Carrara Shader Tree - Order Help + Alpha Background Scene

SileneUKSileneUK Posts: 1,975
edited December 1969 in Carrara Discussion

Sorry if I am being really thick about this stuff... :red:

I tried to search the forum for these...maybe it's the way I'm searching. But can someone show me the order of using these texture maps?

I know where to put the main texture in the shader tree, eg of a leather/buckskin fabric or of a stone colouring. But often it does not have the depth of texture I want if the actual model is smooth. I have done some displacement "painting" is the model room... with decent results. But it's not always an option on some things.

To get around that, I thought if I use a Bump utility creator, I would get several files that will help with the illusion. Are they all appropriate to Carrara? I saw Dot_Bat had CrazyBump in her screenie when she was asking for help with something else. So I decided to bite.

Where do you put files that have these extensions:

NRM (that is the purple-pink one)
DISP (Do you have to enable Displacement and put it there? Or is it used as a Bump? Or where?
OCC Light grainy fine texture look
SPEC Dark grainy fine texture look


Also... whilst I am at it. I know about Alpha when doing, eg eyelashes or fringe on clothing, uneven hems, etc.

But I have seen it mentioned that you can do a render pass with an alpha background. So that only the items you have modelled show up and the background comes out transparent in the rendered image.

How do you do that?

If there is a tutorial or a post on any of these... I'd be happy to see them!

Cheers, xx :) Silene

Comments

  • evilproducerevilproducer Posts: 9,050
    edited December 1969

    SileneUK said:
    Sorry if I am being really thick about this stuff... :red:

    I tried to search the forum for these...maybe it's the way I'm searching. But can someone show me the order of using these texture maps?

    I know where to put the main texture in the shader tree, eg of a leather/buckskin fabric or of a stone colouring. But often it does not have the depth of texture I want if the actual model is smooth. I have done some displacement "painting" is the model room... with decent results. But it's not always an option on some things.

    To get around that, I thought if I use a Bump utility creator, I would get several files that will help with the illusion. Are they all appropriate to Carrara? I saw Dot_Bat had CrazyBump in her screenie when she was asking for help with something else. So I decided to bite.

    Where do you put files that have these extensions:

    NRM (that is the purple-pink one)
    DISP (Do you have to enable Displacement and put it there? Or is it used as a Bump? Or where?
    OCC Light grainy fine texture look
    SPEC Dark grainy fine texture look


    Also... whilst I am at it. I know about Alpha when doing, eg eyelashes or fringe on clothing, uneven hems, etc.

    But I have seen it mentioned that you can do a render pass with an alpha background. So that only the items you have modelled show up and the background comes out transparent in the rendered image.

    How do you do that?

    If there is a tutorial or a post on any of these... I'd be happy to see them!

    Cheers, xx :) Silene


    NRM is probably a Normal map. The crib notes explanation is that you could create one from a high density and detailed mesh, and then use it on a lower density and less detailed mesh to give it the appearance of the higher detailed mesh. There's more to it than that, but I use C7.2 and can't use them. In C8 and later, you would use it in the Bump Channel. From the problem you describe with details, this may go a long way to solving it.

    I would think the DISP is a displacement map. You could try it in the displacement part of the shader, but sometimes in Carrara it's like using a sledge hammer when all you need is ball peen hammer. I've had good luck using them in the Bump channel as Carrara's bump system seems fairly robust. It may be an option if the Normal map doesn't give the desired effect.

    The SPEC map would be used in the Highlight Channel. In Carrara, the Highlight channel is used to define the brightness of a specular light effect. The brighter the color or value, the brighter the effect. If you were to use a color or a color image map, then the highlight would take on that color.

    In reality, the specular light effect and a reflection are all one thing, but in many 3D programs, they're separate to provide more control over the effects.

    Shininess is used in conjunction with Highlight. Shininess defines how soft or hard edged the specular effect becomes. If you were to use a 1-100 numerical slider, the low end of the slider would create a softer, more diffuse edge to the highlight, and higher on the slider would make a harder edge. A dark color would do the same as a low numeric setting and a brighter color will do the same as a higher value on the slider.

