Poll: Assuming you won't upgrade for about 3 years, would you get 2x512mb or 4x512mb?

[Duvie]

I can show you examples in video encoding where it made 1 sec in a 2 min clip....abiout 1%......In most cpu intensive apps it doesn't come close to 6%....

I suggest you test it yourself and let us know....Dont believe what you hear...most ppl are idiots and spew out what they hear and only further fuel the misinformation.....

 

[Duvie]

Quick test: everything the same except 1t vs 2t

cinebench 2003 = 0%

superpi 2mb = <3%

prime95 benc 2048K = >1%

Povray 3.5 chess2 = 0%

3dsmax7 rendering = 1%

winrar (file compression) = 7%


Definition of cpu dependent apps are like rendering and that appears to be int he 0-1% range...so for me no real hit....


I may do some rendering in viz as a test as well....

the TMPGenc was 1%

I dont see it ppl...tell me apps this will show up....

 

[stevty2889]

I wasn't sure if going to 2gig would make much of a differance, but I tried it out, because while playing star wars galaxies, I would get noticable lag, when loading areas with a lot of players or buildings, because it was paging. Upped it from 1gig to 2gig, and the lag dropped greatly. No more huge lag, no more paging. Oddly enough, not that I pay much attention to it anyway, my 3dmark03 score went up by 500 points.

 

[Zebo]

Originally posted by: Naustica
Future proofing is usually waste of money. Tech gets cheaper and faster all the time. Buy what you need now.

 

[Zebo]

This is so last year but Command rate is huge w/ A64s: For the life of me I cant find the link where 1T vs. 2t is tested for 939

Here's a qualitative summery

The Importance of Command Rate
Socket 754 Single-Channel motherboards performed best with a memory Command Rate setting of 1T in BIOS, but that generally was a stable option with only one DIMM. 2 or more DIMMs normally required a 2T Command Rate setting for most stable performance. There was a performance increase at the 1T Command Rate setting, but the real performance increase was very small.

Socket 939 Dual-Channel motherboards were found to exhibit a very wide performance difference between a Command Rate setting of 1T and a setting of 2T. The impact on memory bandwidth is dramatic between these 2 settings. In SiSoft Sandra 2004 standard buffered Memory Benchmarks, a 1T command rate showed a Sandra bandwidth of 6000 Mb/sec, while a 2T rate with the same 2 DIMMs in Dual-Channel mode was only 4800 Mb/sec. This is a huge difference in memory bandwidth and the Command Rate setting definitely impacts performance test results on Socket 939 motherboards. All AnandTech benchmarks were run at a Command Rate setting of 1T. This includes all benchmarks that were run in the CPU tests, as all benchmarks were rerun in the CPU tests as soon as we had verified the performance impact of Command Rate settings.

http://www.anandtech.com/printarticle.aspx?i=2069

I just did a simple test cuz it's late

winrar 3.4

1t = 721 KB/s
2t = 675 KB/s

 

[Blain]

Originally posted by: Naustica
Future proofing is usually waste of money. Tech gets cheaper and faster all the time. Buy what you need now.

:laugh:

 

[jdogg707]

Originally posted by: Gerbil333

Originally posted by: jdogg707
2 x 1GB? That way you can run 1T, and at full speed.

Expensive.

Not Really

And Again

 

[Blain]

Originally posted by: jdogg707

Originally posted by: Gerbil333

Originally posted by: jdogg707
2 x 1GB? That way you can run 1T, and at full speed.

Expensive.

Not Really

And Again

Those aren't low latency DIMMs. :shocked:

 

[jdogg707]

Originally posted by: Blain

Originally posted by: jdogg707

Originally posted by: Gerbil333

Originally posted by: jdogg707
2 x 1GB? That way you can run 1T, and at full speed.

Expensive.

Not Really

And Again

Those aren't low latency DIMMs. :shocked:

Who cares, the performance impact from going to 4 512MB sticks and going to either 2T or DDR333 will have a much bigger impact than the latency of the DIMMs themselves, especially on the A64 platform.

 

[Gerbil333]

I can't find anything specifying if that RAM is 1T/2T. I know that some of Kingston or Corsair's 1GB Value memory is 2T...

Edit:

I just pulled a stick of Corsair 256mb XMS2700LLPT from one of my machines and stuck it in my computer to run alongside my Kingston 2x512mb HyperX PC3200, bringing the total RAM to 1.25 GBs. I couldn't feel a difference. I went back to 1 GB and tried the same programs, and that didn't feel any slower than 1.25 GBs.

So, it looks like I can't utilize 2 GBs just yet...at all. The most RAM I'd ever used was about 1002 MBs according to the Peak Commit Charge. I know that includes Page File usage too, but usually when the Peak approaches the amount of physical memory, that suggests it's time for more RAM. After trying 1.25GB, WinXP managed to set a new record peak of 1070 MBs, but I'm going to attribute that to Windows XP's ability to take advantage of as much RAM as possible.

And since the majority vote that I shouldn't upgrade to 2GB yet, here's my plan:

I'll keep my 2x512mb of HyperX PC3200 for now and replace the generic 2x512mb PC2100 in my junk computer with a set of $83.90 Corsair 2x512mb Value PC3200 (SPD timings of 2.5-3-3-8-1T). Then by the time I really do need 2GB, I can throw throw that in my socket 939 system, lower the timings, and be happy.

Edit #2: Now that I've been reading up on nF4 boards, it looks as though the DFI nF4 Ultra-D may not like my HyperX. I'm basically idling right now and only 585,740kb of my physical RAM is free. That Corsair 4x512mb Value PC3200 at $169.46 shipped is looking very attractive.

 

[Gerbil333]

Bump. The thread started out with a unanimous vote for not upgrading, but somehow nearly evened out after I kept posting, so I think I need more votes.