Caustics quality
Wouter,
Thank you once again 8)
You are a great help here and you also help indirectly via your tutorials on your website. I appreciate it !
Caustics quality
Wouter,
I have a question about the blurry reflections on the coins…
in my tests, it seems the only thing that affects the quality of the coin blurry reflections is the material editor under Reflection-->Subdiv.
However, when I bump the material subDiv higher (about subDiv=35) the render times go up exponensially. Is this the only way to improve blurry reflections.
(I have been playing with the QMC sampler, but it doesn’t seem to affect that aspect)
Caustics quality
Ok, first of all, this scene seems to be a very difficult case. As you can see in my image and even better in your ppt example, there is something going on on the coins surfaces.
The left of that coin is smooth glossy reflection. But the right is not. I think this is because of the very strong small light, which creates problems for the QMC sampler.
I also can’t get rid of this noise.
But usually you indeed control blurry reflection quality by the subdivs in the material, or with the QMC sampler settings. To be able to use the QMC sampler to the max, you should use QMC AA, then you notice the effect best. QMC AA is also faster than adapt subdiv AA if your scene has lots of glossies, area shadows, DOF, etc… Actually I always use QMC AA at min/max=1/4. I never change it!
Then you control the AA, glossies, DOF, MB, Area shadows, … with the QMC sampler settings, together with material settings of course. The QMC sampler is a global quality control, while the subdivs on materials, or dof subdivs etc are controls for the quality of these individual effects.
In the QMC sampler, lower noise th is higher quality. There is a global subdivs multiplier too, altough I prefer to do it locally, gives me more control. Note that the QMC sampler settings also affect IR map calculations! The only things that are not affected are lightcache, photon map and photon mapped caustics.
Good quality for glossies is smth between 20 and 30 subdivs, same for area lights altough this depends on the size of the light too. If you then set noise th=0.002, you get very smooth results. For test renders, simply set noise th= 0.2, and global subdivs multi to 0.3 or something. That will render very fast so you can check lighting or material colors etc… I also change the ir map min/max for test renders, just lower them both. The hsph is lowered for your test renders by the global subdivs multi=0.3, so you can leave this at a high value if you want.
The global subdivs multiplier is tricky. If you have 30 subs on a glossy material, and you use a global subdivs multiplier of 5, this doesn’t mean that your material will now render with 5*30=150 subdivs. 30 subdivs actually means 30*30=900. So globally multiplying by 5 means 5*30*30=4500. If you want the same quality for your glossy, but with global subdivs at 1 again, you should set material subs to square root of 4500=67!!!
Keep this in mind, because 150 or 67 subdivs is a really big difference 
Same goes for hsph subdivs and all other subdivs values that are controlled by the QMC sampler.
Caustics quality
Wouter,
Excellent info again Thank You.
This is the culprit, I would not have guessed this is how it multiplies. Makes more sense now. Thank You !
Caustics quality
That’s why I don’t use it, I only use it to downgrade quality.
If for example you want all your glossy materials to look a bit better, you would increase the global subdivs value. But you don’t want hsph subs to increase, so you have to calculate the new hsph subs value to get same quality here. Same for area shadows, DOF, MB, etc… I would like to see global subdivs multiplier for seperate things like area lights, reflection glossy, refraction glossy, etc… That would be the easiest way imo if you want to increase all you glossies, or all your area shadow subs, all at once without affecting all other settings.
Caustics quality
Wouter,
I have found that the time stamp does not include the caustic-map calculation phase. In my test a map having lights of 8000 photons costs me 1hour just for the photon calc phase… so … when you said 1.5 hour of render then you only meant the IR+LC … but with your settings a 12500 photons a scene would have taken over 2hours (maybe 3) just for the photons alone and the memory usage would be enormous (so your total time should have been well over 3hours unless you are on a renderfarm).
This makes me curious because in my first image (beginner’s lack) in only 2hours the caustic was getting good by IR alone and without shooting photons. So I still wonder if the caustics tab does any good after all.
I will try to replicate the settings of that first image (if I can remember them 
) (Now I wish we had EXIF data in the images)
Caustics quality
Hi Thomas,
That rendertime was all inclusive. I didn’t use the frame stamp, I read it from the command line, or is this wrong too?
