Texturing Question.

[leigh]

Now I tried to save the text off as a giff so there was no background to it, but when I plug that into the colour channel the transparent part of that file makes my cube transparent in that area. Do I need to plug this into another channel for this way to work?

If I recall correctly, this is because when you bring a texture into Maya that has an alpha channel in it, and apply it to a shader, it automatically applies the texture to the transparency of the shader as well. You just have to break the connection in the Hypershade.

Incidentally, using a TIFF with an alpha channel is probably better than a GIF.

[SkynetUK]

Hi, thanks for the response and advise.

I tried the Tiff and saved that with transparency set on, but it still showed with a background. So I found out that I had to make the alpha in PS and save as a .Tga but even when I do that it just then makes what ever is black on the alpha transparent in Maya.

I guess what Im really wanting to know is there a way to project an image onto an object with out going through the colour channel? Seems to me that would be an obvious feature, I even tried projecting the image with a light, but didn't really work as well as I would have liked. Also it wouldn't seem like such a good thing to have lights all over the place etc etc.

So yah really would like to know how to do that, I know I'm very new to this field but having a whole texture using up the colour channel for just adding txt or a small logo/symbol to an object seems rather long. If I change my mind and want to tweak with the colour on that object I have to jump back and forth between Maya and PS. That can't be right, can it? :shrug:


Cheers...

[phix314]

I'm a little confused. If I understand, you want to have some sort of bitmap image mapped onto an object, without connecting it to the shader?

Someone correct me if I'm wrong, but that's the way this 3D thing works. The model is created, you texture it, then you render it.


having a whole texture using up the colour channel for just adding txt or a small logo/symbol to an object seems rather long


What else would you want to use up the color channel for?

I would make sure you know what you want to do beforehand so you dont end up bouncing back and forth between Maya and PS. But, yea, if I understand your question correctly, that's the way its done.

[SkynetUK]

It seems Im not doing a very good job of explaining or maybe even understanding certain workings of 3D.

This is as straight as I can put it...

Lets say I have a simple cube object. On the front face I wanted to put some text or a logo. Now that txt has to be on a image file correct? If so, then that image file is going to have a background colour to it, right? Unless I make the background transparent even still I have to connect that image to the colour channel. Which would then leave me no option to change the shader colour, then having to make all changes in PS.

What if I wanted to just use the colour channel in Maya to colour the actual object, hence that's what the colour channel would be for??? And then just project the txt onto the front face of the cube with out having to change the whole colour of the image file in PS each time I want to make a slight adjustment to colour or texture.

Like what you can do with a light using a texture map, Batman kinda projection lets say. But as mentioned before this method does not quite provide the desired affect. Then that way would leave the colour channel open to edit and tweak around at ease rather than jumping back and forth. Uknow?

I know I could bake the shader or create a texture in PS and then add the text over the top etc etc..ZzzzZz but this all seems way to long and gratuitous to me for the simple task I am trying to do, not mention how much of a vibe & creative killer that all seems. I don't mean this in any negative way, just saying surely I must missing something here.


I hope that makes better sense and thank you for your time.

[phix314]

Ok, I think I see where you're going.

In this, there's the text, which is simply a targa file with an alpha channel, then the color, which is a single color ramp. The ramp is on the bottom layer of a layered texture node, with the text image on top, and the alpha of the text image feeding into the alpha of that layer in the layered texture node.

So...

Image File + Alpha (multiplied onto...)
Solid Ramp Color

http://i126.photobucket.com/albums/p104/roaphotosharing/Maya/deleteme.jpg

Each are seperate from each other, and can be edited as such.

Generally, though, you don't go back and forth so much that it becomes a problem or hurdle.

[SkynetUK]

Thanks Steve, I think this is what I was looking and trying to go for and also looks like a straight forward node connection. I'll give that a go and see how I get on.


Again thanks for taking the time to help out. :beer:

Cheers!

[SkynetUK]

I think I am missing something here.

The thing Im a little unclear on is plugging the alpha in the layered shader. The txt file node has Alpha out but the layered node does not have an alpha channel that I can see.

This is what I did....I put the ramp node into the layered node, then the file node into the layered node. Making sure to place the file node on top (To the left side) with in the layered node. as in picture below.

Cheers

[sundialsvc4]

No matter what the tool is... approach the problem in this way:


Use nodes. You are going to construct a simple node setup with two distinct inputs (the existing shader, and the imagery) being processed in some way (e.g. color grading, blurring), combined together, and maybe post-processed once more. Give the computer some way to know what is "the imagery you want" and what is "background." That is to say, make sure that your source image-file contains an alpha channel. Alpha should be 1.0 for everything that is image; 0.0 for everything that is not. You can use a specific RGB color-value for this purpose as well. Basically, any sort of "bright-line rule" that the computer can apply to say, "keep this, get rid of that" will do. You'd rather not have to use a garbage-matte also, but... any port in a storm. Now, in your compositing node-network, "tweak it" until you get the effect that you want.
Now for the head-twist: while you naturally may envision this effect as "I gotta do all this work to get the texture just right before I apply it to this object's surface," actually that is not the only way you can do this. You can also accomplish the effect in 2D, and this would allow for a lot more ease of tweaking.

Render the object with the blank texture. Then, separately, render with the (alpha-equipped) image texture. Produce two strips with only that object. They will, of course, register precisely. The computer has done all the fancy geometric work of calculating how the image's pixels should map onto the 3D surface. Now you can blend them to your heart's content, as pure flat-images, in 2D.

But there are shadows, you say? Fine. Gimme a third strip with "just the shadows, ma'am." Blend that in, mixing it in "downstream" of the first two components. (How much shadow would you like? I can twist the knob and turn it up, or turn it down...)

"Shot breakdowns" like this take a little getting used to, and they produce a lot of files and things that you have to carefully track, but when you finally get to the "mix-down" and you can push this knob up a little bit here, pull that one down there (and if you really wanna get funky, animate the knob-movements ... the sky's the limit here), and you're not freakin' out because "it's gonna take X days to re-render all this garbage" (nope... doesn't take any time at all), you'll see what all the fuss is about and why all the "extra" work was worth it.

It is equally important to study two-dee as well as three-dee. Get to know the entire process, at least to the point where you're not bluffing your way through conversations on the topics and tryin' to look like you're still sober. :surprised As they say, "to become a really good photographer, you gotta spend time in the darkroom. And if you're gonna get your work published, volunteer for a summer at a magazine, 'cuz the easier your image is to print on paper in little-biddy colored dots, the more likely it is they'll buy it."

[phix314]

This is what I did....I put the ramp node into the layered node, then the file node into the layered node. Making sure to place the file node on top (To the left side) with in the layered node. as in picture below.


You have to manually connect the out alpha of the file node into the layered node.

http://i126.photobucket.com/albums/p104/roaphotosharing/Maya/network.jpg

Out Transparency doesn't work as well.

[SkynetUK]

Thanks again Steve, but I don't seem to have alpha on the input 2 connection in the layered node. Input two being the ramp node correct, So maybe the ramp node i have is not set up right?

The only options I can see for Input 2 is "Colour", "Transparency" & "Glow Colour".

I attached a screen shot of it too. I feel so close! hahaha

Cheers

[SkynetUK]

My Bad!!

I was trying to use a Layered Shader rather than a layer texture! LOL What a muppet. :rolleyes:

I guess I need to try getting some real sleep sometime soon.

Okay so now that all work just fine. Any suggestions on how to use a projection node with this set up?

Cheers!!

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.