Der allgemeine Benchmark-Thread

**Blue**

glaub das liegt an irgendein sse

Sje · seit dem 26.10.2006 dabei

**Alpha**

Jemand mal ein RPi3 mit nbench als Vergleich? Würde mal wissen, ob die Werte in etwa passen?

Code

BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)


TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :          551.46  :      14.14  :       4.64
STRING SORT         :           84.32  :      37.68  :       5.83
BITFIELD            :      1.6799e+08  :      28.82  :       6.02
FP EMULATION        :           84.83  :      40.71  :       9.39
FOURIER             :          8224.1  :       9.35  :       5.25
ASSIGNMENT          :            10.8  :      41.09  :      10.66
IDEA                :          2826.6  :      43.23  :      12.84
HUFFMAN             :          905.11  :      25.10  :       8.01
NEURAL NET          :          11.051  :      17.75  :       7.47
LU DECOMPOSITION    :          366.35  :      18.98  :      13.70
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 31.057
FLOATING-POINT INDEX: 14.661
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 : 4 CPU ARMv7 Processor rev 4 (v7l)
L2 Cache            : 
OS                  : Linux 4.4.3-raspberrypi-v7+
C compiler          : armv7a-hardfloat-linux-gnueabi-gcc
libc                : /usr/lib/gcc/armv7a-hardfloat-linux-gnueabi/5.3.0/libgcc_s.so.1
MEMORY INDEX        : 7.206
INTEGER INDEX       : 8.185
FLOATING-POINT INDEX: 8.131
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.

Alles anzeigen

**Pain**

Code

matthias@alarmpi ~/nbench-byte-2.2.3 % ./nbench


BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)


TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          515.86  :      13.23  :       4.34
STRING SORT         :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          61.388  :      27.43  :       4.25
BITFIELD            :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :      1.5217e+08  :      26.10  :       5.45
FP EMULATION        :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          109.07  :      52.34  :      12.08
FOURIER             :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          7667.5  :       8.72  :       4.90
ASSIGNMENT          :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          6.8335  :      26.00  :       6.74
IDEA                :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          1822.3  :      27.87  :       8.28
HUFFMAN             :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          859.17  :      23.82  :       7.61
NEURAL NET          :
** WARNING: The current test result is NOT 95 % statistically certain.
** WARNING: The variation among the individual results is too large.
                    :          9.1083  :      14.63  :       6.15
LU DECOMPOSITION    :          210.15  :      10.89  :       7.86
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 26.238
FLOATING-POINT INDEX: 11.157
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 : 4 CPU ARMv7 Processor rev 4 (v7l)
L2 Cache            :
OS                  : Linux 4.1.19-4-ARCH
C compiler          :
libc                :
MEMORY INDEX        : 5.385
INTEGER INDEX       : 7.581
FLOATING-POINT INDEX: 6.188
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.

Alles anzeigen

Irgendwas ist doch hier faul. oO

thosch97

Code

BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)


TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :          1368.6  :      35.10  :      11.53
STRING SORT         :           494.9  :     221.14  :      34.23
BITFIELD            :      4.4264e+08  :      75.93  :      15.86
FP EMULATION        :          404.44  :     194.07  :      44.78
FOURIER             :           27452  :      31.22  :      17.54
ASSIGNMENT          :          47.663  :     181.37  :      47.04
IDEA                :          8520.5  :     130.32  :      38.69
HUFFMAN             :          3885.9  :     107.76  :      34.41
NEURAL NET          :       0.0059394  :       0.01  :       0.00
LU DECOMPOSITION    :          2320.1  :     120.19  :      86.79
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 116.500
FLOATING-POINT INDEX: 3.296
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 :
L2 Cache            :
OS                  : OpenBSD 5.8
C compiler          : gcc version 4.2.1 20070719
libc                :
MEMORY INDEX        : 29.448
INTEGER INDEX       : 28.792
FLOATING-POINT INDEX: 1.828
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
./nbench  17346.46s user 48.32s system 99% cpu 4:50:24.92 total

Alles anzeigen

niwax

thosch97

Code

BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)


TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :          816.32  :      20.94  :       6.88
STRING SORT         :          187.44  :      83.75  :      12.96
BITFIELD            :      3.6046e+08  :      61.83  :      12.92
FP EMULATION        :           187.2  :      89.83  :      20.73
FOURIER             :           23036  :      26.20  :      14.71
ASSIGNMENT          :          27.363  :     104.12  :      27.01
IDEA                :          5487.6  :      83.93  :      24.92
HUFFMAN             :          1843.1  :      51.11  :      16.32
NEURAL NET          :           50.84  :      81.67  :      34.35
LU DECOMPOSITION    :          1297.8  :      67.23  :      48.55
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 63.899
FLOATING-POINT INDEX: 52.395
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 : Dual GenuineIntel Common KVM processor 1862MHz
L2 Cache            : 4096 KB
OS                  : Linux 3.2.0-95-generic
C compiler          : Konfiguriert mit: ../src/configure -v --with-pkgversion='Ubuntu/Linaro 4.6.3-1ubuntu5' --with-bugurl=file:///usr/share/doc/gcc-4.6/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --program-suffix=-4.6 --enable-shared --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.6 --libdir=/usr/lib --enable-nls --with-sysroot=/ --enable-clocale=gnu --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-gnu-unique-object --enable-plugin --enable-objc-gc --disable-werror --with-arch-32=i686 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
libc                :
MEMORY INDEX        : 16.536
INTEGER INDEX       : 15.516
FLOATING-POINT INDEX: 29.060
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
./nbench  252,32s user 0,11s system 99% cpu 4:13,75 total

Alles anzeigen

tk1908

**Pain**

Zitat von Alpha

Jemand mal ein RPi3 mit nbench als Vergleich? Würde mal wissen, ob die Werte in etwa passen?

Das Netzteilkabel war faul, die Werte sind deutlich besser und kommen gut an Conis ran.
/edit: Jetzt aber.

Code

BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)


TEST                : Iterations/sec.  : Old Index   
                    :                  : Pentium 90* 
--------------------:------------------:-------------
NUMERIC SORT        :          656.74  :      16.84  
STRING SORT         :          75.741  :      33.84  
BITFIELD            :      1.9976e+08  :      34.27  
FP EMULATION        :          139.41  :      66.89  
FOURIER             :          8571.7  :       9.75  
ASSIGNMENT          :          10.581  :      40.26  
IDEA                :          2055.4  :      31.44  
HUFFMAN             :          993.67  :      27.55  
NEURAL NET          :          11.543  :      18.54  
LU DECOMPOSITION    :          386.17  :      20.01  
==========================ORIGINAL BYTEMARK RESULTS==
INTEGER INDEX       : 33.317
FLOATING-POINT INDEX: 15.349
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, W
==============================LINUX DATA BELOW=======
CPU                 : 4 CPU ARMv7 Processor rev 4 (v7
L2 Cache            : 
OS                  : Linux 4.1.19-5-ARCH
C compiler          : 
libc                : 
MEMORY INDEX        : 7.316
INTEGER INDEX       : 9.151
FLOATING-POINT INDEX: 8.513
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, g
* Trademarks are property of their respective holder.

Alles anzeigen

thosch97

SSD im T460p.

Und wo wir grad dabei sind, die im Desktop auch noch.

**Alpha**

Irgendwie ist der 4K-Thread ziemlich seltsam bei der SSD im T460p?

akino_germany

Zeit für ne SSD.

mrshadowtux

850 Pro schafft mehr

13:58 afeld@feldibook /home/afeld
% sudo hdparm -t /dev/sda
[sudo] Passwort für afeld:

/dev/sda:
Timing buffered disk reads: 1570 MB in 3.00 seconds = 523.19 MB/sec

Chrisse

Zitat von mrshadowtux

850 Pro schafft mehr
13:58 afeld@feldibook /home/afeld
% sudo hdparm -t /dev/sda
[sudo] Passwort für afeld:
/dev/sda:
Timing buffered disk reads: 1570 MB in 3.00 seconds = 523.19 MB/sec

Code

/dev/sda:
 Timing buffered disk reads: 1542 MB in  3.00 seconds = 513.76 MB/sec

Ich muss ja sagen, die 2% mehr rechtfertigen für mich den Aufpreis gegenüber einer 850 Evo nicht. Interessant wäre mal, wie hoch der Filesystem overhead ist:

Code

man hdparm


This measurement is an indication of how fast the drive can sustain sequential data reads under Linux, without any filesystem overhead.

**Pain**

Code

matthias@kirino ~ % sudo hdparm -t /dev/sda


/dev/sda:
 Timing buffered disk reads: 1554 MB in  3.00 seconds = 517.75 MB/sec

850 EVO 500 GB.

**Xaar**

Code

root@THINKPADX300:~$ hdparm -t /dev/sda


/dev/sda:
Timing buffered disk reads: 394 MB in  3.01 seconds = 130.77 MB/sec

Samsung PM800 128 GB (MMCRE28G8MXP-0VBL1).

**Pain**

Zwar kein Bench, aber ist doch ganz brauchbar, das olle 08/15 FreeNAS (immer noch i3-530).

thosch97

T460p, mittlerweile mit Bitlocker. Das könnte man sogar noch mit dem Wetter erklären.

niwax

850 EVO

**winfreak**

Das kann ich auch

Hier die 850 Evo von Dezember 2015:

Und hier meine 830 von September 2011:

Der allgemeine Benchmark-Thread

Jetzt mitmachen!

Benutzer online in diesem Thema