Here's a little teaser....

[KTE]

I can get you a few 8347's

Strangely enough out of the 9 vendors he did contact the 8347 was the exact CPU 1 of them said he might be able to get hold of late October. Maybe it was you he contacted afterall (jk)

Thanks Ste but he's far up the ladder from me at the corp and the best I get to speak to is his secretary. They order batch servers directly from the like of HP, Sun, Dell and other server builders and never go to typical retail outlets for single CPU buys. Its gear for a research lab at a government funded hospital BTW, so the order would be quite large.

[s7e9h3n]

Strangely enough out of the 9 vendors he did contact the 8347 was the exact CPU 1 of them said he might be able to get hold of late October. Maybe it was you he contacted afterall (jk)

Thanks Ste but he's far up the ladder from me at the corp and the best I get to speak to is his secretary. They order batch servers directly from the like of HP, Sun, Dell and other server builders and never go to typical retail outlets for single CPU buys. Its gear for a research lab at a government funded hospital BTW, so the order would be quite large.

If you caught me a couple weeks ago, I could have supplied him as many 8350's as he needed

[gallag]

I never claimed to have seen any record breaking fast k10's at all. I said that from benchmarks I've been priveledged (lucky enough) to see, that k10 cpu's will run SPi 1M in less than 26seconds...and when OC'd will manage to bring that down to 17seconds. I never said which K10 cpu's so I'm sorry if any of you managed to slide an exact model number/revision/stepping in there ...I certainly didn't mean the first one's out the factory with no bios.
What are you guys on......10yr old child pills?

As for now, we have around 6 users on XS with a k10...all running bugged to the max in server boards/AM2 boards with ECC mem/no oc not to mention most guys have trouble trying to get theirs to boot and run properly.......yet you guys already seem to decide these are the final product????

Come on....

Where did you witness these k10s running? And how come they were not also bugged? sorry if this sounds like i am interrogating you, I am genuinely interested.

[JohannesRS]

Btw: Soldner, didn't intend to say that you lied. Thats why said it "in between these signs"...

[BeardyMan]

I never claimed to have seen any record breaking fast k10's at all. I said that from benchmarks I've been priveledged (lucky enough) to see, that k10 cpu's will run SPi 1M in less than 26seconds...and when OC'd will manage to bring that down to 17seconds. I never said which K10 cpu's so I'm sorry if any of you managed to slide an exact model number/revision/stepping in there ...I certainly didn't mean the first one's out the factory with no bios.
What are you guys on......10yr old child pills?

As for now, we have around 6 users on XS with a k10...all running bugged to the max in server boards/AM2 boards with ECC mem/no oc not to mention most guys have trouble trying to get theirs to boot and run properly.......yet you guys already seem to decide these are the final product????

Come on....

a difference of nearly 20secs is just to much to coverup with faster mem.

[zpdixon]

I just do the openssl. But i dont think it run well : only one cpu run. It should run on Linux. Please let me know how?

type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
md2               1449.00k     3054.78k     4214.85k     4665.85k     4804.30k
mdc2                 0.00         0.00         0.00         0.00         0.00
md4              15350.40k    54126.60k   157606.54k   298261.62k   405246.76k
md5              14174.73k    49422.88k   141059.09k   262657.00k   312424.88k
hmac(md5)        21801.33k    71048.50k   180837.68k   291207.91k   249707.40k
sha1             14510.02k    47249.78k   119463.93k   194800.77k   239247.29k
rmd160           11639.53k    34359.30k    76082.83k   109708.79k   125589.71k
rc4             224624.66k   256885.87k   263172.02k   267750.02k   269124.41k
des cbc          53687.09k    55342.95k    55489.39k    55784.59k    55924.05k
des ede3         19719.34k    19884.10k    19994.27k    20051.53k    19993.81k
idea cbc         36091.68k    38692.84k    39475.80k    39765.86k    39770.57k
rc2 cbc          20258.67k    20911.40k    21074.26k    21115.37k    21136.65k
rc5-32/12 cbc        0.00         0.00         0.00         0.00         0.00
blowfish cbc     81029.78k    85576.21k    86592.08k    86928.58k    86951.11k
cast cbc         75352.42k    79531.72k    80737.32k    81029.78k    81029.78k
aes-128 cbc      52766.84k    56660.64k    57723.09k    58032.57k    58042.61k
aes-192 cbc      45497.53k    49483.01k    50526.17k    50770.82k    50770.82k
aes-256 cbc      40900.09k    43919.41k    44644.00k    44924.93k    44930.95k
camellia-128 cbc        0.00         0.00         0.00         0.00         0.0
camellia-192 cbc        0.00         0.00         0.00         0.00         0.0
camellia-256 cbc        0.00         0.00         0.00         0.00         0.0
sha256            9979.01k    24648.36k    45642.97k    57999.97k    58674.42k
sha512            3135.02k    12525.45k    19688.10k    27898.67k    31767.51k
                  sign    verify    sign/s verify/s
rsa  512 bits 0.000535s 0.000044s   1869.0  22676.5
rsa 1024 bits 0.002409s 0.000118s    415.1   8472.5
rsa 2048 bits 0.013613s 0.000386s     73.5   2589.7
rsa 4096 bits 0.087500s 0.001354s     11.4    738.5
                  sign    verify    sign/s verify/s
dsa  512 bits 0.000397s 0.000478s   2517.8   2090.3
dsa 1024 bits 0.001103s 0.001329s    906.8    752.3
dsa 2048 bits 0.003602s 0.004389s    277.6    227.8

