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.
Vlado, if I set lightmap to screen, will it be the exactly the same for a 800*600 or 1600*1200 image? or are there other things that influence the LM result?
So in example 1 I used less subdivs for LM, LM went faster but IR map slower. Results of GI quality look the same, except under the curtains. Probabely using 2000 subdivs without any filtering will work best, need to test that too maybe.
When using adaptive QMC as image sampler, you need to use high subdivs for glossies, which influences rendertime much less than with adaptive subdivision. If you use ad subdivision, lower all glossies to 10 and subdivs for area light also, otherwise it'll take forever to render!
My initial test with qmc in second bounce were 3H rendertime, and glossies were worse and GI looked terrible... So this is a pleasing result for me.
Nice tests Wouter. So it seems that for these type of scenes adaptive qmc is the way to go. (that's also in the manual, but it is good to see a real example)
Side note: the first image shows clearly why I dislike HSV colormapping so much. Parts that ought to be white become fully saturated. Exponential corrects this with the disadvantage of desaturating the whole scene. I like the linear version the most, despite the blown out area. It feels more natural.
P.S. to Vlado: wouldn't it be possible to make a color mapping mode that preserves saturation in not blown out area's, or would that look weird. It just seems that neither of the current color mapping modes can adjust the image to what a human eye would see
You can contact StudioGijs for 3D visualization and 3D modeling related services and on-site training.
I fully agree, I'm doing some tests on that. When set to linear, if you use 2 for dark and 1 for light, you get a perfect result! But you need to desaturate the GI solution a bit more, which by the way are excellent options
straight out of max, linear color mapping Dark:2/bright:1
Good to see you figured it out, Wouter; I'm testing something slightly different now: instead of increasing all the subdivs, I'm trying to use the Global subdivs multiplier and using low subdivs everywhere else.
I must say this has been one of the most useful threads for me since I started using vray. Thanks Wouter for starting it and vlado for participating in it in such depth.
The lightmaps for this build are much better than before, in my opinion. Shadows are much more defined and the softness of the image seems to be gone as well. Great job vlado! There worth the little extra time needed.
I was even able to get clean glossy reflection in this scene as well as others because of what was explained here. As you guys know, this had been driving me nuts. It had nothing to do with vray, it was the operator that was in error! Sorry for bitching about that one.
The gamma application to an image is really helping to speed things up. I can set the lights much lower than I normally would, which saves a great deal of time in calculations, and still get a nice clean and bright image. Great tip!
Anyway, looking forward to vlado's post and tutorial of the scene as I'm sure others are.
Comment