AMD’s Athlon 64 4000+ and FX-55 processors

AS A CHILD, I loved to play Archon, the chess-like action/strategy game for 8-bit computers. I played often on my Atari 800, and once, I tried playing over at a friend’s house on his Commodore 64. He popped the disk into the drive, booted up the computer, and started loading the game. We waited.

And waited.

We went outside, played in an abandoned field, caught some crawdads in the creek, and came back inside. Still loading.

This was the Commodore 64, the computer outselling my beloved Atari 800 by some obscene ratio, the computer my friend had bragged about repeatedly, the computer helping run Atari’s home PC business into financial ruin. It must have taken 40 minutes for that game to load via the C64’s terminally lethargic 5.25″ disk drive. At about minute 39, something deep inside my head must have, silently and imperceptibly, snapped. You are now reading what I believe is the 40th CPU review that I’ve written for The Tech Report over the course of the last five years. These reviews have been dedicated to the task of providing the reader, with concrete certainty, information about how the most central component in a PC performs across a range of different applications. I suppose it’s my way of helping prevent another atrocity like the C64 disk drive situation—or at least of documenting it properly so people know what they’re getting into.

The folks at AMD must understand my dysfunction, because they have been producing ever-faster processors to feed my reviewing habit since day one. Today is no different; AMD is lifting the curtain on a pair of new CPUs that are faster than anything else they’ve ever produced, the Athlon 64 4000+ and Athlon 64 FX-55. As usual, we’ve run them through a broad spectrum of benchmarks and compared them to their closest competitors, and I can assure you, these things could play one very mean game of Archon, no waiting required.

AMD’s new top models
The faster of these two CPUs is the Athlon 64 FX-55. For the uninitiated, the Athlon 64 FX is AMD’s image part, one of those super-expensive products that exist to demonstrate to the world what a company is capable of producing. As the new Corvette is to the Chevy Cobalt, so the Athlon 64 FX is to the Athlon 64 3000+. If the FX series has the longest line on the benchmark graphs, the marketers reason, folks will want to purchase the cheaper Athlon 64s. The Athlon 64 FX’s foil is the Intel Pentium 4 Extreme Edition processor, Intel’s competing image product. Neither of these chips sells in any kind of volume, and they both cost more money than a French diplomat can siphon off of a U.N. oil-for-food program on a slow afternoon. They do, however, cast a halo effect on the rest of their makers’ products.

The Athlon 64 FX-55

AMD has committed to providing only one model of FX processor at a time, so the FX-55 supercedes the Athlon 64 FX-53. The difference between the two is all of 200MHz: the FX-53 ran at 2.4GHz, and the FX-55 runs at 2.6GHz. Otherwise, little has changed. The FX-55 still has 1MB of L2 cache, still drops into Socket 939 motherboards, and is still produced on AMD’s 130nm fabrication process.

All of that seems pretty simple. The real mind bender is the Athlon 64 4000+, which is basically the same darned thing as the Athlon 64 FX-53: an Athlon 64 processor clocked at 2.4GHz with dual memory channels and 1MB of L2 cache. The only real difference between the 4000+ and the FX-53 is the fact that the 4000+ doesn’t share the FX series’ unlocked bus speed multiplier for easy overclocking. That’s it.

Since the Athlon 64 3800+ has a 512K L2 cache and a 2.4GHz heartbeat, you could be forgiven for expecting the 4000+ to have the same size cache and purr along at 2.6GHz. In fact, AMD may someday introduce just such a product under the 4000+ moniker—AMD has been known to monkey around with its model numbers like that—but today is not that day. If I try to explain all the logic behind AMD’s model numbering scheme, not to mention the gaps in logic, I’m liable to start hyperventilating, so instead, I’ve worked up a quick table that attempts to capture all of the Athlon 64 model numbers to date. It may be incomplete, but I think I’ve got ’em all. Have a look:

Notice how different bits are traded off as equivalent, so that the move from a single memory channel (on Socket 754) to dual channels (on Socket 939) is considered equivalent to a 200MHz clock speed increase or an additional 512K of L2 cache. Except when it isn’t.

No doubt AMD will be making some hay out of the fact that it’s introducing a chip rated “4000+” just days after Intel announced it was dropping plans to take the Pentium 4 to 4GHz. However, the Athlon 64 4000+ is obviously just a renamed version of an existing product, so gloating would seem a little inappropriate at this point. Also, you should have a look at the cooler that AMD supplied with our review samples.

