Time for faster memory!

[Crashman]

Well, with PC3200 official, don't you guys think it's time they started using 4.5ns chips on the overclocker's modules? Or even 4ns? Or better yet, 3.6ns like so many video cards use?

I know the main limits to DDR right now are the modules themselves, but it seems they could squeze just a hair more out of them, to make memory stable at the fastest settings, at LEAST at stock PC3200 speed.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[slvr_phoenix]

I completely agree, but I don't think that anyone will be offering these super-fast chips in a dimm yet for two reasons:
1) The signal noise is a much bigger problem in a PC than in a graphics card so these kinds of speeds may not be possible not because of the ns of the memory chips but because of the DIMM sockets and longer bus lengths.
2) They haven't finished taking us for all that we're worth yet. Memory manufacturers are finally making good profits again. I doubt they'll want to release noticably faster speeds and kill their margins on PC3200 and PC2700.

That aside, I'd love to get my hands on some 4ns chips in a stick rated as PC3200 for a fair price.

"<i>Yeah, if you treat them like equals, it'll only encourage them to think they <b>ARE</b> your equals.</i>" - Thief from <A HREF="http://www.nuklearpower.com/daily.php?date=030603" target="_new">8-Bit Theater</A>

 

[Crashman]

Yes well, my thoughts are that if they used 4.5ns memory, the latency problems we're seeing right now could be isolated to the module itself, and they would be encouraged to work on that module until PC3200 was just as stable as everything else.

<font color=blue>Watts mean squat if you don't have quality!</font color=blue>

 

[JP5]

I guess in 6 months we will be saying, "Remember when PC3200 was hell of expensive and wouldn't run stable at low latency. I just bought 2 512 sticks for $50"

 

[slvr_phoenix]

Yes well, my thoughts are that if they used 4.5ns memory, the latency problems we're seeing right now could be isolated to the module itself, and they would be encouraged to work on that module until PC3200 was just as stable as everything else.

My guess would be that even if we did that we would only find that it is the memory bus itself that needs fixing. I think we need a system that is designed to run on lower voltages. Maybe we just need to move the memory sticks closer to the northbridge...

"<i>Yeah, if you treat them like equals, it'll only encourage them to think they <b>ARE</b> your equals.</i>" - Thief from <A HREF="http://www.nuklearpower.com/daily.php?date=030603" target="_new">8-Bit Theater</A>