Amiga.org

Amiga News and Community Announcements => Amiga News and Community Announcements => Amiga Hardware News => Topic started by: Argo on March 09, 2007, 02:45:15 AM

Title: Sam440ep Memory Test
Post by: Argo on March 09, 2007, 02:45:15 AM
ACube Systems has published the Sam440ep memory test  on its products page.

http://www.sam440.com/eng/products.html
Title: Re: Sam440ep Memory Test
Post by: derringer3 on March 09, 2007, 08:20:18 AM
Not bad, but a today athlon x2 with ddr2 800 runs more than 2200 MB/s, and a core2 with ddr2 800 runs more than 4000 MB/s. I curious its real performance. (Using OS4 on SAM440)
Title: Re: Sam440ep Memory Test
Post by: AmigaPapst on March 09, 2007, 08:44:55 AM
@
Please notice, that Sam is a embedded board not a desktop system.
I think the benchmarks are very fast, for such board.
I hope we will see AmigaOS 4 on it, because without OS4 it is unuseable. :/
Title: Re: Sam440ep Memory Test
Post by: Jupp3 on March 09, 2007, 12:05:15 PM
Quote
but a today athlon x2 with ddr2 800 runs more than 2200 MB/s, and a core2 with ddr2 800 runs more than 4000 MB/s.

But for those who care, it doesn't run OS4. Neither does Samantha, but Sam running it some day is a lot more probable.
Title: Re: Sam440ep Memory Test
Post by: Colin_Camper on March 09, 2007, 02:40:46 PM
@AmigaPapst

Quote
I hope we will see AmigaOS 4 on it, because without OS4 it is unuseable. :/


That's a bit strong! It is after all an embedded board and, as such, is great for Linux (esp uClinux) etc etc.

I hope they do really well with it because, like with genesi, good sales outside the Amiga market will inevitably lead to more choice inside the Amiga market.  :-)

These guys are successful in leveraging the embeddded market whereas Eyetech were unlucky/hapless.  8-)
Title: Re: Sam440ep Memory Test
Post by: humppa on March 09, 2007, 04:20:10 PM
@AmigaPapst

Quote
Please notice, that Sam is a embedded board not a desktop system.


But 99.9% of all Amigans will use it as a desktop system. They are interested in OS4 on desktop platforms, not embedded devices. So what's so wrong about comparing SAM's performance to desktop hardware?

Quote
I hope we will see AmigaOS 4 on it, because without OS4 it is unuseable. :/


Well, it can be still used "for all your embedded ideas" without OS4. :-P
Title: Re: Sam440ep Memory Test
Post by: Elektro on March 09, 2007, 04:21:08 PM
Quote
Not bad, but a today athlon x2 with ddr2 800 runs more than 2200 MB/s, and a core2 with ddr2 800 runs more than 4000 MB/s.


athlons 64 do a lot more than 2200 MB/s.
Title: Re: Sam440ep Memory Test
Post by: LawlessPPC on March 09, 2007, 11:17:42 PM
put it one way 600 and something mhz running os4 its not going to be slow. Look how big linux can get
Title: Re: Sam440ep Memory Test
Post by: DamageX on March 10, 2007, 01:28:19 AM
Quote
Quote

Not bad, but a today athlon x2 with ddr2 800 runs more than 2200 MB/s, and a core2 with ddr2 800 runs more than 4000 MB/s.

athlons 64 do a lot more than 2200 MB/s.

How are you measuring this? The sam board doesn't look too bad to me.

On my 1583MHz Sempron w/ DDR333:
REP MOVSD (mem copy) = 330MB/S
REP LODSD (mem read) = 650MB/S
REP LODSD (mem read in L1 cache) = 3.0GB/S
Title: Re: Sam440ep Memory Test
Post by: AmigaPapst on March 10, 2007, 08:19:44 AM
@humppa
>But 99.9% of all Amigans will use it as a desktop system. They are interested in OS4 on desktop platforms, not embedded devices. So what's so wrong about comparing SAM's performance to desktop hardware?