The cooler for the FX-55 (left) and AMD’s stock cooler for older models (right)

Is this what it takes to get the Athlon 64 FX-55 running comfortably at 2.6GHz? Perhaps not; it may be a little bit of overkill. Still, the outsized heatsink with integrated heatpipes certainly makes one wonder.


Test notes
Much has changed since we last did one of these big comparative CPU reviews. New games like Doom 3 and Counter-Strike: Source have arrived. Microsoft has released Service Pack 2 for Windows XP, and ultra-low-latency DDR400 DIMMs have debuted. Intel has transitioned the focus of its CPU lineup to new-look Prescott-core chips with fancy model numbers and LGA775 packaging, and AMD has begun introducing chips made on its 90nm fab process into the mix, including lower end Socket 939 processors. Naturally, we’ve assembled a field of competitors, test rigs, and benchmarks that reflects these changes.

One of the most notable additions is a 90nm version of the Athlon 64 3500+. We’ve already done preliminary testing on its power consumption, and now we’ll look at its performance versus the 130nm version of the same product. We’re also curious to see how the new Socket 939 version of the Athlon 64 3200+ stacks up against the Pentium 4 540 running at 3.2GHz.

Please note that several of our test CPUs are actually underclocked versions of other products. Specifically, the Pentium 4 model 540 and 550 entries are actually our Pentium 4 560 3.6GHz engineering sample, which came with an unlocked multiplier for testing at different speeds, running at 3.2 and 3.4GHz. Similarly, the 130nm version of the Athlon 64 3500+ is a down-clocked Athlon 64 3800+, and our Athlon 64 3200+ results were achieved by testing the 90nm Athlon 64 3500+ at 2.0GHz. For most intents and purposes, save perhaps for our power consumption tests, these underclocked processors should perform just like the real McCoys.

Unfortunately, we weren’t able to squeeze in any results from an Athlon 64 system with PCI Express. We’ve already looked briefly at the VIA K8T890 chipset in action, and PCI-E motherboards for the Athlon 64 based on chipsets from a number of different companies are reportedly very close. Fortunately, based on our experience with the K8T890, we anticipate that performance with PCI Express will track very closely with that of the AGP-based system we used for testing.

Our testing methods
As ever, we did our best to deliver clean benchmark numbers. Tests were run at least twice, and the results were averaged.

Our test systems were configured like so:

Processor Athlon 64 3200+ 2.0GHz (S939)
Athlon 64 3500+ 2.2GHz (90nm)
Athlon 64 3500+ 2.2GHz (130nm)
Athlon 64 3800+ 2.4GHz
Athlon 64 4000+ 2.4GHz
Athlon 64 FX-55 2.6GHz
Pentium 4 540 3.2GHz
Pentium 4 550 3.4GHz
Pentium 4 560 3.6GHz
Pentium 4 Extreme Edition 3.4GHz
System bus 1GHz HyperTransport 800MHz (200MHz quad-pumped)
Motherboard Asus A8V Deluxe Abit AA8 DuraMax
BIOS revision 1008 beta 1 1.4
North bridge K8T800 Pro 925X MCH
South bridge VT8237 ICH6R
Chipset drivers 4-in-1 v.1.11 beta (9/7/04) INF Update
Memory size 1GB (2 DIMMs) 1GB (2 DIMMs)
Memory type OCZ PC3200 EL DDR SDRAM at 400MHz OCZ PC2 5300 DDR2 SDRAM at 533MHz
CAS latency 2 3
Cycle time 5 10
RAS to CAS delay 2 3
RAS precharge 2 3
Hard drive Maxtor MaXLine III 250GB SATA 150
Audio Integrated VT8237/ALC850 with 3.64 drivers Integrated ICH6R/ALC880 with drivers
InGraphics GeForce 6800 GT 256MB AGP with ForceWare 66.81 drivers GeForce 6800 GT 256MB PCI-E with ForceWare 66.81 drivers
OS Microsoft Windows XP Professional
OS updates Service Pack 2, DirectX 9.0c

All tests on the Intel systems were run with Hyper-Threading enabled.

Thanks to OCZ for providing us with memory for our testing. If you’re looking to tweak out your system to the max and maybe overclock it a little, OCZ’s RAM is definitely worth considering.

The test systems’ Windows desktops were set at 1152×864 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

We used the following versions of our test applications:

The tests and methods we employ are generally publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

Memory performance
We begin with synthetic memory tests, because they’re just so intriguing. As always, these results don’t straightforwardly predict performance in real-world applications, but they do have some bearing on the performance of memory-bound tasks.

