Chaos Corona Forum

Chaos Corona for 3ds Max => [Max] I need help! => Topic started by: mrjojo on 2015-03-06, 08:41:05

Title: Strange material behaviour
Post by: mrjojo on 2015-03-06, 08:41:05
Hi,
Im getting really weird (reflections?) along the edges of my objects. If you look closely at the coffee mug or the eggs youd see what I mean.Any ideas what might be going wrong?
Title: Re: Strange material behaviour
Post by: maru on 2015-03-06, 09:16:58
Unfortunately it looks like this, so this would be expected:
https://forum.corona-renderer.com/index.php/topic,4542.45.html

Solution is to change material settings, for example reduce reflectivity.
Title: Re: Strange material behaviour
Post by: GestaltDesign on 2015-03-06, 09:33:58
Could another option be to revert back to the old default BRDF by ticking the 7. 2 legacy checkbox in the advanced rollout, you may get more predictable results.
If you are desperate for GGX without significantly changing material behaviour you could try the nested blend approach as described in this thread:
https://forum.corona-renderer.com/index.php?topic=4260.0 (https://forum.corona-renderer.com/index.php?topic=4260.0)
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 10:41:05
Unfortunately it looks like this, so this would be expected:
https://forum.corona-renderer.com/index.php/topic,4542.45.html

Solution is to change material settings, for example reduce reflectivity.

You mean reduce reflectivity level, or the reflection color to a darker value?
Title: Re: Strange material behaviour
Post by: maru on 2015-03-06, 10:53:04
Unfortunately it looks like this, so this would be expected:
https://forum.corona-renderer.com/index.php/topic,4542.45.html

Solution is to change material settings, for example reduce reflectivity.

You mean reduce reflectivity level, or the reflection color to a darker value?
It should have the same result.
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 11:29:47
I set the reflection level to 0.8 and heres the result after 95 passes which took bout 27 minutes. I also put my render setting next to my render.Tell me if these settings are any good for a clean and noiseless render? if not what would you suggest?
Title: Re: Strange material behaviour
Post by: maru on 2015-03-06, 11:37:35
Is this 1.0 version? Use UHD cache indstead of HD.
Is your problem too much noise or that it takes too much time? I can't see much noise in this image.
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 12:01:17
Is this 1.0 version? Use UHD cache indstead of HD.
Is your problem too much noise or that it takes too much time? I can't see much noise in this image.

Yea im using 1.0 demo version. Im new to Corona so I wouldnt know what the best balance between quality and time?
Title: Re: Strange material behaviour
Post by: maru on 2015-03-06, 12:09:27
Generally, if you have some goal rendering time you would like to achieve, and you get bad results in it (for some reaseon), you can:
-ignore noise for now and see how many passes produce good edges of geometry and textures (or example if it's 16 passes, or 32, or 64?)
-then use the lowest possible number of passes (let's say 16)
-render 16 passes and see if there are any other problems (noise?)
-if there is noise, this most likely means you need to increase sampling settings, either GI/AA value or LSM
This is just an example. If you don't care so much about rendering time, you can simply set no time/pass limit and just wait for the image to refine.

Maybe this guides will be helpful:
https://coronarenderer.freshdesk.com/support/solutions/articles/5000516729
https://coronarenderer.freshdesk.com/support/solutions/articles/5000516731
Title: Re: Strange material behaviour
Post by: romullus on 2015-03-06, 12:15:53
Something is wrong with your image, there's almost no contact shadow. Check your material settings and or HDRI quality.
Render settings are also wrong. If i'm right, there's no enclosed enviroment in your scene. In that case you don't need such high GI samples. Set GI vs. AA to 8 and LSM to 4.

Your liquids are modeled incorrectly. Look here how to properly do this in Corona: https://coronarenderer.freshdesk.com/support/solutions/articles/5000515618-how-to-model-a-liquid-in-glass-in-corona-

Also you're getting too lowperformance out of your CPU. Might be due to bitmap paging. Read this: https://forum.corona-renderer.com/index.php/topic,1662.msg12232.html#msg12232
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 13:20:13
Something is wrong with your image, there's almost no contact shadow. Check your material settings and or HDRI quality.
Render settings are also wrong. If i'm right, there's no enclosed enviroment in your scene. In that case you don't need such high GI samples. Set GI vs. AA to 8 and LSM to 4.

Your liquids are modeled incorrectly. Look here how to properly do this in Corona: https://coronarenderer.freshdesk.com/support/solutions/articles/5000515618-how-to-model-a-liquid-in-glass-in-corona-

Also you're getting too lowperformance out of your CPU. Might be due to bitmap paging. Read this: https://forum.corona-renderer.com/index.php/topic,1662.msg12232.html#msg12232

Thanks a bunch man very informative. However, what is there to be checked in my materials tho? what might be preventing the contact shadows?
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 19:38:47
Generally, if you have some goal rendering time you would like to achieve, and you get bad results in it (for some reaseon), you can:
-ignore noise for now and see how many passes produce good edges of geometry and textures (or example if it's 16 passes, or 32, or 64?)
-then use the lowest possible number of passes (let's say 16)
-render 16 passes and see if there are any other problems (noise?)
-if there is noise, this most likely means you need to increase sampling settings, either GI/AA value or LSM
This is just an example. If you don't care so much about rendering time, you can simply set no time/pass limit and just wait for the image to refine.

