alShader 2.0 ?

[Stephan Wagemann]

Hi, there is no sign on the bitbucket source for any update related to arnold 5.
You are working behind the curtain on this?

[karas]

I think yes!!

在 2017年4月14日星期五 UTC+8下午9:03:32，Stephan Wagemann写道：

[vasilis2...@gmail.com]

at least the alLayer and Textures (noises)  .... not change the flow of our work ;

please !

[vasilis2...@gmail.com]

https://docs.chaosgroup.com/display/OSLShaders/Additional+procedural+noises.

similar with alshaders textures, works wirh arnold 5

[Anders Langlands]

I'm working on an update to arnold 5 for most of the library. This will not include alSurface, alHair or alCel, and I'm not sure whether alLayer will really work properly. The intention is to make it easier for people to transition existing setups without having to rebuild things entirely from scratch and should be seen as a stop-gap only. 

Most of what alShaders offered for arnold 4 is now available out of the box in 5, so it makes sense to just switch to in-built nodes where appropriate as soon as possible since those will be officially supported by SA.

I may well release something specifically targeted at arnold 5 further down the line, but that will look very different (and much smaller) than what the current version looks like, and it remains to be seen what things need writing to supplement the existing toolset.

Cheers,

Anders

On Mon, 17 Apr 2017 at 02:01 <vasilis2...@gmail.com> wrote:

https://docs.chaosgroup.com/display/OSLShaders/Additional+procedural+noises.

similar with alshaders textures, works wirh arnold 5

--
You received this message because you are subscribed to the Google Groups "alshaders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Kopa Kopatych]

please alSurface for mtoa 2.0, aiStandardSurface dont have directional sss model, and  he has the attributes of ior and roughness common for specular and transmittance.Besides in 4 arnold core alSurface render faster more aiStandard
sorry for my English

[Bor Arroyo]

Hi! I join this proposal but I wanted to comment on something. I'm having a problem .. there is no version of arnold where you can open an alsurface shader at the same time as an aiStandartSurface. For all the people that made a library with alSurface is a little problematic to try to transform it. Also it complicates a lot the creation of a script to be able to do it.
I understand the argument that the correct thing is to make a transition from one shader to another for many reasons, but do not think that the ideal would be to have a last version of alshaders that coexisted in a same version of arnold to try to make these conversion scripts and from there discontinued it?

Thank you very much Anders

[Anders Langlands]

Hmm I could make an "empty" alSurface shader that just has the parameters on it but doesn't actually do anything. Would that help in conversion?

[Bor Arroyo]

That sounds perfect for me Anders, with that it would be much easier to make that conversion, sure many would appreciate it.
Thx!

[Jean Lim]

Hi,Anders.Is that means alSurface will not continue develop base on arnold 5，it will no longer  exist in “alShader” later release？I really like the alSurface shader，it still batter than the aiSurface standard shader，I think.

在 2017年4月17日星期一 UTC+8下午1:26:49，Anders Langlands写道：

[Anders Langlands]

Due to changes in Arnold 5 I can't just "upgrade" alSurface. I'm keen to try out the new standard_surface shader for a bit as it seems to do most of the important stuff alSurface did. I may decide to write something new later, but it will be different from both.

[Tino]

that would be great - also the everywhere used alRemapcolor would be nice to have and make the conversion very simple

Am 20.04.17 um 22:32 schrieb Anders Langlands:

[Trẫm Sup]

- Physical sky in Arnold 5:    Thin Walled's working with Sun but it is NOT working with Sky    -> Look Terrible !

- Physical sky in Arnold 4 :   Alshader Backlighting work very well with both Sun and Sky       -> Look Realistic !

-> I'm unbearable when looking result of Thin walled in Arnold 5.  So I hope Anders Langlands can re-write the Alshader or other new shader for C4DtoA with Arnold core 5 soon. It will be really amazing!

