|
The comments are owned by the author. We aren't responsible for their content.
Author | Thread |
---|---|
joerg |
Published: 2007/3/9 8:24 Updated:
2007/3/9 8:24
|
Just can't stay away
Joined: 12/01/2006
From: Germany
Comments: 1927
|
Re: Sam440EP memory tests published
For comparison here are the results of stream_d on an AmigaOne XE with 750FX/800 (executed on AmigaOS4, but of course the OS makes no difference for such benchmarks):
-------------------------------------------------------------
Please note that the stream benchmarks aren't optimized, with functions optimized for PPC like the newlib.library memcpy() and memset(), or the AmigaOS4 final IExec->CopyMem[Quick]() and IUtility->SetMem(), you get much faster speeds than with trivial code like in the stream benchmarks or the clib2 memcpy()/memset() functions. |
ikir |
Published: 2007/3/9 12:30 Updated:
2007/3/9 12:30
|
Not too shy to talk
Joined: 11/27/2006
From: SYS:Prefs/
Comments: 374
|
Re: Sam440EP memory tests published
It seems much more faster than A1.
|
COBRA |
Published: 2007/3/9 15:10 Updated:
2007/3/9 15:23
|
Not too shy to talk
Joined: 11/24/2006
From:
Comments: 370
|
Re: Sam440EP memory tests published
Faster memspeed is important for Sam440 to compensate for the lack of L2 cache. Some dnetc benchmark results would be interesting, it'd give us a better idea of how overall performance compares to a G3/800 in some computing-intensive tasks.
Some more comparison benchmarks can be found here. |
srupprecht |
Published: 2007/3/9 16:03 Updated:
2007/3/9 16:03
|
Just popping in
Joined: 01/06/2007
From:
Comments: 11
|
Re: Sam440EP memory tests published
That's a G4 (7455 I believe) @ 800mhz (A1):
------------------------------------------------------------- This system uses 8 bytes per DOUBLE PRECISION word. ------------------------------------------------------------- Array size = 1000000, Offset = 0 Total memory required = 22.9 MB. Each test is run 10 times, but only the *best* time for each is used. ------------------------------------------------------------- Your clock granularity/precision appears to be 2 microseconds. Each test below will take on the order of 65684 microseconds. (= 32842 clock ticks) Increase the size of the arrays if this shows that you are not getting at least 20 clock ticks per test. ------------------------------------------------------------- WARNING -- The above is only a rough guideline. For best results, please be sure you know the precision of your system timer. ------------------------------------------------------------- Function Rate (MB/s) RMS time Min time Max time Copy: 186.0894 0.0862 0.0860 0.0867 Scale: 166.2200 0.0964 0.0963 0.0969 Add: 173.7731 0.1383 0.1381 0.1386 Triad: 171.0790 0.1404 0.1403 0.1408 |
Antique |
Published: 2007/3/11 1:25 Updated:
2007/3/11 1:25
|
Home away from home
Joined: 11/30/2006
From: Norway
Comments: 2628
|
Re: Sam440EP memory tests published
Looks promising,but would love the see the sam at +1ghz.
|
LiveForIt |
Published: 2007/3/13 18:24 Updated:
2007/3/13 18:24
|
Home away from home
Joined: 11/20/2006
From: Norway
Comments: 3870
|
Re: Sam440EP memory tests published
Sam440 is starting to look interesting now
I hope the sam440 will come whit faster CPU at some point? |
joerg |
Published: 2007/3/16 16:59 Updated:
2007/3/16 16:59
|
Just can't stay away
Joined: 12/01/2006
From: Germany
Comments: 1927
|
Re: Sam440EP memory tests published
Quote:
Sam440 is starting to look interesting now |