Long gone are the times when users (and testers themselves) were looking forward to the release of each new processor from Intel or AMD. The database of test results is so large that the results of a new model, which differs from previous processors only in clock frequency, are quite expectable and the ballpark figures are known beforehand. Of course, sometimes we have to deal with something new: a new core, a new untried cache size... but today is not the case.

Despite its intriguing name, AMD Athlon 64 FX-60 is nothing more than a higher-clocked modification of our old acquaintance — Athlon 64 X2 4800+. The difference comes down to the clock: two cores in the Athlon 64 X2 4800+, equipped with 1 MB L2 Caches, ran at 2400 MHz, while the new Athlon 64 FX-60 has the same cores running at 2600 MHz. The number of transistors (about 233 mln), die surface area (about 199 mm2), maximum heat dissipation (110 Watts) — all these characteristics in Athlon 64 X2 4800+ and new Athlon 64 FX-60 are the same. There is actually nothing surprising about it, as practically all the other characteristics match too, except for the above mentioned clock (but including the process technology — 90 micron). To all appearances, designers managed to "squeeze" FX-60 into the X2 4800+ heat dissipation due to the voltage reduction by 0.05 V: the corresponding range for X2 is within 1.35—1.40 V, for FX-60 - within 1.30—1.35 V.

Thus, the new Athlon 64 FX takes up a strange position in the updated line of AMD processors: it equals FX-55 in terms of a single core clock (that is slower than FX-57), but it has two such cores instead. Of course, AMD's main objective is evident: to make up for the recent announcement of Intel Pentium XE 955. But it's not quite clear why the company decided (for the first time in the history of this series) to make a dual core processor out of FX. We shall discuss AMD's motive and our conclusions in the end of this article. As for now, let's have a look at the test results. Be warned: you will see nothing surprising on the diagrams for the above mentioned reasons.

Testing

Testbed configurations

• Processors
  • AMD Athlon 64 FX-60 (2 x 1 MB L2, 2 x 2.6 GHz core)
  • AMD Athlon 64 FX-57 (1 MB L2, 2.8 GHz core)
  • AMD Athlon 64 FX-55 (1 MB L2, 2.6 GHz core)
  • AMD Athlon 64 X2 4800+ (2 x 1 MB L2, 2 x 2.4 GHz core)
  • Intel Pentium eXtreme Edition 955 (2 x 2 MB L2, 1066 MHz FSB, 2 x 3.46 GHz core)
  • Intel Pentium eXtreme Edition 840 (2 x 1 MB L2, 800 MHz FSB, 2 x 3.2 GHz core)
  • Intel Pentium 4 670 (2 MB L2, 800 MHz FSB, 3.8 GHz core)
  • Intel Pentium 4 eXtreme Edition 3.73 GHz (2 MB L2, 1066 MHz FSB)

• Motherboards
  • Intel D955XBK Desktop Board (Intel 955X chipset)
  • ASUS A8N SLI Deluxe (NVIDIA nForce4 chipset)

• Memory
• 2x512 MB PC5400 (DDR2-533) DIMM 3-3-3-8 (Corsair)
• 2x512 MB PC3200 (DDR-400) DIMM 2-2-2-5 (Corsair)
• Video card: ATI Radeon X800 (256 MB)
• HDD: Samsung SP1614C (SATA), 7200 rpm, 8 MB Cache
• AC power adapter: FSP 550-60PLN (500-550W)
• Windows XP Professional SP2, DirectX 9.0c
• ATI CATALYST 5.4 (Display Driver 6.14.10.6525)

Test results

Diagrams with all test results (64 all in all) are published on a separate web page — without comments, just as is. The article provides only summary diagrams that calculate the results of entire test groups into average scores. This approach appeases curiosity of the most inquisitive readers, who are against cutting down the number of test results published in our articles, and still makes the article less motley and graphics-intense.

SPECapc for 3ds max 6 + 3ds max 7.0

A complete set of diagrams

This diagram shows no outstanding results. Athlon 64 FX-60 is expectedly faster than Athlon 64 X2 4800+ (due to higher core clocks) as well as Athlon 64 FX-57 (due to two cores, instead of a single core in the FX-57). It's also natural that the Athlon 64 X2 4800+ is less slow than the FX-57: 3ds max has always been well optimized for SMP, especially in rendering (you can see it well on the detailed diagrams). On the whole, we can congratulate AMD, whose dual core processor reached another clock threshold, but the diagram contains no revelations: we could have drawn something like that on a sheet of paper with a soft-tip pen, knowing only the specs of the new processor. The defeat of Pentium XE 955 is also expected: it was outperformed by Athlon 64 X2 4800+ in this test.

Maya 6.5

A complete set of diagrams

Our strength is in logic and the knowledge of software and test peculiarities. Sometimes, it can even replace scrupulousness of our testers, who will not stir a finger without launching a couple of benchmarks :). SPEC for Maya does not contain a render subtest, that's why the FX-60 is just a little faster than the FX-55 in this benchmark (probably due to SSE3 support and further optimization of the memory controller in Toledo core); the first place is taken by the FX-57 — it's a single core processor, but it has the maximum clock. The situation is quite the contrary in a separate render test: it favours dual core processors, FX-55/57 CPUs are significantly outperformed. Attentive readers of the iXBT.com CPU section knew that before. We can just confirm that the test results conform with the assumptions that could have been made relying on trivial logic.

