Onepagebook, coolaler and Superpi

[guess2098]

Thx bro (my best clock-for-clock >5GHz is 50252: 5025.2*10.000)

I'm feeling happy not to have broken the 10s yet...

try to boot with 400 then pull
you will find something better

[shimmishim]

interesting thread.

the most interesting thing is the cinebench scores which show a 4 second gain from a simple "tweak"

[metro.cl]

this is looking bad

there have been several problems lately here on XS, not guilty until prooven is the smart way to go i believe

One thing is that Super pi some times give you lower time because of weird things i remember benching with my opty148 i got a better time with worst cpu speed, worst ram & worst FSB and it was just simply because some times you get weird (legit) runs when runing on the edge of stability

[zbogorgon]

Lets say that he really has an Ultimate Tweak, that could show us and super improved our times.
He would proove all of us, that we are lame, that we were so wrong. He would proove us that he really is the best Superpi GURU at this time.
He would have a tweak, named after him that would be used by many generations.
He would recived the highest respect a overclocker can have.

If he doesn't, he has nothing, his next Superpi/3DMark result would always be marked by this, by possibility of a cheat and will therfore never be widelly excepted as a undisputed WR.

the way I see it, he has no choice

[jmax_oc]

OPB faked one time during the 28GP. So he can do it again without problem... 0.6sec fater with TCCD whereas others get BH5 (= bench at higher frequencies)
Link : http://www.biwa.ne.jp/~yok/28GP/28GP.htm

Coolaler faked during the 600MHz FSB. So he can also do it again. 17.4sec at superpi 1m with Conroe 4Mb cache at 3600MHz (= quartz removal)
Link : http://www.xtremesystems.org/forums/...d.php?t=114600

[Vapor]

interesting thread.

the most interesting thing is the cinebench scores which show a 4 second gain from a simple "tweak"

Not only that, but if you search Coolaler's results, you'll notice the Console (official) timer suddenly stopped working as well.

[alexio]

Lets say that he really has an Ultimate Tweak, that could show us and super improved our times.
He would proove all of us, that we are lame, that we were so wrong. He would proove us that he really is the best Superpi GURU at this time.
He would have a tweak, named after him that would be used by many generations.
He would recived the highest respect a overclocker can have.

If he doesn't, he has nothing, his next Superpi/3DMark result would always be marked by this, by possibility of a cheat and will therfore never be widelly excepted as a undisputed WR.

the way I see it, he has no choice

Yeah you should be proud of what you have achieved together with Vapor and everybody else that wanted this to be posted in public.

[afireinside]

Did OPB give you the tweak file for you to test or what?

No. The people who told me about it are now telling me it's nothing special. Different story. They also told me they would send it to me even though it was "Secret" but that never happened. I really don't know what to think at this point.

i remember benching with my opty148 i got a better time with worst cpu speed, worst ram & worst FSB and it was just simply because some times you get weird (legit) runs when runing on the edge of stability

I experienced this to, infact my two best runs per time (25.2 @ 3.2 and 23.1 @ 3587) happened this way and could never be matched. ALSO this was a .2s or smaller difference AT OVER 20 SECONDS A RUN. We're now under 10 seconds a run. Huge difference.

[charlie]

Isn't anyone HERE CURIOUS why Team Japan has been "off" the sPi WR's after owning them for 7+ years?? Noone in the WORLD understands superPi benching and O/S tweaking better than Team Japan. Hmmmm????

Picoo has the most fortuitous superPi runs yet from Team Japan and they're not even close to the "WR"...

Noone is screaming CHEAT, but there seems to be SOME sort of an ABNORMALITY.

[mrlobber]

BTW...TaPaKaH = HaKa PaT = Hacker Pat.

Just to point out, TaPaKaH = Tarakan (if Russian letters are converted to Latin alphabet), and Tarakan in Russian means (C)ockroach in English.

[NiCKE^]

If he doesn't, he has nothing, his next Superpi/3DMark result would always be marked by this, by possibility of a cheat and will therfore never be widelly excepted as a undisputed WR.

the way I see it, he has no choice

That would mean bad PR for TeamGroup, OCX and OPB himself and what could the intention of this thread being more then showing that XS Staff believes there's something bogus with OPB and Coolalers scores?
No I'm not saying they want to make TG, OCX and OPB look bad, because they clearly stated they just want answers

[Gautam]

It'd do us all better to keep politics out of this.

