Chaos Corona Forum

Chaos Corona for 3ds Max => [Max] Feature Requests => [Max] Resolved Feature Requests => Topic started by: Siahpoosh on 2014-09-08, 13:16:44

Title: XYZ bias for AO
Post by: Siahpoosh on 2014-09-08, 13:16:44
it would be nice to have XYZ bias for AO map.
Title: Re: XYZ bias for AO
Post by: Ondra on 2014-09-08, 14:18:55
What would that feature do?
Title: Re: XYZ bias for AO
Post by: romullus on 2014-09-08, 14:45:05
Something similar like richdirt (http://www.enrichpro.com/en/richdirt/fea-sill-edge.html) does, i guess. But instead XYZ, wouldn't it be more logical to use UVW?

p.s. i'm not advocating this request at all ;]
Title: Re: XYZ bias for AO
Post by: Ludvik Koutny on 2014-09-08, 15:02:34
Something similar like richdirt (http://www.enrichpro.com/en/richdirt/fea-sill-edge.html) does, i guess. But instead XYZ, wouldn't it be more logical to use UVW?

p.s. i'm not advocating this request at all ;]

It's harder with UVW afaik, cause you need to have your UV's always oriented certain way. With XYZ bias in VrayDirt, you can do it in wordlspace regardless of UV space.

I remember i used it to make burned train wagon. I just took unwrapped clean model of train wagon, created material of burned metal, and then blended these two materials using VrayDirt modulated by noise map stretched in Z axis upwards. It created those nice black streaks as if flames where burning out of the wagon windows :) And i could then use that blend material and make any material i plug as a base layer appear burned :)

In same manner, if you want to do weathering leaks on concrete for example, you want it based on world axis, as rain falls from top to bottom... not UVs ;)
Title: Re: XYZ bias for AO
Post by: Siahpoosh on 2014-09-08, 20:13:30
using UVW is wrong way for calculating AO .
Title: Re: XYZ bias for AO
Post by: maru on 2014-09-08, 20:23:15
I remember i used it to make burned train wagon. I just took unwrapped clean model of train wagon, created material of burned metal, and then blended these two materials using VrayDirt modulated by noise map stretched in Z axis upwards. It created those nice black streaks as if flames where burning out of the wagon windows :) And i could then use that blend material and make any material i plug as a base layer appear burned :)
And you can do the same in Corona by plugging a stretched noise map into AO distance slot, right?
Title: Re: XYZ bias for AO
Post by: Siahpoosh on 2014-09-08, 20:43:32
I remember i used it to make burned train wagon. I just took unwrapped clean model of train wagon, created material of burned metal, and then blended these two materials using VrayDirt modulated by noise map stretched in Z axis upwards. It created those nice black streaks as if flames where burning out of the wagon windows :) And i could then use that blend material and make any material i plug as a base layer appear burned :)
And you can do the same in Corona by plugging a stretched noise map into AO distance slot, right?
yes but you will get dirt in all directions  . with bias option we can change the direction of dirt :
http://help.chaosgroup.com/vray/help/150SP1/examples_vraydirt.htm#ex7 (http://help.chaosgroup.com/vray/help/150SP1/examples_vraydirt.htm#ex7)
http://www.peterguthrie.net/blog/2009/10/vray-dirt-tutorial/ (http://www.peterguthrie.net/blog/2009/10/vray-dirt-tutorial/)
Title: Re: XYZ bias for AO
Post by: maru on 2014-09-09, 14:26:29
Understood.
Title: Re: XYZ bias for AO
Post by: Ludvik Koutny on 2014-09-09, 15:02:44
Yes, exactly... say you have concrete with some details on facade. You want leaks going only downwards. If you just use stretched noise without directional bias, you will get streaks from both sides, looking wrong :)
Title: Re: XYZ bias for AO
Post by: vkiuru on 2014-09-29, 10:21:57
Yes, this would be a greatly appreciated feature.
Title: Re: XYZ bias for AO
Post by: Ondra on 2015-02-02, 16:46:19
implemented
Title: Re: XYZ bias for AO
Post by: pokoy on 2015-02-02, 18:41:27
That's a great feature, thanks for implementing it.

One question: is the direction bias aligned to the world or local object's coordinates? Local object coordinates are important so it 'sticks' with the object when it's animated/rotated.

Another useful option would be to make the bias depend on an arbitrary object's local coordinates in the scene. That way, you could assign the map to a bunch of objects where all of them could have different local coordinates, with the directional bias being controlled by one object only. That would be really useful.
Title: Re: XYZ bias for AO
Post by: CiroC on 2015-02-02, 22:09:42
implemented

Great news indeed. This is very useful. Many thanks
Title: Re: XYZ bias for AO
Post by: romullus on 2015-02-02, 23:12:59
Yeap, really cool!
Now it's time to implement anisotropy support for object space in CoronaMtl *wink, wink*
Title: Re: XYZ bias for AO
Post by: Christa Noel on 2015-02-03, 05:39:42
implemented
bravo!
Title: Re: XYZ bias for AO
Post by: racoonart on 2015-02-03, 11:58:11
Love it! Combined with Ray directionality it's really easy to use :)

Title: Re: XYZ bias for AO
Post by: Ludvik Koutny on 2015-02-03, 12:16:41
That is one dirty building!
Title: Re: XYZ bias for AO
Post by: Javadevil on 2015-02-05, 12:37:25
implemented

Holly shit ! thats awesome, your knocking out the features !!
Title: Re: XYZ bias for AO
Post by: Siahpoosh on 2015-02-09, 13:27:01
Woow , great job , thank you