Can someone explain bit for me i know its 8bit= byte (1 or 0) but nothing more, and I canīt figure out how a craphic card or a processor could work in 16bit. Would that mean that it can handle 2 numbers at the same time or 32bit = 4 numbers at the same time. Sorry for bad english, please help! /Henrik

I think your confusion is due to not understanding CPU and BUS architecture. Very simply put: At the machine level each bit of a computer instruction will turn on or off a different switch (gate) in a CPU/memory array. If you only have an 8 bit CPU/BUS structure, there are 255 possible combinations (memory addresses or commands) handled 8 bits at a time. 1. Let's take a simple Load Accumulator command. This will fetch from memory an 8 bit value and load it into the A register (within the CPU). This will take at least 2 clock cycles, one to deliver the instruction (load A) and one for the memory address (everything is manipulated via memory). 2. If you move to a 16 bit CPU/BUS structure, you can now accomplish this in one clock cycle as the instruction and memory address could both be delivered at the same time. Notice that both 1 and 2 could only access memory locations 0-255. In order to access more memory, you need more bits (or more clock cycles). Moving to a 32 bit structure would allow for the same command structure (8 bits - 255 possible commands) and leave 24 bits for a memeory address ( 0 - 16,777,215 (16MB)memory locations). Thus you could access anywhere in a larger memory array per clock cycle. Not to confuse things, but as CPUs grew in power, they have more paths (gates) availible to do more things at the same time (during the same clock cycle). Moving to a 16 bit computer instruction structure (CPU) allows 65535 different combinations of gate switchings to handle things. For the 16 bit Bus structure, we're back to 2 clock cycles for our Load A instruction. The first clock cycle for the instruction, the second to access memory. For the 32 bit Bus, our possible memory locations has gone down to 65,535 (64K)locations, but it can be done in one clock cycle. To access more memory will require more bits/more clock cycles. Hope this helped more than confused :)

Damn michael, you are on the ball !!!

Dave, That's because I started in this biz when computers were built using RDL technology. I once met a guy (in the states) that worked on the older tube type (like radio tubes)computers!

to put the over technical posts in perspective: when talking about 8 bit, 16 bit, 32 bit, 64 bit, etc., they are most likely talking about the width of the data bus. The wider the data bus, the greater the throughput, which is very important in data processing. If you compare a cpu with 16 bit internal and 8 bit data bus with a cpu that is both 16 bit, performance of the latter is at least double. This is given the same clock speed on both cpus. The data bus is the pipe that the data flows through, thus a clogged pipe (narrow data bus) lets far less through than one that isn't clogged with results you all know too well. [grin]