As expected, the new Athlon 64 chips are able to take good advantage of their dual channels of DDR400 memory. They do especially well in the Sandra test, which uses extensive buffering, tuning, and black magic to squeeze as much bandwidth out of a system as possible. The Pentium 4s have more memory bandwidth available to them, in theory, thanks to the 925X chipset’s dual channels of DDR2 533MHz memory, but the results don’t work out that way.

Cachemem is a little more relaxed, and probably more representative of many real-world apps. Here, the Prescott-based Pentium 4s do relatively better, probably due to Prescott’s very aggressive speculative pre-fetching of data from memory into the L2 cache.

Linpack lays bare the first few stages of the memory hierarchy for all to see, starting with the L1 cache, moving to the L2 cache, and then into L3 cache or main memory. The higher the number of MFLOPS, the faster the CPU is able to do a calculation with a certain data set size. You can see how the scores for all of the processors tend to drop as the data sets get larger and more of the work spills over into a slower part of the memory hierarchy. Also, note the difference between the Athlon 64 chips with 512K and 1MB of L2 cache, which is quite obvious here.

I want to take a quick detour to point out one really notable difference. Have a look at this:

These results were consistent across multiple benchmark runs, and I’m confident they are for real. The 90nm version of the Athlon 64 appears to have a slightly faster L2 cache than the 130nm version. Since performance is the same inside the Athlon 64’s 64K L1 data cache, I don’t believe this is a difference in the CPU core.

AMD has stated the 90nm and 130nm versions of the Athlon 64 are essentially the same, so I asked them about these results. All they would say is that for the 90nm parts, “some small optimizations were made in the memory controller and also in the way instructions execute.” I think this looks more like a change in the way the L2 cache is organized. AMD and Intel both pack their cache transistors in ever tighter over time, and such a change could result in higher performance, as well. Whatever the case, the difference in L2 cache performance appears to result in ever-so-slightly higher performance all around for the 90nm 3500+, as you’ll see.

Finally, we have memory access latency numbers, which are pretty clearly defined by CPU type and memory subsystem. The 80ns times for the Pentium 4 DDR2 systems are fairly decent in the grand scheme of things, but the Athlon 64s with integrated memory controllers and ultra-low-latency DDR400 are extremely quick at fetching data from RAM.

Memory performance (continued)
Next up is a slightly indulgent look at memory access latencies in more detail. If the following intimidates you, just skip to the next page with the gaming results. Remember, though, to flip back here if the boss is looking over your shoulder.

I’ve colored the data series below according to how they correspond to different parts of the memory subsystem. Yellow is L1 cache, light orange is L2 cache, and orange is main memory. The red series, if present, represents L3 cache. Of course, caches sometimes overlap, so the colors are just an interesting visual guide.

I won’t linger too long on these results, except to say that they confirm the Athlon 64 4000+ and FX-55 are exceptionally quick at getting out to main memory, especially compared to the competing Pentium 4 chips.


Doom 3
Let’s get right down to the gaming results now with Doom 3. We tested using a custom-recorded demo that should be fairly representative of most of the single-player gameplay in Doom 3.

Yikes. The Athlon 64s trounce the Pentium 4s in Doom 3, and not by a little bit. Our slowest A64, the 3200+, outruns the mega-expensive Pentium 4 Extreme Edition.

Far Cry
Our Far Cry demo takes place on the Pier level, in one of those massive, open outdoor areas so common in this game. Vegetation is dense, and view distances can be very long.

Relative performance in Far Cry is similar to Doom 3. Again, the Athlon 64s roll.


Counter-Strike: Source
This is a final, release version of Counter-Strike: Source that we’re using, available for purchase via Valve’s Steam distribution system. Our demo game takes place on the cs_italy map.

Three makes a trend, as CS: Source shows results similar to Doom 3 and Far Cry. The Athlon 64s are dominating performance in the latest games, and the FX-55 and 4000+ are always at the top of the pile.

Unreal Tournament 2004
Our UT2004 demo shows yours truly putting the smack down on some bots in an Onslaught game.

Once again, the Athlon 64 wipes up.

We decided to try something a little different with UT2004 and test CPU performance using its software renderer, just to see what happened.

For once, the Pentium 4 runs a game faster than the A64.



The newest 3DMark is largely GPU-bound rather than CPU-bound, as was 3DMark03. However, the Athlon 64 FX-55 still manages to produce the top score by a hair.

