DDR Status Report

Where Is It?

Good question, bad answers.

Motherboards

There’s two types of mobos out there, one based on the Ali Magik chipset, the other based on the AMD 760 chipset.

The first isn’t terribly good. The second is meant to get the ball rolling rather than be a real competitor. There’s persistent rumors out there that AMD is not producing too many chipsets with which to make the mobos. This makes AMD 760
boards rare and pretty expensive ($180 is a pretty good deal for an Asus A7M266). Pretty hard to recommend a mobo without
a multiplier adjustment to overclockers, especially with DDR RAM sky-high.

That brings us to Via.

Via is expected to come up with the chipsets that drive the DDR mobos the average person is actually going to end up buying,
just as the AMD 750 chipset quickly gave way to the KX133.

Problem is Via’s been having problems. Major problems. A couple samples have been snuck out and tested and they were just no good. Now the just-no-good boards basically reflected the state of the art a while back. In the last couple days, a prototype MSI board did get tested, and it was OK. The good news is that Via’s figured out their problems. The bad news is we’re not going to even start seeing the results until May at earliest, and given what we see from Via boards now, one should
be rather queasy about how the first out of the shoot are going to be like. I know I am.

Memories . . . .

GamePC just did a review of DDR RAM. The Crucial PC1600 memory did extremely well; actually much
than the PC2100 memory offered by Mushkin and Samsung. Great for Crucial, bad, really bad, for everyone else. There simply shouldn’t be that kind of difference in performance.

Waiting for Via

My sources are telling me Micron is ready, but they don’t want to start cranking these things out until there is big demand, and big demand in all likelihood means Via boards.

Skip This Dance?

2002 will mean a new form of DDR, which will not be compatible with current motherboards. There’s not going to be any miraculous improvement from that, either. PC2600 will probably give you 5% more oomph than PC2100, and PC3200 about 10% over
PC2100. So the overall performance improvement will probably be 15% for PC2600 over PC133, and about 20% for PC3200.

If you buy a Clawhammer in 2002; odds are you won’t have a choice in the matter, it’s mobo will probably require one of these new species of DDR.

Thoroughbred may be a different story, though I think PC133 is likely to start crippling performance with 2Ghz+ processors.

A lot of you have decided to skip this dance already, and I’m beginning to think that was the right move.

If you haven’t, you may not dance until summer.

Email Ed


Tags:

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>