    So, the weird one to me is the OCC map. I assume it means Occlusion and I'm not sure where it would go. I have a possible use for it though. You could try putting it in the Shininess channel to drive the spread of the highlight. If that doesn't work for you, try setting a numeric slider in Shininess channel and set it to around 3% to see if that gives the desired effect.

    Alpha options can be found in the Render Room under the File Format section of the Output Tab. I know you use Photoshop so you may wish to try rendering without the pre-multiply option. You're mileage may vary. There are good reasons to use the multiply option, but I'm not sure PS (or at least older versions of PS which I use) handle pre-multiplied alphas correctly. Try both and see which works better. Best yet, check PS's documentation to see what Adobe suggests.

    Picture_1.png
    255 x 478 - 44K
  • SileneUKSileneUK Posts: 1,975
    edited December 1969


    NRM is probably a Normal map. The crib notes explanation is that you could create one from a high density and detailed mesh, and then use it on a lower density and less detailed mesh to give it the appearance of the higher detailed mesh. There's more to it than that, but I use C7.2 and can't use them. In C8 and later, you would use it in the Bump Channel. From the problem you describe with details, this may go a long way to solving it.

    I would think the DISP is a displacement map. You could try it in the displacement part of the shader, but sometimes in Carrara it's like using a sledge hammer when all you need is ball peen hammer. I've had good luck using them in the Bump channel as Carrara's bump system seems fairly robust. It may be an option if the Normal map doesn't give the desired effect.

    The SPEC map would be used in the Highlight Channel. In Carrara, the Highlight channel is used to define the brightness of a specular light effect. The brighter the color or value, the brighter the effect. If you were to use a color or a color image map, then the highlight would take on that color.

    In reality, the specular light effect and a reflection are all one thing, but in many 3D programs, they're separate to provide more control over the effects.

    Shininess is used in conjunction with Highlight. Shininess defines how soft or hard edged the specular effect becomes. If you were to use a 1-100 numerical slider, the low end of the slider would create a softer, more diffuse edge to the highlight, and higher on the slider would make a harder edge. A dark color would do the same as a low numeric setting and a brighter color will do the same as a higher value on the slider.

    So, the weird one to me is the OCC map. I assume it means Occlusion and I'm not sure where it would go. I have a possible use for it though. You could try putting it in the Shininess channel to drive the spread of the highlight. If that doesn't work for you, try setting a numeric slider in Shininess channel and set it to around 3% to see if that gives the desired effect.

    Alpha options can be found in the Render Room under the File Format section of the Output Tab. I know you use Photoshop so you may wish to try rendering without the pre-multiply option. You're mileage may vary. There are good reasons to use the multiply option, but I'm not sure PS (or at least older versions of PS which I use) handle pre-multiplied alphas correctly. Try both and see which works better. Best yet, check PS's documentation to see what Adobe suggests.

    Thanks EP, Great explanations. I know one size does not fit all, but these will really help me a lot so I appreciate the time you took to explain them. I will try and sort out some renders using them over the next few days and post any results.

    Will mess about with the Alpha channel as well... I didn't realise it was controlled in the Render area, though it was something I had to do with the background.

    Cheers, xx :) Silene

  • evilproducerevilproducer Posts: 9,050
    edited December 1969

    If you're compositing, another thing to consider is using a shadow catcher. They catch shadows, are invisible (except for the shadow) and if an object is behind and partially obscured by one, the shadow catcher acts as a mask.

    To create one, go to the Texture Room, and choose Edit--> Create Master Shader. A new shader window will open. Look for the Lighting Models pulldown menu and choose Shadow Catcher. All done, unless you want to put a color in the color channel which seems to create a very slight color hue in the shadow. It's pretty subtle really. None of the other shader settings seem to do anything.

  • evilproducerevilproducer Posts: 9,050
    edited December 1969

    I forgot to add, that to use the shadow catcher, apply it to any object just as you would any other shader. If the object has multiple domains, a quick way to apply it would be (assuming it's saved to the Shader Browser) is to drag it from the browser and drop it on the multi-colored ball at the top of the object's shading domain list, and it should change all the domains to the shadow catcher.

  • SileneUKSileneUK Posts: 1,975
    edited December 1969

    I forgot to add, that to use the shadow catcher, apply it to any object just as you would any other shader. If the object has multiple domains, a quick way to apply it would be (assuming it's saved to the Shader Browser) is to drag it from the browser and drop it on the multi-colored ball at the top of the object's shading domain list, and it should change all the domains to the shadow catcher.

