A few general questions

[frank1331]

I feel embarassed to ask this, but here are a few things I've never really understood.

 

1. What exactly is an alpha channel/map, and why would I want to use it?

 

2. What is texture baking? Is it used for single scenes or just animations?

 

3. Is it possible using Max/Viz and vray, to render out a model with a basic material on it, with correct lighting and shading, but then to paint a true materail to individual surfaces (similar to Piranessi)? If possible how?

 

Sorry but I've always wondered these things and know this si the best place to ask.

 

Thanks

[carloscristerna]

hi there

 

an alpha chanel is just a black and white image in wich you can decide what is transparent and what is not....is like a mask...when you render a building against nothing, and you save your alpha chanel is going to be easier to cut your image in photoshop so that you can put a sky and stuff like that...they are very useful .....

 

texture backing, is what they use in 3d games, you see that in reallity there are no lights in the scene but everything is "painted" on the walls or flors...get it??? in that way its going to be easier for the computer to render everything with movement...the scene is translated to textures all the efects of the light and all the stuff and then they use it as simple textures...

 

it is possible to paint on it, it will be like using color pencils or watercolor over the basic rendering, you can do it in photoshop, but the quality of the rendering is going to depend in your drawing abilities...

 

hope this helps you....read the help files in max, they are very useful..

 

Carlos Cristerna

[Hector]

 

1. What exactly is an alpha channel/map, and why would I want to use it?

 

You may already know part of this story but I have to tell it from the begining:

 

A digital image is usually described as a grid of square pixels, having each pixel a certain color.

 

A more detail description would say that you have in reality is three grids (or say, "channels") on top of each other and perfectly aligned and that each pixel is composed of three squares in the same position but each on a different channel. (technically it is a bit different but it is easier to understand this way).

 

Now lets consider a single pixel in isolation and its three stacked components (or channels). Imagine you take the first channel an give it 8 bits of memory to use and asign one color: RED. Say that those 8 bits of memory allows that channel to have 256 potential intensities of red from wich to choose only one at any given time. Repeat exactly the same process with the other two chanels using color GREEN for one and BLUE for the other.

What you have now is three stacked channels, each with 256 variations of its asigned color. If you consider the final "color" of the pixel as the "blending" of its three channels, you have actually more than 16 millions of possible colors to choose from for the pixel to show, due to the possible combinations of different values of Red Green and Blue of its channels (256x256x256). What you have is a "24 Bits Pixel". If you think that each

pixel in the image does the same, you have what is commonly knonw as a "24Bit Image".

 

NOW, going back to the subject, imagine that you take one pixel and give it an "extra" channel and the usual 8 bits of memory for it. But this time you say these bits gives the channel a range of 256 grays (pure black and pure white included) to choose from. Instead of "blending" the chosen gray with the other three colors (channels) it is used to define a level of tranparency of that particular pixel from completely transparent to completely opaque depending of what gray you choose. This means you now have a "32 Bit pixel" and accordingly a "32Bit Image". This extra channel is called the "ALPHA CHANNEL".

 

This concept of using values of grays to define the value of a characteristic is widely used in 3d, for instance in a bumb map. However in this case you just use a "8 bit image", this is, a greyscale image which has only one channel where each pixel has only 8 bits of memory in total.

 

If you multiply the number of pixels contained in the image by the amount of bits asigned to each pixel (24 or 32 in this particular case) you get a good approximate of the size of the file (without compression).

 

Not all the file formats support 32 bits, wich means that not all file formats support alpha channels.