Vào 20:03:32 UTC+7 Thứ Sáu, ngày 14 tháng 4 năm 2017, Stephan Wagemann đã viết:

[Frederic Servant]

Hi everyone, hi Anders,

I'm from the Arnold team.

@vasilis: We will add more noises soon. Generally we will add more production ready shaders in the near future.

@Kopa: We believe the new empirical SSS that is used in standard surface is as good and faster than the directional SSS. We previously used a cubic SSS that was waxier. If you still think directional is better, can you post images? Also, can you compare the render times?

@Kopa: I think having the same IOR for spec and transmittance makes sense. Do you have a case where you need a different one? There is an extra roughness parameter  for transmittance in standard surface.

@Tino: The color_correct shader does everything alColorRemap does.

@Tram: Thin walled works fine for me with physical sky. Be aware that skydome lights are directly visible in Arnold 5, so there's no need to use a background anymore. You can revert to the old behavior with the "camera" parameter on the skydome. 

To everyone: We gladly welcome any feedback you have on Arnold 5, for example:

- what's do you miss the most in the standard surface compared to alSurface?

- which shaders are missing in Arnold 5?

- how does it compare to alShaders in terms of performance?

- any other feedback

Cheers,

--

Fred

On Sat, Apr 22, 2017 at 10:09 AM, Trẫm Sup <supsu...@gmail.com> wrote:

- Physical sky in Arnold 5:    Thin Walled's working with Sun but it is NOT working with Sky    -> Look Terrible !

- Physical sky in Arnold 4 :   Alshader Backlighting work very well with both Sun and Sky       -> Look Realistic !

-> I unbearable when looking result of Thin walled in Arnold 5.  So I hope Anders Langlands can re-write the Alshader for C4DtoA with Arnold core 5 soon. It will be really amazing!

Vào 20:03:32 UTC+7 Thứ Sáu, ngày 14 tháng 4 năm 2017, Stephan Wagemann đã viết:

Hi, there is no sign on the bitbucket source for any update related to arnold 5.

You are working behind the curtain on this?

--

You received this message because you are subscribed to the Google Groups "alshaders" group.

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+unsubscribe@googlegroups.com.

[Tino]

i know that - and thats great

my problem is if alColorRemap disappears how do i figure out the
existing connections and replace it by aiColorCorrect

Am 22.04.17 um 11:54 schrieb Frederic Servant:

>
>
> @Tino: The color_correct shader does everything alColorRemap does.
>
>

> Cheers,
> --
> Fred
>

[Anders Langlands]

I'll have a build up this weekend you can use to load existing networks and upgrade from.

--
You received this message because you are subscribed to the Google Groups "alshaders" group.

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Tino]

Thanks much appreciated!

Am 22.04.17 um 12:18 schrieb Anders Langlands:

[vasilis2...@gmail.com]

Add "direct" sss from alsurface shaders.

 

Random Color per shell polygon (not per face or object id).

Bevel edges in rendering like Vray 

Denoiser like Renderman or Vray (To be minimized the noise from interior scenes)

The textures noises released from Anders(We really thank you for the effort you make for us)

These are some of what we all would like.

However Arnold 5  - Is excellent .

with respect Bill

[Trẫm Sup]

@Frederic Servant : Thanks for explain. But Thin walled doesn't have Radius for up-scale intensity of translucency. It only has "Weight" and it's too weak. So I have to up skydome intensity to 20 for seeing more clearly Translucency. But that thing Impact on diffuse ( when turn up skydome intensity ). Then I have to turn down Sun color more black... And finally it's anoyying. 

Vào 20:03:32 UTC+7 Thứ Sáu, ngày 14 tháng 4 năm 2017, Stephan Wagemann đã viết:

Hi, there is no sign on the bitbucket source for any update related to arnold 5.

[Ilyas DIZDAR]

Hi everyone !

Little question about cryptomatte. 

Without alShader how we will be able to continue to use it in production with arnold 5 ?

