Trace: Accueil FLOPS

FLOPS

flops.c

gcc -O3 -DUNIX -o flops

MODULE FADD FSUB FMUL FDIV
1 50% 0% 43% 7%
2 43% 29% 14% 14%
3 35% 12% 53% 0%
4 47% 0% 53% 0%
5 45% 0% 52% 3%
6 45% 0% 55% 0%
7 25% 25% 25% 25%
8 43% 0% 57% 0%

Summary of results

pcfarm cosinux pluton micron Meta charon MacBook kant zarathoustra tanker
OS Debian Etch Debian Etch FreeBSD 7.0 gcc4.2 64bits Win32 Win32 Win32 OpenSolaris MacOS X 10.5 Debian Etch NetBSD 4.0 FreeBSD 7.0
Compiler gcc 4.1.1 gcc 4.1.1 gcc 4.2 VC 6.0 VC 6.0 VC 6.0 SS12 gcc 4.2 gcc 4.1.1 gcc 4.x gcc 4.2
CPU Type Intel Pentium 4 Intel Xeon AMD Opteron 250 Intel Mobile Core 2 Duo L7500 Intel Core 2 Duo E8500 Intel Pentium M 750 Intel Core 2 Duo T7500 AMD Athlon XP 1900+ Transmeta Crusoe TM5800 AMD Athlon™ 64 X2 5200+
CPU Freq 3.00 GHz 3.20 GHz 2.40 GHz 1.60 GHz 3.16 GHz 1.86 GHz 2.20 GHz 1.60 GHz 1.00 GHz 2.60 GHz
Mod. 1 901.8858 965.6640 1405.9542 502.7353 1961.1491 762.0668 838.2656 1023.7240 754.1616 151.8459 1448.2092
Mod. 2 359.0649 365.3225 865.0993 294.5431 1224.4619 441.1077 440.1029 438.7116 465.4254 136.5294 835.8749
Mod. 3 978.5660 1039.8141 1932.7696 841.5353 1961.6858 1123.0968 1149.0999 1760.4868 1311.5515 206.7945 1859.5648
Mod. 4 907.2960 962.9143 1739.9996 857.7172 2051.2821 1061.5066 1276.0525 567.4052 1103.3793 205.0616 1677.9128
Mod. 5 972.0663 1035.2940 1729.8708 912.5438 1186.9854 621.0993 1015.0879 612.4561 1063.8487 234.4722 1665.0349
Mod. 6 979.3746 1055.4554 1892.5986 952.7109 1206.4679 684.6498 1262.1811 827.7106 1212.5997 314.7081 1822.0041
Mod. 7 258.9085 275.1750 502.3997 152.1809 542.3729 195.8247 365.3252 280.9953 300.3919 80.3831 484.0393
Mod. 8 946.4228 1013.4135 1816.1883 899.1921 2197.4249 941.5803 1097.0867 809.7110 1220.1334 312.1205 1750.6512
  • As of 2008/12, AMD Opteron 250 (pluton) beats them all (see also Johan De Gelas' benchmark page)
  • MacBook's Intel Core 2 Duo 2.20 GHz behaves surprisingly badly w.r.t. the lower-cadenced one on micron for modules 4, 5, 6 and 8. OS issue ?
  • Note: This benchmark does not take into account memory accesses and cache performance, thus it does not reflect real world performance of the CPU and must be considered as an assessment of the (kind of) theoretical floating point unit performance.

Detailed results

flops std output + `cat/proc/cpuinfo` or equivalent + `uname -a`

pcfarm/Debian GNU Linux Etch

FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
   1     -8.1208e-11      0.0155    901.8858
   2      1.4704e-15      0.0195    359.0649
   3     -3.8213e-15      0.0174    978.5660
   4      6.1151e-14      0.0165    907.2960
   5     -4.4419e-14      0.0298    972.0663
   6      7.7002e-15      0.0296    979.3746
   7     -6.6161e-13      0.0463    258.9085
   8      2.2789e-14      0.0317    946.4228

   Iterations      =  512000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =   452.7733
   MFLOPS(2)       =   516.5398
   MFLOPS(3)       =   781.0876
   MFLOPS(4)       =   955.7414


processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      : Intel(R) Pentium(R) 4 CPU 3.00GHz
stepping        : 3
cpu MHz         : 2992.621
cache size      : 2048 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 c
lflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc pni monitor ds_cpl est ci
d cx16 xtpr
bogomips        : 5989.07
clflush size    : 64
processor       : 1
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      : Intel(R) Pentium(R) 4 CPU 3.00GHz
stepping        : 3
cpu MHz         : 2992.621
cache size      : 2048 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 c
lflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc pni monitor ds_cpl est ci
d cx16 xtpr
bogomips        : 5984.99
clflush size    : 64

