ScienceMark thread (with score list)

the fact that the test is ram depenmdent would give my built in s754 ram controller the adv which maybe could be what made my test results better i cant think of nething ????

cdawall said:

is there any progream you want me to rnu to prove 100% stable it runs pcmark05 3dmark2001se aquamark3 and superpi so umm.... it could be a glitch but hey i am happy about 2nd place maybe if i tweak the ram some more i can get 1st im not that far away ohh and i think the prob is not the prog i think it is amd s754 setup i am running the htt at over 1000mhz and the ram at 325.3mhz but the primordia score is kinda outta whack so watever ive run it 3x and got sim results so what now we email sciencemark about glitches oh and temp wise (i was watching my temp prog the whole time) was never over 54C

Click to expand...

I'd just test & see what you can post here on this test (it IS a good test of Cpu/Memory stability)... take your time doing so, because this test helped me get the VERY LAST DROP of performance out of this beast!



* Just by taking the time to eck out what I could from o/c's...

I also (if you noted the photo I posted of some outrageous score) saw some "glitches" in this test, & they looked like the one you saw (except WAY more outrageous results)...

What it told me was it could complete THIS benchmark test, finally consistently enough, & thus meaning I was getting close to where I had to be mhz-wise on the CPU!

AND the memory ratio (I went from 184/200mhz unstable on RAM, to 197/200mhz STABLE)...

You're probably very close to where you are stable now, & the only things that helped me so far have been this benchmark, & running Folding@Home (both memory/cpu intensive)... when they got "nutty" (if & when they did)? I knew, back down a notch, & try again!

* Those other tests you note ought to be good too!

APK

P.S.=> Good luck & HOPE you can beat me... it is a painstaking hassle, especially in this heatwave out there today! apk

Alec§taar said:

I'd just test & see what you can post here on this test (it IS a good test of Cpu/Memory stability)... take your time doing so, because this test helped me get the VERY LAST DROP of performance out of this beast!



* Just by taking the time to eck out what I could from o/c's...

I also (if you noted the photo I posted of some outrageous score) saw some "glitches" in this test, & they looked like the one you saw (except WAY more outrageous results)...

What it told me was it could complete THIS benchmark test, finally consistently enough, & thus meaning I was getting close to where I had to be mhz-wise on the CPU!

AND the memory ratio (I went from 184/200mhz unstable on RAM, to 197/200mhz STABLE)...

You're probably very close to where you are stable now, & the only things that helped me so far have been this benchmark, & running Folding@Home (both memory/cpu intensive)... when they got "nutty" (if & when they did)? I knew, back down a notch, & try again!

* Those other tests you note ought to be good too!

APK

P.S.=> Good luck & HOPE you can beat me... it is a painstaking hassle, especially in this heatwave out there today! apk

Click to expand...

i am pretty sure i did beat you but yeah i guess ill notch my stuff back and run again but remember on both of my 2.44ghz test i got extremely high scores so lets see wat i can do this winter

cdawall said:

i am pretty sure i did beat you but yeah i guess ill notch my stuff back and run again but remember on both of my 2.44ghz test i got extremely high scores so lets see wat i can do this winter

Click to expand...

Well, test her out stable (if you have tested stable to your satisfaction, then so be it)...

However, I would test more...

See, I say that, because that result with the glitch on Primordia section in your result I saw, just made no sense!

It inflated your score, it happened to me as well!

Again: Do see the photo I had with one of those in it (Molecular test result was outrageous by comparison to all others run, first key point!)

Second key point noting err/abend in test program was:

Progress bar outside of groupbox border also KEY THING TO NOTE IS THAT

(Much like yours was, except in the Primordia test on your system)

Most of these, if you note? All score around the SAME relative numbers in each test, not HUGE variances, & certainly not those that go outside groupbox border by progress bar (definitely a glitch)...

The photo I posted with the SAME 'glitch'?

It would set the world record probably (& I KNOW this machine won't do that!)

Then get back to us... put that pic up & all that!



* It's possible you beat me, but I had to play around with my system a great deal to get what I did out of this testscore, but moreso for stability out of the settings I used to get it!