Cryptomatte was one of the key point for us to switch from vray to arnold.

Its just a great feature that really improved our productivity.

I hope we wont have to change our pipeline about id/matte with arnold 5 and be back to the old way ...

[Ark]

Hi Frederic,

Congratulations with the new release!

After using alSurface I'm missing these things in new aiStandardSurface:

1) per lobe trace_sets - I often make a reflection pass where only character reflects in the objects of the environment, but those objects are still being reflected in that character (via raySwitch) - and I use trace_sets for specular lobe to achieve that.

2) per lobe extra_samples - just recently I had a heavy sss-ed character (with some specularity) with a brushed-metal knife - used less AA, more SSS samples and just a few Gloss, that worked fast enough for the character, but knife was very noisy, so, extra_samples for specular lobe helped to clean this up without slowing down the rest of the image.

3) splitting specular and coat into separate AOVs

Any chance there are workarounds for that with the new aiStandardSurface?


Best regards,
Ark

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Jean Lim]

Hi，Frederic

After did some test on Arnold5.0， compare with the Arnold4,alShaders and other offline renderer，I think there's something can be improve in the future release.

1）directional SSS mode.(I think the directional SSS mode is batter than empirical SSS mode in some cast，I understand solidangle team concern about the rendertime,but I also think puting this mode in the shader will give user more choise.)

2）3 layer sss.(I think this can be a SSS mode also,a lot of artist did the skin material base on the 3 layer aiSkin shader，now thay are confusing about the new shader just have one layer SSS color.so maybe have a 3 layer SSS mode will help those artist shift to the new aistandard surface shader more easier)

3）specular and coat layer separate into buil-in AOVs(I try to separate them by using LPE，but I only get the specular separated.)

4）Denoiser like Renderman or Vray.(I think it is make a  sense that some people don't wanna spend too much time on a super clean render image,a denoiser can help them get a clean image when useing a lower sample setting)

English not my native language,if I type something wrong and confuse you,my apology.Thanks for showing up here,is there have any website that people can post freeback or advice to help improve the Arnold render？


在 2017年4月22日星期六 UTC+8下午5:54:50，Frederic Servant写道：

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Jean Lim]

That's cool！ vray exciting to hear this,thanks for the reply,cheers！

[Jonah Friedman]

Hi Ilyas,

Cryptomatte is going together available and thanks to some new features Solid Angle is adding in 5.0.1, I'm confident it will be better than it was in Arnold 4 both in terms of functionality and workflow. 

We have a passthrough shader ready but since it'll be obsolete so soon I think we should wait for 5.0.1 instead. 

--

You received this message because you are subscribed to the Google Groups "alshaders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

JONAH FRIEDMAN

	PSYOP NYC | LA OFFICE: +1 (212) 533-9055 MOBILE: +1 (347) 873-3992 45 HOWARD ST, 5TH FLOOR, NEW YORK NY 10013

[Jonah Friedman]

That should have said, Cryptomatte is going to be available.. (autocorrect)

[Trẫm Sup]

Yes. I show you my test here to prove that:  "same translucenty level, much different diffuse/skydome level. Same diffuse/skydome level -> much different translucenty level " . 

And I just mention Translucency that affected by Sky, not by Sun

 * Test 1:

- With the same skydome level ( intensity = 2.3 ), Translucency by Sky ( not by Sun ) of Arnold 5 is very very weak. ( Although "Subsurface Thin wadded weight "turned up maximum )

- The only solution is increase really much Skydome intensity to have the same translucenty level with Alshader arnold 4 ( from 2.3 to 30.3 ). But increase Skydome intensity change all too much to the Diffuse :

* Test 2 :

 Here is another test, same translucenty level, different diffuse/skydome level.

Thank you for listening !

Vào 16:54:50 UTC+7 Thứ Bảy, ngày 22 tháng 4 năm 2017, Frederic Servant đã viết:

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Ilyas DIZDAR]

thanks Jonah !

