WinBench 99 Scores

	Intel Coppermine Pentium III 733MHz	AMD Athlon 700MHz
CPUMark 99	65.6	61.7
FPUMark 99	3850	3840
3D WinBench 99, Transform & Lighting	105	95.6
3D WinBench 99, Transform	206	186
3D WinBench 99, 1024x768x16	1220	1170
3D WinBench 99, large textures	16.6	16.4

3D GameGauge Scores

Jane's WW2 Fighters	45.49	43.95
Quake II Crusher	64.20	59.80
Expendable Average	58.77	53.39
Descent 3 Average	48.92	47.08
Powerslide Average	153.00	148.00
Madden 99	68.10	59.17
Battlezone II	34.00	30.00
Half-Life	49.20	48.70
Unreal	62.41	54.60
3D GameGauge	64.90	60.52

Quake III Test Scores

Q3Test 1.08 Demo1, 1024x768x32	31.70	30.80
Q3Test 1.08 Demo1, 512x384x16 "Fastest"	110.00	101.00
Q3Test 1.08 Demo1, 1600x1200x32	10.30	9.80

32-bit Scores

Jane's WW2 Fighters, 1024x768x32	37.99	37.44
Expendable, 1024x768x32	33.70	33.29
Quake II, 1024x768x32	52.10	50.80
Battlezone II, 1024x768x32	24.00	24.00
Powerslide, 1024x768x32	61.04	59.90

The results of the synthetic tests were quite interesting. We used Ziff-Davis' WinBench 99 CPUMark and FPUMark to get a gauge of raw CPU performance. The Coppermine CPU ran a little less than 10 percent faster than the Athlon, but the two processors were essentially dead even in FPUMarks. The Intel CPU was running at a slightly higher clock rate, which indicates the Athlon FPU is a bit better. However, it also illustrates the huge L1 cache on board the Athlon may have been partly responsible for the disparity in results between the 600MHz Athlons and the 600MHz Pentium IIIs as shown in an earlier Athlon feature article. (see http://www.gamespot.com/features/competition/index.html). Next on deck were the 3D WinBench tests. In both cases, the Pentium III results were roughly 10 percent better than the Athlon results, clearly, it wasn't just the clock rate difference at work here. If, as the FPUMark score indicates, the Athlon floating-point unit is a bit better, then the Intel CPU got its boost from the throughput gains using RDRAM and the 133MHz frontside bus. The larger cache of the Intel processor may have been a factor as well. The 3D WinBench WinMark scores didn't differ much, but many of the tests pegged at 100fps through much of the testing, so it's hard to draw any reasonable conclusion here. Interestingly, the 3D WinBench large-texture test scores, run at 1600x1200x32, were nearly identical, so even when stressing the memory subsystem, the Athlon didn't suffer much. Coppermine outpaced the Athlon in the 16-bit GameGauge scores across the board. Some of the differences may be attributable to the more efficient throughput of the Coppermine/Camino combination. The faster L2 cache may have also played a small role, but games tend to be less cache-dependent than other applications. Finally, some of the differences were no doubt due to the roughly 5 percent clock rate difference. Still, all of it indicates the Athlon core may not be that much better than the Pentium III, as previously thought. The 32-bit game scores were somewhat inconclusive, with the Pentium III showing small, but not statistically significant, differences. What seems to have happened is that Intel has come from behind and achieved parity with AMD. This is no doubt a relief to Intel, but it's still very significant that Athlon can stand up to Intel's latest and greatest and hold its own. Once AMD shrinks Athlon to 0.18 micron, AMD may once again have the opportunity to pull ahead. As before, AMD's biggest Achilles' heel continues to be the motherboard and core-logic chipset problems. I would have liked to run the AMD tests on a shipping off-the-shelf motherboard, but neither the FIC or Asus motherboards could complete all the benchmark tests. The Camino-based system, despite its beta nature, has been rock solid, even with the supposed RAMBUS problems (of course, we were only using two RDRAM sockets, not all three). So stability is still Intel's strong suit, and it's not a trivial one by any means. If I were a large OEM looking to minimize support calls, I'd look long and hard before choosing an Athlon solution.