Nuclearus - New Build - Benchmarks

[SOLDNER-MOFO64]

Got myself a small boost....7676

--------------------------------------

ALL UPDATED

[tictac]

AB9 Pro
e6320
2 x 1GB Team Value 667
6200tc 256mb
Stock board
Bios 17 official

Vcore : 1.43V
Vdimm : 2.3V
VMCH : 1.45V
VICH : 1.25V

FSB : 457MHz
Strap : 1033
Dram ration : 1:1
Timing : 4-4-4-8-18-36-2T

12272

[tictac]

so need ver 2 to validate? ok will download it.

[Tsunami Fall]

Heres my second attempt, Going to try and break 8k soon

[ReaperX]

One from me,@400x9 with 1.42V.

[evoic]

I'd like to add my name to the list.
If the list has been updated recently, I think I'm the top E6400 / E6420.

11,149

[ceemic]

evoic, look at my e6400 score: http://ceemic.pri.ee/c2d/nmc11445.JPG

[evoic]

evoic, look at my e6400 score: http://ceemic.pri.ee/c2d/nmc11445.JPG

ceemic....look at my score and look at my VERSION.

[ceemic]

There's a mistake on "score list": I have e6400, not e6600!

[evoic]

There's a mistake on "score list": I have e6400, not e6600!

You're still not understanding properly.
You scored 11,445 using Nuclearus version 1.2.1, the new version is the one that is required to be added to the list.

Version 2.0.0

That's the version I used when I hit 11,149.

You may still score higher, but you have to score higher with the correct version to make it count.

[Lane-k]

Not tweaked.. just for try @4.2ghz

[xtant25]

Here is mine Pentium D 820 @ 3850 using a gigabyte GA-965P-DS3 (rev1.3)

[N00B]

My bench on the Asrock 4coreDual.

E6420@2440Mhz
Asrock 4coreDual-VSTA
Team Xtreem 8500
6800le Agp
Lc Power 550w

Got to get my E6420@Ds4.

[Kashelz]

Ok, here is update from me, 12.7k is what I get max for now I just changed onother CPU to test with and went 4+ Ghz!



And I cant understand, how can get E6600 @ only 3.463GHz and get 12,722??? Its 587 Mhz lower Mhz but scores more by 11 points? WTF? What I do wrong?

[freakBYnature]

And I cant understand, how can get E6600 @ only 3.463GHz and get 12,722??? Its 587 Mhz lower Mhz but scores more by 11 points? WTF? What I do wrong?

Here's some mind food.... try to play a game... say S.T.A.L.K.E.R. for about 30 minutes, exit, take the bench... write down the score... then reboot, wait until all your services has started... take the bench again... see a difference? 3DMark atleast works like that... always a better score on a fresh reboot.

[Kashelz]

Here's some mind food.... try to play a game... say S.T.A.L.K.E.R. for about 30 minutes, exit, take the bench... write down the score... then reboot, wait until all your services has started... take the bench again... see a difference? 3DMark atleast works like that... always a better score on a fresh reboot.

I always run benchmark on fresh load of windows, I think its better and more memory then is free, and 4.05Ghz on water is not so stable to play hard games Any more suggestions what tweaks to do?

[freakBYnature]

Sounds like you have a bottleneck-problem to me, but may be many things. If your cpu don't pass orthos it simply means that it don't calculates correct, and that is all a cpu is ment to do.... calculate stuff... so, if the wrong calculations get bottled up all the time then .... you see? That may be the reason for you clocking down getting more juice, since your RAM and CPU works more at an optimum at the lower clock + the cpu calculates correctly every cycle instead of every 1000nd cycle

I'm not sure how you are supposed to tweak it any more than what I just explained, I'm not familiar with neither your board, cpu or ram.... maybe it's time to go intel on my ass?

[evoic]

.... maybe it's time to go intel on my ass?

I had an Opty 146 @ 2.4 - 2.6Ghz that I loved in an extremely solid system.

I sold it to go dual core and I stuck with my beloved AMD by purchasing an X2 4400+.

I noticed absolutely zero performance increase and I was pissed.
Pissed enough that I decided to join the dark side and bought and Intel system.

The difference is night and day.
It pained me to say that for the first few weeks, but these days it's a point of pride.
Intel made an amazing CPU line and people get WAY more than they pay for when they plunk down the greenbacks for a C2D processor. (IMHO)
.
.

[Cheffy]

this benchmark will complete fine (and higher) for me at much above whats actually even slightly stable.

i can get 3100mhz stable for orthos... but even wow bails in a few mins.
this will on a good day and with flowing wind complete at 3220mhz with the same voltage, even while im getting explorer errors and random program lockups...

[Kashelz]

Sounds like you have a bottleneck-problem to me, but may be many things. If your cpu don't pass orthos it simply means that it don't calculates correct, and that is all a cpu is ment to do.... calculate stuff... so, if the wrong calculations get bottled up all the time then .... you see? That may be the reason for you clocking down getting more juice, since your RAM and CPU works more at an optimum at the lower clock + the cpu calculates correctly every cycle instead of every 1000nd cycle

I'm not sure how you are supposed to tweak it any more than what I just explained, I'm not familiar with neither your board, cpu or ram.... maybe it's time to go intel on my ass?

No, it cant be correct! I have Xeon 3060 @ 3700 Mhz for 24/7, its with 1.4v under TAT 30min load, no errors, all is completly stable like Orthos 8+ hours, TAT full load and other software that I have tested! And all is fast, maybe my OS is slow, I have not reinstaled it quit long time... And even my CPU @ stable 3.7Ghz is much slower than E6600 @ 3.4GHz, cant understand why, basically they both are the same CPUs

[pmp]

kiełbasa squad results below.
seems that nuclearus is picky.
better results was done using ram@1200 while lower was done on 1000/4.




Hmm, i compared two version - one which i used for above, and current one. this bench sucks as results are pretty different ;(

[Jaan]

Locki, I've tested now v2.0RC1 with 1/2/4/8 cpu system and I can tell you, that the less cores means more stable MultiThr bench result.
For 8-core sytem I have up to 2000 points delta- this is not good.
And ALU test result is allways better for the first time bench, MT is usualy not. FPU result stability is OK.
So imo bench is still need impromens for result stability (MultiThread part of it)

A few result from me too:
singlecore
old ES CedarMill 631@6510MHz/1,712V/LN2 -181,7- -180,0*C during bench (no cold bug ) Forget remove maxmem=600, so maybe it is a reason for slow result:


and 8-cores result- Intel V8 system, all default. (8cores@3GHz)

This FB-RAM is so slooow

[Locki]

Jaan! expressly for yor powerfool system, I made new build "NMCv2 RC2!" with tabsheet "options" where you can increase duration of all subtest... by 1..100 times
http://cp.people.overclockers.ru/cgi...clearMCrc2.rar

[aspms*]

Here's my first 13K run:

Rig:

C2D E6600 L629F @ 4.12Ghz 1.63V
CPU coling: Thermalright Ultra 120 eXtreme + Sharkoon Silent Eagle 2000
Gigabyte P35-DS4 @F3
2x1Gb Crucial 10th Anniversary edition @ 572 5-5-5-8 2.2V (I'll try later to reach the same frequency with higher voltage and CL4 timings )

[Lancer33]

Second AMD?