[Xtreme Addict]

Maybe purpose abnormality... Facts aren't lieing

[metro.cl]

Isn't anyone HERE CURIOUS why Team Japan has been "off" the sPi WR's after owning them for 7+ years?? Noone in the WORLD understands superPi benching and O/S tweaking better than Team Japan. Hmmmm????

Picoo has the most fortuitous superPi runs yet from Team Japan and they're not even close to the "WR"...

Noone is screaming CHEAT, but there seems to be SOME sort of an ABNORMALITY.

you have a point, but didnt intel rule CPUs until AMDs K8? there is always someone that could do something better than you

[zbogorgon]

That would mean bad PR for TeamGroup, OCX and OPB himself and what could the intention of this thread being more then showing that XS Staff believes there's something bogus with OPB and Coolalers scores?
No I'm not saying they want to make TG, OCX and OPB look bad, because they clearly stated they just want answers

Did you leave the rest of my message out on purpose? If OBP prooves his way than that would mean a huge promotion for OBP, TeamGroup and OCX.

And if you were on holiday last three days, It was I who started the debate and I have no relationship with XS owners nor I was influenced to do so.

[Lithan]

you have a point, but didnt intel rule CPUs until AMDs K8? there is always someone that could do something better than you

Spi hasnt changed, windows hasn't changed. That comparison is moot.

[afireinside]

you have a point, but didnt intel rule CPUs until AMDs K8? there is always someone that could do something better than you

What? TJ has always kicked pi ass on Intel

[cpulloverclock]

as it was stated, spi times with freqs doesnt scale linearly, when in 3GHz range, below 50000 is not a problem, in 4.5GHz it's really hard in 5.5GHz it is imposible, just read the 1st topic please

so u confirm what I say, sayin that

[zbogorgon]

so u confirm what I say, sayin that

heh, so sorry, I misunderstood your post

[Vapor]

I'd like to straighten out the argument over performance product scaling.

1) 50500 is a solid generalization for a VERY well tweaked run at ~4.5+GHz.

2) The higher the MHz, the higher the 'ideal' performance product is.
2a) The lower the MHz, the lower the 'ideal' performance product is.

3) The reason why it slightly scaled with MHz is because two other attributing factors to performance: FSB and RAM, don't scale with CPU MHz. They're just as achievable at low CPU speeds, probably even easier to achieve to be frank.

[metro.cl]

What? TJ has always kicked pi ass on Intel

i was refering that even if you are the best in something (intel was the best in cpu) someone can beat you (amd been much smaller and young beat them in cpu arch), i wasnt refering to performance in super pi between amd and intel.

[zbogorgon]

I have a question.
If Coolaler and Onepagebook don't show up or either of them doesn't provide a satisfactory explanation, what kind of actions from XS staff and other OC community, like hwbot, can we expect?

[zakelwe]

When this thread kicked off tonight I had no idea we would be seeing results from other overclocking people that supported this claim, I thought it would just be to do with Conroe SuperPi, but it seems to go back to AMD days as well and some of the links shown how do you explain those ?


http://www.biwa.ne.jp/~yok/28GP/28GP.htm

Look at the names on that table and how far they are behind. They are all grouped but OPB is way ahead. How ?

Regards
Andy

[cpulloverclock]

A good tweak for a nice ratio :

- Start an 1m run at 4GHz for example
- Change the frequency with clockgen, setfsb or Systool and screen

It's easy and everyone can do it


---> OF COURSE IT'S

[boblemagnifique]

Isn't anyone HERE CURIOUS why Team Japan has been "off" the sPi WR's after owning them for 7+ years?? Noone in the WORLD understands superPi benching and O/S tweaking better than Team Japan. Hmmmm????

Picoo has the most fortuitous superPi runs yet from Team Japan and they're not even close to the "WR"...

Noone is screaming CHEAT, but there seems to be SOME sort of an ABNORMALITY.

Just 1/1.5 year

The majorty a people is a S.H.I.T in your eyes