If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
Exciting News: Chaos acquires EvolveLAB = AI-Powered Design.
To learn more, please visit this page!
New! You can now log in to the forums with your chaos.com account as well as your forum account.
Just FYI, so that there are less surprises, we are probably going to do that in stages; where the first stage would be ability to change materials, the camera, and light properties but without moving geometry, and the second stage where we can move everything.
Well that would be fine as long as everything in the shading network updates properly, without any nasty surprises (of course I expect there to be some bugs from the start), because shading and lighting is the main point of using IR. It would certainly help a lot if hide/unhide/isolate selection worked at least.
What would be unfortunate though, is if there would be some restrictions on shading graph updates, for example mental ray had a limitation that color mapping curve in bitmaps and output maps did not work, etc... That would again harm the usability a lot.
So it's fine when there is clearly established simple set of rules, where you exactly know what to expect. Shading/lighting/camera working, but geometry not is a nice example of it. That would be just fine. But things like "falloff map does not work in mode X or Y when IR is used, RGB level in output map does not work, bitmaps can not be changed during rendering" etc... that kind of stuff, that would put us back to square one.
Even if it was just shading on it's own, without lighting or camera, but working perfectly 100% reliably, that would be still better than having everything working but just for 80% of the features. There's no 80/20 rule in the interactive rendering I am afraid.
Or to put it more simply, win scenario here would be if you could do everything in IR mode that you can do in regular rendering mode without need to ever restart/re-initialize IR rendering as long as you don't touch geometry. Meaning given that geometry does not change at all, you could spend working in IR mode for say three hours, and once you finish and run production renderer frame, it would look the same to the pixel
Regarding activeshade, i honestly don't see any advantages for it... Dunno if RT GPU can only work in that mode for IR purposes, but if vray could completely discard it, we could avoid all this "mess" between different settings in production or active shade mode.
Ideally _in my opinion_ we could just set production renderer = "Vray Renderer", then on the main settings you could have "CPU" and "GPU" mode, each with their settings and that's it. Both would be always interactive and the whole process would be much simpler to deal with. I'm sure that there would be a lot of implications and issues with this method, but personally i can't think of a simpler way to work with vray.
Ideally _in my opinion_ we could just set production renderer = "Vray Renderer", then on the main settings you could have "CPU" and "GPU" mode, each with their settings and that's it. Both would be always interactive and the whole process would be much simpler to deal with.
That's exactly the way we have done things in Maya in the latest service pack from a few weeks ago. We'll see how it goes [FYI in Maya there is no difference between the final frame renderer and the IPR renderer, they produce the exact same results.]
Comment