[edit: I updated my percentages after learning that linhvndiy was running his 2347's at 1950 MHz to obtain the above scores]

Thanks linhvndiy ! These scores are very interesting.

I have compared them with the scores of K8 at the same speed to do a clock-for-clock comparison between K8 and K10. Technically I ran openssl on a dual Opteron 280 (2.4 GHz), but I scaled its scores down to simulate a 1950 MHz K8 (all of the openssl speed tests scale linearly with the clock frequency, so that's a pretty good estimation, see the end of this post for my K8 results.) For the tests using different buffer sizes (16 to 8192 bytes), I only took into account the three larger buffer sizes (256, 1024 and 8192 bytes), because the overhead of the OpenSSL API with the 2 smaller buffer sizes (16 and 64 bytes) is too high and cripples the results on both K10 and K8 which makes any sort of direct comparison difficult.

So here is how K10 fares against K8 in the most popular encryption/hashing algorithms:

o md4: K10 is between 3% and 6% faster than K8 (+6%, +4%, +3% for the three different buffer sizes)
o md5: the throughput varies too much between the three buffer sizes (the K10 scores vary by +6%, +5%, -8% compared to K8), are you sure your machine was idle ?
o hmac(md5): the throughput varies too much here too (+7%, +5%, -18%)
o sha1: K10 seems to have a negligible advantage over K8 (+7%, +5%, and +4%)
o rc4: K10 is as fast as K8 (within 1% of each other)
o blowfish: K10 is consistently 5% faster than K8
o aes-128: K10 is between 16% and 17% faster than K8 with the three different buffer sizes (for example: 8192-byte test: 58042 kB/s vs. 49500 kB/s)
o aes-192: K10 is exactly 18% faster than K8 no matter what the buffer size is
o aes-256: K10 is also 18% faster
o sha256: the throughput varies too much here (+11%, +11%, +3%), weird
o sha512: K10 is between 4% and 5% faster than K8
o rsa 1024-bit: K10 is +12% faster than K8 on sign operations (415 vs. 370 sign/s)
o rsa 1024-bit: K10 is +10% faster than K8 on verify operations (8473 vs. 7730 verify/s)
o dsa 1024-bit: K10 is +9% faster than K8 (907 vs. 830 sign/s)
o dsa 1024-bit K10 is +9% faster than K8 (752 vs. 690 verify/s)

Overall, K10 is, clock-for-clock, 0% to 18% faster than K8 on these 32-bit (100% ALU) OpenSSL speed tests. (It would have been interesting if the guys who ported OpenSSL to Windows enabled the SSE2 assembly implementation of sha512...)

linhvndiy, you said you wanted to benchmark the 8 cores at the same time. This is possible with "openssl speed -multi 8" under Linux/*BSD/Solaris... But you should know that these tests all scale linearly with the number of cores and the frequency clock (they all fit in the L2 cache), so just multiplying your scores by 8 gives a very precise estimation.

Now, if I can have one more wish ( ) I would ask you to run the same benchmark under 64-bit Linux/*BSD/Solaris. The RSA scores would jump by about x3 (the BN lib just loves 64-bit archs), and the RC4 and MD5 throughput would increase by 15-30% (at least that's what is observed with K8). Running openssl speed in 64-bit mode (and with -multi 8) is how the guys at http://www.tecchannel.de/server/proz...28/index9.html obtained their excellent RSA scores.

