Squeeze your computer AOA EXIBITION

[Samuknow]

Finally getting close to 15000

ORB - Compare

[Samuknow]

ORB - Compare

Another

Still getting a little better. I think the 6400 is holding me back.


168228 on AM3.

[ccperf721p]

Good scores for the E6400 though. AM3 seems to like high memory speed and high FSB. Might be due to it not being so demanding on todays hardware so speed trumps power..

I grabbed my Dominators out of the closet and threw them in the board just to see how far I can take this Q9450 on the FSB. I have a feeling I will be reimaging my windows drive this weekend.

[ThunderRd]

Sam- what tool do you use to OC the card? if it's Rivatuner I have found a few settings that may help...

@Cliff- I'm tempted to see if I can get in the 12s on SuperPi, but I'm scared of the volts I had to use...what do you think? Should I leave well enough alone, or do you think I can raise the voltage long enough to get over 4G and get a good run in, say 2 or 3 minutes from boot to print screen? [see my last post]

[Samuknow]

I am using the EVGA tool. It works really well. I use ntune to make the card adjustments.

[ThunderRd]

Originally Posted by Samuknow

I use ntune to make the card adjustments.

I've used nTune before; you might want to try Riva for a laugh and tell me what you think of it. It's the most easily understandable and has the most options IMO. It also has separate control for shaders and core clock which you can also link if you want. (BTW shaders have the most positive effect on your folding speeds)

[Daniel ~]

Originally Posted by ThunderRd

Sam- what tool do you use to OC the card? if it's Rivatuner I have found a few settings that may help...

@Cliff- I'm tempted to see if I can get in the 12s on SuperPi, but I'm scared of the volts I had to use...what do you think? Should I leave well enough alone, or do you think I can raise the voltage long enough to get over 4G and get a good run in, say 2 or 3 minutes from boot to print screen? [see my last post]

Forgive my butting in here TB.

I just happen to have a rule of thumb for you!

How will you feel about this when you have to explain to your wife why 3.6 just didn't cut it, but 3.9 would have if only....":O}

I think a guy could probably make himself happy for quite a while with 3.6...but that's just a poor boy talking...":O}

[Daniel ~]

You guys are turning in some REALLY nice scores! congrats!

[ccperf721p]

Originally Posted by ThunderRd

@Cliff- I'm tempted to see if I can get in the 12s on SuperPi, but I'm scared of the volts I had to use...what do you think? Should I leave well enough alone, or do you think I can raise the voltage long enough to get over 4G and get a good run in, say 2 or 3 minutes from boot to print screen? [see my last post]

At 1.65v+ you are pushing voltage already. It's doubtful it would kill the CPU even at 1.7v, but it will shorten it's life some. If I am in a position to replace the unit, I don't give voltage much thought.

You are so close to the 12's now, I would start to look at memory timings and reducing latencies. Might try dropping down a divider and tightening up some timings. I get the same times from DDR2-1000 4 4 4 8 as I do from DDR2-1200 5 5 5 15, but I only have to run 2.2v to keep the 5 5 5 timings rather than 2.45 to run 4 4 4 12 up to 1000mhz.

[ThunderRd]

Originally Posted by ccperf721p

At 1.65v+ you are pushing voltage already. It's doubtful it would kill the CPU even at 1.7v, but it will shorten it's life some. If I am in a position to replace the unit, I don't give voltage much thought.

You are so close to the 12's now, I would start to look at memory timings and reducing latencies. Might try dropping down a divider and tightening up some timings. I get the same times from DDR2-1000 4 4 4 8 as I do from DDR2-1200 5 5 5 15, but I only have to run 2.2v to keep the 5 5 5 timings rather than 2.45 to run 4 4 4 12 up to 1000mhz.

Yeah, when I get a rainy day to concentrate I'll work on the memory a bit. I know I can get it to 7-6-6-6-18-1T stably, even when running DDR3-17000+ at 1:2, so I know there's some room there. I think you're right, I'd rather finesse it at this point than throw risky voltages around. Using a bigger hammer is rarely the best method

Originally Posted by Daniel ~

Forgive my butting in here TB.

I just happen to have a rule of thumb for you!

How will you feel about this when you have to explain to your wife why 3.6 just didn't cut it, but 3.9 would have if only....":O}

I think a guy could probably make himself happy for quite a while with 3.6...but that's just a poor boy talking...":O}

true...but a big difference 'tween that scenario and mine:

3.978G/13.016 secs looking for 4.0G/any sub-13 secs run. The difference between 442FSB, which ran, and 444, which probably needed more volts than I was willing to try. :\ Could you have happily walked away from that challenge? lol

Still, a nearly 1.6G overclock ain't bad even if it's only for a benchmark.

[chrisbard]

Its still too hot here to OC my GFX card. I have to wait a bit before going for that 06 score.

[Daniel ~]

Originally Posted by ThunderRd

Yeah, when I get a rainy day to concentrate I'll work on the memory a bit. I know I can get it to 7-6-6-6-18-1T stably, even when running DDR3-17000+ at 1:2, so I know there's some room there. I think you're right, I'd rather finesse it at this point than throw risky voltages around. Using a bigger hammer is rarely the best method





true...but a big difference 'tween that scenario and mine:

3.978G/13.016 secs looking for 4.0G/any sub-13 secs run. The difference between 442FSB, which ran, and 444, which probably needed more volts than I was willing to try. :\ Could you have happily walked away from that challenge? lol

Still, a nearly 1.6G overclock ain't bad even if it's only for a benchmark.

Not with a smile on my face ..no.

Especially as it's quite likely you'll hit your mark...I was was being overcautious.

I think it's because of recent hardware troubles of my own...in short...never mind...":O}