Universal V-Ray settings (well, almost)

[studioDIM]

Since IR map is resolution dependant it won't make much sense imo. Settings that work well at 800*600 will take forever on 3200*2400 for example (if you don't tweak the min/max rate)

Hence the LC/PPT calculator to keep them consistent across resolution changes...

Lele

[Metin_7]

Sounds good. Where can I find that calculator?

[andrewjohn81]

This is great, but what about for animations?

I don't have much use for compositing often as we don't usually even have the time, and when I do have time the project is usually too busy.

I'd rather not have 40+ minute rendertimes though.

[vlado]

What kind of animation is that (e.g. what kind of objects will be moving and in what environment)?

Best regards,
Vlado

[rpc212]

Sounds good. Where can I find that calculator?

Ya, I'd like to know this too!

Thanks!

[highpoly]

i am trying this setting, but for some reason, i am getting TONS of noise..i even lowered the adpative amount to 0.85 in rQMC and 0.001 but still noise. I played with the globla subs and it did help a little, but i thoght the 100 max subs in QMC AA was overiding everything??

[voltron7]

@ highpoly

I've also had this issue trying QMC for first bounce. It seemed to help raising subdivisions and rqmc thresh, but as explained earlier, is very slow, at least on my machine, hehehe So for now, i'm still using IR/LC

[Micha]

Are there any IRR/LC universal settings ?

Micha posted something along these lines over at the V-Ray for Rhino forum, I will try to find it.

Best regards,
Vlado

Today I have seen this old post. My experience of the last months is, that an adaptive amount 1 cause IR artefacts sometimes (at areas with less direct light I got strong blotches and no IR setup tuning helps). But Vlados universal setup works quite good with IR too, if adaptive amount is lowered at 0.95 or less.

My basic rules are - higher IR subdivs against the typical IR blotches and higher max rate for better shadow details. I start with ~40 subdivs and max rate -1 most and refine this two settings, if necessary. IR interpolation samples are at 20 and I use 30...40, if a high subdiv count (around 90) dosn't help against a blotchy look. So, most I use IR+LC.

I'm not sure about the influence of the adaptive amount. What would be, if we use an adaptive amount of 0.85 allways? Why touch this option and jump btween 0.85 and 1? For me, this option is the biggest Vray mystery.

[m_hinks]

I've seen a similar thing when using IR, and changing the adaptive amount to 1. any thing below, as you say .95 is ok, no splotches. Just at 1 is a problem.

I also wonder if someone could post a universal setting for IR and LC not Brute and LC?

[m_hinks]

Im doing an animation with DMC AA for the first time, and trying to use the universal settings where possible. However, as above, I am using IR and LC, not brute force LC. I now see that the adaptive amount must be .95 and no better to stop the IR going crazy blotchy, but have as much adaptive nature.

I think as most people seem to be using IR for primary as brute is just too slow in most cases, a universal settings for IR and LC would be very useful from the people in the know, ie Vlado!

[Jynks]

Provided that you start with a default V-Ray renderer:
(*) In the Image sampler rollout, switch to "Adaptive QMC".
(*) Set the "Max. subdivs" to 100 (one hundred). Leave the "Min. subdivs" to 1.
(*) Set the primary GI engine to QMC GI. Do not change the subdivs.
(*) Set the secondary GI engine to Light cache.
(*) Set the light cache interpolation samples to 5.
(*) Set the QMC Sampler "Adaptive amount" to 1.0.
(*) Un-check the "Clamp output" option in the Color mapping rollout.[*] Typically you will need to adjust the Noise threshold as the default produces too much noise. A good value is 0.001 for V-Ray 1.4x, and 0.005 for V-Ray 1.5.
(*) You are done.

In vray 1.5 these settings are not in the lists anymore....

Adaptive QMC = Adaptive DMC?
QMC GI = Brute Force?

So how tyo do translate these values into the 1.5 setting options?

[rivoli]

adaptive dmc. which is deterministic monte carlo, instead of quasi monte carlo I guess.

[zeronove]

Adaptive QMC = Adaptive DMC? -> CORRECT
QMC GI = Brute Force? -> CORRECT

[vlado]

We've corrected the original post in the thread to reflect the new names for V-Ray 1.5 and later.

Best regards,
Vlado

[Jynks]

<--- edit

opps posted in wrong thread.. sorry .. I could not find teh delete button.