Windows XP 64-Bit Preview: First Look at Athlon 64 Performance
by Wesley Fink on February 7, 2004 12:30 AM EST- Posted in
- Systems
System and Memory Benchmarks
SiSoft Sandra 2004 64-Bit
The 64-bit
version of Sandra 2004 has been available for a while, but we did not have an Operating System to reliably run with the 64-bit version. Sandra 64-bit runs fine on the XP64 preview.
While Sandra is a Synthetic Benchmark, we were curious to see if there
would be any performance difference in memory, CPU Arithmetic, and Multimedia
benchmarks between the 32-bit and 64-bit versions. Everything was kept the same; we even used
nVidia drivers close to the same version number. The only difference is Sandra 2004 tests were
run on XP Pro, while Sandra 2004 64-bit tests were run on XP 64-bit Preview
Edition.
|
SiSoft Sandra 2004 - Athlon 64 FX51 Performance |
|||
|
|
32-Bit (Windows XP SP1) |
64-Bit (XP64 Preview Edition) |
% Change 32 to 64-bit |
|
Sandra 2004 Standard Buffered |
INT
5722 FLT 5660 |
INT
5910 FLT 5831 |
+3.2% |
|
Sandra 2004 UNBuffered |
INT
2588 FLT
2682 |
INT
2811 FLT
2791 |
+6.3% |
|
Sandra 2004 CPU Arithmetic |
9161 mips 3470/4534 mflops |
10121 mips 3881/4105 mflops |
+10.5% mips -0.2% mflops |
|
Sandra 2004 CPU Multimedia |
INT 16404 FLOAT 21642 |
INT 16598 FLOAT 22869 |
+1% INT +5.7% FLOAT |
The 32-bit
vs. 64-bit results in Sandra are very interesting. Even in this pre-release version of XP64, the
Athlon 64 CPU and Memory Performance is higher than in 32-bit Windows XP. Mips, which is based on ALU tests, is more
than 10% faster, and Integer and Float tests in the Sandra 2004 Multimedia
benchmark is 1% to 6% faster. The only
area without increased performance in 64-bit is the mflops component of the
Arithmetic benchmark. If we look closer,
this benchmark is a combination FPU performance and iSSE2 performance. While Floating Point increases some 11.6% in
the move from XP to XP64 Preview, the Intel SSE2 results decrease by about the
same amount. The net result is
virtually no change in the composite mflops.
We do not know if this is because Intel SSE2 is penalized by 64-bit
operation or whether XP64 and/or Sandra 2004 64-bit benchmark require some
optimizations for 64-bit performance.
Super Pi
Super PI is
very simple - it calculates the value of pi.
In the benchmark you can select the number of placed for calculation,
and we used 2 million places as used in memory tests at AnandTech.
|
Super Pi - Athlon 64 FX51
Performance |
|||
|
|
32-Bit (Windows XP SP1) |
64-Bit (XP64 Preview Edition) |
% Change 32 to 64-bit |
|
Super Pi 2M Places |
88 seconds |
88 seconds |
0% |
As you can
see, Super Pi was exactly the same result in both 32 and 64-bit.
Facebook
Twitter
RSS
42 Comments
View All Comments
Staples - Saturday, February 7, 2004 - link
I really hope those game scores are due to premature video drivers. As you see, Halo did almost as well as the 32bit platform and as you should know, DX9 games are almost solely based on the GPU. So if Halo did almost as well on both platforms, it says that the video drivers can't be that premature, either that or explanation 2 is that we can expect a huge increase in DX9 games.Corsairpro - Saturday, February 7, 2004 - link
Too bad there weren't any decent video drivers. Every one who just glances at the numbers is going to claim "The message is clear x86-64 has failed" when it comes to games. Oh well, more supply for me to buy!buleyb - Saturday, February 7, 2004 - link
Not that I'm not excited, but you should point out Wes that this isn't just a 64bit OS, but an AMD 64bit OS, meaning that the performance improvement has a lot to do with the new general purpose registers and such. I don't want people thinking that 64bit is a pure performance improvement, because it really isn't by itself.But still, nice work :)
KristopherKubicki - Saturday, February 7, 2004 - link
Skol. Well done Wes.saechaka - Saturday, February 7, 2004 - link
boy am i glad i just bought this athlon 64 notebook. huurraaayy for meWesley Fink - Saturday, February 7, 2004 - link
There are times editing would be useful in this comments section. XP, and not Halo, had about the same performance.Wesley Fink - Saturday, February 7, 2004 - link
Halo was the game that was close to the same performance in XP and XP64, and not Halo as #4 pointed out. Since X2 is DirectX 8.1 with heavy use of transform and lighting effects, it has little relevance to the Halo performance. Corrected in the article.Emma - Saturday, February 7, 2004 - link
"It is very interesting that the DirectX 9 game Halo is already very close to 32-bit performance at only 4% slower than 32-bit performance. This means the newest 32-bit games, or at least the newest games from Microsoft, may be as fast on 64-bit as 32-bit at the launch of XP64, or possibly even faster."Can you clarify this please. The table shows there being a -19.1% change...
Boonesmi - Saturday, February 7, 2004 - link
by the way ive read in several threads of guys using pcmark 2004 and getting incredible fps in divx encodingEcmaster76 - Saturday, February 7, 2004 - link
Very interesting. That 15% increase in media encoding should have the AMD execs laughing maniacally. That might end up getting them a 15% increase in market share.