APK

_33 said:

BTW Alec, if you want to test how hot is your CPU CORES, I have *THE* program for you. It is called Core Temp Beta 0.9.0.91. It will tell you precicely the core temp reading from each of the cores in your X2 proc! It may be a scary view. Not only that, but the censors might be wacky in your cores and not report right temps (that could happen, not the most reliable part of the CPU).

Click to expand...

Thanks man... I will use that!

EDIT PART:

I have, works GREAT (nice simple & to-the-point miinimalist design, & yet fully feedback featured too)...

Works on AMD as well!

(Lists Intel stuff, iirc, ONLY in the readme though, got me a bit spooked)...



* "TEMPS MATTER HUGE!"

(Especially in this heatwave... ugh!)

APK

P.S.=> 33, did you write this one? apk

cdawall said:

the fact that the test is ram depenmdent would give my built in s754 ram controller the adv which maybe could be what made my test results better i cant think of nething ????

Click to expand...

Nope...as far as I know.....basically ALL AMD64 CPU's have the Onboard memory controller.

Lets look at the facts.....your on the Oldest 64bit AMD platform...your running PC2100 memory.......nothing In your setup eclipse's the Hardware from most any other tester....yet you think your score Is legit?....If you check back a few pages like Alec recommended....then you would read that Thats the way you see If your system Isn't stable at the current configuration.

If you honestly think your 2.4gig with PC2100 memory Is going to spank a FX60 Primordea score........slide the pipe over here would ya.

fyi at 2.44 i have a higher clock both htt and cpu than an fx60 so umm... yeah im just going to drop it back some and then we will see i will start at 2.4 and move up i am sure that will ake u'all happy andyes i read alec super score i posted my first score around there and you know wat the time diff between the tests when i ran the primordia test were cut in half from stock to 2.44ghz so glitch or no it was some major improvement


edit- the 1st test was at 2.4ghz the 2nd 2.35ghz still high in primordia?? no idea what is wrong there

Yet again - Achilles said it best in the film "TROY":

Achilles from the film "TROY" said it best:

Alec§taar said:

"Is there NO ONE ELSE!?!"

Click to expand...

APK

i am going to say it again is there any program thst i can run to prove complete stability i just ran cpu burn in 4 30min. no issue at any speed up to 2.44ghz maybe thru some fluke amd s754 a64 3000+ just has primordia beat

hello alecstaar,
i beat ya lol



this was at 4.3ghz with the mem at 833.look at the bas/flops its pretty high eh?.can you jump over me now?.(metaphorical gauntlet thrown lol)

here ya go-

cdawall said:

i am going to say it again is there any program thst i can run to prove complete stability i just ran cpu burn in 4 30min. no issue at any speed up to 2.44ghz maybe thru some fluke amd s754 a64 3000+ just has primordia beat

Click to expand...

More than likely your Fluke results are a result of you running your HTT Bus past the Limit of 800 that the 754 platform has......back your HTT bus down to 800 and then re-run the test.

sorry i thought you was asking me.i think i can get higher,but it just wont run the test at 4.4ghz.it does the first coupla tests then bsods.i'm not sure if it is the ram been too high or not enuff vcore.i had it at 1.438.it might need more but my cooler is really been pushed to the max at that.

i've seen some guys 930 at 4.8ghz,i just cant seem to get mine to windows at 300fsb or more.it'll post at 4.5 but will not do windows.more tweaking i guess.i would love to try it will water cooling.

tigger69 said:

hello alecstaar, i beat ya lol

Click to expand...

Yes, you did... from 1317.07. -> 1395 -> 1424.36!

All of your progress bars & such are w/ in their groupbox border, + all scores are right around one another consistently...




Looks legit: Good Job! You did put me into 4th place, imo... & you certainly do NOT "give up" either... lol!

tigger69 said:

this was at 4.3ghz with the mem at 833.look at the bas/flops its pretty high eh?.can you jump over me now?.(metaphorical gauntlet thrown lol)

Click to expand...

Probably not today @ least - we are due for an even hotter day than yesterday, & that tipped in @ 100F! It's 7:30 a.m. & I am drinking my morning cup of coffee & can feel this much already, & the forecast of the weather is seconding it...

Plus, I was pretty near my limits on my first score 1390 -> 1399 -> 1401 imo, @ least, especially if I want to keep it this way & run stable over relatively long durations (very important to me, & part of WHY I did this test no less - helped me eck out that LITTLE BIT MORE outta my machine & proved it's a GOOD stability test too!)

My RAM - it kills me on overclocks, trying to get higher than what I can now!

(Tatty One, Ketxxx, & POGE all feel the same: That I need to purchase better than stock oem DDR-400 & go @ it w/ out having to use "divider speeds" like DDR-333 as I do (but, I am @ 197/200mhz rated memory speed, so, not all bad - fairly "tight" result))

They were my "mentors/tutors" in doing my "Ram timing-chain re-tighten" phase of my overclock & came to that conclusion, as I did... & I definitely think those guys know what they're doing in this regard...

Fact is, I even tried to "volt it up" yesterday (Corsair 2x256mb sticks ordinary DDR-400)... however, it won't do more than 2.6v stock (e.g.-> 2.65V-2.7V) for very long thru this benchmarks' tests which also function (bonus) as a stability test of an overclock!

* So, there's not much point in me trying that tack to increase RAM performance here, nor "push for more" @ least not today, as it will be TOO hot - & I definitely will need better RAM to go @ my overclocks @ 12 mult. (rather than 11x I use now) & DDR-400 speed as a starting point (rather than DDR-333 as I use now).

APK

P.S.=> I really don't know what to make of cdawall's results, other than to point out the fact his Primordia results progressbar shooting past their groupbox's borders (much like my "super-result" photo did, albeit doing that on the molecular test #1) & being WAY outside of the group of his other results... he does seem fairly genuine & adamant his results are "straight" & that his system is stable, & we have NO way of knowing what is what there, unless we sat @ his machine testing w/ him w/ other apps for longterm stability (which is what I am after outta this - my highest possible o/c, & a stable result)... apk

when i had my amd,it was the generic ram letting me down.the ram i have now is'nt too bad.it was only £100 for a matched pair.

i dont know why it wont do 4.4 stable.i dont know if it is the ram or the vcore or the board that just wont do 300fsb.i guess it must be quite stable as it is now tho' to pass the science mark at 4.3.i guess more tweaking is in order when i can be arsed.

the combination of this board and chip seems to be ok.i am quite impressed with this 930.i really thought a 4800 x2 would beat my chip.i guess it thrashes mine on games tho'.

i just dont see that the chip cdawall has can beat mine and yours,it just doesnt seem right,his parts are'nt exactly high end either.

am i seeing things or was there a reply by tatty one here a second ago?

tigger69 said:

when i had my amd,it was the generic ram letting me down.the ram i have now is'nt too bad.it was only £100 for a matched pair.

Click to expand...

What kind do you use now? Both yourself & Steevo have better RAM now, & I would like to start getting some "feedback" on the BETTER & more overclocker-ready RAM out there... thanks for info.!

(Might as well start inquiring now imo)

tigger69 said:

i dont know why it wont do 4.4 stable.i dont know if it is the ram or the vcore or the board that just wont do 300fsb.i guess it must be quite stable as it is now tho' to pass the science mark at 4.3.i guess more tweaking is in order when i can be arsed.

Click to expand...

Yup, I am "experimenting" right now in fact, using 12x multipliers by 231mhz (limit of what I can do & complete ScienceMark 2.0) FSB, & getting 1393 scores on it... lower score on this test & lower RAM-mhz return (185/200mhz) too.

A 232mhz FSB fails out & will not finish the Primordia section here (caught in endless loop on iteration #2, & noting scientifically exponents for results numbers) indicating instability imo @ the level of FSB here.