If you don't know what 64-bit distro to run, I recommend the 64-bit Ubuntu 7.10 (release candidate), the OpenSSL version they distribute is very recent (package named "openssl", version 0.9.8e).

- Z

"openssl speed" using the 32-bit Windows port of OpenSSL 0.9.8e running on:
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 33
model name      : Dual Core AMD Opteron(tm) Processor 280
stepping        : 2
cpu MHz         : 2405.476
cache size      : 1024 KB

OpenSSL 0.9.8e 23 Feb 2007
built on: Wed Feb 28 01:35:20 2007
options:bn(64,32) md2(int) rc4(idx,int) des(idx,cisc,4,long) aes(partial) idea(int) blowfish(idx)
compiler: cl  /MD /Ox /O2 /Ob2 /W3 /WX /Gs0 /GF /Gy /nologo -DOPENSSL_SYSNAME_WIN32 -DWIN32_LEAN_AND_MEAN -DL_ENDIAN -DDSO_WIN32 -D_CRT_SECURE_NO_DEPRECATE -D_CRT_NONSTDC_NO_DEPRECATE -DBN_ASM -DMD5_ASM -DSHA1_ASM -DRMD160_ASM -DOPENSSL_USE_APPLINK -I. /Fdout32dll -DOPENSSL_NO_CAMELLIA -DOPENSSL_NO_RC5 -DOPENSSL_NO_MDC2 -DOPENSSL_NO_KRB5 -DOPENSSL_NO_DYNAMIC_ENGINE
available timing options: TIMEB HZ=1000
timing function used: ftime
The 'numbers' are in 1000s of bytes per second processed.
type             16 bytes     64 bytes    256 bytes   1024 bytes   8192 bytes
md2               1597.75k     3568.29k     4965.69k     5497.35k     5674.21k
mdc2                 0.00         0.00         0.00         0.00         0.00
md4              17908.11k    62660.00k   183608.38k   352740.42k   485592.36k
md5              16226.33k    56910.50k   163840.00k   308830.48k   417473.49k
hmac(md5)        24636.15k    81670.76k   208121.77k   341955.99k   425412.77k
sha1             16008.79k    52702.61k   137180.83k   228728.23k   283938.50k
rmd160           13354.47k    39887.58k    89472.52k   130245.25k   150081.32k
rc4             256140.70k   306881.58k   320910.79k   326595.60k   327808.05k
des cbc          60699.04k    63226.74k    63937.56k    64071.86k    64305.16k
des ede3         22338.35k    22840.12k    22980.88k    23171.21k    23027.67k
idea cbc         43123.55k    46596.91k    47527.52k    47798.34k    47798.34k
rc2 cbc          23512.32k    24422.76k    24637.96k    24690.53k    24692.35k
rc5-32/12 cbc        0.00         0.00         0.00         0.00         0.00
blowfish cbc     93077.48k   100013.21k   101526.27k   102144.39k   102144.39k
cast cbc         87267.70k    92820.01k    94121.83k    94626.15k    94920.60k
aes-128 cbc      57367.81k    59726.65k    61030.25k    61030.25k    60919.45k
aes-192 cbc      50541.39k    51949.89k    52675.72k    52849.95k    52849.95k
aes-256 cbc      44792.99k    46154.65k    46661.70k    46863.73k    46929.28k
camellia-128 cbc        0.00         0.00         0.00         0.00         0.0
camellia-192 cbc        0.00         0.00         0.00         0.00         0.0
camellia-256 cbc        0.00         0.00         0.00         0.00         0.0
sha256           10971.24k    27588.43k    50805.41k    64589.86k    70135.20k
sha512            3692.13k    14766.89k    23183.76k    32873.14k    37448.10k
                  sign    verify    sign/s verify/s
rsa  512 bits 0.000492s 0.000041s   2032.8  24127.0
rsa 1024 bits 0.002308s 0.000108s    455.2   9512.6
rsa 2048 bits 0.012288s 0.000333s     83.4   3055.7
rsa 4096 bits 0.077100s 0.001137s     13.0    879.7
                  sign    verify    sign/s verify/s
dsa  512 bits 0.000361s 0.000429s   2769.9   2331.6
dsa 1024 bits 0.000980s 0.001177s   1020.4    849.9
dsa 2048 bits 0.003127s 0.003701s    319.8    270.2

[linhvndiy]

I did test at 1950mhz. And i feel this 32bit ssl is not the test for barcelona. I think barce will perform real in Linux and multi threads. But i have to research how to test under Linux.

