What students are saying about Joe Crawford
Founder of Teaching 3D
"Joe exemplifies what a teacher should be." - Brian Donley, Sony Entertainment
Color bit depth:
bitmap, 1bit depth:
(black/white)
0 - black
1 - white
Greyscale:
every pixel gets a single value, typically 8 bits.
black = 00000000
white = 11111111
in hex:
black: 00
white: FF
black --- dark grey grey light grey white
0 1 2 3 4 5 6 7 8 9 A B C D E F
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
00 44 AA FF
0 255
first digit is multiplied by 16
second digit is multiplied by 1
F * 16 = 240
F * 1 = 15
240 + 15 = 255
Two hex digits represents an 8 bit number:
Color is often stored as R B G numbers
one for red, one for green, one for blue
R G B
blue is: 00, 00, FF
0, 0, 255
0, 0, 1 (normalized like in maya)
yellow: FF, FF, 00
grey: 7F, 7F, 7F ( or 80, 80, 80 )
Most of our rendered images and photoshop files use
8bit per channel RGB color (and possibly alpha)
24 bit targa is 8 for red, 8 for green, 8 for blue
32 bit adds 8 for alpha
png works the same, but can go higher
48 is 16 for red, 16 for green, 16 for blue.
(change bit depth to 16 in photoshop)
64 adds 16 for alpha
OpenExr (recommended for rendering) can go high for HDR:
32 bits for red, 32 bits for green, 32 bits for blue, 32 for alpha
is compressed losslessly and intelligently for small file size!
has an option for half precision which is still super high.
Video is often stored in other color spaces, like HSV, YUV, etc.
Print is often stored in CYMK. (Looks these up in wikipedia!)
We don't need them though, because we can just convert at the end
RGB works best during production.
End result can be 8 bit per channel, will be good enough.
Human eye can't see any better unless there is ultra bright stuff.
But almost no current monitor can show this.
We need to store RGB 24bit video for demo reel.
Earlier in the process, higher numbers are useful.
Always use square pixels until export for a particular device that needs
other pixel aspect.