I use a dual xeon 2.8 system.
Your initial test looks decent, but not clean at all. The small caustics patters are good, but the big reflections are really noise (big blotchy noise) and these are the hardest to clear up.
I agree that IR map is good enough, I would never need such sharp caustics anyway. Maybe I’ll do another test, a more balanced one between quality and speed. I did the 12500 photons test just to get maximum quality out of it.
Caustics quality
Indeed the command line is the correct one. hmm now I am puzzled.
This means that i am doing something wrong when setting up the caustics tab because with only 8000 subdivs on the lights I get 1 hour of caustic map calculation and I think I used conservative caustic settings:
- MaxPhotons = 10
Max Density = 0.03
Multiplier = 1
Search Dist = 1
Well, I will look into it some more.
Material and DOF subdivs are 10 and multiplied by 30 (this is equivalent to 55 subdivs). Even at 55 subdivs the coins are not clean.
Caustics quality
This time I wanted to find out if it is possible to do nice cautics using GI only (no photons) and how far would the settings would have to go to achieve this. I like GI caustics because they are easier to test with region render.
It turns out the most consuming part are the caustics that fall on the blurry parts of the coins. Those coins had to be boosted to 80 subdivs and the noise is still slightly there. It takes 15min/bucket to finish these coins :shock:
Overall the quality came out ok but the render time is not too good (18hours :shock:). I do not think the use photon mapped caustics would improve the time by match… when I use Wouter’s method (at 12500 subD) it takes my PC 4hours just for the photon map and then we have to go through the cautics on the blurry coins again (and we still have to boost the metals to 80 subDvs or more). So basically by using the photon method we are gaining some speed from the GI calculation but we are loosing it again on the photon map (in both cases we need lightcache and in both cases we need high definition materials for the blurry/caustic effects and in both cases the caustics on the coins will be the most time consuming operation).
Caustics quality
It seems, this scene is my right of passage to Vray. If I can get this down to 2hours or less (at “milky smooth” quality) then I feel I can do anything with Vray…
Caustics quality
What is your QMC noise threshold? I hope to finally get some time to play with the scene this weekend.
Best regards,
Vlado
Caustics quality
Image Sampler: Adaptive QMC [1,4] Antialiasing [Area 1.5]
QMC Sampler: Ad.Am=0.02 Noise=0.002 MinSmpl=5 Mult=1
Irradiance Map: MinR=-6, MaxR=0, SubD=80, Samples=20, Colr=0.3, Nrml=0.1, Dist=1, Delone, DensityB 10
Light Cache: SubD=2500, sampleSize=0.002, Scale=Screen, ph=1, Reconstr=None
Vlado, if you do get around to try this scene, is it possible to try it on a single processor PC having similar stats to this --> PC specs = AMD 64 3500 1GB.
Caustics quality
Adaptive amount 0.02 is just crazy. That means you’re almost not using adaptive sampling, you’re making some kinf of brute force rendering by using that low value (probably not the right explanation but I guess you know what I mean )
I also can’t get the noise on the blurry coins to go away, I’m not sure if it is related to the material, or to the caustics falling on it.
-6/0 is also a bit extreme. I don’t think the first 2 or three passes wil actually help. Remember how it works, your first pass will be computed on a 800/2/2/2/2/2/2=12.5px wide image. Don’t think vray will get much usefull info out of that. It will not improve this image much, it’s just a general tip
One thing to try, to avoid blurring your ir map solution too much, is decreasing the ‘samples’ value in IR map controls. The bigger this number, the more surrounding samples get interpolated together, resulting in smoother images, but less detailed GI. So for this type of caustics, lowering it from 20 to for example 12, will keep your GI very sharp. Try it with 4 or smth on a simple scene, you’ll see the drastic impact it has
For Vlado, this image is extremely difficult if you keep the ‘clamp colors’ turned off. It’s hopeless to get clean AA on the very bright reflections.
Caustics quality
Adaptive amount 0.02 is just crazy. That means you’re almost not using adaptive sampling, you’re making some kinf of brute force rendering by using that low value (probably not the right explanation but I guess you know what I mean Smile )
You are right. I should have used 0.2 or something, but other higher values would introduce blotchy GI. I have found that high values (like 0.85) also decrease the caustic density.
