Chaos Corona Forum

Chaos Corona for 3ds Max => [Max] Feature Requests => [Max] Resolved Feature Requests => Topic started by: racoonart on 2012-09-06, 13:21:18

Title: deadclowns wishlist
Post by: racoonart on 2012-09-06, 13:21:18
There are some minor and major wishes I have for future versions of Corona:

- a button to save beauty and renderelements in one go (as in vray)
- AO map - distance mapslot (great thing to make those rotten buildings look)
- glossy reflections values linear. Currently it looks like the glossiness curve is more of an exponential function. It would be a lot easier to "guess" the right values if it would be a more linear way (vray & mr are linear, finalRender is like corona here)
- anisotropy with world axis (it's uvw coordinates right now, right?)
- diffraction/obstacle map/colored aperture maps
- matte-shadow material/ object properties
- visible in reflection switch in object properties
- pflow support
- world position renderelement
- keep continuity for displacement
- reflection dim distance
- spherical (360*180°) camera
- aperture, exposure, white balance settings in camera
- affect GI / reflections in corona lights
- direct lights / ies lights
- Endless plane primitive object
Title: Re: deadclowns wishlist
Post by: Ludvik Koutny on 2012-09-06, 14:10:21
I would like glossiness to remain the way it is. I actually really dislike Vray's glossiness curve as its way too sensitive and putting glossiness maps in takes a lot of color corrections to get right amount of glossiness variance.

BTW MentalRay (Arch&Design mat) and Vray (VrayMTL) have very different glossiness curves, and one in Corona is actually very close to one in MR Arch&Design material ;)
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-06, 14:38:45
mhh.. okay, I could have sworn the arch&design material would behave like the Vray mat. But I don't use it really often so I'm sure you're right.
I'm pretty happy with vray handling it's glossiness but I think it's all a matter of habit. I didn't expect you would want to change it anyways ;)
Title: Re: deadclowns wishlist
Post by: Ludvik Koutny on 2012-09-06, 15:10:01
There will be several levels of user interface in future releases...

Basic - with presets and speed/quality sliders for various aspects (sampling, GI, etc...)

Advanced - where there will be still a lot less clutter than currently is (current UI is debug one), which will allow user to set up some advanced parameters and tweak things

Debug - will contain all the controls which you would not normally want to touch.

It would be no problem to to add a value that would control exponent of glossiness value curve in Debug level of UI for expert users ;)
Title: Re: deadclowns wishlist
Post by: Ondra on 2012-09-06, 15:11:55
The problem with glossiness mapping is, that it is VERY dependent on concrete situation, so what seems fine in one scenario might be quite off in another. Also changing it would affect existing scenes, so I prefer not to mess with it, unless its really clear, that it is off.

Anisotropy is (should be ;)) relative to world coordinates

Displacement should never make holes, if the base geometry is sane and it has single smoothing group. Try using single smoothing group for your model, or at least the displaced parts

Why do you want reflection dim distance? That is one fake I do not want to implement, because I see no use for it. The overall attitude is that you should be able to get by in Corona with minimum fakes, because the renderer should handle most situation without them.

Other things are doable and will eventually get implemented. Unfortunately, I dont have time for it right now (1-2weeks)
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-06, 16:17:25
Quote
Displacement should never make holes, if the base geometry is sane and it has single smoothing group. Try using single smoothing group for your model, or at least the displaced parts
That's right, there are no holes on areas which are sharing the same smoothing group. My request is to keep the boundaries between those groups closed. Example: Simple Box primitive with a noise map in the displacement slot. Each of the 6 sides will be offset in their normal direction and the borders will break. Vray & finalrender have an option to keep those borders closed.

Quote
Why do you want reflection dim distance? That is one fake I do not want to implement, because I see no use for it. The overall attitude is that you should be able to get by in Corona with minimum fakes, because the renderer should handle most situation without them.
Personally, I don't like it either, but some of my clients do. I had several "shiny white apple-look" product animations lately with an object on a white reflective plane and some of my clients like strong reflections near to the object which are fading very fast to white (or whatever color the background has). It's just kind of tricky to do that later if you're not able to set this up in your materials.
Title: Re: deadclowns wishlist
Post by: Ludvik Koutny on 2012-09-06, 17:04:14
Reflection distance cutoff is way too much of a fake IMHO...  The effect you are talking about is usually a simplified perception of not perfectly sharp reflection, where sharpness and therefore also clarity of reflection falls off as the reflection get's more distant from contact point.

