1 gb vs 2 in overclocking

[Acanthus]

Originally posted by: Aries64

Originally posted by: Acanthus

Originally posted by: Aries64

Originally posted by: thanasi
why would I wanna start over when Im at 2.8 doing great I could sell those 2 sticks and get another 2 512s so I can have 2 gbs run it in 2t i heard that and 1t is no difference

Actually, there is a very discernable (i.e., realworld) difference in speed and performance between 1T and 2T timing.

Don't know what your system and RAM specs are, but if your RAM is capable of 1T try multiple-renderings of a large MB image with various filters at both 1T and 2T settings and I think you'll see what I mean. Its more noticeable with ultra low-latency RAM (2-2-2-5) but you should notice the difference even if you don't have ULL RAM. I've seen the difference in rendering time in Photoshop myself.

The difference in gaming is very pronounced too. We're talking about 8+ FPS on my system at default speed. Difference is greater at higher clocks.

Test it yourself, if its not an artificial bench, youre looking at 2% or less in games going from 1t to 2t.

Acanthus, those are MY observations from PERSONAL EXPERIENCE in real world applications and gaming - not artificial bench numbers. Thats why I said "I've seen the difference in rendering time in Photoshop myself". I have literally seen multiple redraws after edits in various layers in Photoshop.

And the 8+ FPS I stated earlier is also from personal experience on my system. In Halo v1.07 I get 78+ FPS at 2T, but I get 84+ FPS at 1T (CPU speed and all other settings being equal).

Obviously, every system and each game is different, but now that you know the numbers I got at 1 and 2T settings, maybe you should try yours? Out of curiosity, where did you get that 2% figure? Personal experience while gaming, running real world apps like Photoshop, or scores from synthetic benchmarks off "trusted" web sites?

Personal experience, and the massive thrad in the CPU forum that directly disputes what you just said.

weee

 

[Aries64]

Originally posted by: Acanthus

Originally posted by: Aries64

Originally posted by: Acanthus

Originally posted by: Aries64

Originally posted by: thanasi
why would I wanna start over when Im at 2.8 doing great I could sell those 2 sticks and get another 2 512s so I can have 2 gbs run it in 2t i heard that and 1t is no difference

Actually, there is a very discernable (i.e., realworld) difference in speed and performance between 1T and 2T timing.

Don't know what your system and RAM specs are, but if your RAM is capable of 1T try multiple-renderings of a large MB image with various filters at both 1T and 2T settings and I think you'll see what I mean. Its more noticeable with ultra low-latency RAM (2-2-2-5) but you should notice the difference even if you don't have ULL RAM. I've seen the difference in rendering time in Photoshop myself.

The difference in gaming is very pronounced too. We're talking about 8+ FPS on my system at default speed. Difference is greater at higher clocks.

Test it yourself, if its not an artificial bench, youre looking at 2% or less in games going from 1t to 2t.

Acanthus, those are MY observations from PERSONAL EXPERIENCE in real world applications and gaming - not artificial bench numbers. Thats why I said "I've seen the difference in rendering time in Photoshop myself". I have literally seen multiple redraws after edits in various layers in Photoshop.

And the 8+ FPS I stated earlier is also from personal experience on my system. In Halo v1.07 I get 78+ FPS at 2T, but I get 84+ FPS at 1T (CPU speed and all other settings being equal).

Obviously, every system and each game is different, but now that you know the numbers I got at 1 and 2T settings, maybe you should try yours? Out of curiosity, where did you get that 2% figure? Personal experience while gaming, running real world apps like Photoshop, or scores from synthetic benchmarks off "trusted" web sites?

Personal experience, and the massive thrad in the CPU forum that directly disputes what you just said.

weee

Yeah, interesting thread. However, that thread still does not change any of the results I have seen myself. I know what the real numbers are on my system.

Go ahead and call me a liar if it makes you feel better. Would you like me to email you the timedemo results from my system? I'll be happy to.

Unfortunately theres' no way I can show you the difference in gaming and the Photshop example I spoke about earlier in the thread, as you'd have to be sitting in front of my computer to see the speed difference in action.

 

[d33pblue]

Originally posted by: d33pblue

Originally posted by: Otter

Originally posted by: d33pblue
I'm running 4x512MB Corsair Value Select and I absolutely cannot get over 2.45Ghz stable with a Venice 3000+ on a DFI Ultra-D motherboard. I, like he, have treid everything, but I get BSODs (on bootup) like crazy over 2.45Ghz. Once in windows, though, its prime stable for 8+ hours at 2.45.Go figure.

Have you tried removing two of the sticks to see if it will overclock better?

Over on the DFI-Street.com forum people say to stay away from Corsair's value RAM.

What stepping and week is your Venice? I"ve seen a few reeports of LBBLE 0516 chips topping out around 2.5GHz.

Update: With only 1GB ram (2x512 CVS), I can run 2.61Ghz. Go figure

I have the LBBLE stepping 3000+

And whether or not it'll clock higher with 1GB of ram is irrelevant to me. I very frequently use more than 1GB of memory and that is more useful than an extra 100Mhz on the overclock. If I get bored one day, I might try it, but I seriously doubt I'll return to 1GB of ram.

 

[imported_Otter]

D33pblue, it occurs to me you just might not have enough power. At boot, everything runs at the same time, and hence, you can have power supply problems at boot that won't affect stability later on if you get lucky and make it through boot. What is your PSU?