Chaos Corona Forum
Chaos Corona for 3ds Max => [Max] Daily Builds => Topic started by: maru on 2019-04-05, 17:01:44
-
Hi!
We have just added new image filtering option to Corona:
https://forum.corona-renderer.com/index.php?topic=22529.msg146930#msg146930
Daily builds info and download: https://coronarenderer.freshdesk.com/support/solutions/articles/5000570015
It is available in Render Setup > System.
We would be super grateful if you could experiment with this option in your scenes (simple tests, wip, production - anything) and share your results here. We need any feedback: image quality, rendering time, how it works with high quality denoising... anything you spot.
So far our tests confirm that it helps in situations such as:
-Fine patterns, which are mixed with noise (patterns are easier to see)
-GI noise in general
-DOF, especially when there are bright spots blurred by the DOF
-Sharp edges of bright objects
-Moire patterns
I am attaching some simple test images below.
More on the helpdesk:
https://coronarenderer.freshdesk.com/support/solutions/articles/12000056115
https://coronarenderer.freshdesk.com/support/solutions/articles/12000056143
NOTE: currently it cannot be used together with the AI denoisers!
-
Looks promising! Will have to test this out: )
-
I didn't have much time to play with it, but from what i saw so far, i'm very impressed. But please, please make it compatible with real time denoiser - i can't live without it :/
One thing i noticed, it looks that HQ filtering doesn't like refractive materials, to be more specific, not filtering itself, but its combination with denoising. Look how much denoising has smeared the pixels behind refractive lens in render with HQ filtering.
-
Had to re check my denoiser activation, to be sure it was on w/o Hi-quaity filtering. Hi-quality filtering makes a noticeably significant difference in most situations.
critical, not obvious edit: 'Hi-quality filtering'
-
This thread is not about HQ denoising, but HQ image filtering.
And I can't wait to test this out tonight!
-
i rendered the same frame with the HQ Image filter turned on in the newest build, compared to the same frame from the previous build and i experienced longer render times with the new HQ image filter on: 42 minutes vs 1 hour 54 minutes. But I do think the quality is a little better though. But sadly, at this time, not better enough to warrant using it because of the increase in render time, it seems.
-
Hi, if it's possible could you maybe send the scene for testing? Such slowdowns should not happen and I'd like to track it down :) Thanks!
-
i rendered the same frame with the HQ Image filter turned on in the newest build, compared to the same frame from the previous build and i experienced longer render times with the new HQ image filter on: 42 minutes vs 1 hour 54 minutes. But I do think the quality is a little better though. But sadly, at this time, not better enough to warrant using it because of the increase in render time, it seems.
did you use uhd cache?
-
Supposedly the new filtering can change the adaptive noise level, so if you have used noise level to stop your rendering, then that could explain such a big difference. Other than that, I think you've rendered your images way too clean to spot a difference. The best way to compare results is to set a time limit, not a noise limit or pass limit. You want to see how different the images look after the exact same duration of rendering.
-
With ai denoising =D
5hrs on twin x5650
Obvious artifacts at dof areas.
-
I apologize did not notice this entry.
Some kind of weird quality "Use high quality filtering"
So it should be?
-
With ai denoising =D
5hrs on twin x5650
Obvious artifacts at dof areas.
High quality filtering does not work with AI denoisers, you should get warning message about that. If you rendered with one of AI denoisers enabled, then HQ filtering was automatically disabled.
-
With ai denoising =D
5hrs on twin x5650
Obvious artifacts at dof areas.
High quality filtering does not work with AI denoisers, you should get warning message about that. If you rendered with one of AI denoisers enabled, then HQ filtering was automatically disabled.
Before the test, I read this information. Denoise is disabled
Maybe I do not understand something, but when I click on this parameter, it only gets worse.
It becomes noticeably more noise and stops working with him, denoise
-
Feodor, i was refering to Nekrobul's post. As for your results, are you comparing tent filtering with denoiser versus high quality filtering with no denoiser?
-
With ai denoising =D
5hrs on twin x5650
Obvious artifacts at dof areas.
High quality filtering does not work with AI denoisers, you should get warning message about that. If you rendered with one of AI denoisers enabled, then HQ filtering was automatically disabled.
Yes, i think i misunderstood controls. Will run another test in this scene.
-
Feodor, i was refering to Nekrobul's post. As for your results, are you comparing tent filtering with denoiser versus high quality filtering with no denoiser?
Sorry mix up the message)
So I have not checked.
I was a little puzzled by the function that degrades image quality. When using noise, "Using high quality pixel filtering with AI denoising is not supported."
At this stage of development of this function, for the time being it seems to me useless, harmful
-
One thing i noticed, it looks that HQ filtering doesn't like refractive materials, to be more specific, not filtering itself, but its combination with denoising. Look how much denoising has smeared the pixels behind refractive lens in render with HQ filtering.
It looks that same can be said about reflections - without denoising HQ filtering gives very good results, but if denoising is enabled, it just smears details badly. I think that enabling HQF by default, is a little bit premature at this point.