[zpdixon]

I did test at 1950mhz. And i feel this 32bit ssl is not the test for barcelona. I think barce will perform real in Linux and multi threads. But i have to research how to test under Linux.

Given that the OpenSSL assembly code is absolutely not optimized for the K10 architecture (yet) and that it is exclusively using the ALU unit which is the one that changed the less between K8 & K10 (compared to, say, the FPU and SSE units), I think that 5%-18% of perf improvements across the range of algorithms is pretty good (I'll edit my post with updated percentage numbers later now that I know the clock was 1950 MHz).

Doing 64-bit OpenSSL speed tests is very easy. Just follow my link to download the Ubuntu 7.10 64-bit DVD image. Burn it. Install it. Then in a terminal: "sudo apt-get install openssl", type in your user passwd, then "openssl speed".

- Z

[Dresdenboy]

Doing 64-bit OpenSSL speed tests is very easy. Just follow my link to download the Ubuntu 7.10 64-bit DVD image. Burn it. Install it. Then in a terminal: "sudo apt-get install openssl", type in your user passwd, then "openssl speed".

- Z

I hope, there won't be K10 related problems while running Linux. Might sudo apt-get also work in a live system? Would make things easier I suppose.

However, Prime95 benchmarking won't cause such troubles

[linhvndiy]

i will have supermicro board fully support barcelona today. I will try to do some bench tonight.
And the prime test too.

[linhvndiy]

