The Close spi1m to 8.xxxs from Coolaler

Why not a dual PI screen shot?

Such a stupid question nn_step ;)

Still weird SPI bug.

Quote:

---

Originally Posted by nn_step

Why not a dual PI screen shot?

---

same question here.

edit
use superpi 1.4mod
it's faster ;)

Quote:

---

Originally Posted by afireinside

Man so close :( Still very impressive coolaler :toast: Could you share details on the cooling? Is it solid? Brazed? Pics of the base?

---

i am wondering this as well

coolaler
Congrats!! It's a little more to 8.xxx.:toast:

Very interested to that calculation time.
I think that is usual even if he wondered seems from the last calculation time and value output time.
I don't understand why it is...
Though I also do not have the ulterior motive.:)

Quote:

---

Originally Posted by nn_step

Why not a dual PI screen shot?

---

Someone badly failed to catch the XS spirit :rolleyes:

Incredible job one more time Coolaler :clap:

Quote:

---

Originally Posted by LowRun

Someone badly failed to catch the XS spirit :rolleyes:

---

Yeah something about not being able to get excited by something that seems a hair off...

:slobber: that's one hell of a time
flat 9.000s is insane :woot:
(again) great result coolaler :toast:

Quote:

---

Originally Posted by nn_step

Yeah something about not being able to get excited by something that seems a hair off...

---

Share with us your wisdom nn_step.

Dam guys why is it always rib NN day?
Will he ever get off the chopping block?
I was thinking id also like to see a dual shot to see just how much it would raise the scores, say 9.25 9.4?
But if I remember correctly he has one of the cores disabled thru bios I think?

Nice work man..;)

Tras at 4 or 8 instead of 1, aint that a performance gain??

Coolaler, force the dog to look at the screen while making the run, that'll do:lol2:

nice coolaler :) let the 8.xxx sec fall :)

Next run, vcore to 1.8v and more and an easy sub 9 :fact:

Quote:

---

Originally Posted by nn_step

Yeah something about not being able to get excited by something that seems a hair off...

---

This is how you understand compliments?.. :shakes:

Lol! Go for a us english version of windows and you're gonna break the 9.000 barrier ...

great job! another WR ! 8.xxx is coming very soon ...

Quote:

---

Originally Posted by Sadolf

Lol! Go for a us english version of windows and you're gonna break the 9.000 barrier ...

---

Welcome to Paradise (XS)

Coolaler, send your time there please http://www.biwa.ne.jp/~yok/JustTime.html
Time(s)：
■Name：
■CPU：
■Clock(MHz)：
■FSB(MHz)：
■Date：
■Screen Shot：（リンク先）
■E-Mail：

Quote:

---

Originally Posted by cpulloverclock

Welcome to Paradise (XS)

Coolaler, send your time there please http://www.biwa.ne.jp/~yok/JustTime.html
Time(s)：
■Name：
■CPU：
■Clock(MHz)：
■FSB(MHz)：
■Date：
■Screen Shot：（リンク先）
■E-Mail：

---

Hmmm, lookin at that thread, it took Kaz-n just over 5ghz to get 10 flat. Theres no way a 500mhz increase is needed to take another second off, or it that true?? With kaz-n's tweaks and Coolaler's cpu, I think mid 8's would happen.

Wow, congrats!! You have truly become the Wayne Gretsky of overclocking!! Would you be willing to post a screenshot of your memset? it took me about 10 hours of memset tweaking to finally break 30s on my 805 (still waiting for newegg to get conroe!) I'm sure you've tweaked the hell out of your ram, but try bringing the 3rd ram timing down to 2 (3-3-2-1) I hit 3-4-2-3 on my PQI turbo cheapo ram. (little lower frequency however) Also, do you have hyper path 3 enabled or disabled? Ram throttling enabled/disabled?

Look here for spi glitch by stephen http://www.xtremesystems.org/forums/...ad.php?t=83366

I think It happens when your system is highly unstable but just barely stable to pass 1m spi.

Quote:

---

Originally Posted by kyosen

Hi cpulloverclock,

> Sorry Coolaler but something is weird in your 9.000s run
>
> Here is your 9.156s score
> Between the loop 19 and 20 there is a difference of 0.406s (normal)
>
> Your 9.000 screenie
> The difference is now at 0.140s... impossible except with a BUG
>
> I want an explanation... Bug for me or awesome tweak...
--
Interesting...but look carefuly again between loop 18 and 19.
0.642s consumed there.
In ~10s 1M run, 1 loop takes ~0.4s as you pointed out,
and the gap may vary longer or smaller in each loop.
I think that final 0.140s gap is derived/compensated from preceding longer calculation,
i.e. 0.642 - 0.4 = 0.142, and 0.142 + 0.140 = 0.282...hmm still smaller, but within feasible, I feel.

Of course I have no intention of blaming for anyone, you and coolaler:)
...If coolaler has special tweak for shortening final loop, please private mail to me;)

---

0.642 - 0.4 = 0.242

0.242 + 0.140 = 0.382

So, it is definately correct and not dodgy ;)

Quote:

---

Originally Posted by coolaler

some day i hope 8.999s come true:woot:

---

Try running the benchmarks in Windows' safe mode and/or with a higher priority in the Windows Task Manager. It should increase the amount of CPU time the process gets and give you a sub-9 second time.

Quote:

---

Originally Posted by ShiningArcanine

Try running the benchmarks in Windows' safe mode and/or with a higher priority in the Windows Task Manager. It should increase the amount of CPU time the process gets and give you a sub-9 second time.

---

lol he already uses realtime

The chances of a .000-time are like... one in thousand, and the chance of hitting a time as low as 9.000s is like... even smaller?

smooth :cool: