Amiga.org
Amiga computer related discussion => Amiga Hardware Issues and discussion => Topic started by: on January 05, 2003, 04:25:11 PM
-
How much bogomips get the AOne in linux?
-
What's Bogomips?
I know what mips are, just never heard of bogomips.
-
it's just a test speed of the machine in Linux and it's printed before X-window system starts.
Used for compare systems independant of which CPU is used
-
I wonder how many BOINGmips that is? :-D
-
You should be able to get the bogomips by doing:
cat /proc/cpuinfo
It works on ia32, ia64 and sun4u so I imagine it would work with PPC ^_^
-
I get 1266 on my A1-G3@599Mhz.
Marginally more per Mhz than on my P3 laptop (1394 @701Mhz)
-
My AMD Duron @750Mhz does 1642 bogomips in SUSE linux
-
To answer my own question:
From whatis.com (http://www.whatis.com)
"Bogomips is a measurement provided in the Linux operating system that indicates in a relative way how fast the computer processor runs. The program that provides the measurement is called BogoMips. Written by Linus Torvalds, the main developer of Linux, BogoMips can indicate when you boot a computer whether the system options have been specified for optimum performance. You compare the bogomips for your computer with what they ought to be for your computer's particular type of processor. Torvalds named the program BogoMips (for "bogus (or fake) MIPs") to suggest that performance measurements between two computers can be misleading because not all contributing factors are stated or even understood. Although MIPS (millions of instructions per second) has been frequently used in computer benchmarks, it's agreed that the variation of context tends to make the measurement misleading. Bogomips measures how many times the processor goes through a particular programming loop in a second.
BogoMIPS is built into some versions of Linux. It also exists as a stand-alone application program that you can download from certain Web sites. In October, 2001, Wintermute, a self-declared hacker site, reported that its server had achieved 2785.28 bogomips on a computer with an AMD Athlon microprocessor operating at 1,396 MHz"
-
@yoodoo: "1200 on A1-G3 @ 600 MHz"
On my Duron @ 600 MHz, using SuSE, I get 1196. I think the Duron is probably fairly similar to the G3 in state-of-development?
Your G3 beats my Duron by 5.8%
tony
-
I guess itll take some time till we reach values like 4 x 5570 from this little beauty here (quad 2,8ghz) :-D
-
I don't know anybody who actually uses bogomips for anything. They're pretty useless except for sometimes finding that one of your CPUs isn't running at the right clock speed for some reason.
-
I dont think bogomips say anything about the perfomance at all.. From what i heard, its same no matter what cpu you have, a 1ghz pIII and a 1ghz athlon has same...
Im not really sure about this though, but thats what someone told me.
-
my k6-2 450: bogomips : 901.12
-
strobe wrote:
I don't know anybody who actually uses bogomips for anything. They're pretty useless except for sometimes finding that one of your CPUs isn't running at the right clock speed for some reason.
Yes exactly what i thought.
-
I dont think bogomips say anything about the perfomance at all.. From what i heard, its same no matter what cpu you have, a 1ghz pIII and a 1ghz athlon has same...
Correct... For Pentium or greater x86 architecture, or PowerPC/Power4, bogomips's just actual measured mhz*2. So according to that, my AMD 1.25ghz (which runs at 1.2611 for some strange reason) would score 2522.2 bogomips. Hooray for useless numbers!
-
PS: Linus can suck my balls
-
you must be a big fan then.
-
I have a friend who claims a G3@600 will only get 350 bogomips, someone please tell me how wrong he is :-)
-
Sorry it gets 596.5 bogomips, and its a mos build of bogomips running on a pegasos, both the 604 and the G3 measurments were done under MOS.
(http://www.ryusworld.webhop.net/grab.jpg)
(http://www.ryusworld.webhop.net/G3vs604e.jpg)
I have theses images thanks to naTmeg, I would request that the links to them are not posted to anywhere else without my permission.
-
Well a Pegasos under Linux got also 1200Bogomips (It's logical btw ;) ).
So apparently AmigaMARK doesn't compute the Bogomips the same way.Or maybe it's not a native PPC version... Well according to the readme, AmigaMARK use the 68k emulator of MorphOS to run under the Pegasos so it's not PPC Native.
-
yoodoo wrote:
you must be a big fan then.
lol, now thats funny.
-
this is done on the mos-native version of amigamark.
i dunno why it get 1200 bogomips on linux.. :)
anyway..
-
Here's some 68k bogomips ratings for comparison:
68030/16 Atari Falcon 3.95
68030/25 Amiga 3000 6.21
68030/50 Amiga 1200 12.36
68040/25 Amiga 4000-040 16.61
68040/40 Amiga 1200 26.52
68060/50 Amiga 1200 99.53
68060/50 Amiga 4000 99.74
68060/66 Amiga CS MkII 132.71
(http://www.hobby.nl/~clifton/bogomips-3.html#ss3.11)
-
Oh, and this one:
Commodore 64 0.0033
:-D
-
strobe wrote:
PS: Linus can suck my balls
It can?
Now *that's* a feature I'd like to see implemented in AOS & MOS! :lol:
-
naTmeg wrote:
this is done on the mos-native version of amigamark.
i dunno why it get 1200 bogomips on linux.. :)
anyway..
