OGR-NG Client Benchmark Налаштування

[A1ex01]

вот попался в руки комп: Phenom 9550-2.2ГГц, windows xp-64

dnetc v2.9104-510-GTR-09042813 for Win32 (WindowsNT 5.2).
Using email address (distributed.net ID) 'укr.nеt'

[Jul 17 15:21:02 UTC] Automatic processor detection found 4 processors.
[Jul 17 15:21:02 UTC] Loading crunchers with work...
[Jul 17 15:21:03 UTC] Automatic processor type detection found
an AMD Phenom processor.
[Jul 17 15:21:03 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[Jul 17 15:21:03 UTC] OGR-NG #a: Loaded 27/26-16-3-22-49*
Packet was from a different user/core/client cpu/os/build.
[Jul 17 15:21:03 UTC] OGR-NG #b: Loaded 27/26-16-3-24-47*
[Jul 17 15:21:03 UTC] OGR-NG #c: Loaded 27/26-16-3-25-46*
[Jul 17 15:21:03 UTC] OGR-NG #d: Loaded 27/26-16-3-27-44*
[Jul 17 15:21:03 UTC] OGR-NG: 144 packets remain in buff-in.og2
[Jul 17 15:21:03 UTC] OGR-NG: 0 packets are in buff-out.og2
[Jul 17 15:21:03 UTC] 4 crunchers ('a'-'d') have been started.
[Jul 17 15:21:03 UTC] Paused... ('taskmgr.exe' active)
[Jul 17 15:21:28 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.34 [23,678,976 nodes/sec]
[Jul 17 15:21:35 UTC] RC5-72: using core #0 (SES 1-pipe).
[Jul 17 15:21:55 UTC] RC5-72: Benchmark for core #0 (SES 1-pipe)
0.00:00:16.53 [4,523,231 keys/sec]
[Jul 17 15:21:55 UTC] RC5-72: using core #1 (SES 2-pipe).
[Jul 17 15:22:14 UTC] RC5-72: Benchmark for core #1 (SES 2-pipe)
0.00:00:16.46 [7,603,057 keys/sec]
[Jul 17 15:22:14 UTC] RC5-72: using core #2 (DG 2-pipe).
[Jul 17 15:22:33 UTC] RC5-72: Benchmark for core #2 (DG 2-pipe)
0.00:00:16.82 [7,949,656 keys/sec]
[Jul 17 15:22:33 UTC] RC5-72: using core #3 (DG 3-pipe).
[Jul 17 15:22:52 UTC] RC5-72: Benchmark for core #3 (DG 3-pipe)
0.00:00:16.82 [7,949,656 keys/sec]
[Jul 17 15:22:52 UTC] RC5-72: using core #4 (DG 3-pipe alt).
[Jul 17 15:23:10 UTC] RC5-72: Benchmark for core #4 (DG 3-pipe alt)
0.00:00:16.32 [8,327,184 keys/sec]
[Jul 17 15:23:10 UTC] RC5-72: using core #5 (SS 2-pipe).
[Jul 17 15:23:29 UTC] RC5-72: Benchmark for core #5 (SS 2-pipe)
0.00:00:16.34 [8,327,184 keys/sec]
[Jul 17 15:23:29 UTC] RC5-72: using core #6 (GO 2-pipe).
[Jul 17 15:23:50 UTC] RC5-72: Benchmark for core #6 (GO 2-pipe)
0.00:00:17.42 [9,538,662 keys/sec]
[Jul 17 15:23:50 UTC] RC5-72: using core #7 (SGP 3-pipe).
[Jul 17 15:24:08 UTC] RC5-72: Benchmark for core #7 (SGP 3-pipe)
0.00:00:16.32 [8,196,096 keys/sec]
[Jul 17 15:24:08 UTC] RC5-72: using core #8 (MA 4-pipe).
[Jul 17 15:24:29 UTC] RC5-72: Benchmark for core #8 (MA 4-pipe)
0.00:00:17.51 [5,703,615 keys/sec]
[Jul 17 15:24:29 UTC] RC5-72: using core #9 (MMX 4-pipe).
[Jul 17 15:24:47 UTC] RC5-72: Benchmark for core #9 (MMX 4-pipe)
0.00:00:16.32 [4,235,280 keys/sec]
[Jul 17 15:24:47 UTC] RC5-72: using core #10 (GO 2-pipe alt).
[Jul 17 15:25:06 UTC] RC5-72: Benchmark for core #10 (GO 2-pipe alt)
0.00:00:16.82 [7,949,656 keys/sec]
[Jul 17 15:25:06 UTC] RC5-72 benchmark summary :
Default core : #6 (GO 2-pipe)
Fastest core : #6 (GO 2-pipe)
[Jul 17 15:25:13 UTC] OGR-NG: using core #0 (FLEGE 2.0).
[Jul 17 15:25:32 UTC] OGR-NG: Benchmark for core #0 (FLEGE 2.0)
0.00:00:16.39 [15,891,367 nodes/sec]
[Jul 17 15:25:32 UTC] OGR-NG: using core #1 (rt-asm-generic).
[Jul 17 15:25:51 UTC] OGR-NG: Benchmark for core #1 (rt-asm-generic)
0.00:00:16.34 [16,125,984 nodes/sec]
[Jul 17 15:25:51 UTC] OGR-NG: using core #2 (rt-asm-mmx).
[Jul 17 15:26:10 UTC] OGR-NG: Benchmark for core #2 (rt-asm-mmx)
0.00:00:16.32 [22,862,451 nodes/sec]
[Jul 17 15:26:10 UTC] OGR-NG: using core #3 (cj-asm-sse2).
[Jul 17 15:26:29 UTC] OGR-NG: Benchmark for core #3 (cj-asm-sse2)
0.00:00:16.54 [29,105,223 nodes/sec]
[Jul 17 15:26:29 UTC] OGR-NG: using core #4 (cj-asm-sse-p4).
[Jul 17 15:26:48 UTC] OGR-NG: Benchmark for core #4 (cj-asm-sse-p4)
0.00:00:16.36 [22,679,552 nodes/sec]
[Jul 17 15:26:48 UTC] OGR-NG: using core #5 (cj-asm-sse-k8).
[Jul 17 15:27:07 UTC] OGR-NG: Benchmark for core #5 (cj-asm-sse-k8)
0.00:00:16.34 [23,678,976 nodes/sec]
[Jul 17 15:27:07 UTC] OGR-NG benchmark summary :
Default core : #5 (cj-asm-sse-k8)
Fastest core : #3 (cj-asm-sse2)
[Jul 17 15:27:07 UTC] Core #3 is significantly faster than the default core.
Please file a bug report along with the output of
-cpuinfo.

и клиент 64

dnetc v2.9104-510-GTR-09042813 for Win64 (WindowsNT 5.2).
Using email address (distributed.net ID) 'укr.net'

[Jul 17 15:28:56 UTC] Automatic processor detection found 4 processors.
[Jul 17 15:28:56 UTC] Loading crunchers with work...
[Jul 17 15:28:56 UTC] Automatic processor type detection found
an AMD Phenom processor.
[Jul 17 15:28:56 UTC] RC5-72: using core #1 (KBE-64 3-pipe).
[Jul 17 15:28:56 UTC] RC5-72 #a: Loaded random 53:9F43AC04:00000000:1*2^32
[Jul 17 15:28:56 UTC] RC5-72 #b: Loaded random 53:98C4E680:00000000:1*2^32
[Jul 17 15:28:56 UTC] RC5-72 #c: Loaded random 53:91DEC5B3:00000000:1*2^32
[Jul 17 15:28:56 UTC] RC5-72 #d: Loaded random 53:9224BE43:00000000:1*2^32
[Jul 17 15:28:56 UTC] RC5-72: 0 packets remain in buff-in.r72
[Jul 17 15:28:56 UTC] RC5-72: 0 packets are in buff-out.r72
[Jul 17 15:28:56 UTC] 4 crunchers ('a'-'d') have been started.
[Jul 17 15:28:56 UTC] Paused... ('taskmgr.exe' active)
[Jul 17 15:29:21 UTC] OGR-NG: Benchmark for core #0 (FLEGE-64 2.0)
0.00:00:16.51 [27,630,953 nodes/sec]
[Jul 17 15:29:29 UTC] RC5-72: using core #0 (SNJL 3-pipe).
[Jul 17 15:29:47 UTC] RC5-72: Benchmark for core #0 (SNJL 3-pipe)
0.00:00:16.32 [8,327,184 keys/sec]
[Jul 17 15:29:47 UTC] RC5-72: using core #1 (KBE-64 3-pipe).
[Jul 17 15:30:08 UTC] RC5-72: Benchmark for core #1 (KBE-64 3-pipe)
0.00:00:16.95 [8,817,747 keys/sec]
[Jul 17 15:30:08 UTC] RC5-72: using core #2 (ANSI 4-pipe).
[Jul 17 15:30:28 UTC] RC5-72: Benchmark for core #2 (ANSI 4-pipe)
0.00:00:17.54 [6,171,759 keys/sec]
[Jul 17 15:30:28 UTC] RC5-72: using core #3 (ANSI 2-pipe).
[Jul 17 15:30:48 UTC] RC5-72: Benchmark for core #3 (ANSI 2-pipe)
0.00:00:17.01 [5,874,736 keys/sec]
[Jul 17 15:30:48 UTC] RC5-72: using core #4 (ANSI 1-pipe).
[Jul 17 15:31:07 UTC] RC5-72: Benchmark for core #4 (ANSI 1-pipe)
0.00:00:16.71 [3,749,530 keys/sec]
[Jul 17 15:31:07 UTC] RC5-72 benchmark summary :
Default core : #1 (KBE-64 3-pipe)
Fastest core : #1 (KBE-64 3-pipe)
[Jul 17 15:32:10 UTC] *Break* Shutting down...

sse2 таки рулит

[A1ex01]

xeon v3@2400, 12core/24HT, W8.1x64

dnetc v2.9112-521-GTR-16020313 for Win32 (WindowsNT 6.2)
[Sep 22 06:33:11 UTC] OGR-NG: Benchmark for core #6 (cj-asm-sse4.1)
0.00:00:17.17 [47,636,927 nodes/sec]
[Sep 22 06:50:08 UTC] RC5-72: Benchmark for core #12 (YK/RT AVX2)
0.00:00:17.26 [21,339,369 keys/sec]

если при бенчах работает одно ядро, то турбо дает 2.6ГГц, т.е. это цифры @2.6
@2.4GHz 24 потока дают ~671 Mnodes/s, а должно быть 12*47,6(@2.6)=571,2. HT добавил больше сотки?