3DMark’s CPU tests are another story. Both of 3DMark05’s CPU tests disable hardware vertex shading, relying on the CPU to handle that load, and DirectX’s software vertex shaders are multithreaded. Interestingly enough, the CPU 1 test is also multithreaded, with a separate thread for an object path-finding calculation as might be used in a game. In that test, the Hyper-Threaded Pentium 4 3.6GHz comes out on top. Overall, though, Athlon 64s take the top three spots.


WorldBench application performance
Next up is a brand-new benchmark for us here at TR, WorldBench 5 from the folks at PC World magazine. With the death of Business and Content Creation Winstone, this seemed like a good time to explore other options, and PC World has just recently made WorldBench 5 available to the outside world. Like Winstone and Sysmark, WorldBench uses scripting to step through a series of tasks in common Windows applications. Also like those benchmarks, WorldBench produces an overall score for comparison. More impressively, WorldBench spits out individual results for its component application tests, allowing us to compare performance in each.

A default WorldBench run, which completes each scripted test at least three times, takes about seven hours to run one of our test rigs. At the end, the results are computed using a harmonic mean, which is essentially just a fancy way of averaging.

In WorldBench 5, an overall score of 100 is equal to the performance of PC World’s reference system, a nicely appointed Athlon 64 FX-51-based PC with a gig of RAM and a GeForce FX 5950 Ultra. I believe our test rigs have slower disk drives than that system, which leads to slightly lower scores than one might expect. There’s very little separation between the different CPU speeds, overall, which isn’t surprising for a benchmark that stresses the entire system, including the hard drive, memory, and graphics card. Regardless, AMD’s newest Athlon 64 chips take the top two spots in our tests. Among the lower end processors, the Intels look relatively strong, with the Pentium 4 550 edging out the Athlon 64 3500+ and the Pentium 4 540 finishing ahead of the Athlon 64 3200+.

Now, prepare for a torrent of individual application test results.

Both of the 3ds max tests focus on using the application for modeling and animation, not doing final scene rendering. As a result, they lean heavily on the graphics card, although the CPU can make a tangible difference, as well.

The Athlon 64 4000+ and FX-55 cement their top-dog status with solid performances across a range of tests. This is only part of the WorldBench suite, but I’ve stopped here to call your attention to a few things. The final three graphs are interesting for various reasons. One of the WorldBench developers told me that memory access latency plays a big part in Mozilla performance, and you can see that he wasn’t kidding. All of the Athlon 64 systems excel there. Windows Media Encoder is another story; the Pentium 4 has long been a strong performer in video encoding applications.

The most interesting result is the multitasking test, which combines Mozilla and Windows Media Encoder. Despite the potential benefits of Hyper-Threading, the Athlon 64 manages to perform relatively well here, no doubt on the strength of its excellent Mozilla performance.


WorldBench application performance (continued)

The Nero test depends greatly on disk controller performance, and the 925X chipset contributes to a big victory for the Pentium 4-based systems. I’ll be curious to see whether other Athlon 64 chipsets can match the 925X in this test.

WorldBench’s MS Office test is another that stresses multitasking, switching between various Office apps as it runs. The Pentium 4s would appear to benefit from Hyper-Threading here, since they perform relatively well.

We’ve been looking for a good Photoshop benchmark for a while now, and it appears we finally have one. Lo and behold, the Athlon 64 looks really strong in Photoshop, with the 3500+ outperforming the Pentium 4 560.

We’ve long assumed that the Pentium 4 is the processor of choice for video editing, but these results show how much performance depends on the application. Adobe’s Premiere runs considerably faster on the Athlon 64, while Roxio’s VideoWave Movie Creator runs best on the Pentium 4.

The final test in the WorldBench suite is WinZip compression, where the Pentium 4 excels.

All in all, the creaters of WorldBench have assembled a very nice set of real-world application benchmarks, and I’m nearly quivering with glee at seeing execution times in seconds for the results. Previously, in order to get results of this quality, we’ve had to use our own set of individual application-based benchmarks. Speaking of which, let’s see how the new Athlon 64 models handle those tests…


Sphinx speech recognition
Ricky Houghton first brought us the Sphinx benchmark through his association with speech recognition efforts at Carnegie Mellon University. Sphinx is a high-quality speech recognition routine. We use two different versions, built with two different compilers, in an attempt to ensure we’re getting the best possible performance.