Clients may perceive and understand it as simple fading of reflection. You can either show them it can be done better, or you will have to use some trickery, or eventually other renderer.

Fakes like this one usually mess up the core, and clean core is one of the reasons Corona is so fast and efficient.

I would simply solve it by using either glossy reflection, or putting falloff map into reflection slot, and then tweaking parallel to perpendicular curve to hit the sweet spot, where angle of camera against surface creates the fade effect you need.

Once i have some spare time, i might be able to pump out an example scene.
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-06, 17:32:22
Don't get me wrong here. It's more than ok to follow the "we don't want to do fakes" agenda. If that will keep the renderer fast and streamlined I'm supporting that idea. I know there are several ways to do this behavior - it's just a little bit of extra work, setting up some extra passes to do it via postwork for example. If you choose to avoid fake techniques then, please, stick to that idea!
Title: Re: deadclowns wishlist
Post by: Ludvik Koutny on 2012-09-06, 17:37:50
There will be some fakes...  that kind of fakes, which are sometimes necessary, like no direct visibility, toggle of shadow casting, matte shadow, etc...  But things like reflection falloff which were originally meant to improve rendering times rather than to be used as a tool would probably not be worth it ;)

I am always ready help you achieve the result by suggesting an alternative way ;)
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-14, 16:01:10
4 more wishes:
- CoronaLight mapslot (since the material emission thing has no directionality)
- 2 Sided material (front-back)
- Region rendering in corona VFB (i know it takes 3dsmax' regions into account but there is no way to set it via the VFB)
- Ability to load the plugins over network. I didn't test it but since the corona_release.dll is located in the max root i'm not sure if this will work (and installing it on several nodes evertime is a bit time consuming)
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-15, 13:19:11
and another 3:
- please do some better RE naming. Lets say I want the global renderoutput filename to be "basename_1234.exr" , then my renderelements will be named "basename__RenderlementA1234.exr"
I also don't like the spaces in names :S
- please add a Velocity renderelement. There are cases where I want to do some little blurring later and 3D moblur is not needed.
- Refraction Glossiness on twosided (thin) mode, currently it seems to be deactivated in this mode
Title: Re: deadclowns wishlist
Post by: Ondra on 2012-09-15, 13:28:40
RE naming: oh, max uses it directly as filenames? Ok, in that case I'll add underscores, or something

Velocity: It will be added sometimes in the future, Unfortunately it wont be as easy as I hoped

Refraction glossiness on twosided glass: because the rays enter and immediately leave the material, they dont have chance to spread (they have 0 length, so it does not matter what the IOR or glossiness is) - therefore the parameter has no effect. To simulate thin frosted glass, either model it with thickness and use solid glass, or use 1 plane geometry with solid glass and IOR 1. In either case, the shadows optimizations cannot be done, so its not very suitable for for example windows.
Title: Re: deadclowns wishlist
Post by: Ludvik Koutny on 2012-09-15, 13:54:42
Low refraction glossiness for 1sided plane is pretty much translucency ;)
Title: Re: deadclowns wishlist
Post by: Ondra on 2012-09-15, 14:03:24
Translucency has the lobe centered around inverse normal, refraction around ideal refraction direction ;). If you dont care which one to use, use always translucency.
Title: Re: deadclowns wishlist
Post by: racoonart on 2012-09-15, 22:45:16
Ok, i can live with that.
Another thing:
- Is it possible to get a separate Reflection pass? Me (any my clients) like to have a separate reflection pass for compositing. Currently it's bundled with the lighting pass :(
Title: Re: deadclowns wishlist
Post by: Ondra on 2012-09-15, 23:35:59
There is Reflection GI pass, which does not include a portion of direct lighting. I am reluctant about making full reflection pass now, because it would inevitably lead to an overhead :/. But I know it will be necessary to add in future