That that great news for us.

Great thanks to Psyop and solid Angle team !

[Kopa Kopatych]

Hi!Thank for you answer.
I fit pictures
first the empirical mode in alSurface, render time 2min 23 sec
https://cloud.mail.ru/public/MnHw/EjA9Ki9eC

second the direction mode, render time 4min 14 sec
https://cloud.mail.ru/public/AySp/8XTWS2i2W

The result is obvious and only 2 minutes difference

@Kopa: I think having the same IOR for spec and transmittance makes sense.

How to make crystal clear glass with a stained surface?

Without the haze of the glass itself, only the spots on the surface.

The idea of coat layer is not entirely clear, this is for varnish paint? What is its meaning if, when you connect texture in coat color, it kills all the other layers, why it was impossible to realize just two speculars  as it was done in alSurface, it's a more artistic

Sorry for my English

суббота, 22 апреля 2017 г., 12:54:50 UTC+3 пользователь Frederic Servant написал:

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Kopa Kopatych]

if you release direction sss mode, bottom  separating as required specular and transmission roughness and bottom use coat layer as secondary specular layer, at will be great
Please treat with understanding

суббота, 22 апреля 2017 г., 12:54:50 UTC+3 пользователь Frederic Servant написал:

Hi everyone, hi Anders,

[Frederic Servant]

Thanks for the example Kopa.

The rendertime is twice as slow, which is not negligible, but I see a difference in look obviously.

Would it be possible to try with the new standard surface in Arnold 5 to compare both the result and the rendertime?

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+unsubscribe@googlegroups.com.

[Kopa Kopatych]

Of course, however, the point is that the empiric model did not change as far as I know

воскресенье, 23 апреля 2017 г., 23:56:50 UTC+3 пользователь Frederic Servant написал:

[Trẫm Sup]

What's about my Translucencyyy?

Vào 03:56:50 UTC+7 Thứ Hai, ngày 24 tháng 4 năm 2017, Frederic Servant đã viết:

[Frederic Servant]

Hi Tram,

We heard you :) There's a lot of feedback as expected, and this changes a lot of things for all users obviously.

Arnold 5 is about new shaders, but also new tech that we enable us to tackle more interesting issues in future versions.

For now we are gathering information, good and bad, to make some adjustments and move forward. I think you'll see changes coming pretty soon.

Keep the (constructive) feedback coming!

Cheers,

--

Fred

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+unsubscribe@googlegroups.com.

[Kopa Kopatych]

In addition, the render time is not the one that should be emphasized first of all, but in some way you are right, you need to understand the competitive time.
But back to the beginning, my position is not the time to deliver to all, but the result at a favorable time.
And sorry for my English
воскресенье, 23 апреля 2017 г., 23:56:50 UTC+3 пользователь Frederic Servant написал:

[karas]

@Frederic Servant:
1.We prefer how alsurface control the metalness under specular,not under diffuse like standard surface;

2.Extra sample control is really helpful for optimize rendering;

[Kopa Kopatych]

A5 surface 1 min 47 sec https://cloud.mail.ru/public/2SJ9/ikuU2iBa8
The result is the same as on A4

воскресенье, 23 апреля 2017 г., 23:56:50 UTC+3 пользователь Frederic Servant написал:

[Frederic Servant]

Hi,

That's a great example scene to show off the directional mode.

I'm not sure I understand for the hair though. On your image the color and radius seem really different.

On Mon, Apr 24, 2017 at 3:31 AM, <readers...@gmail.com> wrote:

Hi, 

I just wanted to add some results to the sss test comparing alShaders (directional) and Arnold 5. While I do like the new surface and hair shaders, the results are quite different and I guess the loss of not having the choice of using alSurface or alHair at all to achieve a different look is a shock to the system. There just seems to be a lot less control overall to tweak (e.g. the hair - I couldn't get the individual strands to look more separated without regooming and color seems more difficult to control).