The new Athlon 64s perform well in Sphinx, but they can’t quite match the P4 Prescott, which has excelled in Sphinx since its introduction. Oddly enough, the Pentium 4 Extreme Edition winds up near the back of the pack here, in part because it’s derived from the very different Northwood core.

LAME MP3 encoding
We used LAME to encode a 101MB 16-bit, 44KHz audio file into a very high-quality MP3. The exact command-line options we used were:

lame –alt-preset extreme file.wav file.mp3

I recently encoded the new Proto-Kaw CD using these same settings, and the resulting MP3s sound excellent.

Any of these CPUs will rip through an MP3 encode much faster than real time, but the FX-55 does it fastest.

DivX video encoding
We used the default settings for the DivX codec to encode a 3000-frame sequence from a DVD-formatted MPEG2 source file.

Even these new Athlon 64s can’t keep pace with the Pentium 4 in DivX encoding. Notice how the Athlon 64 4000+ basically ties the 3800+. The only difference between these two processors is L2 cache size, which doesn’t matter much for encoding a video stream.


Cinebench 2003 rendering and shading
Cinebench is based on Maxon’s Cinema 4D modeling, rendering, and animation app. This revision of Cinebench measures performance in a number of ways, including 3D rendering, software shading, and OpenGL shading with and without hardware acceleration. Cinema 4D’s renderer is multithreaded, so it takes advantage of Hyper-Threading, as you can see in the results.

The combination of Hyper-Threading and SSE2 optimizations is a potent one on the Pentium 4, and it shows here. The new Athlon 64s perform respectably, but can’t quite match the Pentium 4.

The AMD chips take two of the three shading tests in Cinebench, redeeming themselves a little.


POV-Ray rendering
POV-Ray is the granddaddy of PC ray-tracing renderers, and it’s not multithreaded in the least, because it’s designed to be a cross-platform application. POV-Ray also relies more heavily on x87 FPU instructions to do its work, and it contains only minor SIMD optimizations.

This one is just a showcase for the Athlon’s excellent floating-point unit. We keep it around because it’s a useful test of X87 FPU performance, but the Cinebench results should be considered more representative of a modern rendering application.


ScienceMark is optimized for SSE, SSE2, 3DNow! and is multithreaded, as well. In the interest of full disclosure, I should mention that Tim Wilkens, one of the originators of ScienceMark, now works at AMD. However, Tim has sought to keep ScienceMark independent by diversifying the development team and by publishing much of the source code for the benchmarks at the ScienceMark website. We are sufficiently satisfied with his efforts, and impressed with the enhancements to the 2.0 beta revision of the application, to continue using ScienceMark in our testing.

Performance in the three ScienceMark simulations tracks pretty closely with the results we saw in POV-Ray, suggesting these tests use the x87 FPU for heavy lifting.

The BLAS tests use different code paths to handle matrix multiplication. With the proper optimizations for SSE and SSE2, the Pentium 4 is very hard to beat. However, the Athlon 64 performs well across a range of code paths.


picCOLOR image analysis
We thank Dr. Reinert Muller with the FIBUS Institute for pointing us toward his picCOLOR benchmark. This image analysis and processing tool is partially multithreaded, and it shows us the results of a number of simple image manipulation calculations. We’re using a new build of picCOLOR this time out; it removes the video tests, which are highly dependent on the chipset and video card, from the calculation of the overall score.

The new Athlons just barely sneak past the Pentium 4 560, although the 4000+ is tied with it for all practical purposes.


Power consumption
I’m a little disappointed in our power consumption testing because we ran into a big, hairy snag. This problem affected the previous Athlon 64 90nm tests we performed, although we were unaware of it at the time. It seems the CPU voltage option on our Asus A8V Deluxe motherboard isn’t actually functional; we could set the proper voltage for each CPU in the BIOS, but no matter what, both the AsusProbe monitoring software and CPU-Z reported 1.6V for the CPU. It’s possible this problem was caused by the revision 1008 beta 1 BIOS we were using, but that BIOS was necessary for compatibility with the 90nm Athlon 64. This problem was exacerbated by the fact that the A8V Deluxe is the only motherboard we had on hand that would POST with a 90nm Athlon 64 processor. As a result, all of the readings you’ll see below were taken with the CPU voltage at 1.6V (although we did set the proper value on the BIOS for each processor). Generally, 130nm Athlon 64s are supposed to run at 1.5V, and the 90nm flavors expect 1.4V. Take the results as you will, or ignore them if they offend your precise sensibilities.

