Re: Lenovo ThinkSystem HR330A

[Swift Griggs <swiftgriggs%gmail.com@localhost>]

On Tue, 24 Sep 2019, Sad Clouds wrote:
> CPU clock rate can sometimes be a bit misleading. Quite often, the 
> faster the clock rate, the more time CPU is spent waiting for memory 
> access, etc.

Sure, I absolutely agree. It's just when you have nothing else to go by... 
clock speed is better than a totally uninformed guess (kinda). Goes back 
to that whole "MIPS" debate about "a million of WHAT instruction per 
second?"

> I've been developing various performance tools in order to compare 
> different hardware platforms.

Cool man, that's an ambitious goal. Everyone loves to argue and throw 
bombs between ISAs and hardware platforms. I just mention Geekbench 
because for whatever reason, I see a lot of weird hardware in their 
benchmark list and their benchmarks are supposedly somewhat "real world" 
as they do some "real" stuff (encoding, compression, etc).

I love benchmarks (even if they do cause too many arguments). They are 
fun.

> Below is an example where 1.4 GHz RPi3 ARM CPU seems to outperform 2.4 
> GHz x86_64 CPU on integer division.

That's an interesting test. It reminds me of the days when MIPS would 
stomp the guts out of x86 for FP operations. It's always good to check 
assumptions on performance. Things like this keep folks honest (or make 
them exaggerate to the moon, one of the two *grin*).

-Swift