Contour Enthusiasts Group Archives Forums General Common Interests For you Apple gurus!

I used to have to use Apple computers in college for graphic design because everything was tailored to them and all through elementary and high school for that matter. I have never liked them, the layout and operation of the systems just always bugged me for some reason. Once I graduated and went into web design, I never touched them again. Just a personal preferrence.

I like all types of machines equally (except when my Macintosh decides not to crash and the PC decides to crash each time I open IE)

I'm talking more along the lines of processor count and performance based on per processor workload capabilities, not bit architecture...

looks like an AMD commercial to me!

It's still hard to determine which CPU is "better." It's harder to compare an Apple RISC CPU with an Intel/AMD CISC CPU than just by speed alone.

It's also dependant upon which is most important to you. "Better" could mean lower price, in which case Intel/AMD is for you. If it means potential, it's Apple, it goes on and on.

More capable? RISC all the way.

As a serious computer geek, I feel the need to clear the air here.

First off, as Bishop mentioned, Intel/AMD products are CISC (Complex Instruction Set Computer) and current apples are RISC (Reduced Instruction Set Computer)

Computer scientists have long known that RISC chips have inherently better theoretical performance than CISC chips.

However, in order for a RISC chip to reach that level of performance in a real test, it needs a much larger volume of RAM.

Back when ram was very expensive, (When IBM designed the PC) the CISC chips with less memory were at the knee of the price performance curve.

RISC entered the picture well above the knee, and therefore, you are paying vastly more money for slightly improved performance.

there are 2 advantages to upping the "Bit count" of a processor.

Processors communicate with RAM in parallel, therefore, for an 8086 chip, 8 bits are transmitted at a time, a 286 is 16-bit, a 386, 486, Pentium, PMMX, PII, PIII, P4, are all 32 bit, as are AMD competiors, the 5x86, K6, K6-2, K6-III, and anything with the "Athlon" moniker.

every time you up the bus width, you can more more data into and out of the processor on every clock cycle.

the "Bit count" also determines the largest numbers that the processor can compute with on a single clock cycle. The most important numbers that every processor must deal with are Memory addresses, needed to retrive data from RAM.

The way the memory is addressed, a 32-bit processor can handle 4 GB of system memory. Server chips like the Intel Xeon can use up to 8 GB in single processor systems, but processing a memory address requires two clock cycles.

64-bit progessing makes anything that processes the same volume of data (being greater than 4 GB,) repeatedly incredibly fast, because the data can be kept in RAM.

Additionally, 64-bit processors have more 64-bit registers, allowing calculations involving numbers over one million times larger to be processed at once, which is valuable for simulations, and more realistic 3D graphics.

The Apple G5 and the AMD Opteron both do as the x86 has done, and add extentions to expand on the original chip, and code base, thereby preserving the function of all the software that came before.

Maybe X86 isn't the best archetecture out there, but it's what we got, and as long as we can get x86 to fill our computing needs cost effectively, we will continue to use it.

I said that to say this... Intel's Itanuim line of 64-bit processors dropped native x86 functionality, in favor of starting again with a fresh, clean slate. running 32-bit code compiled for an x86 requires emulation, which reduces the effective speed at which the program can be run.

The G5, and the Opteron, (With x86-64) both expand on the original 32-bit origins and theoretically will run 32-bit apps just as fast as 32 bit counterparts.

Lastly, the main reason that P4 chips have faster clocks, than Athlon XP's but cannot break out of a dead heat performance-wise is because they have longer execution pipelines.

the pipeline has a set number of stages, and every instruction must move from one to the next, in line in time with the clock signal. therefore, there are more clock cycles that pass between when a command goes in, and when the result comes out.

Longer pipelines allow for faster chip speeds, but the real determiniation of performance is how many instructions are completed per clock cycle. the Athlon XP's shorter pipeline keeps it competitive with the P4

you forgot to mention that the g4 has a much shorter pipeline, and therefore requires less ghz to obtain the same objective as a pentium