I wouldn’t worry about it 
I have done a lot of test the past weeks on these values and it turns out that they do not really add to the render times. (I can tell with good certaintly that the culprit is not there). Also, based on my tests it turns out that adding those low passes -6, -5, -4 it also adds an extra layer of sample points to the irradiance solution. In other words the irradiance map ends up being more dense than if you choose -3 0. In this case the more dense the map the better, because anything else creates incomplete (irregular) not smooth caustics. These values have been researched quite extensively.
One thing to try, to avoid blurring your ir map solution too much, is decreasing the ‘samples’ value in IR map controls. The bigger this number, the more surrounding samples get interpolated together, resulting in smoother images, but less detailed GI. So for this type of caustics, lowering it from 20 to for example 12, will keep your GI very sharp. Try it with 4 or smth on a simple scene, you’ll see the drastic impact it has
Less samples would make it more grainy (I have tried it
So, the IR settings have been researched for highest IR caustic density. (I have spent over 90 hours on this image)
For Vlado, this image is extremely difficult if you keep the ‘clamp colors’ turned off. It’s hopeless to get clean AA on the very bright reflections.
Yes, the aliasing is very difficult. I have tried all aliasing algorithms and and at various settings (from 1 to 4) … but it doesn’t seem to help. Clamping the colors is simply not an option in this image… I really want to see those burned out areas !
Caustics quality
Hey thomas,
I know the passes don’t help much, that’s whay I said it was just a general tip, not really for this image. There are better ways to add more samples to the map than by using extreme low passes.
Higher values for adapt amount indeed means more blotchy GI. But remember that lowering this value affects everything, also your DOF and glossies for example. So if you want better GI only, this is not the way to go because it will improve quality of a lot of other stuff too, that you don’t even notice.
Less samples, I mean the samples value, not the hsph. It will indeed be more grainy, but grainy means sharp. Increase other values to get rid of the grain, but still retain the sharp GI.
Delone triangulation also means you need higher GI settings, as it is a non blurry method. I believe the samples value from above has less effect when using delone, so maybe if you keep delone, samples doesn’t matter much.
I’m going to dig up this scene again
Caustics quality
Higher values for adapt amount indeed means more blotchy GI. But remember that lowering this value affects everything, also your DOF and glossies for example. So if you want better GI only, this is not the way to go because it will improve quality of a lot of other stuff too, that you don’t even notice.
Exactly ! The coines would not look as good with higher values either… and… despite all that and even at these extreme low adaptive values and at 80 smaples per coin (as well as Adaptive QMC sampling) they are still garinny.
Less samples, I mean the samples value, not the hsph. It will indeed be more grainy, but grainy means sharp. Increase other values to get rid of the grain, but still retain the sharp GI.
Delone is the only acceptable one, I have tried the others but they blur the little caustic spots around the coins and I want those really sharp. I found the other methods harder to produce a good cautic result.
If you do … can you try it on a single processor PC ? 
I mean if your dual can get you a 12500 photon map in 1hr and mine gets it in 4hrs … its not fair :lol:
Caustics quality
Hehe, just multiply my rendertime with 4 then
Caustics quality
Have you tried QMC GI for first engine? Becausein your case, if you want really exact GI, this could be faster than IR map. The use of IR map is to speed up GI, at the cost of some detail. Since detail is all you want here, maybe IR map is not the best engine. (it is better however if you want a really big render, because you can then save the IR map and reuse it on the big render, and that is not possible with QMC GI or PPT)
Caustics quality
hmm, no I haven’t tried QMC GI … I mean.. I did try it briefly, but gave up on it for some reason which I can’t remember. Somehow I do not think QMC will be faster than PPT fot this scene though.
P.S. These kind of heavy caustic scenes are quite important for me, beacuse some of my work involves jewelry.
Caustics quality
Aaaah that explains a lot
QMC was just a though, didn’t try it extensivly too. I did one test, which was quite nice, but just like ppt, to get the caustics ok, it needed much more rendertime.
Here is that old test, 1h50min on my pc: