• 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.

Has Anyone seen a 1T vs. 2T command rate thread for A64's?

Page 5 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: Zebo

No way you can run 1T with 4x512, only ram that can do this is crucial 8T.
Click to expand...

Ok, this thread pretty much found that there's not that big of a difference between running 1T or 2T (with 2x512, yes), so 2T with 4x512 would be....? Anyone done any tests comparing 2x512 vs. 4x512 at 2T? Or 2x512 vs 2x1GB? Preferrably on a Venice or San Diego.

I'm wondering because those of us that will be upgrading soon will be getting Venice, SD, or X2, and with memory prices so low, it'd be nice to know if we can pair up 4 sticks with one of the new AMDs (or a couple 1gb sticks)... and not have the performance hit and overclocking problems we used to get if we wanted to run higher density sticks, or 4 sticks with the old memory controller.

Hope my questions made sense, I'm not quite as hardcore as most of you, but I'd like to know what I can, and should, pair up in my new system.... for the longest time it has seemed like 2x512 2-2-2-5 1T was the "best" setup... now we see that 2T isn't as horrible as some thought, and even some looser timings can be used with the new steppings while still seeing gains in CPU cycles which offset the added latency. But I haven't seen any comparisons of different AMOUNTS of RAM.

Again, this wouldn't really have been much of a concern a month ago, but with memory so cheap now...there's really no reason not to add more, if it isn't going to hurt performance and/or overclockability.
 
Originally posted by: Starman
A few percent here, a few percent there, and all those percentage points quickly add up. Most people who want to see a negligible difference between 1T vs. 2T are also those who want to see negligible differences between low-latency RAM and Value stuff. What goes so often forgotten is that all these factors eventually add up and you're no longer talking about a negligible difference. Value RAM, 2T command rate, buying a slower processor...you've got to pick your poison and be careful not to drop the entire context of system performance variables. Before you know it you're no longer talking about 1-5% but a cumulative 10-15% or worse.
Click to expand...

:thumbsup:

Good info. However, I'd also point out that how one plans to use their system (as well as budget considerations) matters as well.

i.e. for those mainly interested in playing the latest games will see better results from spending the extra $150 to get the 6800GT over the 6600GT rather than spending $100 for more expensive ram, or going from a 3000+ to a 3500+.

For gamers, those little 2% increases here and there may add up, but it won't come close to matching the % increase you'll get from buying the best video card(s) you can afford (as long as it's not handicapped by the rest of the system).
 
Originally posted by: Starman
Originally posted by: zendari
So whats the bottom line? Is the 1T vs 2T thing real or rumor?
Click to expand...

It is real, but it appears the difference only varies from 1-5%. However, it's important to remember that this is just one aspect of achieving the best performance for your PC. A few percent here, a few percent there, and all those percentage points quickly add up. Most people who want to see a negligible difference between 1T vs. 2T are also those who want to see negligible differences between low-latency RAM and Value stuff. What goes so often forgotten is that all these factors eventually add up and you're no longer talking about a negligible difference. Value RAM, 2T command rate, buying a slower processor...you've got to pick your poison and be careful not to drop the entire context of system performance variables. Before you know it you're no longer talking about 1-5% but a cumulative 10-15% or worse.
Click to expand...

Thanks a million. 1-5% isn't bad.

by avalon
"Well, what if it was possible for anyone to overclock their memory enough in 2T mode to bypass that 1-3% performance hit and gain additional performance advantages?
That would be very interesting.
"

Does anyone have an answer to this?
 
Thanks a million. 1-5% isn't bad.

by avalon
"Well, what if it was possible for anyone to overclock their memory enough in 2T mode to bypass that 1-3% performance hit and gain additional performance advantages?
That would be very interesting.
"

Does anyone have an answer to this?
Click to expand...

It really depends on the individual machine and how much overclocking is involved.

If you can get more mhz out of your overclock with 2T, then go for it.

In my situation, putting it at 2T does not yield a better overclock as I believe my cpu has hit a ceiling. I'll definitely be running at 1T for that marginal performance boost.
 
See Review here

http://www.systemcooling.com/mushkin_pc3200_2gb-06.html

For comparison of 2GB Mushkin at 2-3-2-7 vs 1GB & 2GB of Corsair PC3200 LL at 1GB - 1T and 2GB - 2T.

