Sunshine (noise tests)

Discuss or ask critique about your current works
Forum rules
Important notice: All artwork submitted on our public gallery forums gallery forums may or may not be used by OTOY for publication on our website gallery.
If you do not want us to publish your art, please mention it in your post clearly. (put a very red small diagonal cross in the left right corner of the image)
Any images already published on the gallery will be removed if the original author asks us to do so.
We recommend placing your credits on the images so you benefit from the exposure too, and use a minimum image width of 1200 pixels, and use pathtracing or PMC. Thanks for your attention, The OctaneRender Team.


For new users: this forum is moderated. Your first post will appear only after it has been reviewed by a moderator, so it will not show up immediately.

This is necessary to avoid this forum being flooded by spam.
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

Thanks Rob. Your post on the Altus denoising system actually inspired me to look at some alternatives to neat video. The 1000 - 2000 samples range with noise reduction is where I am aiming, but some very exciting technologies indeed. Animations will be the true test...

Looking forward to your test images nuno. Saw your recommendations utilized on another post with some very nice results. Will need to explore more in the future :D
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

Same image/ Post-Processing. 500 vs 1000 samples with PT. Denoised first with neat, then a second pass with denoiser II (helped clean up some of the hot pixels which was a new discovery).
Attachments
Arch09_Vol007_500Samples(01).jpg
Arch09_Vol007_1000Samples(01).jpg
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

500 de-noised.
Attachments
Arch09_Vol007_500Samples(02).jpg
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

Last one tonight, I promise :D

Pt. 1000 samples.
Attachments
Arch13_Vol003_1000Samples(01).jpg
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

PT. 500-1000 samples. Everything normally clean enough for decent denoise at 500, but the spectral highlights are the one thing that takes a while to clean without producing strange patterns. Perhaps octane 3 handles these a bit better?
Attachments
Girl01.jpg
Girl01_500samples01.jpg
Girl01_1000samples01.jpg
djropi
Licensed Customer
Posts: 33
Joined: Sat Aug 09, 2014 3:11 pm

Pt. 500-1000 samples. Caustic blur @ 1. 2nd image had some spectral noise which did not clean very well at 1000. Attempting to keep hot pixel removal at 8 or above, but going lower with a combo of de-noiser sometimes does the trick. Think it is time to start doing some animation tests soon...
Attachments
Bro01_500samples.jpg
Bro02_1000samples.jpg
Bro03_500samples.jpg
Post Reply

Return to “Works In Progress”