AMD Athlon 64 FX 51 23rd September 2003
Before we get started we need to explain the format of our reviews over the past six months. Readers will have noted that an entire review is laid out on one long page instead of 15-20 short pages. While it takes longer for a single page of this length to load we have designed the layout so that text appears first and graphics last so you can start reading before completion of loading. We decided not to split over 15-20 pages as many people find it annoying to have interruptions once they start getting into a review. It also means you can use the search option in the menu bar to go straight to parts that are of interest. We'd like our readers to let us know which method they prefer by providing feedback in our Forums. In any case, by the time you finish reading this paragraph the page should have mostly loaded.
It's been a long time in coming but the world's first consumer 64-bit processor is finally here. Named the Athlon64 FX 51 and running at 2.2GHz, can it win back the performance crown from Intel's flagship P4s and compete with the forthcoming Prescott CPUs?
We see in the above diagram that the AMD64 Core eliminates the 4GB barrier we encounter with today's 32bit systems, good news for those of us who use memory intensive programs.
Here are the manufacturer's specifications:
Here's a summary of the test system:
For comparison we'll use Intel's D875PBZ reference Canterwood board and all the high end Athlon XPs we can get our hands on.
Let's start with the synthetic benchmarks.
Strangely we see the lowest score here and suspect that this benchmark suite needs updating to correctly analyze this chip.
Now the opposite is true and the onboard memory controller blows away all competition even besting the 800MHz FSB and dual channel of Intel's Canterwood motherboard.
Lowest score again but this too is due to the benchmark in need of updating.
Pretty much the same. We'll run these again when the next Sandra update is available.
This is a series of tests and is more comprehensive than any of the Sandra benchmarks.
And in comparison to other machines:
The memory score is in a league of its own. Looks like AMDs decision to have the memory controller on-chip has been vindicated. The CPU score is a nice increase from the XP3200+.
The following results were obtained with our standard graphics card (Nvidia Ti4400):
The P4 is finally dethroned.
And for comparison:
Looking at the design of the chip we believed there may be significant improvements with a high-end graphics card so we threw in a 5900 Ultra and let it spin:
This is absolutely incredible. Let's compare:
There's a difference of over 3000 points compared to the Athlon XP3200+. The architecture certainly helps with CPU limitations.
Now for some real world benchmarks starting with UT2003 Flyby.
Although fill rate limited at higher resolutions we can see the enormous potential of AMD's new CPU.
The Athlon 64 FX 51 is a clear winner at all resolutions. This bodes well for performance with Doom3 and Half Life 2 (very soon now). Speaking of which, you will need a powerful graphics card for these next generation games so we ran the same test with a 5900 Ultra.
The CPU ceiling has been lifted far more than a few speed grades of XP would have done.
Being able to play at a 110fps instead on 85 may not be a big deal here but casting our eyes into the future we can see older architectures being brought to their knees by new generations of software (we wont even mention new Operating Systems).
Let's turn to an area where we know that fast CPUs will make a difference - Audio and Video encoding. This is becoming more and more popular and is very computationally intensive with long processing times (relatively speaking that is, this field is not for those that complain about how long their Outlook Express takes to load).
For consistency we will use Jet Li's The One as our test matter. It is not interlaced and contains a mixture of action types and is not too long. There will be three tests all using Divx 5.02. Audio will be encoded separately. I will try and keep my commentary to a minimum as all configuration information is shown in the images below.
and here are the results
A big increase over the XP3200+ in this bandwidth intensive test thanks largely to the onboard memory controller.
AviSynth and VirtualDub
No serious Divx encoder uses Xmpeg alone and it's just used by the media for benchmarking purposes so let's get serious. We ripped our source material to hard disk and created a DVD2AVI project file using forced film (it was 99% film). Loading this into Gordian Knot we first saved an .avs file with no changes at all (720x480) and no filters of any sort. This was loaded into VirtualDub with the following CODEC parameters :
After encoding we got these results:
The increase is even greater here in this field that the P4s have ruled exclusively.
This is all good stuff but how about a real-world test? To simulate a realistic test we added a neutral bicubic resize filter in the .avs file and used the following CODEC parameters (including two popular Pro settings) which are designed to total 700MB (when the audio is muxed in):
Which resulted in the following.
We never expected to see this. It seems that for everyday Divx use the Athlon64 FX 51 is the best choice.
What about audio? We took the AC3 track from the above sample material and used HeadAC3he to convert it into Vorbis format so our final muxed file could have Ogg containment. There isn't space here to go into the advantages of Ogg Vorbis over MP3 and AVI so let's just say that Vorbis sounds about the same as MP3 for half the file size or twice as good for the same file size (that is subjective though).
Since it is more meaningful to show throughput than time taken (which depends on the length of the source) we display the results thus:
Very little to separate the high-end processors - it seems anything will do for audio encoding.
We waited and waited and now it's finally here. The delay may have been costly but it still beats the cream of the current competition with ease. The 64-bit performance has been neglected despite us having a 64-bit Windows XP beta in our possession. Everything worked fine in 64-bit mode as far as XP was concerned but the availability of 64-bit drivers and lack of a stable 64-bit DirectX means that those tests will have to wait for another day.
It's a milestone in computing and a welcome one as current architectures become somewhat long in the tooth. With excellent all-round performance and making a superb games machine the Athlon64 FX 51 has no trouble winning our gold award. Users will get immediate benefits and long-term security when the eventual migration to 64-bit computing becomes a necessity.
We would like to thank AMD UK for the loan of the test equipment.
All trademarks are the property of their respective owners.