-
Ok. I made a proper test of real life sitiuation this time.
Isues i have noticed.
With new sampler reflections in the mirror (wall detail shadows) are washed out for some reason.
New sampler handles DOF areas great but only at edges of objects. Large surfaces (floor in this specific case) looks horible a lot of speckle noise.
Also reflection\refraction detail being washed out with denoiser (parfume bottle)
For both images rendertiome 40 mins.
Old sampler made 266 pases
New sampler made 230 pases
In both cases native denoiser was used.
I will purge 4k trough the night just to see what will happen.
-
I have an idea, can you try disabling adaptivity for this test?
-
I have an idea, can you try disabling adaptivity for this test?
Good thing i have visited forum before going to sleep, shure yes. And i will make same settings i used for previous just with 4k.
-------------------------------------------------
10hrs 588pases no adaptivity 4GI\AA as pevious
Shadow details on wall present after denoising as on image with old sampling.
And noise on the floor is much lower.
PS - The denoise error was VFB halucinacion. It is not present on saved images(denoised). I compared them using photoshops diference blending and there is only slight differenc in noise stucture. I will attach both images for it will be simplier to compare.
-
Ok, I tried another test with limiting by passes (PT+PT, same as my first test on page 1), but I left the Corona HQ denoiser on. Details in the file name.
The main difference I see in the type of noise, is that with HQ filtering off, there is the 'classic' type of CG noise. But with it on, the noise is a little more 'smudged', which I assume is the HQ denoiser trying to deal with a different type of noise. I may try another test with all denoising off.
-
Btw, Nekrobul, if you are doing difference tests, you should not be saving as .jpg files, as they will introduce their own noise, even on 100%.
-
And you also should not render so clean images. Sure it's good to test how the new image filter contributes to final quality image, but it will be harder for human eye to spot the differences. I'd do one more test, again with the same amount of passes, but lower amount of them (let's say a half) so you can spot the differences better.
From the current test, it appears that the HQ image filter mainly helps to resolve very small highlights/glints.
-
I've tried new filtering in different scenes - exteriors and interiors. I based my test on same render time for pictures rendered with new and old filtering. And have to say that all the time I get the same result - pictures are obviously cleaner with new filtering, but Corona statistic shows less passes and higher noise level. But personally I prefer cleaner picture rather than some numbers from statistic lines :) I like this new filtering. Hope to see that it starts to work with Fast preview denoising soon.
Thank you guys!
-
I will get back to the testing on weekend. I just have cuple of *real life* scenes with a lot of complications on dof and highlights.
-
And you also should not render so clean images. Sure it's good to test how the new image filter contributes to final quality image, but it will be harder for human eye to spot the differences. I'd do one more test, again with the same amount of passes, but lower amount of them (let's say a half) so you can spot the differences better.
From the current test, it appears that the HQ image filter mainly helps to resolve very small highlights/glints.
I get what you are saying, but to me practically, the noise in both of these images (HQ and non-HQ) are the minimum amount of noise I'd be willing to accept in my renders. There isn't any reason for me to make comparisons with more noise/less passes, since I'd never use them any way. My tests are more practical for me to figure out if this feature would benefit me.
Maybe other people can focus more on the academic, less passes/more noise tests.
-
And you also should not render so clean images. Sure it's good to test how the new image filter contributes to final quality image, but it will be harder for human eye to spot the differences. I'd do one more test, again with the same amount of passes, but lower amount of them (let's say a half) so you can spot the differences better.
From the current test, it appears that the HQ image filter mainly helps to resolve very small highlights/glints.
I get what you are saying, but to me practically, the noise in both of these images (HQ and non-HQ) are the minimum amount of noise I'd be willing to accept in my renders. There isn't any reason for me to make comparisons with more noise/less passes, since I'd never use them any way. My tests are more practical for me to figure out if this feature would benefit me.
Maybe other people can focus more on the academic, less passes/more noise tests.
This yes, wanted to mention that too about the logic in testing on final quality renders. It is like we are aiming for final quality in the end right? So that is what we should be comparing.
Just was to sleepy yesterday.
-
I've tested so far and found the quality to be just about equal but the render time to go up a LOT!
-
Hi, after some delay due to illness I'm happy to report that the issue with adaptivity has been fixed, and render times are pretty much the same now. So it should be a pure win :)
It's in review and testing now and should be out ASAP!
-
Does intel denoiser still not work with new noise filter? In my tests it doesn't
-
Correct, it doesn't, as mentioned in both the Max and C4D release blogs:
"Note: Neither of the AI denoisers will work with the new filtering, so it is automatically turned off if either of those is selected. As with all things AI, the software has been “trained” on particular examples of “noisy version of the image vs. clean version of the image” and that training has been with the old, harsh type of noise. We are working with Intel to have their AI denoiser retrained to work with our High Quality Image Filtering, so hopefully that restriction will be removed in the future!"