Chaos Corona Forum

Chaos Corona for 3ds Max => [Max] General Discussion => Topic started by: Ondra on 2014-09-15, 20:31:08

Title: Experiment: Noise levels
Post by: Ondra on 2014-09-15, 20:31:08
Which image has less noise? And how big difference do you see?

1:
(http://i.imgur.com/SgrXj39.png) (http://i.imgur.com/SgrXj39.png)


2:
(http://i.imgur.com/LtWCJTK.png) (http://i.imgur.com/LtWCJTK.png)
Title: Re: Experiment: Noise levels
Post by: zzubnik on 2014-09-15, 20:42:26
By a slim margin, number two feels cleaner to my eyes, after tabbing between them in Photoshop for several minutes.

Title: Re: Experiment: Noise levels
Post by: romullus on 2014-09-15, 20:42:30
I think #1 has a little bit more uniform noise, so it seems like less noisy. But i'm very tolerant to noise, so for me, difference is almost imperceptible.
Title: Re: Experiment: Noise levels
Post by: ecximer on 2014-09-15, 21:13:46
№1.
the difference is noticeable, but not much.
Title: Re: Experiment: Noise levels
Post by: CiroC on 2014-09-15, 21:25:44
Nº1 for me. The render seems to have a better contrast.
Title: Re: Experiment: Noise levels
Post by: Juraj on 2014-09-15, 21:27:06
I also think number one, maybe because the 'grain' seems smaller to me ?
Brother thinks opposite (num2), he's not 3d artist.
Title: Re: Experiment: Noise levels
Post by: Benjamin_F on 2014-09-15, 22:42:05
I think number one is more homogeneous, especially near the window.
Title: Re: Experiment: Noise levels
Post by: johan belmans on 2014-09-15, 22:42:49
for me number one, but hardly noticeable
Title: Re: Experiment: Noise levels
Post by: juang3d on 2014-09-15, 23:33:23
I agree with Benjamin_F, i used the wall near the window to make my choice.
Number one for me :)

Cheers.
Title: Re: Experiment: Noise levels
Post by: Siahpoosh on 2014-09-15, 23:39:17
#1 , is there any difference in lighting ? i can see a different in candle shadow on the wall.
Title: Re: Experiment: Noise levels
Post by: Captain Obvious on 2014-09-16, 15:49:38
"LtWCJTK.png" is marginally noisier. I measured. It's something like 2.4 % more noisy.

edit: So that's image 1 that's cleaner.
Title: Re: Experiment: Noise levels
Post by: Ondra on 2014-09-16, 16:10:08
"LtWCJTK.png" is marginally noisier. I measured. It's something like 2.4 % more noisy.

edit: So that's image 1 that's cleaner.
How did you measure it without the reference image? If you have some reliable metric for estimating noise tell me, I can use it for adaptivity ;)
Title: Re: Experiment: Noise levels
Post by: steyin on 2014-09-16, 16:23:34
They look exactly the same to me.
Title: Re: Experiment: Noise levels
Post by: Stan_But on 2014-09-16, 16:57:23
My vote for first also
Title: Re: Experiment: Noise levels
Post by: pokoy on 2014-09-16, 18:24:04
No 1 feels cleaner, but no 2 seems to have a more uniform noise across the whole image.
Title: Re: Experiment: Noise levels
Post by: Siahpoosh on 2014-09-16, 18:45:22
look at this
(https://forum.corona-renderer.com/index.php?action=dlattach;topic=5216.0;attach=20768;image)
Title: Re: Experiment: Noise levels
Post by: elindell on 2014-09-16, 18:54:59
1 also feels cleaner if I scrub back and forth between them, but honestly just eyeballing it they look of equal noisiness
Title: Re: Experiment: Noise levels
Post by: maru on 2014-09-16, 19:44:40
I think the two images are extremely similar in terms of noise but if I had to choose, I would pick #1. Looks like it has (at least perceptually) less noise in the direct lighted area near the window. But I wouldn't be surprised if Keymaster was testing us and the two images were identically noisy. ;)
Title: Re: Experiment: Noise levels
Post by: Ondra on 2014-09-16, 19:55:54
#1 is less noisy. It uses one trick from recent paper about sampling. It will be probably included in daily build soon
Title: Re: Experiment: Noise levels
Post by: Siahpoosh on 2014-09-16, 20:23:39
"LtWCJTK.png" is marginally noisier. I measured. It's something like 2.4 % more noisy.

edit: So that's image 1 that's cleaner.
How did you measure it without the reference image? If you have some reliable metric for estimating noise tell me, I can use it for adaptivity ;)
Maybe this could help :
http://homepages.inf.ed.ac.uk/rbf/CVonline/LOCAL_COPIES/VELDHUIZEN/node18.html
Title: Re: Experiment: Noise levels
Post by: Chakib on 2014-09-16, 20:29:17
I can see that the first image had a liiiittle bit less noise  than the second one, especially on the back of the chair.
Title: Re: Experiment: Noise levels
Post by: zzubnik on 2014-09-16, 20:38:30
Well, I guess I'm going blind. I really struggled to make out much overall difference. I think I need glasses!
Title: Re: Experiment: Noise levels
Post by: filipskrzat on 2014-09-16, 20:50:35
#1 is less noisy
Title: Re: Experiment: Noise levels
Post by: juang3d on 2014-09-16, 23:11:02
Awesome :D

Will this make the images look cleaner sooner?

Cheers
Title: Re: Experiment: Noise levels
Post by: cecofuli on 2014-09-17, 00:29:38
Maybe... are you working also on "noise pattern flickering" problem?  =^___^=
Title: Re: Experiment: Noise levels
Post by: Captain Obvious on 2014-09-17, 09:25:50
"LtWCJTK.png" is marginally noisier. I measured. It's something like 2.4 % more noisy.

edit: So that's image 1 that's cleaner.
How did you measure it without the reference image? If you have some reliable metric for estimating noise tell me, I can use it for adaptivity ;)
I just measured the maximum deviance between each pixels and its neighbours. It doesn't work for calculating the error in a single image (you need the reference image for that), but it works for checking which of two images have the greater variance.
Title: Re: Experiment: Noise levels
Post by: gabrielefx on 2014-09-17, 09:49:59
try to reduce the noise putting 50 ies or tiny emitters inside the room.
Title: Re: Experiment: Noise levels
Post by: Zom-B on 2014-09-17, 11:17:45
#1 is less noisy. It uses one trick from recent paper about sampling. It will be probably included in daily build soon
There is a very simple way to tell what image has less noise: File size for the PNG!
PNGs are super mega sensible to noise vs filesize ratio.
Title: Re: Experiment: Noise levels
Post by: cecofuli on 2014-09-17, 13:31:03
try to reduce the noise putting 50 ies or tiny emitters inside the room.

Can you explain better?
Title: Re: Experiment: Noise levels
Post by: boumay on 2014-09-28, 15:39:31
try to reduce the noise putting 50 ies or tiny emitters inside the room.
Yes it would be interesting to know more