Linux b06u0009 2.6.22 #1 SMP Mon Oct 29 17:04:05 CET 2007 i686 GNU/Linux

cosinux/Debian GNU Linux Etch

FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1     -8.1208e-11      0.0145    965.6640
     2      1.4704e-15      0.0192    365.3225
     3     -3.8213e-15      0.0163   1039.8141
     4      6.1151e-14      0.0156    962.9143
     5     -4.4419e-14      0.0280   1035.2940
     6      7.7002e-15      0.0275   1055.4554
     7     -6.6161e-13      0.0436    275.1750
     8      2.2789e-14      0.0296   1013.4135

   Iterations      =  512000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =   463.6448
   MFLOPS(2)       =   550.0338
   MFLOPS(3)       =   833.6961
   MFLOPS(4)       =  1022.4030

processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      : Intel(R) Xeon(TM) CPU 3.20GHz
stepping        : 3
cpu MHz         : 3202.403
cache size      : 2048 KB
physical id     : 0
siblings        : 2
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 c
lflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm pni monitor ds_cpl cid
bogomips        : 6340.60

processor       : 3
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      : Intel(R) Xeon(TM) CPU 3.20GHz
stepping        : 3
cpu MHz         : 3202.403
cache size      : 2048 KB
physical id     : 3
siblings        : 2
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 c
lflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm pni monitor ds_cpl cid
bogomips        : 6389.76

Linux master 2.6.8 #2 SMP Wed Nov 16 10:29:26 CET 2005 i686 GNU/Linux

pluton/FreeBSD 7.0

  FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1      4.0146e-13      0.0100   1405.9542
     2     -1.4166e-13      0.0081    865.0993
     3      4.7184e-14      0.0088   1932.7696
     4     -1.2557e-13      0.0086   1739.9996
     5     -1.3800e-13      0.0168   1729.8708
     6      3.2380e-13      0.0153   1892.5986
     7     -8.4583e-11      0.0239    502.3997
     8      3.4867e-13      0.0165   1816.1883

   Iterations      =  512000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =  1055.7636
   MFLOPS(2)       =   980.6289
   MFLOPS(3)       =  1461.9797
   MFLOPS(4)       =  1847.4412

Vendor ID: "AuthenticAMD"; CPUID level 1

AMD-specific functions
Version 00000f5a:
Family: 15 Model: 5 []
Processor name string: AMD Opteron(tm) Processor 250

CPU: AMD Opteron(tm) Processor 250 (2392.55-MHz K8-class CPU)

FreeBSD pluton 7.0-RELEASE FreeBSD 7.0-RELEASE #0: Wed Jun 18 23:09:49 CEST 2008     alarcher@pluton:/usr/obj/usr/src/sys/PLUTON_ULE  amd64

micron/Win32

   FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1     4.0146e-013      0.0278    502.7353
     2    -1.4166e-013      0.0238    294.5431
     3     4.7184e-014      0.0202    841.5353
     4    -1.2557e-013      0.0175    857.7172
     5    -1.3800e-013      0.0318    912.5438
     6     3.2380e-013      0.0304    952.7109
     7    -8.4583e-011      0.0789    152.1809
     8     3.4867e-013      0.0334    899.1921

   Iterations      =  512000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =   374.0219
   MFLOPS(2)       =   342.9836
   MFLOPS(3)       =   608.4027
   MFLOPS(4)       =   896.6207

Processor 1 (ID = 0)
Number of cores		2 (max 2)
Number of threads	2 (max 2)
Name			Intel Mobile Core 2 Duo L7500
Codename		Merom
Specification		Intel(R) Core(TM)2 Duo CPU     L7500  @ 1.60GHz
Package			Socket P (478) (platform ID = 7h)
CPUID			6.F.B
Extended CPUID		6.F
Core Stepping		G0
Technology		65 nm
Core Speed		1596.8 MHz (8.0 x 199.6 MHz)
Rated Bus speed		798.4 MHz
Stock frequency		1600 MHz
Instructions sets	MMX, SSE, SSE2, SSE3, SSSE3, EM64T
L1 Data cache		2 x 32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache	2 x 32 KBytes, 8-way set associative, 64-byte line size
L2 cache		4096 KBytes, 16-way set associative, 64-byte line size
FID/VID Control		yes
FID range		6.0x - 8.0x
max VID			1.113 V
Features		, VT

Windows Version		Microsoft Windows Vista (6.0) Ultimate Edition  Service Pack 1 (Build 6001) 

Build Options		/Ox /Oi /GL /D "WIN32" /D "NDEBUG" /D "_CONSOLE" /D "_UNICODE" /D "UNICODE"
			/FD /MT /W4 /nologo /c /TC

This benchmark is dedicated to the memory of Pandore (2003 - 2007)

charon/Win32

  FLOPS C Program (Double Precision), V2.0 18 Dec 1992

  Module     Error        RunTime      MFLOPS
                           (usec)
    1     4.0146e-013      0.0184    762.0668
    2    -1.4166e-013      0.0159    441.1077
    3     4.7184e-014      0.0151   1123.0968
    4    -1.2557e-013      0.0141   1061.5066
    5    -1.3800e-013      0.0467    621.0993
    6     3.2380e-013      0.0424    684.6498
    7    -8.4583e-011      0.0613    195.8247
    8     3.4867e-013      0.0319    941.5803