[Frederic Servant]

Thanks Kopa!

So A5 is indeed 37% faster (1m47s vs 2m27s), but of course with the empirical model only.

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+unsubscribe@googlegroups.com.

[Kopa Kopatych]

You use only sss layer, or with diffuse?

понедельник, 24 апреля 2017 г., 12:20:25 UTC+3 пользователь Frederic Servant написал:

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[SW]

Thanks Frederic :) 

Regarding the hair, I wanted the individual strands of hair to be more prominent which I could achieve with a ramp on the dye color and randomise features in alHair but found it difficult to replicate a similar look with the new hair shader. The groom is the same with a similar ramp for the color but I couldn’t get ramdomisation or maybe I'm doing it all wrong  - need some tutorials :D

The new A5 Hair shader is great for realistic hair though. 

[SW]

Hi,

I just added the base skin color to the sss color with a deep, mid and shallow to the radius setting. So nothing on the diffuse. The subsurface section in the MtoA user guide was helpful. It was just a quick test and could probably tweak it to look closer to the directional sss result with alShaders.

I just wish it wasn’t a choice between having A5 or alShaders, so many good things in both.

[Miguel González Viñé]

You could increment your diffuse in your aiStandardHair to reduce the hair's SSS. Maybe does it help to what you're looking for?

[SW]

Hi, Yep I did try playing with the diffuse but the result looked like too fake. I'll try again when i learn more :)

[Ark]

I have a good impression from the new sss with much better render times - like x10 times faster. 

The most prominent effect of the directional mode seems to be the contrast in every little detail that shifts the top surface color to the opposite (white surface with directional sss becomes cyan). I've never really understood if that's a feature or a bug (why under the necklace or a sleeve the skin gets cyanish), but it has it's charm :) I was able to get a very similar look with the new sss by setting radii to 1 0 0 (only R goes in) and boosting the opposite colors (i.e. G and B) of the sss texture (making it yellow-greenish) and overall gain x1.5 times. And I have a much better render times in a fraction of time.

Three layers would be nice anyway, since I can control the intensity of the scattering on different depths, while new sss falloff feels kind of linear.

Here are the images:

[Jean Lim]

Hi Ark.really amazing result！can you share a little bit how your new SSS texture looks like？compare to the original one.

[mr.brun...@gmail.com]

Hi Frederic, any chance the "Convert Deprecated Shaders" utility could be updated to support alShaders conversion to Arnold 5 shaders and utility nodes?

Thanks!

[vasilis2...@gmail.com]

I would like to ask something with every reservation .

The empirical method Is the same as Burley/Christensen Normalizer; (Based on <https://www.youtube.com/watch?v=Z4xwrvSdbiI&feature=youtu.be&t=1103)

If it is :

1. Uses Radius (fake) or it's true sss Through Object (Path trace)

2.Why when we have fully red in  sssRadius it will die faster than the other two (green and blue ) https://youtu.be/Z4xwrvSdbiI?t=1550

Translation was done through Google.

I apologize if anything has been translated accurately

With respect

Vasilis

[Brecht Van Lommel]

Hi Trẫm,

So for leaves you would typically enable Thin Walled, and set Subsurface to a value like 0.5, to make half of the light scatter to the front and half to the back. Ray Depth for Diffuse would be set to a value like 4 to get scattering between leaves.

I can only guess what is happening in this specific scene, in my tests it seems to be scattering the right amount of light. However the parameters in alSurface and standard surface are different. In Standard Surface we mix the front and back scattering, while in alSurface the front and back scattering are added together, and then clamped in case energy conservation would be broken. This means that for Standard Surface you'd generally need to use a brighter Base Color and Subsurface Color than alSurface.

[Brecht Van Lommel]

Hi Vasilis,

1. Empirical SSS is similar to the Burley/Christensen method, it's not brute force scattering.

2. I'm not sure what is meant by "die faster" in that video. Higher radius for red means red is scattered more deeply than other colors, i.e. it is blurred more.