Yes, but for the most desktop uses on OS4 you don't need more performance at the moment.

>Well, it can be still used "for all your embedded ideas" without OS4.

I think no, because fot the most embedded uses you need a fast OS, like os4. Linux is too slow on this low performance machine.
Title: Re: Sam440ep Memory Test
Post by: Hammer on March 11, 2007, 09:44:14 AM
@DamageX  
Try using the same stream_d benchmark.

In reference to "sam_memory_test.pdf"
http://www.sam440.com/eng/images/sam_memory_test.pdf

/usr/checkbench/stream # ./stream_d

This system uses 8 bytes per DOUBLE PRECISION word.
...
Function Rate (MB/s) RMS time Min time Max time
Copy: 285.7041 0.1121 0.1120 0.1121
Scale: 270.4713 0.1184 0.1183 0.1185
Add: 256.4075 0.1873 0.1872 0.1875
Triad: 250.5925 0.1916 0.1915 0.1918

----------------------------------------
Using a real AMD64 processor not some Sempron junk.

AMD Opteron 248 @2200Mhz with PathScale EKO complier.

Compiler: PathScale EKO Compiler Suite, Release 1.1
           Model Name: ASUS SK8N Motherboard, AMD Opteron (TM) Model 248
                  CPU: AMD Opteron 248
              CPU MHz: 2200
                  FPU: Integrated
       CPU(s) enabled: 1 core, 1 chip, 1 core/chip
      Secondary Cache: 1024KB (I+D) on chip
               Memory: 4x512MB, DDR400, PC3200, Corsair, CL2
     Operating System: SuSE Linux 9.0 (AMD64) 2.4.21-102-default

> pathcc -Ofast -lm stream_d.c second_wall.c -o 1.1ccOfast
> ./1.1ccOfast
-------------------------------------------------------------
This system uses 8 bytes per DOUBLE PRECISION word.
-------------------------------------------------------------
Array size = 2000000, Offset = 0
Total memory required = 45.8 MB.
Each test is run 10 times, but only
the *best* time for each is used.
-------------------------------------------------------------
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 7885 microseconds.
    (= 7885 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: 4613.5614 0.0070 0.0069 0.0072
Scale: 4520.3330 0.0071 0.0071 0.0071
Add: 4558.4067 0.0106 0.0105 0.0107
Triad: 4554.9003 0.0105 0.0105 0.0106


--------------------
Using SunFire V40z with 4 x Opteron 2.6GHz processors.

Sun Studio 11(MB/s) complier.
Sun Studio: -fast -xarch=amd64a -xvector=simd -xprefetch -xprefetch_level=3
(setenv PARALLEL 1)
Copy
 4658

Scale
 4614
 
Add
 4628
 
Triad
 4627

Sun Studio 11(MB/s) with  4 processor Automatic Parallelization switch.

For Automatic parallelization:
cc -fast -xarch=amd64a -xvector=simd -xprefetch -xprefetch_level=3 -xautopar stream_d.c second.c
(setenv PARALLEL 4)

Copy
 18120
 
Scale
 18108
 
Add
 17758
 
Triad
  17626

-----------------------------------------------
Same machine with GCC4.1 (MB/s)complier.

GCC 4.1: -O3 -funroll-all-loops -ffast-math -fpeephole -m64 -mtune=k8 -fprefetch-loop-arrays

Copy
 2766
 
Scale
 2745
 
Add
 2970
 
Triad
  2969

AMD64 goes well with SUN’s Studio 11 and PathScale's EKO compliers.

The GCC compiler isnt able to exploit this type of scalability since doesn't support automatic parallelization or OpenMP.

Note that there are two main types of Semprons i.e. K7 based (supports only single MCH via FSB) and K8 based(available in single MCH via Socket 754 or Dual MCH via Socket AM2/Socket S1).

SUN’s Studio 11 is avilable as a free download.