| CPU2006 license: | 13 | Test date: | Feb-2008 |
|---|---|---|---|
| Test sponsor: | Intel Corporation | Hardware Availability: | Mar-2008 |
| Tested by: | Intel Corporation | Software Availability: | Nov-2007 |
| Hardware | |
|---|---|
| CPU Name: | Intel Core 2 Quad Q9550 |
| CPU Characteristics: | 2.83 GHz, 1333 FSB |
| CPU MHz: | 2833 |
| FPU: | Integrated |
| CPU(s) enabled: | 4 cores, 1 chip, 4 cores/chip |
| CPU(s) orderable: | 1 chip |
| Primary Cache: | 32 KB I + 32 KB D on chip per core |
| Secondary Cache: | 12 MB I+D on chip per chip, 6 MB shared / 2 cores |
| L3 Cache: | None |
| Other Cache: | None |
| Memory: | 4 GB (4x1GB Micron DDR2-800 CL5) |
| Disk Subsystem: | Seagate 320GB NCQ SATA, 16MB cache, 7200 RPM |
| Other Hardware: | None |
| Software | |
|---|---|
| Operating System: | Windows Vista Ultimate (64-bit) |
| Compiler: | Intel C++ Compiler for IA32 version 10.1 Build 20070913 Package ID: w_cc_p_10.1.011 Intel Fortran Compiler for IA32 version 10.1 Build 20070913 Package ID: w_fc_p_10.1.011 Microsoft Visual Studio 2005 SP1 (for libraries) |
| Auto Parallel: | No |
| File System: | NTFS |
| System State: | Default |
| Base Pointers: | 32-bit |
| Peak Pointers: | 32-bit |
| Other Software: | SmartHeap Library Version 8.1 from http://www.microquill.com/ |
Results Table
| Benchmark | Base | Peak | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | Copies | Seconds | Ratio | Seconds | Ratio | Seconds | Ratio | |
| Results appear in the order in which they were run. Bold underlined text indicates a median measurement. | ||||||||||||||
| 410.bwaves | 4 | 1780 | 30.5 | 1772 | 30.7 | 1765 | 30.8 | 4 | 1780 | 30.5 | 1772 | 30.7 | 1765 | 30.8 |
| 416.gamess | 4 | 1132 | 69.2 | 1131 | 69.3 | 1131 | 69.3 | 4 | 1140 | 68.7 | 1139 | 68.8 | 1139 | 68.7 |
| 433.milc | 4 | 1722 | 21.3 | 1723 | 21.3 | 1723 | 21.3 | 4 | 1711 | 21.5 | 1710 | 21.5 | 1710 | 21.5 |
| 434.zeusmp | 4 | 691 | 52.7 | 693 | 52.5 | 679 | 53.6 | 4 | 645 | 56.4 | 636 | 57.3 | 650 | 56.0 |
| 435.gromacs | 4 | 426 | 67.0 | 426 | 67.0 | 426 | 67.0 | 4 | 430 | 66.5 | 430 | 66.4 | 430 | 66.5 |
| 436.cactusADM | 4 | 829 | 57.7 | 828 | 57.7 | 826 | 57.9 | 4 | 764 | 62.6 | 763 | 62.7 | 762 | 62.7 |
| 437.leslie3d | 4 | 1663 | 22.6 | 1662 | 22.6 | 1663 | 22.6 | 4 | 1684 | 22.3 | 1683 | 22.3 | 1685 | 22.3 |
| 444.namd | 4 | 551 | 58.2 | 551 | 58.2 | 551 | 58.2 | 4 | 550 | 58.3 | 550 | 58.4 | 550 | 58.3 |
| 447.dealII | 4 | 702 | 65.2 | 696 | 65.8 | 699 | 65.5 | 4 | 696 | 65.8 | 695 | 65.8 | 696 | 65.7 |
| 450.soplex | 4 | 1216 | 27.4 | 1205 | 27.7 | 1210 | 27.6 | 4 | 1202 | 27.8 | 1221 | 27.3 | 1208 | 27.6 |
| 453.povray | 4 | 258 | 82.6 | 258 | 82.5 | 258 | 82.6 | 4 | 252 | 84.3 | 253 | 84.2 | 252 | 84.4 |
| 454.calculix | 4 | 648 | 50.9 | 648 | 50.9 | 648 | 50.9 | 4 | 461 | 71.5 | 462 | 71.4 | 465 | 71.0 |
| 459.GemsFDTD | 4 | 2139 | 19.8 | 2131 | 19.9 | 2136 | 19.9 | 4 | 2084 | 20.4 | 2086 | 20.3 | 2081 | 20.4 |