[Brecht Van Lommel]

Hi Kopa,

About the coat, as far as I know this behaves similar to the second specular in alSurface in how it is layered internally, even if the parameter are different. It can be used for the same purposes, varnish, sweat/oil on skin, clearcoat on car paint, etc.

For clear glass with a stained surface, the coat could be used, or the transmission weight could be textured perhaps. From a physics point of view, specular and transmission happen on the same layer with the same IOR, some rays scatter below and some above the surface. So normally it would be two layers or some variation in the surface properties of a single layer.

Regarding texturing, the Coat Weight controls the amount/thickness of coating, and the Coat Color the color of the coat, which if it is a dark color will indeed obscure the layers below. Perhaps in your cases you can connect the texture to the Coat Weight, or to both the Coat Weight and Coat Color?

[Kopa Kopatych]

Hi
Yes, I can do this, but it's already implemented not as in alsurface
But could make it all much easier by simply adding a mod attribute, with a choice of the type of blending

Guys,help please
How in A5 inverse transmission scatter atribut, that it works as an absorption
I can not get the result that I could achieve in alSurface
http://joxi.ru/krD1gBxsE3Vgwm

понедельник, 24 апреля 2017 г., 19:37:07 UTC+3 пользователь Brecht Van Lommel написал:

[Anders Langlands]

Guilty secret time - the single scattering in transmission is the one area where the shader could be not energy conserving. I wouldn't expect to be able to match it in standard.

[Trẫm Sup]

I copy all content of bịn to the C:\Program Files\MAXON\CINEMA 4D R18\plugins\C4DtoA\shaders 

Then I coppy all thes folder of your file to the C:\Program Files\MAXON\CINEMA 4D R18\plugins\C4DtoA

But when turn C4DtoA material network I still dont see Alsurface . Can you help ?!

Vào 07:08:52 UTC+7 Thứ Ba, ngày 25 tháng 4 năm 2017, Anders Langlands đã viết:

[Anders Langlands]

Do you see the other shaders?

[Ark]

Hi Jean,

It's hard to compare, since it was three orange->reddish maps for directional, but here it is:

[Jean Lim]

Big Thanks！  so the “radius” just a color？

在 2017年4月25日星期二 UTC+8下午10:02:00，Ark写道：

[Ark]

Yep, haven't mapped it... yet :)

[Ark]

Here is a big thing that my mind can't figure out - I'm making a skin with aiStandardSurface... I've created diffuse and sss and now I want to make the skin oily. When I enable specularity and increase IOR, the whole skin goes dark (I have just an area light and no environment to reflect). As a result, the skin can be either dry or metallic, the oilier it is - more metallic it gets. My strong belief is that reflections are additive and the fact that there's nothing bright to reflect (or the ability to reflect at all) shouldn't make the surface diffuse/sss darker.

So, am I doing something wrong or aiStandardSurface is not really suited for skin (besides adding specularity on top with another shader)? Mentalray had a separate misss_fast_skin shader with two specularities added on top, that didn't darken diffuse or sss... alSurface has two specularities that don't darken diffuse (though, unfortunately, they darken sss)... aiStandardSurface darkens both diffuse and specularity... Please, advise.

[Kopa Kopatych]

Hi!

Maybe you use metalness or darker coat color, dont use them, metalnes only for metal surface and his need white diffuse, coat layer should always white this is write in arnold help.

среда, 26 апреля 2017 г., 18:40:21 UTC+3 пользователь Ark написал:

[Jean Lim]

Hi,thank you for the reply

I think maybe you increase IOR too much. increase IOR means the material become more reflective,witch means the reflection will cover other effect (diffuse，SSS，refraction). the old skin material also behavior like this too.if you using the coat layer,make sure coat color is white,it also will darken SSS a little bit. 

