Hi,
We work on a feature film in stereoscopic.
I have a question : do you have a plan to add some optimization in stereoscopic ?
For example, Renderman can calculate 2 stereoscopics images at once. And the render time with Renderman is :
- 1 image : 100% (time)
- 2 images (stereoscopics ) : 120% (time)
I don’t know how, but, for example, calculate one time the diffuse in deep pixel/shade ?
Thanks.
To optimize the rendering process when doing stereoscopic or DOF effects you can use Shade map.
For more information please visit:
Yes, but Shade map is in Maya ? (nighlty)
just want to ask the same thing, coz in official built VRayStereoRig doesnt exist…
cheers
The feature will be available in the new V-Ray for Maya release. Which we expect to come out soon.
If you need to use the stereo effect and the shade map urgently please drop us an e-mail to: support@chaosgroup.com
we’ll consider your request and reply.
Thank you Zdravko for the information.
Our dealine is in end of december. So, it depends on when this new Vray for Maya will be released.
Please send us an e-mail so we could discuss that further.
Hey there!
Is the stereoscopic stuff planned for the upcoming SP? It seems that we’ll badly need it for an upcoming project 
It has been there for a while now, but you’d have to create it from the Python callback as we don’t have a GUI for this. Let me know if you need more info and I’ll explain how to turn it on.
Best regards,
Vlado
Wow, that was fast Yes, thank you, an explanation to this would be helpful.
Hi Vlado! Anything on this? You can also mail the explanation to m.pech@abyss-fx.com if the forum isn’t the right place right now. Thank you!
Well, I just found some issues that I need to fix; hopefully they will go into the next night build tomorrow, but in general it goes like this. Use the Python post-translate callback to add a VRayStereoscopicSettings plugin to your scene, like this:```
from vray.utils import *
p=create(“VRayStereoscopicSettings”, “stereoCam”)
p.set(“eye_distance”, 0.5)
The stereoscopic settings have a number of options, for generating shade maps and for using them, but I found a few issues there which I'm looking into now. They should be fine for the next nightly/stable build tomorrow.
In any case, to render a shade map, use the following after the above code:```
p.set("sm_mode", 1)
p.set("shademap_file", "somefile.vrst")
Then you render the animation to generate the shademap, and then change the “sm_mode” to 2 so that V-Ray uses the rendered maps.
Best regards,
Vlado
Thank you very much!
EDIT: But right now nothing gets written. The path in the vrscene is valid, but no file in there. There isn’t any output in the output window/script editor either, so it’s not so easy to debug for me.
Oh, and thanks for adressing those issues you mentioned so fast!
EDIT2: Hmm, I just hit render a few times more and now I got a map. Also in the vrayFB the image aspect ratio changed from time to time. But maybe these are the points you work on anyway, so maybe I just wait for the next update.
Another question would be, if it’s possible to use your own custom stereo rig in combination with the shademap.
The stereo plugin has parameters to specify the left and right cameras, but I’m not quite sure how these would be exported from Maya yet.
Best regards,
Vlado
I continued testing with the build from today (19193) and it really doesn’t work. The shademap doesn’t get written and vray behaves really strange when rendering. Sometimes it renderes just the image, then again the image is rendered and an additional black space of 3 times the size of the image is added to the rendering in VFB, sometimes smaller. Also there are differences if I use the render button in the maya UI or the render button in the VFB.
So: If this feature isn’t really usable, that’s ok, of course. I know that it’s not really supported yet, but my hope was, that it’s nearly finished. If you’d say ‘Don’t use it now!’ I’d stop looking into it and continue with the old-fashioned stereo way.
I’d do that; there is a reason it is not an official feature yet 
Best regards,
Vlado
Yes, that’s what i thought. Thanks anyway!
Hi,
It is possible to calculate, in stereo camera, one time the brute force GI for example ?
The best will be to do not use some “cached files”.
I know Renderman can “separate” some “calculations” which are view dependant (specular for example) and not view dependant (GI, diffuse, etc etc) without to cache something.
It render on the same time both images.
What about VRay ?
The idea is to do not use cached file.
Cached file will be great if we can use it between each frame.
Hi,
Are there any improvements to this?
I can get V-Ray to write out the shademap, but shouldn’t it be rendered without depth of field and such effects?
The image that I get during shademap creation still has those effects applied.
Also when switching to reuse of the shademap I do not see any improvements in rendertime.
Another thing that I recognized is that the adjust_resolution does not seem to work.