AMD Athlon 64 & Athlon 64 FX - It's Judgment Day
by Anand Lal Shimpi on September 23, 2003 1:25 PM EST- Posted in
- CPUs
32-bit vs. 64-bit Performance
Our entire benchmark suite to this point has been on 32-bit applications under a 32-bit OS, mostly because there are no good desktop 64-bit applications at this point in a popular 64-bit OS (not to mention the issues with 64-bit Windows XP we described earlier).
Under Linux however we don't have to wait for applications to be released in a 64-bit version, we can simply recompile them. Linux would thus provide us with an excellent venue to see the tangible performance increases from exposing the additional general purpose registers in 64-bit mode.
We ran all benchmarks on Red Hat Enterprise 2.9.5WS (Taroon), a beta release, booted in single user mode to avoid system services interfering with benchmark results. Neither Red Hat 9 nor 9.0.93 Beta (Severn) supply a 64-bit compiler or libraries, which is why we used Taroon.
The Taroon kernel initially had issues on startup requiring us to disable APIC and ACPI support to get it to install. Once actually running the OS was quite stable however DMA disk access was disabled for some reason.
We used the following compiler that came with Taroon:
gcc 3.2.3 20030502 (Red Hat Linux 3.2.3-16)
And the following kernel:
2.4.21-1.1931.2.393.ent
With this compiler and kernel we ran the following tests:
Whetstone
A simple C loop measuring floating point performance, configured to do double precision calculations.
Compiled with:
-O3 -msse2 -mfpmath=sse (and -m32 for 32bit, -m64 for 64bit)
The performance improvements due to 64-bit are in the 10 - 20% range we mentioned earlier.
Bytemark
An old integer CPU benchmark (FP results were discarded) - for more information on the tests visit this site.
Compiled with:
-O3 -msse2 -mfpmath=sse (and -m32 for 32bit, -m64 for 64bit)
Here we do see a small 2% drop in performance when moving to 64-bit in one test, however the rest of the tests show a 0 - 15% improvement across the board.
Lame 3.93
A MP3 encoder; encoded a 40minute .wav file (403MB).
Lame args: -b 192 -m s -h --quiet <file> - >/dev/null
(192kbps, simple stereo, high quality, output to nothing to avoid disk hits)
Compiled with:
-O3 -fomit-frame-pointer -fno-strength-reduce -malign-functions=4 -funroll-loops
-ffast-math -msse2 -mfpmath=sse (again, -m32 for 32bit, -m64 for 64bit)
The performance improvement here is astounding - in 64-bit mode the Athlon 64 FX managed to finish the encode 34% quicker than in 32-bit mode, if these results are any hint of what could be in store for Windows users, there's a lot of promise behind the Athlon 64...assuming we get software support in time.
We wanted to do a transcode benchmark but that didn't work out - one library found a bug in gcc and transcode refused to compile. It actually forced a compile error because a structure came out padded, meaning they didn't expect anyone to run it on a 64bit machine just yet.
Facebook
Twitter
RSS
122 Comments
View All Comments
Anonymous User - Tuesday, September 23, 2003 - link
#10 Without XP64 only Linux apps can give great performance boost over 32 bits apps. Yes, I'm not a gamer, I don't care about UT2K3 or DX9 benchmarks.Anonymous User - Tuesday, September 23, 2003 - link
#19i agree. good review, and how anyone could see an AMD bias in that article, i have no idea. he fvcking said repeatedly that AMD isn't really cutting it with the A64/FX. the worst type of fanboy is one that cant read.
its sad to see AMD lagging like this... if they fall too far behind, they wont be able to compete with Intel on a high-end level at all - which sucks for competition.
Anonymous User - Tuesday, September 23, 2003 - link
That was a well written review.Anonymous User - Tuesday, September 23, 2003 - link
#15, doesn't the fact that you can buy an FX51 along with an Athlon64 3200+ at a variety of stores now contradict that statement? I mean someone even said they sighted them in retail stores and the sort across the world before even launch.Anonymous User - Tuesday, September 23, 2003 - link
For the non-fanboys out there, I have a couple questions.1. Is dual-channel really necessary for the Athlon 64? What is the bandwidth of the Hypertransport bus?
2. Anand contends: "Ignoring the performance boost Intel gains by going to dual-channel, OEMs demanded a dual-channel solution from AMD simply as a checkbox feature." Do you buy this? I thought CPU manufacturers pushed the mobo guys around, not vice versa.
3. After launching on two different Sockets (754 & 940) - is AMD going to convert everything over to 939 in '04, or are we going to see high-end on 940 and low-end on 754?
4. Where are the UT2K3 64-bit version benches?
5. I know it's kind of hard to judge, but how is the driver support for 64-bit coming along? (As a measure, does anyone know if nvidia and/or ATI will be ready with 64-bit drivers by the new WinXP64 launch?)
6. This is more of a mobo/chipset questions, but where does PCI Express/3GIO fit into AMD's hypertransport plans?
Anonymous User - Tuesday, September 23, 2003 - link
What i really dont understand is this: "AMD has lost a considerable amount of credibility" and again : "AMD has also priced the Athlon 64 and Athlon 64 FX very much like the Pentium 4s they compete with, which is a mistake for a company that has lost so much credibility"What exactly do you mean by that Anand?
Anonymous User - Tuesday, September 23, 2003 - link
#13The FX-51 will not be avaiable in any reasonable quanitities until next year. The availability of any succeeding products is pure speculation.
Anonymous User - Tuesday, September 23, 2003 - link
I think we'll see the athlon 64 outpacing the p4's in divx encoding once we see a 64bit enabled codec.Anonymous User - Tuesday, September 23, 2003 - link
All reviews (with exception of Toms Hardware and x86secret, go figure) show the FX51 outpacing the P4 Emergency Edition in the gross majority of benches. That's not even counting the fact the P4EE isn't even going to be available until November (making this a paper launch) and by it's availability the FX53 and Athlon64 3400+ will have already been out in the market with us looking forward to the FX55 in late 2003 or early 2004.Anonymous User - Tuesday, September 23, 2003 - link
#4 & #7. Please give me a couple of URLs (not thg) so i can validate your inspired response.