SSE2 makes Opterons slower than Athlon XPs

Dom · May 2, 2003, 11:30 PM

A JAPANESE JOURNALIST said that an investigation he's conducted shows that using SSE2 multimedia extensions with the AMD Opteron makes the microprocessor slow to a crawl.
Takeo Noguchi. at ASCII Plus Magazine, analysed the Opteron 242 as a microprocessor for a single chip machine, and said that a 1.6GHz chip, according to AMD< slightly outperform a Xeon/Pentium 4 running at 2.8GHz.

He told the INQUIRER that he measured the execution time to encode DV-format AVI files to MPEG, WMV, Divx and Xvid. He found that for encoding DivX and Xvid, Opteron 242's performance is roughly equivalent to a Pentium 4-2.8GHz, which matches the expectation from SPEC int/fp results.

But, he added, using well known JapaneseMPEG 2 encoding software TMPGenc ((http://www.pegasys-inc.com/), the Opteron's performance is a staggering 30% lower than a Pentium 4-2.8GHz.

He said: "At first I supposed it was because TMPGEnc is thoroughly optimised for the Pentium 4".

But he checked the results from SiSoftware Sandra 2003, and found to his surprise that its multimedia integer performance is much lower than the Athlon XP, the architectural cousin of the Opteron.

His results are as follows:

Sandra Multimedia Int
Opteron 242 (1.6GHz) 6300
Pentium 4-2.8GHz 11148
Athlon XP-2600+ 11614

He claimed that even degrading the Athlon XP to 1.6GHz it will still deliver 8933, meaning, he adds that the Opteron's performance is 30% lower on a clock per clock basis.

He said that the results for the Athlon XP come from using MMX multimedia instructions, while the Opteron uses the newly supported SSE2 instructions.

He said that means that the SSE 128-bit SIMD instructions are 30% slower than the MMX 64-bit SMD int instructions. He concludes that using SSE2 on the Opteron actually drags down its multimedia integer performance.

When he reset the SSE2 checkbox for TMPGEnc, he said, "questioning his own hypothesis", its performance far exceeded the Pentium 4-2.8GHz.

TMPGEnc Encoding Time (seconds)
Opteron 242 (1.6GHz) 424
Pentium 4-2.8GHz 295
Athlon XP-2600+ 262
Opteron 242-SSE2 OFF 291

But it's not all bad news for the Opteron. He said that well known Japanese archiving software GCA (http://www.emit.jp/gca/gca.html) compresses WAV files a staggering 50% faster than P4-2.8GHz, equivalent to a Pentium 4.4GHz chip, if such a beast existed.

Takeo's full story will be posted at the Akiba2go web site, in Japanese, after 14:00 GMT.

----------------------------------------
News source: TheInquirer

Necrosis · May 2, 2003, 11:52 PM

Okay folks lets figure this one out. New processor with new instruction set running software that doesn't really know quite what it is. Does anyone remember the SSE problem with the XP? Um...yeah DUH!

May 2, 2003, 11:30 PM	#1
Dom DriverHeaven Extreme Member Join Date: Jun 2002 Posts: 12,940 Rep Power: 0	SSE2 makes Opterons slower than Athlon XPs A JAPANESE JOURNALIST said that an investigation he's conducted shows that using SSE2 multimedia extensions with the AMD Opteron makes the microprocessor slow to a crawl. Takeo Noguchi. at ASCII Plus Magazine, analysed the Opteron 242 as a microprocessor for a single chip machine, and said that a 1.6GHz chip, according to AMD< slightly outperform a Xeon/Pentium 4 running at 2.8GHz. He told the INQUIRER that he measured the execution time to encode DV-format AVI files to MPEG, WMV, Divx and Xvid. He found that for encoding DivX and Xvid, Opteron 242's performance is roughly equivalent to a Pentium 4-2.8GHz, which matches the expectation from SPEC int/fp results. But, he added, using well known JapaneseMPEG 2 encoding software TMPGenc ((http://www.pegasys-inc.com/), the Opteron's performance is a staggering 30% lower than a Pentium 4-2.8GHz. He said: "At first I supposed it was because TMPGEnc is thoroughly optimised for the Pentium 4". But he checked the results from SiSoftware Sandra 2003, and found to his surprise that its multimedia integer performance is much lower than the Athlon XP, the architectural cousin of the Opteron. His results are as follows: Sandra Multimedia Int Opteron 242 (1.6GHz) 6300 Pentium 4-2.8GHz 11148 Athlon XP-2600+ 11614 He claimed that even degrading the Athlon XP to 1.6GHz it will still deliver 8933, meaning, he adds that the Opteron's performance is 30% lower on a clock per clock basis. He said that the results for the Athlon XP come from using MMX multimedia instructions, while the Opteron uses the newly supported SSE2 instructions. He said that means that the SSE 128-bit SIMD instructions are 30% slower than the MMX 64-bit SMD int instructions. He concludes that using SSE2 on the Opteron actually drags down its multimedia integer performance. When he reset the SSE2 checkbox for TMPGEnc, he said, "questioning his own hypothesis", its performance far exceeded the Pentium 4-2.8GHz. TMPGEnc Encoding Time (seconds) Opteron 242 (1.6GHz) 424 Pentium 4-2.8GHz 295 Athlon XP-2600+ 262 Opteron 242-SSE2 OFF 291 But it's not all bad news for the Opteron. He said that well known Japanese archiving software GCA (http://www.emit.jp/gca/gca.html) compresses WAV files a staggering 50% faster than P4-2.8GHz, equivalent to a Pentium 4.4GHz chip, if such a beast existed. Takeo's full story will be posted at the Akiba2go web site, in Japanese, after 14:00 GMT. ---------------------------------------- News source: TheInquirer

May 2, 2003, 11:52 PM	#2
Necrosis HardwareHeaven Extreme Member Join Date: May 2002 Location: Ohio Posts: 8,782 Rep Power: 106 System Specs	Okay folks lets figure this one out. New processor with new instruction set running software that doesn't really know quite what it is. Does anyone remember the SSE problem with the XP? Um...yeah DUH!