[OT?] MHz vs. MIPS in processors

were

telling

Megahertz (MHz) is just a measure of the frequency of the clocking signal that goes to the CPU. Because of architectural differences, there's no simple relationship that determines how many clock cycles it takes to execute even a single instruction -- and in most CPU's the number of clock cycles required to execute an instruction depends on xeveral things, including the which instriction it is. Therefore MHz really doesn't work as a way of comparing architecturally different CPUs. Therefore the concept of measuring how many instructions per second a CPU could execute was developed that that's the genesis of the Million Instructions per Second (MIPS) measure. Even this isn't adequate and there are several "standardized" benchmark programs that are sometimes used and the time it takes to run one of these programs is a better way to compare different CPUs. Even then there are ways to cheat and, ultimately, there is no absolutely perfect way to compare the speeds of architecturally diverse CPUs.

Norm

Chaos Master wrote: : Hello people.

: Reading some stuff about workstations, I find that the workstation speeds were : measured in MIPS (Million Instructions per Second).

: Q: For someone new in the area, what is the biggest difference between telling : processor speeds in MIPS and in MHz?

The biggest difference is that MHz is a real quantifiable value of clock speed, MIPS is a highly subjective marketing fluff word of little value.

DLC

: []s : -- : ? Chaos Master. | "I've tried so hard to tell : My Evanescence HP is at: | myself that you're gone :

speeds were

telling

So when you are choosing a microcontroller you choose the one with the highest MHz do you? Doesnt matter that some types (eg PICs) divide the clock by 4 internally while others run at the full speed (eg AVRs)?

MIPS will vary depending on the instructions you are executing, but it still gives a much better idea of processing speed than MHz.

-Daniel

Daniel Watman wrote to Dennis:

Problem is, all MIPS aren't created equal. MIPS numbers between different processor architectures/families may not be remotely comparable. That's not to say that comparing clock speeds gives a better comparison; it doesn't. For comparison, better to rely on a benchmark, e.g., 'stones or flops.

Norm Dresner ( snipped-for-privacy@att.net) said those last words:

I understand it now. Thanks.

were

telling

I always thought that MIPS = MHz or something like, so I sometimes said that a

telling

that a

Only if the processor in question executes, on average, one instruction per clock cycle. That's not the norm, and in fact there is no fixed number of clock cycles required per instruction that applies to all processor architectures - so there's no equating MIPS and MHz.

Having said that, though, the notion that "MHz is a real quantifiable value of clock speed," as if that alone told you something about the relative performance of various processors, is equally nonsensical. You can compare performance by looking at clock speeds alone ONLY if the processors in question have basically identical architectures - but not in general. For instance, suppose we have one processor which on average requires four clock cycles to complete an instruction, while another requires only two. Further assuming that the instruction sets of the two are comparable (in simple terms, you "get as much done" per instruction, on average, between the two), then the second processor at 2 GHz will outperform the first at 3 GHz. So...it ain't that simple. And you definitely don't have anything to brag about if you look at clock rates alone.

This whole situation is what leads to performance benchmark testing, in which programs are written to simulate the sort of real- world tasks that the processor will be called upon to do. You run these benchmarks on the processors you want to compare, and the one which finishes the job first is the higher-performance processor. Simple, and not as subject to marketing spin as the more-simplistic MIPS or MHz numbers alone.

Bob M.

Personally I would have said it was the other way round. I know many chips that used a higher frequency oscillator and internally divided it. Quoting the oscillator frequency of these is quite misleading. MIPS on the other hand is clearly quantifiable no matter what the clock speed. OTOH 10 MIPS of crap instructions may be no better than 5MIPS of a highly optimised instruction set.

Ian