<The systems were overclocked only in the sense that the CPU hadn't been certified to run at 133. If it had crashed, you could have blamed it on overclocking. It didn't. It was just faster on PC133. And it demonstrated that, unlike rambus 800, there are advantages to using PC133 over PC100.>
Dan, please, give it up. Tom's benchmarks were on a pre-production Camino chipset. OK? Pre-production. Wait for the official benchmarks to come out, and wait for tests to be run on the final version of the motherboard and chipset. OK?
By the way, how do you know Tom's platform didn't crash? Tom mentioned nothing about how stable his overclocked system was running. On that uberclocked BX platform, not only do the memory and processor bus overclock to 133 MHz, but AGP also overclocks to 88 MHz and PCI overclocks to 44 MHz. That does two things. First, it abnormally increases system performance. So much for PC133 showing the improvement that you hoped for. But second, stability suffers. I wouldn't trust my AGP and PCI peripherals running at that speed. Would you?
One more time, Dan, wait for the final version and the official benchmarks, both from Intel and from the third-party websites like Tom Uberclockermeister. I can assure you that Intel runs a lot of internal performance tests, and they know a lot more about the benefits and drawbacks of DRDRAM than you can ever hope to know. (One of those people happens to be the husband of a fellow coworker on my team.)
Tenchusatsu |
