry tab, these objects seems be much brighter in contrast to their simply copies siblings, which cause an inhomogeneous appearance in a scene
Re: OctaneRender for Rhino 1.32 [TEST] (RhinoSR7 31213)
Posted: Wed Feb 26, 2014 1:55 pm
by v-cube
...
attached please find the 3dm file with the iges file
Andreas