We don't think it's necessary to comment on the results of the top dual core CPU from Intel (Pentium XE 955), when it is outperformed even by the older Athlon 64 X2 4800+, as these comments will be just a variation of the ones we had already made. They will just clutter the material.

Lightwave 8.2, rendering

This situation is identical to that in the 3ds max and Maya rendering subtests (see detailed diagrams). There is nothing surprising about that.

SPECapc for SolidWorks 2003

A complete set of diagrams

Athlon 64 FX-60 sometimes outperforms the FX-55 — even in those applications, which are practically insensitive to dual cores (dual processing). We have already mentioned two possible reasons: SSE3 support and a newer, optimized, fine-tuned memory controller. Note that Athlon 64 FX-55 is now manufactured with SSE3 support and an optimized controller as well, but we tested an older modification: AMD did not change the CPU name, after it had switched to a newer core.

Adobe Photoshop CS (8)

A complete set of diagrams

As before, Athlon 64 FX-57 and FX-60 ratings depend on how well a given test supports SMP. Adobe Photoshop is not ideal from this point of view, but even its features are sufficient to bring the FX-60 to leaders. It still lost one second to the top dual core processor from Intel, so it formally takes the second place.

Adobe Acrobat 6.0

The steady (but very little) advantage of multi-core CPUs over single core processors in this test is still a mystery to us: we don't have information about SMP support or any multi-thread optimizations in Adobe Acrobat Distiller. However, we can see well that the difference even between the best and the worst results is not that significant.

All-purpose data compression (archiving)

A complete set of diagrams

As always, high-clocked single core processors are the first in the WinRAR test, while dual core CPUs lead in the 7-Zip test. The latter processors often have a better average score, as the clock advantage of dual core CPUs is not that high, while the second core gives a noticeable advantage in 7-Zip. Athlon 64 FX-60 is expectedly the fastest processor.

Multimedia lossy compression (MP3/MPEG2-4)

A complete set of diagrams

Dual core processors are often in the lead in this test, because many programs can take advantage of the second core (Canopus ProCoder, Windows Media Video, and even LAME a little). This review is not an exception, as you can see. Pentium XE 955 is the formal leader due to LAME results (see the detailed diagrams). But according to the other test results, the winner is either the Athlon 64 FX-57 or the FX-60.

CPU RightMark 2004B

A complete set of diagrams

Athlon 64 FX-60 breaks a new performance record in CPU RM 2004, even Pentium XE 955 will have problems competing with it...

3D games and graphics visualization
in professional packages

DOOM 3

Far Cry

Painkiller

Unreal Tournament 2004

Total score in games

A complete set of diagrams

SPEC viewperf

A complete set of diagrams

On the average, the new FX-60 is naturally outscored by the FX-57 in games and in SPEC viewperf, as these programs cannot use the second core and the clock of a single core in the Athlon 64 FX-57 is higher. That's actually no good news, as extreme gamers are the target audience for Athlon 64 FX processors. Now gamers will also have to consider the fact that not all FX processors are equally good, a higher model number does not always mean higher performance in their favourite applications.

Conclusion

This processor is quite predictable (as well as the majority of recent processors from Intel and AMD). We could have even skipped out tests, as we already knew the specs: these results could have been extrapolated +/- 3-4% from the existing data. It actually subtracts nothing from merits of AMD engineers, who managed to stuff two 2600 MHz cores into a single package. In other words, we have no gripes with the technical side of the product, only praises. Its competitive level is also excellent: while Athlon 64 X2 4800+ only outpointed Pentium XE 955, Athlon 64 FX-60 can be called an absolute leader. But in consumer terms... it's worse. For some reason, there is a fly in the ointment, though we could have done great without it. That's just our subjective opinion, but...

As soon as it appeared, Athlon 64 FX was associated with maximum possible clock, mastered by this manufacturer, which made it an ideal choice for gamers rather than professionals. Professionals would buy workstations on Opteron (with two or more processors), and there has recently appeared an alternative - dual core Athlon 64 X2 processors. It fully complies with AMD's official positioning strategy, positioning was logical and transparent. And here it is — Athlon 64 FX-60. Dual core processor (it makes no sense for current games) with the maximum model number in the FX series... but not with the maximum clock!

It's quite easy to understand why Athlon 64 FX-60 is made as it is:

1. AMD seems to be not ready for launching 3-GHz processors (even if with a single core) on a production scale. Judging by the price of Athlon 64 FX-57, the yield of effective dice, which can run even at 2.8 GHz, is still low.
2. It's no secret that Pentium eXtreme Edition 955 will be launched in the nearest future — dual core processor from Intel. It should have been opposed by something (or still better to make a move earlier). The XE processor from Intel had better be opposed by an AMD processor from the similar FX series.

But these explanations are convenient for AMD marketing. And what about us, users and buyers? Why violate a model range together with CPU series positioning strategy, when you are on your peak and outscore the competitor almost everywhere? Just for one red-letter day, for a momentary and actually not very relevant objective? Why not call the unfortunate FX-60, for example, Athlon 64 X2 5200+ to make it much more easy to place? A strange move. Strange and even absurd in long terms.

Perhaps, that's the only our gripe with the Athlon 64 FX-60: it's got the wrong name. It will hardly impact popularity and sales of this CPU. But both Intel and AMD are too frivolous with positioning their products. We have an impression that they deliberately jumble their convoluted model lines in order to confuse customers. We cannot really welcome that...

Write a comment below. No registration needed!