Originally posted by: mechBgon
If you're looking for high memory bandwidth, might want to wait a bit longer for nForce2 to show up. Check out Evan's findings here: http://www.anandtech.com/chipsets/showdoc.html?i=1719&p=4 Looks like about 30% more bandwidth in TwinBank mode using DDR333, compared to KT400 using DDR333. And the latency is way lower (this is good ). Arrrgh, I want one already!
Keep in mind that the TwinBank feature requires two sticks of RAM. I notice that the review didn't test nForce2 in TwinBank mode, so its actual performance should be higher than what the graphs show... it was compared to KT333 with just one stick of RAM in each board.
(from ftp://download.intel.com/design/chipsets/specupdt/29824705.pdf)5. Sustained PCI Bandwidth
Problem: During a memory read multiple operation, a PCI master will read more than one complete cache
line from memory. In this situation, the MCH pre-fetches information from memory in order to
provide optimal performance. However, the MCH cannot provide information to the PCI master
fast enough. Therefore, the Intel® 82801BA terminates the read cycle early to free up the PCI bus
for other PCI masters to claim.
Implication: The early termination limits the maximum bandwidth to ~90 MB/s.
Workaround: None
Status: There are no plans to fix this erratum.
Originally posted by: cmdrdredd
well, I run Intel anyway so it makes no difference. I was looking to see if the KT400 chipset had some flaws in it