在 2017年4月26日星期三 UTC+8下午11:40:21，Ark写道：

[Anders Langlands]

On Thu, 27 Apr 2017 at 03:40 Ark <sag...@gmail.com> wrote:

Here is a big thing that my mind can't figure out - I'm making a skin with aiStandardSurface... I've created diffuse and sss and now I want to make the skin oily. When I enable specularity and increase IOR, the whole skin goes dark (I have just an area light and no environment to reflect). As a result, the skin can be either dry or metallic, the oilier it is - more metallic it gets. My strong belief is that reflections are additive and the fact that there's nothing bright to reflect (or the ability to reflect at all) shouldn't make the surface diffuse/sss darker.

Definitely sounds like you've increased the IOR too much. Glossy reflections are not purely additive as the more reflective the glossy layer the less energy reaches the interior of the medium, so the darker the diffuse will be. alSurface works very similarly to standard_surface in this regard.

 

[Ark]

Indeed, I had big ior values and small weights - problem pretty much disappears after setting this the opposite way. Thanks, guys!

The fact is, I just re-plugged values from alSurface, so, the difference is quite drastic (especially considering that I used beckmann).

[Kopa Kopatych]

And why do you use ior = 12 if it does not give a value above 10 and it equals to the metalness = 1?

The gloss of human skin is 1.33, the glass 1.5

четверг, 27 апреля 2017 г., 16:18:48 UTC+3 пользователь Ark написал:

[readers...@gmail.com]

Hi, 

Have you tried doing without the diffuse color? I believe in the user guide for A5 it mentions to plugin in the Base skin color (Diffuse or albedo) into the subsurface color rather than using diffuse color. Most importantly it's best to work to real world scale, this will solve a lot of issues. The setup for the new shader is actually a lot simpler than most and quite nice for skin though I would still prefer having directional sss as an option and the separate lobes in alsurface seem more intuitive.

Have you tried something like this for skin?

Base skin -> Subsurface color

Deep map -> Radius.R  - 1.0

Mid map -> Radius.G - 0.25 - 0.4

Shallow -> Radius.B - 0.15 - 0.25

(sss scatter maps should be textures and not just colors for best results)

Spec/Roughness map -> Specular roughness (so you have better control where the skin should be more shiny)

I usually stick to an IOR of 1.33-1.4 depending on the look I'm trying to achieve and lighting I'm using. 

Hope this helps :)

[Ark]

    I think I just wanted to get more spec on areas facing camera.

[tyle...@googlemail.com]

Am Freitag, 14. April 2017 15:03:32 UTC+2 schrieb Stephan Wagemann:

Hi, there is no sign on the bitbucket source for any update related to arnold 5.
You are working behind the curtain on this?

I really hope Anders will update the alSurface and alHair soon. I appreciate the new shaders and will definitely give it a go for future projects, but all shader libraries I created so far rely on alSurface exclusively and I cant match them up with the new standard surface. Mostly because I used a combination of directional sss, single scattering transmission with carefully dialed in absorption and attenuation colors. Think about Milk for instance, currently not doable properly with an energy conserving approach, as it needs some fake caustic brightening.
Regarding caustics, these work in alSurface with ibl and in standard they dont..

[Kevin Tsui]

Hi Jonah,

So glad to hear that arnold will support Cryptomatte officially,good job! Can we get cryptomatte api from arnold 5.0.1 sdk package,I used alShaders cryptomatte to create our own shaders and it worked fine,so I want to know if we can use this awesome feature to write our own arnold shaders?

Kevin Tsui.

On Saturday, April 22, 2017 at 11:25:32 PM UTC+8, Jonah Friedman wrote:

That should have said, Cryptomatte is going to be available.. (autocorrect)

On Sat, Apr 22, 2017 at 10:48 AM Jonah Friedman <jfri...@psyop.tv> wrote:

Hi Ilyas,

Cryptomatte is going together available and thanks to some new features Solid Angle is adding in 5.0.1, I'm confident it will be better than it was in Arnold 4 both in terms of functionality and workflow. 