    Something totally new to me!! Will have to swot up. xx :) Silene

  • argus1000argus1000 Posts: 701
    edited December 1969

    You want to put the SPEC map in the Highlight channel.

    You want to put the NRM map in the Bump channel. But first, you have to set it to the Normal Map channel.

    As for DISP maps, you want to click on the Displacement tab, then find the appropriate Texture Map. On the Displacement mapping tab, you want to Enable displacement.. I usually set the amplitude to .1 or .02. I never use Subdivision button because it takes to long to render in animation. But if you're doing stills, you may want to play with it.

    DISP_2.jpg
    655 x 473 - 44K
    DISP_1.jpg
    694 x 517 - 49K
    NRM.jpg
    736 x 524 - 72K
    SPEC.jpg
    640 x 464 - 65K
  • SileneUKSileneUK Posts: 1,975
    edited December 1969

    Thanks, Argus! These maps are all for inanimate objects, probably mostly in natural settings, like rocks, ground textures, and such. So I will try the Disp Mapping. I have terrible luck with that with characters so have stayed away from it... it buggers up facial morphs in a nasty way sometimes.

    OK, I feel ready to start working with these tools now... back to vertex modelling first, though!

    xx :) Silene

  • WendyLuvsCatzWendyLuvsCatz Posts: 38,192
    edited December 1969

    carrara is oddly fussy about which png files show an alpha channel too
    not all work so rendering a black and white mask to put in the alpha texture is always a good idea I find
    carrara rendered ones work of course but not ones done using Irfanview for example and I have found a few others that fail yet work in say iClone or Poser
    GIMP ones work ok too and I assume photoshop and of course Daz studio rendered ones

  • evilproducerevilproducer Posts: 9,050
    edited August 2014

    I have never had a problem with Carrara rendering the alphas.

    Wendy, I wonder if the other renderers default to rendering a pre-multiplied alpha channel, which some image editors interpret a bit off in that they display the background? That's what happens in my version of PS and also in my video editors. It makes it pretty useless for me.

    I never tested to see if Carrara displays a pre-multiplied alpha the same way as PS and my video editors. It may be worth an experiment if I have the time.

    Post edited by evilproducer on
  • argus1000argus1000 Posts: 701
    edited August 2014

    I use displacement maps mainly with characters. The first pic is the character without displacement. The second pic is with a wrinkles displacement map added, as well as some freckles and skin detail added in the color channel. I think it gives substance to the character.

    Fig_2.jpg
    810 x 604 - 52K
    Fig_1.jpg
    732 x 520 - 64K
    Disp.jpg
    640 x 360 - 12K
    No_disp.jpg
    640 x 360 - 12K
    Post edited by argus1000 on
  • WendyLuvsCatzWendyLuvsCatz Posts: 38,192
    edited December 1969

    well here is an image with an alpha channel which does not work in carrara

    my_flower.png
    640 x 400 - 123K
  • evilproducerevilproducer Posts: 9,050
    edited December 1969

    well here is an image with an alpha channel which does not work in carrara

    Drawn/painted in what program? Straight Alpha or Pre-Multiplied Alpha? I'm thinking some programs save the alpha as a pre-multiplied alpha instead of a straight alpha and Carrara may not recognize those when loaded into a shader's color channel for instance.

  • WendyLuvsCatzWendyLuvsCatz Posts: 38,192
    edited December 1969

    I think it was either Photobie or PaintNet cannot recall, I use GIMP now

  • SileneUKSileneUK Posts: 1,975
    edited August 2014

    I got my transparent backgrounds to render fine from my background-less scene.

    But I had to use TIFF for rendering at 300. PSD didn't want to remove the black, it just brought along another black layer to include! With TIFF or with PNG--- when I did my save it was doing the bcg save as a separate item which I didnt' save. So I got my tsp bcg in the one image and am very happy. Now I can layer up my render passes and stop being so bogged down! And I can use my 2D backgrounds more effectively.

    Will mess about with the shader tree items later in the weekend.

    Thanks, again! xx :) Silene

    Post edited by SileneUK on
Sign In or Register to comment.