11x SOLID multiplier & 11.5x FRACTIONAL multiplier in BIOS by 248-250mhz FSB yielded HIGHER ScienceMark 2.0 scores (my high-mark of 1401 was achieved in that manner on ScienceMark 2.0 results) & also a HIGHER RAM-mhz return (195-197/200mhz possible).

Nearly 10 points higher using a LOWER multiplier x HIGHER FSB here on ScienceMark 2.0 is what I am seeing... actually "better" using that & this test's results bear this out for me.

Tough call, either way (but, I now know where the limits I hit are, so it's all good! A man's gotta know his limits in stability testing lol... either that, or cook your system's eggs good & ruin it!)

APK

P.S.=>

tigger69 said:

the combination of this board and chip seems to be ok.i am quite impressed with this 930.i really thought a 4800 x2 would beat my chip.i guess it thrashes mine on games tho'.

Click to expand...

I have no idea either way, but that Intel rig of yours is a GOOD overclocker imo! I gotta get me some BETTER more-overclock ready RAM... by ALL means!

Still, I think I can "tighten" up my memory timing-chain (as I call it) & see if I can get MORE "ROI" from my RAM-mhz speeds using 12x multiplier instead of 11-11.5x...

tigger69 said:

just dont see that the chip cdawall has can beat mine and yours,it just doesnt seem right,his parts are'nt exactly high end either.

Click to expand...

It's not imo, like many of yours, but... he may be straight on his opinion of it being stable & it has found SOMEKIND of 'glitch' that allows test completion, but showing a radical "err" outlier result in his Primordia result (just like that "Super-Result of mine had in "Molecular" test section I posted - insane #'s results like 95294.20 scores, impossible, & progressbar bounds exceeding the groupbox container)... apk

ok here you go i tweaked my ram got it at 42s at 1m super pi with pics to prove it these are my super pi scores 2.35ghz (2T) 2.44ghz (1T) and 2.4ghz(2T) and shortly i will have new science mark 2 score at 1T on the ram

edit-new high score!!!! 1592.84 @2.44ghz and ram at 2.5-3-3-6-9 1T

cdawall said:

ok here you go i tweaked my ram got it at 42s at 1m super pi with pics to prove it these are my super pi scores 2.35ghz (2T) 2.44ghz (1T) and 2.4ghz(2T) and shortly i will have new science mark 2 score at 1T on the ram

edit-new high score!!!! 1592.84 @2.44ghz and ram at 2.5-3-3-6-9 1T

Click to expand...

Looks stable for the most part, & you MAY indeed, have a legit score...

What makes me question it, & others too?

Well, MAINLY that Primordia result being SO far "out of whack" by comparison to your other scores!

It is really the ONLY thing making me "question" your resultset... that, & its progressbar blowing by the bounds of its groupbox container (lesser issue here though imo), but that's it though!



* At this point, regarding YOUR results (especially the Primordia section):

I would write the coder of this program & run your results by him, directing him to this thread no less...

(That is the SURE way to get an answer imo @ least, & also possibly alert him to problems in this program (like my "Super-Result" back a few pages, lol, because it IS appearing to be a LOT like it, & in my case @ least in that one? I KNOW this system can't pull that much of a score off!)).

APK

there that should make you happy made my primordia drop happy now i changed some of the settings and ran it at 2.44ghz and all my other settings what ever the prob was fixed when i changed the element thingy in primordia

cdawall said:

there that should make you happy made my primordia drop happy now i changed some of the settings and ran it at 2.44ghz and all my other settings what ever the prob was fixed when i changed the element thingy in primordia

Click to expand...

It's NOT about "me being happy", but more YOU being happy & knowing your result is indeed, legit... others & yes, myself, aren't convinced of it because of what I noted above is all.



* I do think you ought to write the coder of this program, & run it by him, as to what I state above may be "wrong" w/ your results earlier...

They DO appear to be stable though!

STILL do write him:

Mainly about Primordia being WAY out of sync w/ your other results (plus, the progressbar flooding out of its frame/groupbox signals something is amiss as well, albeit not as much as how much the Primordia result "differs" from your other scores... do note, ours tend to stay ROUGHLY @ par with the other test results, this is a key point).

APK