Well there is no native MorphOS version of AmigaMARK according to the OFFICIAL readme I found on the OFFICIAL AmigaMARK website for the LATEST version of this software.
It says:
- Hardware: - ยท Amiga/Pegasos[JIT]
JIT == Just In Time
That's mean it use the JIT 68k emulator.
So I doubt a lot he has used the MorphOS PPC Native version AmigaMARK as this version doesn't exist apparantly ;) Or point me to a link where I can DL one. The only one I found was this one:
AmigaMARK Beta (http://member.ycn.com/~hausrup/downloads/AmigaMARK.905beta.lha)
Which is clearly not a MOS PPC Version, this archive contain only one binary which is a pure 68k binary. So please check your information before spreading them.
This would explain the huge difference between AmigaMARK result and Linux result because this kind of test can't really benefit of the JIT as it quite never reproduce the same operation 2 times and as you may know JIT compile at the first run of an operation (for the first run it's interpreted) and the benefit of this compilation appear only if it does this same operation a second time as it'll then use the dynamically compiled version. But in the case of BogoMIPS test this type of repetitions of an operation is not happening oftenly.
-
My Dual P100 system gets 492 bogomips, My AMD Athlon 1.4Ghz gets 5870 bogomips.
-
I dont think bogomips say anything about the perfomance at all.. From what i heard, its same no matter what cpu you have, a 1ghz pIII and a 1ghz athlon has same...
Of wich the athlon is way faster than the p3...
If so, why are you guys still making comparisons?
Isn't there a better way of measuring speeds?
-
LOL the funny thing is as far as I am aware naTmeg wrote AmigaMARK, heck I betatested it for him under 68k here.
-
Ryu wrote:
LOL the funny thing is as far as I am aware naTmeg wrote AmigaMARK, heck I betatested it for him under 68k here.
Well in that case ok ;) But then it must not be the same algorithm than the one use in Linux 2.4.18 or maybe Linux use some G3/G4 optimisations (compiling or in the algorithm) to get the best result but maybe these optimizations are not yet implemented in the AmigaMARK MOS version of this test...
At least I know that on PC the Bogomips test is performed using different algorithms that are each optimised for MMX, SSE or other CPU specific instructions and keep the best result.
-
Right, the Morphos version exists and is available for a limited number of betatesters. ;-)
Benchmarking is very difficult because there are a lot of different things to test and different ways to do it. I remember that Bytemarks were always better for PPC and SPARC, that SPEC (sponsored by Intel) was better for Intel (how strange!) ...
It really depends what compiler your use, what optimization are done etc.
That's why fist you need to compare apples with apples: execute the same code in the same conditions.
For example I don't think that comparing a system speed and an emulated system speed specialy with JIT) make any sense to have an idea of general speed difference.
Even in general, you can not say system X is 4 X faster than system Y.
For what ? CPU intense stuff ? My CPU is idling 99% of the time anyway !
I remember that I had to do a search replace in a huge text file at work and that the only tool I had was MS word. It completely froze my computer for hours. I took the same file and did the search / replace on my amiga with CED and it was finished in a few minutes.
Who in the PC world could think that some people still use 50 Mhz machine to browse internet, do mail etc ?
What counts is the user experience on the machine. Does it do the job ? Is it responsive. Does it get me mad or not ?
BTW Natmeg is doing an excelent job. Amigamark is going to be THE Amiga benchmark. :-D
Benchmark are fun but never make any general statement after looking at a benchmark. For example look at the bogomips test and the G3/604 and then look at dhrystones. What does it tells you ?
-
to be exact here the routines used in the two version.
its a assembler code build-in in the c-source.
here the functions:
m68k:
static void delay(ULONG loops)
{
__asm__ __volatile__ ("\n\tmovel %0,%/d0\n1:\tsubql #1,%/d0\n\t"
"bpls 1b\n"
: /*no outputs*/
: "g" (loops)
: "d0");
}
ppc:
static void delay(ULONG loops)
{
__asm volatile ("cmpwi 0,%0,0\n\t"
"mtctr %0\n\t"
"beqlr\n\t"
".loop: bdnz .loop"
:
: "r" (loops)
: "cc");
}
hope this is clear now :)
naTmeg,
-
hum, it looks like the 604 results are comparable to what we get on Linux/Apus.
The G3 results are more than suspicious. :-?
-
lorddef wrote:
My Dual P100 system gets 492 bogomips, My AMD Athlon 1.4Ghz gets 5870 bogomips.
You got some super-CPUs there? :)
One P100 gets about 40 BogoMIPS, and my dual AthlonMP@1752MHz get 3500 BogoMIPS per CPU.
Check out the BogoMIPS ratings listed in the BogoMIPS mini-HOWTO (http://tldp.org/HOWTO/mini/BogoMips-3.html).
I can't believe we're discussing busy-loop timings under Linux! :)
-
Has any one compared JIT-68K speeds between PPC-JIT-68K VS X86-JIT-68K?
-
Just a side note; my old AthlonXP1800+/WinXP-SP1**/JIT-WinUAE(0.8.22.R4, JIT-020/881) delivers ~725.2 Bogo MIPS(via 68K Amiga Mark 2003 beta).
**with XP Pro's bloatware services turned on.