This is prime test with supermicro mainboard, cpu df at 1900mhz. I think this main run fastest, mem controller run at 1600mhz instead 1400mhz on tyan.
[Tue Oct 16 20:24:36 2007]
Compare your results to other computers at http://www.mersenne.org/bench.htm
Quad-Core AMD Opteron(tm) Processor 2347
CPU speed: 1899.93 MHz, 8 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 48
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 28.360 ms.
Best time for 896K FFT length: 33.302 ms.
Best time for 1024K FFT length: 37.527 ms.
Best time for 1280K FFT length: 47.124 ms.
Best time for 1536K FFT length: 59.104 ms.
Best time for 1792K FFT length: 68.229 ms.
Best time for 2048K FFT length: 78.122 ms.
Best time for 2560K FFT length: 103.002 ms.
Best time for 3072K FFT length: 128.072 ms.
Best time for 3584K FFT length: 151.523 ms.
Best time for 4096K FFT length: 171.648 ms.
Best time for 5120K FFT length: 221.236 ms.
Best time for 6144K FFT length: 270.341 ms.
[Wed Oct 17 02:45:58 2007]
Compare your results to other computers at http://www.mersenne.org/bench.htm
Quad-Core AMD Opteron(tm) Processor 2347
CPU speed: 1962.65 MHz, 8 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 48
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 28.068 ms.
Best time for 896K FFT length: 33.749 ms.
Best time for 1024K FFT length: 36.768 ms.
Best time for 1280K FFT length: 41.592 ms.
Best time for 1536K FFT length: 53.454 ms.
Best time for 1792K FFT length: 69.192 ms.
Best time for 2048K FFT length: 78.336 ms.
Best time for 2560K FFT length: 104.301 ms.
Best time for 3072K FFT length: 127.017 ms.
Best time for 3584K FFT length: 150.510 ms.
Best time for 4096K FFT length: 170.562 ms.
Best time for 5120K FFT length: 220.077 ms.
Best time for 6144K FFT length: 269.423 ms.
Best time for 7168K FFT length: 326.853 ms.
Best time for 8192K FFT length: 373.289 ms.
Timing FFTs using 2 threads.
Best time for 768K FFT length: 16.063 ms.
Best time for 896K FFT length: 23.582 ms.
Best time for 1024K FFT length: 26.356 ms.
Best time for 1280K FFT length: 31.061 ms.
Best time for 1536K FFT length: 37.191 ms.
Best time for 1792K FFT length: 43.552 ms.
Best time for 2048K FFT length: 49.355 ms.
Best time for 2560K FFT length: 66.939 ms.
Best time for 3072K FFT length: 79.434 ms.
Best time for 3584K FFT length: 92.848 ms.
Best time for 4096K FFT length: 104.647 ms.
Best time for 5120K FFT length: 130.025 ms.
[Wed Oct 17 02:50:18 2007]
Compare your results to other computers at http://www.mersenne.org/bench.htm
Quad-Core AMD Opteron(tm) Processor 2347
CPU speed: 1909.87 MHz, 8 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 512 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 48
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 27.935 ms.
Best time for 896K FFT length: 33.206 ms.
Best time for 1024K FFT length: 37.916 ms.
Best time for 1280K FFT length: 46.703 ms.
Best time for 1536K FFT length: 58.223 ms.
Best time for 1792K FFT length: 69.253 ms.
Best time for 2048K FFT length: 78.543 ms.
Best time for 2560K FFT length: 103.744 ms.
Best time for 3072K FFT length: 126.878 ms.
Best time for 3584K FFT length: 148.956 ms.
Best time for 4096K FFT length: 170.245 ms.
Best time for 5120K FFT length: 219.297 ms.
Best time for 6144K FFT length: 267.964 ms.
Best time for 7168K FFT length: 324.853 ms.
Best time for 8192K FFT length: 370.556 ms.
Timing FFTs using 2 threads.
Best time for 768K FFT length: 16.163 ms.
Best time for 896K FFT length: 23.021 ms.
Best time for 1024K FFT length: 26.109 ms.
Best time for 1280K FFT length: 31.032 ms.
Best time for 1536K FFT length: 37.103 ms.
Best time for 1792K FFT length: 43.269 ms.
Best time for 2048K FFT length: 49.260 ms.
Best time for 2560K FFT length: 66.193 ms.
Best time for 3072K FFT length: 79.358 ms.
Best time for 3584K FFT length: 92.148 ms.
Best time for 4096K FFT length: 105.078 ms.
Best time for 5120K FFT length: 130.060 ms.
Best time for 6144K FFT length: 161.614 ms.
Best time for 7168K FFT length: 195.442 ms.
Best time for 8192K FFT length: 221.939 ms.
Timing FFTs using 3 threads.
Best time for 768K FFT length: 13.809 ms.
Best time for 896K FFT length: 21.811 ms.
Best time for 1024K FFT length: 24.318 ms.
Best time for 1280K FFT length: 28.503 ms.
Best time for 1536K FFT length: 33.089 ms.