  Iterations      =  512000000
  NullTime (usec) =     0.0000
  MFLOPS(1)       =   550.3669
  MFLOPS(2)       =   398.1093
  MFLOPS(3)       =   635.2573
  MFLOPS(4)       =   879.3432

Processor 1 (ID = 0)
Number of cores		1 (max 1)
Number of threads	1 (max 1)
Name			Intel Pentium M 750
Codename		Dothan
Specification		Intel(R) Pentium(R) M processor 1.86GHz
Package			Socket 479 mPGA (platform ID = 5h)
CPUID			6.D.8
Extended CPUID		6.D
Brand ID		22
Core Stepping		C0
Technology		90 nm
Core Speed		1862.1 MHz (14.0 x 133.0 MHz)
Rated Bus speed		532.0 MHz
Stock frequency		1866 MHz
Instructions sets	MMX, SSE, SSE2
L1 Data cache		32 KBytes, 8-way set associative, 64-byte line size
L1 Instruction cache	32 KBytes, 8-way set associative, 64-byte line size
L2 cache		2048 KBytes, 8-way set associative, 64-byte line size
FID/VID Control		yes
FID range		6.0x - 14.0x
max VID			1.356 V
Features		XD

Windows Version		Microsoft Windows XP Professional  Service Pack 3 (Build 2600) 

MacBook/MacOS X Leopard

   FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1      4.0146e-13      0.0137   1023.7240
     2     -1.4166e-13      0.0160    438.7116
     3      4.7184e-14      0.0097   1760.4868
     4     -1.2557e-13      0.0264    567.4052
     5     -1.3800e-13      0.0474    612.4561
     6      3.2380e-13      0.0350    827.7106
     7     -8.4583e-11      0.0427    280.9953
     8      3.4867e-13      0.0371    809.7110

   Iterations      =  512000000
   NullTime (usec) =     0.0005
   MFLOPS(1)       =   581.4247
   MFLOPS(2)       =   501.6895
   MFLOPS(3)       =   688.9703
   MFLOPS(4)       =   841.1967

    Hardware Overview:

      Model Name: MacBook
      Model Identifier: MacBook3,1
      Processor Name: Intel Core 2 Duo
      Processor Speed: 2.2 GHz
      Number Of Processors: 1
      Total Number Of Cores: 2
      L2 Cache: 4 MB
      Memory: 1 GB
      Bus Speed: 800 MHz

kant/Debian GNU Linux Etch

   FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1     -8.1208e-11      0.0186    754.1616
     2      1.4704e-15      0.0150    465.4254
     3     -3.8213e-15      0.0130   1311.5515
     4      6.1151e-14      0.0136   1103.3793
     5     -4.4419e-14      0.0273   1063.8487
     6      7.7002e-15      0.0239   1212.5997
     7     -6.6161e-13      0.0399    300.3919
     8      2.2789e-14      0.0246   1220.1334

   Iterations      =  512000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =   589.8248
   MFLOPS(2)       =   593.5499
   MFLOPS(3)       =   907.7888
   MFLOPS(4)       =  1212.3734


processor	: 0
vendor_id	: AuthenticAMD
cpu family	: 6
model		: 6
model name	: AMD Athlon(tm) XP 1900+
stepping	: 2
cpu MHz		: 1599.764
cache size	: 256 KB
fdiv_bug	: no
hlt_bug		: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 1
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mp mmxext 3dnowext 3dnow ts
bogomips	: 3202.09

Linux kant 2.6.18-6-486 #1 Sat May 24 09:50:27 UTC 2008 i686 GNU/Linux

zarathoustra/NetBSD 4.0

   FLOPS C Program (Double Precision), V2.0 18 Dec 1992

   Module     Error        RunTime      MFLOPS
                            (usec)
     1     -7.6739e-13      0.0922    151.8459
     2     -5.7021e-13      0.0513    136.5294
     3     -2.4314e-14      0.0822    206.7945
     4      6.8501e-14      0.0731    205.0616
     5     -1.6320e-14      0.1237    234.4722
     6      1.3961e-13      0.0921    314.7081
     7     -3.6152e-11      0.1493     80.3831
     8      9.0483e-15      0.0961    312.1205

   Iterations      =  256000000
   NullTime (usec) =     0.0000
   MFLOPS(1)       =   153.5907
   MFLOPS(2)       =   143.3250
   MFLOPS(3)       =   205.9856
   MFLOPS(4)       =   264.8262

uname -a from a Linux install :

processor       : 0
vendor_id       : GenuineTMx86
cpu family      : 6
model           : 4
model name      : Transmeta(tm) Crusoe(tm) Processor TM5800
stepping        : 3
cpu MHz         : 998.658
cache size      : 512 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr cx8 sep cmov mmx longrun lrti
bogomips        : 1985.74
 
Recent changes RSS feed Creative Commons License Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki