Elmer and Steve - RE: GamePC
Steve is right about GamePC removing the press release from THEIR website. The Athlon press release is the newest one now instead of the "Coppermine" press release.
Turns out I was correct that this PC maker had no idea what Coppermine was. Their description of the PIIIB used to say ".25 Micron Coppermine architecture" but now it says .25 Micron architecture." (Can you even call a process size an architecture?)
And yes, Chris Connolly, the guy who said "Coppermine" has a "significant" performance advantage over the PIII and Athlon, thought the PIIIB was "Coppermine". Here is the proof from his PIIIB review -
gamepc.com
"With a Voodoo3 3500 and the Pentium III at 4.5 x 133 Mhz (600 MHz), performance was incredible. We clocked Quake3 Test High Quality at 90.2 FPS (Athlon clocked at 78 FPS), and Quake3 Fastest at 140 FPS (Athlon clocked at 119 FPS), more than 20 FPS higher than an identical Athlon 600 testbed we setup right next to it. Quake2 Crusher scores were also much higher. We haven't had the chance to run a full WinBench suite, we'll save that for part II of the review."
His results are amazingly different from Anand's which show an Athlon 500 is faster than a PIII 600B and 600 in ALL Quake 3 benchmarks.
Chris DOES say AGP was overclocked 23MHz. And he didn't use DRDRAM but PC133 HSDRAM instead. (He didn't use a i820 board.)
Chris used a Voodoo 3 and Anand used a TNT2. Can anyone come up with a hypothesis to explain the disparity between the Voodoo 3 and TNT2?
If you can, look at this page gamepc.com you will see that with the Voodoo 3 3500 the Athlon 600 has a smaller lead over the PIII 600 compare to with a TNT2.
With the 3500, the Athlon is 9.4% faster in the Fastest demo than the PIII 600. In the PIIIB review, the 600B is 17.6% faster than the AThlon in the same Fastest demo w/the same video card. The resolutions of the two benchmarks aren't told, so I shouldn't be doing this - If you add the performance lead of the Athlon 600 over the PIII 600 to the supposed lead of the PIIIB 600 over the Athlon 600 you get the PIIIB 600 is 27% faster than the PIII 600! Again I know I can't just add these two percents together, but it gives an idea of what the "B" allegedly does.
Now, I ask, How in the world does the PIIIB 600 give such a performace benefit over the PIII 600?
Could it be soley because of the overclocked AGP and PC133 HSDRAM?
I again question this guy's credibility, this time because I am skeptical. ;) |