Best time for 1792K FFT length: 37.643 ms.
Best time for 2048K FFT length: 42.105 ms.
Best time for 2560K FFT length: 59.491 ms.
Best time for 3072K FFT length: 68.819 ms.
Best time for 3584K FFT length: 78.486 ms.
Best time for 4096K FFT length: 87.776 ms.
Best time for 5120K FFT length: 96.075 ms.
Best time for 6144K FFT length: 115.049 ms.
Best time for 7168K FFT length: 134.383 ms.
Best time for 8192K FFT length: 152.169 ms.
Timing FFTs using 4 threads.
Best time for 768K FFT length: 12.659 ms.
Best time for 896K FFT length: 20.397 ms.
Best time for 1024K FFT length: 22.734 ms.
Best time for 1280K FFT length: 26.702 ms.
Best time for 1536K FFT length: 30.707 ms.
Best time for 1792K FFT length: 34.792 ms.
Best time for 2048K FFT length: 38.922 ms.
Best time for 2560K FFT length: 55.094 ms.
Best time for 3072K FFT length: 63.479 ms.
Best time for 3584K FFT length: 72.257 ms.
Best time for 4096K FFT length: 80.775 ms.
Best time for 5120K FFT length: 87.718 ms.
Best time for 6144K FFT length: 104.854 ms.
Best time for 7168K FFT length: 121.390 ms.
Best time for 8192K FFT length: 136.810 ms.
Timing FFTs using 5 threads.
[Wed Oct 17 02:55:20 2007]
Best time for 768K FFT length: 12.210 ms.
Best time for 896K FFT length: 20.615 ms.
Best time for 1024K FFT length: 22.624 ms.
Best time for 1280K FFT length: 26.663 ms.
Best time for 1536K FFT length: 31.017 ms.
Best time for 1792K FFT length: 34.802 ms.
Best time for 2048K FFT length: 38.889 ms.
Best time for 2560K FFT length: 55.187 ms.
Best time for 3072K FFT length: 64.065 ms.
Best time for 3584K FFT length: 72.070 ms.
Best time for 4096K FFT length: 79.979 ms.
Best time for 5120K FFT length: 87.083 ms.
Best time for 6144K FFT length: 103.475 ms.
Best time for 7168K FFT length: 120.259 ms.
Best time for 8192K FFT length: 135.915 ms.
Timing FFTs using 6 threads.
Best time for 768K FFT length: 11.827 ms.
Best time for 896K FFT length: 20.350 ms.
Best time for 1024K FFT length: 22.470 ms.
Best time for 1280K FFT length: 26.510 ms.
Best time for 1536K FFT length: 30.388 ms.
Best time for 1792K FFT length: 34.276 ms.
Best time for 2048K FFT length: 38.393 ms.
Best time for 2560K FFT length: 55.001 ms.
Best time for 3072K FFT length: 62.357 ms.
Best time for 3584K FFT length: 70.534 ms.
Best time for 4096K FFT length: 78.571 ms.
Best time for 5120K FFT length: 85.249 ms.
Best time for 6144K FFT length: 101.560 ms.
Best time for 7168K FFT length: 117.729 ms.
Best time for 8192K FFT length: 132.915 ms.
Timing FFTs using 7 threads.
Best time for 768K FFT length: 11.377 ms.
Best time for 896K FFT length: 20.030 ms.
Best time for 1024K FFT length: 22.347 ms.
Best time for 1280K FFT length: 26.278 ms.
Best time for 1536K FFT length: 30.107 ms.
Best time for 1792K FFT length: 33.871 ms.
Best time for 2048K FFT length: 37.850 ms.
Best time for 2560K FFT length: 53.885 ms.
Best time for 3072K FFT length: 61.956 ms.
Best time for 3584K FFT length: 70.095 ms.
Best time for 4096K FFT length: 77.744 ms.
Best time for 5120K FFT length: 84.498 ms.
Best time for 6144K FFT length: 100.272 ms.
Best time for 7168K FFT length: 116.577 ms.
Best time for 8192K FFT length: 131.545 ms.
Timing FFTs using 8 threads.
Best time for 768K FFT length: 11.269 ms.
Best time for 896K FFT length: 19.927 ms.
Best time for 1024K FFT length: 22.092 ms.
Best time for 1280K FFT length: 25.985 ms.
Best time for 1536K FFT length: 29.836 ms.
Best time for 1792K FFT length: 33.735 ms.
Best time for 2048K FFT length: 37.738 ms.
Best time for 2560K FFT length: 53.270 ms.
Best time for 3072K FFT length: 61.192 ms.
Best time for 3584K FFT length: 68.896 ms.
Best time for 4096K FFT length: 77.015 ms.
Best time for 5120K FFT length: 83.931 ms.
Best time for 6144K FFT length: 99.895 ms.
Best time for 7168K FFT length: 115.865 ms.
Best time for 8192K FFT length: 130.881 ms.
Best time for 58 bit trial factors: 6.102 ms.
Best time for 59 bit trial factors: 6.116 ms.
Best time for 60 bit trial factors: 6.104 ms.
Best time for 61 bit trial factors: 6.127 ms.
Best time for 62 bit trial factors: 11.402 ms.
Best time for 63 bit trial factors: 11.376 ms.
Best time for 64 bit trial factors: 11.220 ms.
Best time for 65 bit trial factors: 11.166 ms.
Best time for 66 bit trial factors: 11.154 ms.
Best time for 67 bit trial factors: 11.135 ms.