The Pentium 4 is more complicated, because voltage specs for Prescott processors are set at the factory and may vary from one CPU to the next. The general expectation is about 1.4V, and we used the closest manual setting on our Abit AA8 DuraMax mobo, which was 1.3875V. The P4 Extreme Edition 3.4GHz ran at 1.575V.

We measured the power consumption of our entire test systems, except for the monitor, at the wall outlet using a Watts Up PRO watt meter. The test rigs were all equipped with OCZ PowerStream 470W power supply units. Unlike our last set of tests, ambient temps in Damage Labs were down in the low 70s, much closer to a sane room temperature. The idle results were measured at the Windows desktop, and we used Cinebench’s rendering test to load up the CPUs. For P4s, we used the multithreaded version of Cinebench to take advantage of Hyper-Threading.

Even with a handicap, the Athlon 64s consume much less power than the Pentium 4 Prescott chips. The battle at 90nm is worth a second look: the Athlon 64 3500+ pulls less power under load than the Pentium 4 550 does at idle. Yow.

That said, the Athlon 64 4000+ and FX-55 aren’t exactly miserly when it comes to electricity. The FX-55 system pulls 190W at the outlet when it’s really cranking, and that’s without the GPU going full bore.


High-end CPUs like these generally don’t have lots of headroom left for overclocking, so I was pleasantly surprised by what I got from these two processors. The FX-55 was the real star, because its unlocked multiplier allows for very low effort overclocking—just set the multiplier a little higher and go. In a few tries, I was able to determine that the FX-55 would run just fine at 2.8GHz, but no more. The system wouldn’t POST with the CPU set for 2.9GHz.

The 4000+ suffered from a weakness of our Asus A8V Deluxe mobo. This is an earlier revision of the board that lacks an AGP/PCI lock, so overclocking the system bus makes the whole system run out of spec. As a result, we were limited to about a 220MHz bus, at which point the 4000+ seemed very happy at 2.64GHz. I was able to get it to POST and boot into Windows on a 225MHz bus at 2.7GHz, but the system crashed and corrupted my Windows installation. I suspect this 4000+ would handle at least 2.7GHz just fine in a mobo with a proper AGP/PCI lock.

The test results below aren’t as stellar as they could be for the overclocked 4000+ because I was playing it conservative with the RAM, which was clocked at 366MHz with 2-2-2-5 timings. A little extra tuning might have yielded better performance with 440MHz RAM.

Like I said, any overclocking one can squeeze out of high-end CPUs like these is a nice bonus. In this case, both of our review samples showed some promise. Your mileage may, of course, vary.

The Athlon 64 4000+ and FX-55 are easily the fastest x86-compatible PC processors available, and we haven’t even gotten into testing 64-bit code that could widen the gap. The new Athlon 64 models are faster overall than anything Intel has to offer, and their gaming performance is particularly strong. As a wave of highly-anticipated new games has hit the market over the past few months, AMD’s lead in gaming performance has grown even more pronounced. Right now, Intel is struggling to get to the next rung on the performance ladder, and AMD is stepping on Intel’s fingers.

That said, AMD is definitely proud of its new babies. The 4000+ will list at $729, and the FX-55 weighs in at an eye-popping $827. Only Intel’s ridiculous P4 Extreme Edition 3.4GHz costs more, at a cool $1K. I don’t recommend that you buy any of these processors, unless you have a government grant. The 3500+ and 3200+ models we tested are much more reasonable options, and I’d half expect a price cut on these models to go alongside the FX-55 and 4000+ intro.

In fact, once you take price into account, the Pentium 4s start looking quite a bit more attractive than they might otherwise. Throw out the P4 Extreme Edition; the Prescott-based P4 560 is often faster than the Extreme Edition, and the 560 lists for only $417. The 560 is also much cheaper than the Athlon 64 3800+, currently priced at $643. AMD may be beating Intel in performance, but the value proposition isn’t there until you work your way down to the 3500+. Even among mid-range processors, outside of gaming performance, the Pentium 4 and Athlon 64 offerings are generally competitive, as the WorldBench scores testify. The P4 550 3.4GHz outscores the Athlon 64 3500+, and the P4 540 at 3.2GHz edges out the Athlon 64 3200+. So Intel may be down, but it’s not completely out of contention, especially for non-gamers.

For gamers, though, the choice is simple: get an Athlon 64, and don’t look back. I feel fairly certain you wouldn’t have any trouble running Archon 2005, if such a thing were to magically appear. 

Comments closed

Pin It on Pinterest

Share This

Share this post with your friends!