What do the 256 Levels in 8-Bit Bit Depth Mean ?

[manishbjain]

Can someone help me with this - What does the 256 Levels in 8 Bit actually mean - Is it 256 levels along the X-Axis from hue to hue (which is not possible when they say 256 Levels of Red Channel coz hue would move from red to orange and so on ) , or its the 256 levels along the Y Axis from Pure Hue to Grey or 256 levels along Z-Axis (Brightness).

 

[gordo]

Can someone help me with this - What does the 256 Levels in 8 Bit actually mean - Is it 256 levels along the X-Axis from hue to hue (which is not possible when they say 256 Levels of Red Channel coz hue would move from red to orange and so on ) , or its the 256 levels along the Y Axis from Pure Hue to Grey or 256 levels along Z-Axis (Brightness).

Levels of grey. Nothing to do with color.

 

[Stephen Marsh]

Try these for further info:

Bit depth

or

http://www.cambridgeincolour.com/tut.../bit-depth.htm



Hope this helps,


Stephen Marsh

 

[mastegman]

Stephen, good link.

Gordo, while they might be often described as 'levels of grey' they do have something to do with colour in the context of a colour reproduction system.

Manishbjain, you are confusing the 256 'levels', or shades, of a specific channel (R, G, or B) with the graphic representation of the spectrum produced by their combination. You have to remember that there is a difference between using your own eyes to experience the world and using a man-made system to reproduce this experience which are still, by comparison, no where near as sophisticated.

As separate channels they are described as 'levels of grey' or, more appropriately, levels, shades or tones, of red, green or blue.

Sorry if I'm going over the basics especially as Stephen has already provided a link to a more graphic explanation.

Computers store information as either a positive or negative charge on an electromagnetic device (like a hard disk or USB). This charge can be interpreted as either a 1 or a 0 and is defined as a bit - the smallest single entity in the computer's universe. This is what's known as binary code as there are only two possible states. This 1/0 can be interpreted by a program as OFF/ON, IN/OUT, UP/DOWN, LEFT/RIGHT or, in the case of graphics, BLACK/WHITE. We can store the information about a 1 bit, black and white, image as a matrix of very small pixels which are individually invisible to the naked eye and perceived as linework when viewed together. If we just look at one pixel, or picture element, and add another bit to describe it we can increase the number of possible outcomes from 2 to 4 (1/1; 1/0; 0/1, 0/0). This means we can describe four (4) shades: black, white and 2 shades of grey. To cut a long story short, when we use 8 bits to describe the tonal characteristics of this pixel we have 2 to the power of 8 which equals 256. This is known as Bit Depth.

For more info...https://en.wikipedia.org/wiki/8-bit_color

If we use this system to store information about the three additive primary colours (R, G, B) we have 256 shades of each colour. When you combine all three channels to re-create the visual spectrum you have 256 x 256 x 256 or 16,777,216 possible outcomes. More theoretical colours than the human visual system can differentiate.

In other words, you can increase the number of possible colours that might be reproduced by a colour reproduction system by using more bits to describe the characteristics of each pixel. This is why high end graphic systems and applications are described as '16 bit' processing (or 48 bit for RGB as this sounds even better). This allows more complex calculations that produce more refined and, hopefully, more accurate results. Of course, this requires more disk space on the electromagnetic storage device you are using, more computing power to process the information, and faster networks to send this information to devices and/or to other recipients of the file(s).

Problems occur in graphic reproduction when devices, e.g. monitors and printers, do not have the necessary hardware to do this. Standard monitors rely on the computer's graphics card and even though the file might be 16 bit you might see 'banding' on screen, or gradients with lines or 'steps, between the tones rather than smooth transitions from one tone or colour to another. High-end monitors often have their own high bit processor built in. This might also happen with printers if the driver is not 16 bit compatible and is why RIPs are preferred for the controlled reproduction of high-end graphics.

When you are looking at the X and Y axis you are looking at a visual tool for representing the relationship between the different colours produced in combination by the primary colours of that particular colour model or system be it RGB, CMYK, HSL/B. The y-axis you talk about represents Saturation in the Hue, Saturation and Brightness or HSB colour model. It is the result of combining the primaries in relative proportions. The more 'equal' they are the more neutral the result. Brightness is plotting the amount of light. Hue is the specific colour. In theory, there are 16,777,216 different colours within the bounds, or gamut, of any colour model made up of of 3 primaries.

How accurately those colours are reproduced is the problem.

I hope that helps.

Mark

 

[gordo]

Gordo, while they might be often described as 'levels of grey' they do have something to do with colour in the context of a colour reproduction system.

[snip]
As separate channels they are described as 'levels of grey' or, more appropriately, levels, shades or tones, of red, green or blue.

I think I disagree. ¯\_(ツ)_/¯

The OP asked "What does the 256 Levels in 8 Bit actually mean"

A greyscale image in 8 bit mode has 256 levels of grey (or "tone") You can assign any color to that image (e.g. a PMS ink) and the 8 bit image will still have 256 levels of grey. The grey will be assigned the PMS ink in printing or be represented on a computer display using RGB. Even though it's in color the 8 bit image will still only have 256 levels of grey.

An RGB image is typically made up of 3, 8 bit images each of which still only have 256 levels of grey. Each of the three greyscale images are assigned a Red, Green or Blue hue for display purposes. The hue assigned to each of those greyscale channels is variable from device to device. They have no intrinsic color themselves. I.e. If I say that a given image is an RGB image that does not tell you what the color of the image is. You don't know the Red hue, nor the Green hue nor the Blue hue and as a result you don't know the color of their resulting composite is.

 

[Danny Whitehead]

Looks like a mere semantic disagreement to me, but I wouldn't describe levels of R, G and B as 'grey levels'... but I might with the levels on a plate. Maybe it's just an intuition thing, because there's generally less need to see and control individual RGB channels than there is with ink.

You don't know the Red hue, nor the Green hue nor the Blue hue and as a result you don't know the color of their resulting composite is.

You do if it's assigned a profile.

 

[gordo]

Looks like a mere semantic disagreement to me, but I wouldn't describe levels of R, G and B as 'grey levels'... but I might with the levels on a plate. Maybe it's just an intuition thing, because there's generally less need to see and control individual RGB channels than there is with ink.


You do if it's assigned a profile.

That wasn't the point.

 

[Glenn McDowall]

the OP's question was just What does the 256 Levels in 8 Bit actually mean, it was pretty vague about colour or grey, its not surprising the answers have gone off in different directions. Mastegman explains it all pretty well. I'd just say that 2 to the power 8 = 256, and leave it at that. You can allocate the 256 to whatever you want.

 

[Chasfinch]

I always think of halftone when I am talking about a specific channel. Only way my mind can handle it