• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD gets emotions

Originally posted by: Keysplayr
Wow, they must really need hits.

Every little bit helps right now probably. They aren't exactly scrooge swimming in a sea of cash. 🙂

Originally posted by: Idontcare
I thought DDR3 was the lowest Watt/GB/s memory technology out there, thanks in part to its lowered Jedec spec Vcc 😕 What is even lower power than jedec spec DDR3?

They do run at higher speeds than DDR2 though...maybe that has an effect? I'm not really sure.
 
Originally posted by: yusux
What does THIS have to do with VIDEO CARDS?

You're right. Is the CPU forum a better place for this? Although I think the same question would be asked if it was moved there.

To any mod: could you move this to CPU?
 
Originally posted by: thilan29
Originally posted by: Idontcare
I thought DDR3 was the lowest Watt/GB/s memory technology out there, thanks in part to its lowered Jedec spec Vcc 😕 What is even lower power than jedec spec DDR3?

They do run at higher speeds than DDR2 though...maybe that has an effect? I'm not really sure.

http://ixbtlabs.com/articles2/...d/ddr3-rmma-page1.html

DDR3 is supposed to be "less power at same bandwidth...or more absolute bandwidth at same power" kind of a trade-off relative to DDR2.

Sure you could construct the scenario where the DDR3 bandwidth is jacked so high as to cause the power consumption to be higher than a much slower lower clocked DDR2 system...but what is the point of making that argument?

I'm a tad baffled where the AMD guy was going with his statement. Sure DDR3-1600 consumes slightly more power than DDR2-800...but presumably your system is equipped with DDR3-1600 memory because you need the bandwidth to support an even higher IPC/Watt computing solution (which is what Nehalem EP does) over what you would have otherwise had were your system equipped with DDR2-800 memory (and thusly a lowered IPC/Watt situation).

A DDR2-1600 memory interface would provide the same bandwidth, and thus the same system-level IPC, but at markedly higher power consumption, and thusly higher IPC/Watt.

Surely this guy is not oblivious to such an obvious unavoidable consequence of the math, so what was the motivation to make this kind of easily falsified statement about DDR3 power consumption?
 
Back
Top