[mAJORD]

thanks for posting lin,

rough Comparison of !mb windsor K8 @ 3.15Ghz DDR2-800

[Wed Oct 17 20:50:49 2007]
Compare your results to other computers at http://www.mersenne.org/bench.htm
AMD Athlon(tm) 64 X2 Dual Core Processor 6000+
CPU speed: 3149.60 MHz, 2 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 23.977 ms.
Best time for 896K FFT length: 28.758 ms.
Best time for 1024K FFT length: 32.070 ms.
Best time for 1280K FFT length: 40.833 ms.
Best time for 1536K FFT length: 49.830 ms.
Best time for 1792K FFT length: 60.381 ms.
Best time for 2048K FFT length: 67.271 ms.
Compare your results to other computers at http://www.mersenne.org/bench.htm
AMD Athlon(tm) 64 X2 Dual Core Processor 6000+
CPU speed: 3150.12 MHz, 2 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 23.929 ms.
Best time for 896K FFT length: 28.854 ms.
Best time for 1024K FFT length: 32.007 ms.
Best time for 1280K FFT length: 40.859 ms.
Best time for 1536K FFT length: 50.156 ms.
Best time for 1792K FFT length: 60.028 ms.
Best time for 2048K FFT length: 67.450 ms.
Best time for 2560K FFT length: 88.718 ms.
Best time for 3072K FFT length: 108.427 ms.
Best time for 3584K FFT length: 130.906 ms.
Best time for 4096K FFT length: 146.278 ms.
Best time for 5120K FFT length: 191.597 ms.
Best time for 6144K FFT length: 233.971 ms.
Best time for 7168K FFT length: 284.569 ms.
Best time for 8192K FFT length: 325.225 ms.
Timing FFTs using 2 threads.
Best time for 768K FFT length: 15.164 ms.
Best time for 896K FFT length: 18.115 ms.
Best time for 1024K FFT length: 20.450 ms.
Best time for 1280K FFT length: 27.264 ms.
Best time for 1536K FFT length: 32.885 ms.
Best time for 1792K FFT length: 39.207 ms.
Best time for 2048K FFT length: 43.853 ms.
Best time for 2560K FFT length: 59.363 ms.
Best time for 3072K FFT length: 71.331 ms.
Best time for 3584K FFT length: 84.593 ms.
Best time for 4096K FFT length: 94.509 ms.
Best time for 5120K FFT length: 104.298 ms.
Best time for 6144K FFT length: 131.939 ms.
Best time for 7168K FFT length: 169.880 ms.
Best time for 8192K FFT length: 211.001 ms.
Best time for 58 bit trial factors: 3.828 ms.
Best time for 59 bit trial factors: 3.845 ms.
Best time for 60 bit trial factors: 3.853 ms.
Best time for 61 bit trial factors: 3.847 ms.
Best time for 62 bit trial factors: 7.028 ms.
Best time for 63 bit trial factors: 7.048 ms.
Best time for 64 bit trial factors: 8.894 ms.
Best time for 65 bit trial factors: 8.823 ms.
Best time for 66 bit trial factors: 8.844 ms.
Best time for 67 bit trial factors: 8.849 ms.


Wondering why the 1 thread - 2 thread scaling is so poor on my sys, may be the overhead of other apps I had running, anyone else got some K8 results? certainly makes K10 look impresive and in Core 2 league.

[Dresdenboy]

This is prime test with supermicro mainboard, cpu df at 1900mhz. I think this main run fastest, mem controller run at 1600mhz instead 1400mhz on tyan. [...]

Thanks for these results! I've expected significantly lower times. Looks like cache architecture is working differently than what Prime95 currently is optimized for (first SSE2 code written for P4, later optimized for other caches and a bit for C2, K8). I'll make a more thorough analysis later.

Wondering why the 1 thread - 2 thread scaling is so poor on my sys, may be the overhead of other apps I had running, anyone else got some K8 results? certainly makes K10 look impresive and in Core 2 league.

Thanks for your numbers too. There are not many results around for K8 > 3 GHz.

Your 2 thread scaling is actually better than Barcelona's for certain FFT sizes - in case of the 1024K FFTs you are closer to 60% time using 2 threads, while linhvndiy's K10 are closer to 70%.

The problem is in the code - Prime95's calculations (special FFT variants) have a lot of serial code compared to parallelizable code. There are diminishing returns when increasing the number of threads as you can see.

Watch this thread, where I also posted linhvndiy's results:
http://mersenneforum.org/showthread.php?t=59&page=3

Two Barcelonas at 1.9 GHz behave like a pair of 3 GHz Netburst based 65nm DC Xeons at smaller FFT sizes, but lose at the bigger sizes:
http://mersenneforum.org/showpost.ph...&postcount=284

[xPliziT]

Here are my Results with a Athlon 64 X2 6400 @ 3500 MHz and CellShock DDR2 1000 @ 4-4-4-12