| 465.tonto | 4 | 659 | 59.7 | 660 | 59.6 | 663 | 59.4 | 4 | 639 | 61.6 | 640 | 61.5 | 638 | 61.7 |
| 470.lbm | 4 | 2689 | 20.4 | 2689 | 20.4 | 2689 | 20.4 | 1 | 567 | 24.2 | 567 | 24.2 | 567 | 24.2 |
| 481.wrf | 4 | 1067 | 41.9 | 1068 | 41.8 | 1067 | 41.9 | 4 | 1067 | 41.9 | 1068 | 41.8 | 1067 | 41.9 |
| 482.sphinx3 | 4 | 1424 | 54.7 | 1428 | 54.6 | 1425 | 54.7 | 4 | 1416 | 55.1 | 1414 | 55.2 | 1418 | 55.0 |
General Notes
Tested systems can be used with Shin-G ATX case, Antec NeoPower 480W power supply Product description located as of 03/2008: http://www.intel.com/products/motherboard/DQ35JO/index.htm The system bus runs at 1333 MHz System was configured with Asus EN8800GTX discrete graphics card Binaries were built on Windows Vista Ultimate (32-bit) The following VS 2005 SP1 updates were applied: KB926601 and KB932232 The start command with the /affinity switch was used to bind processes to cores
Base Compiler Invocation
C benchmarks:
| icl -Qvc8 -Qc99 |
C++ benchmarks:
| icl -Qvc8 |
Fortran benchmarks:
| ifort |
Benchmarks using both Fortran and C:
| icl -Qvc8 -Qc99 ifort |
Base Portability Flags
| 436.cactusADM: | -Qlowercase /assume:underscore |
| 444.namd: | -TP |
| 447.dealII: | -DDEAL_II_MEMBER_VAR_SPECIALIZATION_BUG |
| 453.povray: | -DSPEC_CPU_WINDOWS_ICL |
| 454.calculix: | -DSPEC_CPU_NOZMODIFIER -Qlowercase |
| 481.wrf: | -DSPEC_CPU_WINDOWS_ICL |
Base Optimization Flags
C benchmarks:
| -fast /F1000000000 |
C++ benchmarks:
| -fast -Qcxx_features /F1000000000 shlw32m.lib -link /FORCE:MULTIPLE |
Fortran benchmarks:
| -fast /F1000000000 |
Benchmarks using both Fortran and C:
| -fast /F1000000000 |
Peak Compiler Invocation
C benchmarks:
| icl -Qvc8 -Qc99 |
C++ benchmarks:
| icl -Qvc8 |
Fortran benchmarks:
| ifort |
Benchmarks using both Fortran and C:
| icl -Qvc8 -Qc99 ifort |
Peak Portability Flags
| 436.cactusADM: | -Qlowercase /assume:underscore |
| 444.namd: | -TP |
| 447.dealII: | -DDEAL_II_MEMBER_VAR_SPECIALIZATION_BUG |
| 453.povray: | -DSPEC_CPU_WINDOWS_ICL |
| 454.calculix: | -DSPEC_CPU_NOZMODIFIER -Qlowercase |
| 481.wrf: | -DSPEC_CPU_WINDOWS_ICL |
Peak Optimization Flags
C benchmarks:
| 433.milc: | -fast -Qunroll2 -Oa /F1000000000 |
| 470.lbm: | -fast -Qunroll2 -Qscalar-rep- -Qprefetch /F1000000000 |
| 482.sphinx3: | -fast -Qunroll2 /F1000000000 |
C++ benchmarks:
Fortran benchmarks:
| 410.bwaves: | basepeak = yes |
| 416.gamess: | -fast -Qunroll2 -Ob0 -Qansi-alias -Qscalar-rep- /F1000000000 |
| 434.zeusmp: | -QxT -O2 -Qprec-div- -Qunroll0 -Qscalar-rep- /F1000000000 |
| 437.leslie3d: | -fast -Qprefetch /F1000000000 |
| 459.GemsFDTD: | -fast -Qunroll2 -Ob0 -Qprefetch /F1000000000 |
| 465.tonto: | -fast -Qunroll4 -Qauto /F1000000000 |
Benchmarks using both Fortran and C:
| 435.gromacs: | -fast -Oa -Qprefetch /F1000000000 |
| 436.cactusADM: | -fast -Qunroll2 -Qprefetch /F1000000000 |
| 454.calculix: | -fast -Qunroll-aggressive /F1000000000 |
| 481.wrf: | basepeak = yes |