zarx on 13 December 2012
You people do realise that IBM make CPUs based on 10 different architectures that span 2 different instruction sets. All of which vary wildly in terms of performance...
And AMD have at lest 3 different architectures as well, all that while all based on x86 vary wildly in terms of performance.
The comparison is flawed from the beginning and everyone who answers ether way is wrong. Because there are so many variables.
But here are some benchmarke for you guys to look at
GMP repo [2012-03-22] GMPbench 0.2 results
| CPU |
| A B I | Compiler/Compilation flags |
|
| GMP bench | Score/ GHz | Possib. Score/ GHz |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Opteron/Athlon64 K10 6MB L3 | 3200 | 64 | "gcc 4.2.1" -O2 -m64 -mtune=k8 | 45414 | 42779 | 7679 | 5119 | 6216 | 45.6 | 3500 | 1094 | 1300 | ||||||
| Core i5 2500 (Sandy Bridge) | 3300 | 64 | "gcc 4.6.2" -O2 -m64 -march=corei7 | 41412 | 41741 | 7628 | 5292 | 5213 | 41.8 | 3222 | 976 | |||||||
| POWER7 | 3550 | 64 | "gcc 4.6.1" -O3 -mtune=power7 | 31225 | 30122 | 5080 | 3657 | 3741 | 34.8 | 2398 | 675 | |||||||
| AMD FX (Bulldozer) | 3600 | 64 | "gcc 4.2.1" -O2 -m64 | 29983 | 30588 | 5379 | 3559 | 3943 | 31.9 | 2373 | 659 | |||||||
| Core i7 920 (Nehalem) | 2667 | 64 | "gcc 4.2.1" -O2 -m64 | 26436 | 25056 | 4439 | 2873 | 3330 | 27.4 | 2006 | 752 | 900 | ||||||
| AMD Bobcat | 1600 | 64 | "gcc 4.2.1" -O2 | 12193 | 11470 | 2533 | 1595 | 1709 | 12.5 | 979 | 612 | |||||||
| POWER6 | 3500 | 64 | "xlc" -O2 -qarch=pwr6 | 10561 | 11401 | 2110 | 1292 | 1133 | 13.0 | 841 | 240 | |||||||
| Arm Cortex-A15 | 1700 | 32 | "gcc 4.6.3" -O2 -fomit-frame-pointer | 7352 | 7453 | 1467 | 980 | 936 | 8.78 | 605 | 356 | |||||||
| Athlon32 | 1826 | 32 | "gcc 4.2.1" -O2 -fomit-frame-pointer | 5280 | 6150 | 1299 | 866 | 577 | 7.19 | 458 | 251 | |||||||
| Intel Atom 330 | 1600 | 64 | "gcc 4.4.1" -O2 -m64 | 4592 | 5223 | 986 | 589 | 500 | 5.66 | 374 | 234 | |||||||
| Arm Cortex-A9 | 1000 | 32 | "gcc 4.4.5" -O2 -fomit-frame-pointer | 3715 | 3575 | 773 | 514 | 495 | 3.38 | 288 | 288 | |||||||
| z990 | 1200 | 64 | "gcc 4.4.5" -O2 -fomit-frame-pointer | 3037 | 3838 | 747 | 438 | 317 | 4.01 | 259 | 216 | |||||||
| Alpha 21164 | 600 | 64 | "gcc 3.4.6" -O2 | 1415 | 1789 | 383 | 217 | 167 | 1.85 | 126 | 210 | |
One showing SPARC CPUs crushing all in the server space (because IBM only make specialist embedded CPUs outside of servers these days that is the only data you will find)
| TPC-H @3000GB, Non-Clustered Systems | ||||||
|---|---|---|---|---|---|---|
| System Processor P/C/T – Memory | Composite (QphH) | $/perf ($/QphH) | Power (QppH) | Throughput (QthH) | Database | Available |
| SPARC Enterprise M9000 3.0 GHz SPARC64 VII+ 64/256/256 – 1024 GB |
386,478.3 | $18.19 | 316,835.8 | 471,428.6 | Oracle 11g R2 | 09/22/11 |
| SPARC T4-4 3.0 GHz SPARC T4 4/32/256 – 1024 GB |
205,792.0 | $4.10 | 190,325.1 | 222,515.9 | Oracle 11g R2 | 05/31/12 |
| SPARC Enterprise M9000 2.88 GHz SPARC64 VII 32/128/256 – 512 GB |
198,907.5 | $15.27 | 182,350.7 | 216,967.7 | Oracle 11g R2 | 12/09/10 |
| IBM Power 780 4.1 GHz POWER7 8/32/128 – 1024 GB |
192,001.1 | $6.37 | 210,368.4 | 175,237.4 | Sybase 15.4 | 11/30/11 |
| HP ProLiant DL980 G7 2.27 GHz Intel Xeon X7560 8/64/128 – 512 GB |
162,601.7 | $2.68 | 185,297.7 | 142,685.6 | SQL Server 2008 | 10/13/10 |
@TheVoxelman on twitter
