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

[meelk]

best thread ever. I looked all over for exactly this information. many thanks.

 

[imported_Soldier]

Originally posted by: proxopspete
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!

Well of course your going to see an increase in top speed because you relaxed the timings and lossened up the memory top end. You could probably go 2.5-3-3-7 1T and hit the same speed. The problem is you may be getting worse results with that faster speed. I had systems give much better results with slower cpu clocks and tighter timings. In my experience there needs to be a significant boost in ram speed before 2T is worth while. Too many people get hooked into thinking top cpu speed is all that matters and may not even realize there is still performance improvements to be had but at lower top speeds...


As far as 2T vs 1T I get studdering when using 2T under gaming so I wont touch it....Video encoding, Folding and SETI all give better results for me while running 1T. I have all 3 slots filled with Corsair XMS Pro 3200LL chips.

 

[CheesePoofs]

While skimming through the overclockign guide anandtech posted, I noticed that Jared continually talked about the disadvantages of 2T performance and how its worth running 100-200mhz slower to avoid using it. I asked him:

from what I've read from Zebo's thread in the CPU forums, 2T really doesn't have a significant impact on performance. Could you clarify this?

and he responded with:

2T command rate, as you can see in quite a few instances, really killed performance. Perhaps tweaking other special timings beyond CL, tRCD, tRP, and tRAS might make the impact less, but you could likely tweak the same things with 1T at a lower memory speed. Command rate comes into play on every single memory access, so doubling that delay will certainly have an impact on performance.

I don't get it .... nobody seems to see performance hits but Anandtech's benches. Maybe just synthetic tests take a hit? Also, does the 2T command rate really double the memory access delay?

 

[1Dark1Sharigan1]

My system (while not gaming) seems to run better overall when I use 310Mhz 2.5-4-3-7 @ 2T but I've noticed stuttering in HL2 that didn't happen before . . . I may have to switch back to the 250Mhz 2.5-3-3-7 @ 1T . . .

 

[Axon77]

Hey guys, I stumbled across this forum after buying some RAM on Ebay today. So, I think this is the main question:

Which is better 2X512 @1T vs. 4X 512 @2T?

As soon as I get it, I'm going to throw these puppies in and test them against the benchies I've run up to this point.

Here is my 3dMark03 with only 1 gb OC: 11523
http://service.futuremark.com/compare?2k3=4311749

Stats:

AMD 3500+ @2599
Thermalright XP-90 + Panaflo (60 CFM)
2X 512 OCZ platinum PC 4200 DDR RAM 1T
X800XL @ 432/525
Abit 3rd Eye

 

[Axon77]

Found It!


http://www.xbitlabs.com/articles/cpu/display/athlon64-e3-mem_4.html