[Wed Oct 17 18:32:13 2007]
Compare your results to other computers at http://www.mersenne.org/bench.htm
AMD Athlon(tm) 64 X2 Dual Core Processor 6400+
CPU speed: 3499.87 MHz, 2 cores
CPU features: RDTSC, CMOV, Prefetch, 3DNow!, MMX, SSE, SSE2
L1 cache size: 64 KB
L2 cache size: 1024 KB
L1 cache line size: 64 bytes
L2 cache line size: 64 bytes
L1 TLBS: 32
L2 TLBS: 512
Prime95 32-bit version 25.5, RdtscTiming=1
Best time for 768K FFT length: 21.030 ms.
Best time for 896K FFT length: 25.262 ms.
Best time for 1024K FFT length: 27.967 ms.
Best time for 1280K FFT length: 35.763 ms.
Best time for 1536K FFT length: 43.718 ms.
Best time for 1792K FFT length: 52.707 ms.
Best time for 2048K FFT length: 58.854 ms.
Best time for 2560K FFT length: 77.593 ms.
Best time for 3072K FFT length: 94.625 ms.
Best time for 3584K FFT length: 114.018 ms.
Best time for 4096K FFT length: 127.410 ms.
Best time for 5120K FFT length: 165.333 ms.
Best time for 6144K FFT length: 201.588 ms.
Best time for 7168K FFT length: 244.321 ms.
Best time for 8192K FFT length: 278.076 ms.
Timing FFTs using 2 threads.
Best time for 768K FFT length: 12.993 ms.
Best time for 896K FFT length: 15.601 ms.
Best time for 1024K FFT length: 17.517 ms.
Best time for 1280K FFT length: 23.488 ms.
Best time for 1536K FFT length: 28.120 ms.
Best time for 1792K FFT length: 33.641 ms.
Best time for 2048K FFT length: 37.514 ms.
Best time for 2560K FFT length: 50.270 ms.
Best time for 3072K FFT length: 60.128 ms.
Best time for 3584K FFT length: 71.829 ms.
Best time for 4096K FFT length: 79.863 ms.
Best time for 5120K FFT length: 89.756 ms.
Best time for 6144K FFT length: 111.599 ms.
Best time for 7168K FFT length: 139.406 ms.
Best time for 8192K FFT length: 167.504 ms.
Best time for 58 bit trial factors: 3.403 ms.
Best time for 59 bit trial factors: 3.436 ms.
Best time for 60 bit trial factors: 3.420 ms.
Best time for 61 bit trial factors: 3.436 ms.
Best time for 62 bit trial factors: 6.236 ms.
Best time for 63 bit trial factors: 6.233 ms.
Best time for 64 bit trial factors: 7.895 ms.
Best time for 65 bit trial factors: 7.840 ms.
Best time for 66 bit trial factors: 7.858 ms.
Best time for 67 bit trial factors: 7.845 ms.

[linhvndiy]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

[Sparky]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

Some good scaling for at least that bench.

[informal]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

It seems IMC is benefiting from the OCed speed as well,so the latency is lower.

[KTE]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

We might be able to see a trend if you could just push the clocks higher

[tictac]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

What is your highest overclock?

[mstp2009]

I can't understand why cinebench R10 df clock score is around 11600, but at 2020mhz core is 13400. More than 15% increase but only 7% more clock?

Didn't someone (I think at AT) post that the IMC runs at different speeds (i.e. ratios) depending on the overall CPU speed?

So it is possible that the IMC just "switched gears" when you went to 2.0GHz.

Kinda like moving to a different strap on C2D overclocking.

[mongoled]

Didn't someone (I think at AT) post that the IMC runs at different speeds (i.e. ratios) depending on the overall CPU speed?

So it is possible that the IMC just "switched gears" when you went to 2.0GHz.

Kinda like moving to a different strap on C2D overclocking.

Oh no you cant say that....

the stake bring the staaaaaaaaake



whos got the matches??



hehehhehehee

interesting findings good that more results are coming in

[linhvndiy]

What is your highest overclock?

about 2030mhz, got restart at 214mhz htt.
I am sure mainboard still have to tweak a lot to fully support barcelona.

[Viper666]

doesnt work in ASUS KFSN4-DRE 1.7 BA's no post

[Viper666]

LOL Im sure that ASUS made sure it works !

They didnt

[PhilDoc]

Didn't someone (I think at AT) post that the IMC runs at different speeds (i.e. ratios) depending on the overall CPU speed?

So it is possible that the IMC just "switched gears" when you went to 2.0GHz.

Kinda like moving to a different strap on C2D overclocking.

I think it has more to do with whether the mb has dual power planes or not. Without a split power plane board the MC of a 2350 will run at 1600 mhz. With a dual power plane board the MC will run at 1800 mhz