Early Benchmarks of dual channel DDR for P4
" target="_new">In a clickable format</A>
Interesting and promising. For a Beta chip with Alpha drivers, it was up there with everything else.
I was quite impressed with the P4X333 showing too. Intel will have to be able to truely squash that chipset in performance before they can secure a mainstream market with it though, as well as the I850E with PC1066. The only advantage it has is it is the only officialy Intel product (The VIA issue as well as PC1066 not being supported by intel). Why pay double for less than 5% speed increase?
English is phun.
looks promising. Opinions across the net are mixed. Some say they were a little disappointed. Others claim its alpha stuff so it should improve a lot (as it matures).
I believe it will get better with time. Intel has been executing real well lately. There isnt much to indicate they will miss here.
Benchmarks are like sex, everybody loves doing it, everybody thinks they are good at it.
Intel can't afford any more miss and hit operations with AMD around, it's either do well, or lose some more market share. Intel has a greater leeway for making mistakes though, and I think with some of their designs *cough* *willamette* *cough* they could have done a bit more tweaking before delivering. Still, dropping Rambus the first indication of the right direction, and the DC DDR, be it Via or Intel, sounds like the way to go for the Pentium 4. Besides, we have yet to see a release design, optimization may boost it beyond 1066 RDR.
"When there's a will, there's a way."
Ya, that's what I was thinking. With dual channel DDR SDRAM, the limitation is again with the processor's FSB, not with the memory. Of course, it allows you a lot of overclocking headroom without having to worry about running into memory limitations. 166MHz FSB may be an option without going over spec in the memory, oh joy.
Quote:Doesn't DC PC-2100 DDR already max out the Pentium 4's mem bandwidth at 4.2 Gb/s? Seeing as PC-2700 gives 2.7 Gb/s for single channel, wouldn't DC be too much at 5.4? Or would saturating the memory bus help even further?
Oversaturating the Athlon's memory bus (with nForce or KT333) gains it a few percentage points; it's reasonable to expect the same with the P4. Mostly negligible, but with results so close, it could have been a deciding factor.
Looks like RDRAM doesn't have long for this world...at least in PCs. :wink:
<i>I can love my fellow man...but I'm damned if I'll love yours.</i><P ID="edit"><FONT SIZE=-1><EM>Edited by kelledin on 07/04/02 10:31 PM.</EM></FONT></P>
VIA can never make things right for those of us who have been here for a while simply by releasing improved products. A public apology is in order, an admission of guilt. Perhaps Intel will extract some revenge by hitting them where it hurts-the bottom line, through lawsuites.
<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>