We have a passthrough shader ready but since it'll be obsolete so soon I think we should wait for 5.0.1 instead. 

On Sat, Apr 22, 2017 at 8:08 AM Ilyas DIZDAR <dizdar...@gmail.com> wrote:

Hi everyone !

Little question about cryptomatte. 

Without alShader how we will be able to continue to use it in production with arnold 5 ?

Cryptomatte was one of the key point for us to switch from vray to arnold.

Its just a great feature that really improved our productivity.

I hope we wont have to change our pipeline about id/matte with arnold 5 and be back to the old way ...

--
You received this message because you are subscribed to the Google Groups "alshaders" group.
To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

JONAH FRIEDMAN

	PSYOP NYC | LA OFFICE: +1 (212) 533-9055 MOBILE: +1 (347) 873-3992 45 HOWARD ST, 5TH FLOOR, NEW YORK NY 10013

--

JONAH FRIEDMAN

	PSYOP NYC | LA OFFICE: +1 (212) 533-9055 MOBILE: +1 (347) 873-3992 45 HOWARD ST, 5TH FLOOR, NEW YORK NY 10013

[Jonah Friedman]

You are slightly mistaken- Arnold 5 will not support Cryptomatte directly. Instead they're adding a feature that allows users to get Cryptomattes from their renders without doing anything to the beauty shaders they use. This feature is being tested with Cryptomatte by them, since that's the first use case for it. 

Your shaders should work with Cryptomatte without you needing to actually do anything. If you needed to for some reason, you could also use this to create your own version of Cryptomatte that would then work on all built-in and third party shaders as well. 

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+unsubscribe@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Kai Xu]

That sound awesome👍

2017-06-02 22:46 GMT+08:00 Jonah Friedman <jfri...@psyop.tv>:

You are slightly mistaken- Arnold 5 will not support Cryptomatte directly. Instead they're adding a feature that allows users to get Cryptomattes from their renders without doing anything to the beauty shaders they use. This feature is being tested with Cryptomatte by them, since that's the first use case for it. 

Your shaders should work with Cryptomatte without you needing to actually do anything. If you needed to for some reason, you could also use this to create your own version of Cryptomatte that would then work on all built-in and third party shaders as well. 

[Kopa Kopatych]

Hi Guys!
in mtoa 2.0 don't have this nodes https://monosnap.com/file/mSuZNyvoi6XyxrjO5Dp4CwpdIUpHdp
Any ideas why is in the documentation for A5 and in MTOA not?
Sorry for my english

пятница, 14 апреля 2017 г., 16:03:32 UTC+3 пользователь Stephan Wagemann написал:

[Frederic Servant]

Hi Kopa,

I believe this shader was removed. You can use the AiUserDataColor instead. Conversion RGB to/from vector conversion is handled automatically by Arnold.

I could not find it in the A5 docs, could you give me the URL where you found them?

Cheers,

--

Fred

--

[Kopa Kopatych]

No, I meant aiUserDataColor, I do not have all the nodes that are on the screen,

But the funniest thing  if i load in from Shortcut  the  then they are, but there are no variable environments path , and if load from batch file then are not load

среда, 7 июня 2017 г., 16:12:19 UTC+3 пользователь Frederic Servant написал:

To unsubscribe from this group and stop receiving emails from it, send an email to alshaders+...@googlegroups.com.

[Kopa Kopatych]

but in mtoa 1421 and older works perfct

[Kopa Kopatych]

Decided
If the variable ARNOLD_PLUGIN_PATH has a path to mtoadeploy\2017\shader, A5 don't load this nodes https://monosnap.com/file/mSuZNyvoi6XyxrjO5Dp4CwpdIUpHdp

[karas]

@Frederic Servant

There a "debug" mode on AlLayer shader allow user to switch either layer1 or layer2, thats really convenience for checking output, really hope aiMixShader could add this feature.Thanks