Maybe this guides will be helpful:
https://coronarenderer.freshdesk.com/support/solutions/articles/5000516729
https://coronarenderer.freshdesk.com/support/solutions/articles/5000516731

Thanks for the links. Tho, do you have any idea why im not getting contact shadows out of my materials?
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 20:18:33
Hey guys, heres a quick update, I remodeled the liquid based on the link that romullus provided but the milk renders kind of odd looking as if its outside of the glass.

P.S: I rendered the scene with a different HDRI which is quite high res (200mb in size) but still im not getting any contact shadows! :(
Title: Re: Strange material behaviour
Post by: maru on 2015-03-06, 20:59:05
but the milk renders kind of odd looking as if its outside of the glass
This is correct. See http://upload.wikimedia.org/wikipedia/commons/9/9d/Presidents_Choice_-The_Decadent,_chocolate_chip_cookie,_with_a_glass_of_milk.jpg Your glass is very thick, that's why it's even more apparent.

Contact shadows - I think it's more dependent on light than on materials unless your are using an incorrectly set up bump map.
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-06, 21:31:16
but the milk renders kind of odd looking as if its outside of the glass
This is correct. See http://upload.wikimedia.org/wikipedia/commons/9/9d/Presidents_Choice_-The_Decadent,_chocolate_chip_cookie,_with_a_glass_of_milk.jpg Your glass is very thick, that's why it's even more apparent.

Contact shadows - I think it's more dependent on light than on materials unless your are using an incorrectly set up bump map.

well as far as light goes, im using an HDRI in the environment slot, thats pretty much it. Is there anything to do to get decent contact shadows without doing it in post?
Title: Re: Strange material behaviour
Post by: PROH on 2015-03-06, 22:29:22
Hi. Try this:

1) Check how HDRI behøves with default corona materiale. Make a small box on top of a bigger box, both with default material. Are there any contact shadows? If not, you need more light. If yes, then go to next point.

2) How is the table material made? Shadow catcher or a Wood-material? How does it behave without bump/normal map/displacement? Any kind of self illumination?

Hope it helps
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-07, 07:46:51
Ive ran a test with two boxes with corona and standard materials applied to each one, contact shadows on corona material are a bit stronger than the standard but still not strong enough. Also I dont see boxes making any contact shadows with the table. Thoughts?
Title: Re: Strange material behaviour
Post by: romullus on 2015-03-07, 08:32:55
There is shadow. But if you want it to be more distinct, make sure that yor object don't touch ground very firmly. Either lift them a notch, or give a small chamfer to outer edges of those boxes.

P.S: I rendered the scene with a different HDRI which is quite high res (200mb in size) but still im not getting any contact shadows! :(

File size of HDRI tells almost nothing about its quality.
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-07, 08:42:05
There is shadow. But if you want it to be more distinct, make sure that yor object don't touch ground very firmly. Either lift them a notch, or give a small chamfer to outer edges of those boxes.

P.S: I rendered the scene with a different HDRI which is quite high res (200mb in size) but still im not getting any contact shadows! :(

File size of HDRI tells almost nothing about its quality.

So how do I tell if an HDRI is high quality or not? Any personal favorite website you may wanna suggest where I can get decent HDRI ?
Title: Re: Strange material behaviour
Post by: GestaltDesign on 2015-03-07, 14:00:53
This will get you going:-

http://noemotionhdrs.net/

Perhaps when setting up lighting / playing with HDRIs, start with all objects using the same simple Corona material.
This will help you understand light and shadow interactions better.
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-07, 15:34:42
This will get you going:-

http://noemotionhdrs.net/

Perhaps when setting up lighting / playing with HDRIs, start with all objects using the same simple Corona material.
This will help you understand light and shadow interactions better.

I couldnt find any indoor hdri on their website. Is there any decent and free hdri that i can get my hands on?
Title: Re: Strange material behaviour
Post by: Ondra on 2015-03-07, 17:47:53
the ones from Paul Debevec are really good, try those
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-07, 18:23:15
the ones from Paul Debevec are really good, try those

Are those free or paid and where can I acquire them?
Title: Re: Strange material behaviour
Post by: Ondra on 2015-03-07, 18:31:58
http://www.pauldebevec.com/Probes/ - I see that the resolution is low by todays standard, but the dynamic range is excellent. They are still good enough for experimenting
Title: Re: Strange material behaviour
Post by: mrjojo on 2015-03-07, 18:49:24
http://www.pauldebevec.com/Probes/ - I see that the resolution is low by todays standard, but the dynamic range is excellent. They are still good enough for experimenting

Thanks a bunch! :)