Am looking for best 2GB RAM so far I've found Muskin 2-3-2-7 PC3200 and Crucial PC4000 3-4-4-8. I know latency is much more important in AMD systems that intel.

Which would be best to use in an X2 system? Or recommend somethign else preferably LL and fast as intending on overclocking X2 and is multi locked so need to overclock FSB equivalent. Any recomendations???

Bluemars
 
This thread should really get a sticky.

Someone with an X2 should run tests seeing if the dual core makes ram bandwidth increases more pronounced.
 
Originally posted by: Zebo
Isaneteks is flawed as you implied. He uses different mem speeds and processor speeds and then shoots for similar performance. I'm not really even sure the point there??? It certainly does'nt tell you how much would be gained by using 1T or 2T all things being equal. Maybe if I think about it I can interpolate...


EDIT:
No his results are all fubard. To many variables thoughout the three tests. Differnt processor speeds, mem speeds and different mem timings is three variables that make any kind of guestimates impossible.
Click to expand...

That honestly sounds like the most retarded testing procedure ever. He must be high to have that many variables. Scientific method out the window. In fact, there's probably no method whatsoever.
 
i had read in a prerelease article a few months back that the x2 series of cpu would allow 4x512 at 1t. What what i am reading a: this is not possible, b: doesn't really matter anyway. has anyone tried 1t 4x512? just wondering
 
My quick results:
winchester 3200 @2310
2x512@231, 2.5-3-3-10

1T:
source video stress test: 47.12 fps
winwar: 588kb/s
3dmark03: 5346

2T :
source video stress test: 46.88 fps
winwar: 530kb/s
3dmark03: 5330

%change 1T to 2T
source video stress test: -.51%
winwar: -9.86%
3dmark03: -.3%

i should add that my system is very cpu heavy and i'm not sure if the results would relate well to a more cpu bound situation in the game/synthetic benchmark.

Originally posted by: The Pentium Guy
Zebo once again, you've done it. You've convinced us to use value ram, and stick with 2T. Thanks for the information
-TPG
Click to expand...
thats a poor conclusion, if i've ever head one. As long as you're not using 3 or more dimms, there is as far as i can tell no reason to run 2t, and it has been shown to consistantly slower.
 
Wait, so which is better?
4x512MB low latency high fsb with 2T or 2x1GB low latency with low FSB and 1T?
 
Originally posted by: Fox5
Wait, so which is better?
4x512MB low latency high fsb with 2T or 2x1GB low latency with low FSB and 1T?
Click to expand...

Well I think the point of this thread is to prove that running 4x512 doesn't hurt real world preformance much at all.
 
Hello. New here and a very interesting thread.

Originally posted by: miketheidiot

Originally posted by: The Pentium Guy
Zebo once again, you've done it. You've convinced us to use value ram, and stick with 2T. Thanks for the information
-TPG
Click to expand...
thats a poor conclusion, if i've ever head one. As long as you're not using 3 or more dimms, there is as far as i can tell no reason to run 2t, and it has been shown to consistantly slower.
Click to expand...

Wouldn't it depend on whether you had a) value RAM already and you didn't want to fork out for more expensive memory and b) whether you were overclocking other parts of your system or not?

Dropping to 2t has allowed me to achieve a totally stable 800Mhz+ o/c on a A64 3000+ where my cheap memory, and the lack of working dividers on my Dfi Lanparty nF4 SLI-D mobo were limiting my overclock.

I'm glad to hear that in real world situations dropping to 2t appears to have a very small impact on overall system performance. 🙂
 
Well, I've been having issues with my Chaintech VNF3-250 and Mushkin Performance Series memory (2-3-3) as I couldn't go over 2.25 GHz w/o being unstable. But I never chagned the command timing to 2T.

After reading this thread, I decided to give that a try... and voila! I'm running stable at 2.45 GHz (245x10) and a very happy man! 🙂 Thanks again guys! I'm a 2T believer now!
 
Originally posted by: cronic
i had read in a prerelease article a few months back that the x2 series of cpu would allow 4x512 at 1t. What what i am reading a: this is not possible, b: doesn't really matter anyway. has anyone tried 1t 4x512? just wondering
Click to expand...

Alright I have 4x512mb on OCZ Gold running in 2T. How should I go about testing to see the preformance hit on 2T? I don't have any 1gb sticks, although I might be able to borrow some from a friend. What programs should I use to get realistic preformances differences?

And once again this thread needs a sticky.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top