Onepagebook and his 3D benchies...

[hipro5]

Hello all....:)

First off I would like all of you to be polite...:)

READ WELL the whole post before start posting and compare....Use your mind and don't let others to use it for you....

Now...

I show his 2k5 bench with a single 8800GTX clocked at 653MHz core and 1145MHz rams - Actually it seems that his VGA rams are clocked at 1053MHz and NOT at 1145MHz as he says -
His CPU power is 5302MHz on an EVGA 680 mobo.....
Here's the links: http://service.futuremark.com/compare?3dm05=2450489 and here: http://www.hwbot.org/compare.do?resultId=563812 , http://www.hwbot.org/result.do?resultId=563812

I ran the 2k5 bench at 5405MHz on an ABIT AW9D-MAX mobo with a SINGLE VGA too at 899MHz core and 1215MHz rams.....
Here: http://service.futuremark.com/compare?3dm05=2660983

http://www.thelab.gr/images/Hipro5/CONROE_X6800ES/ABIT_AW9D_MAX/PNG/2k5_27804_5408MHz_SINGLE_8800GTX_415_1_1_891_1215_ 1.png

http://www.thelab.gr/images/Hipro5/CONROE_X6800ES/ABIT_AW9D_MAX/PNG/2k5_27804_5408MHz_SINGLE_8800GTX_415_1_1_891_1215_ 2.png

I have 100MHz more in CPU PLUS I ran it on 975 chipset and 246MHz more on VGA's core and 70MHz more on VGA's rams.....
QUESTION 1: How come and I get a lower score with higher clocks overal?....
QUESTION 2: Let's say he plays his rams at 500MHz with Cas 3-3-3-x.....and he beats me on GT1 and on GT2(which actually is both CPU and VGA)....NATURE TOO?.....
I get 150.3fps with the VGA at 899MHz/1215MHz and he gets 151.7fps at ONLY 653MHz core?.....
Sorry but I just don't bite it....:confused: :rolleyes: :stick:

SECOND:
His Aquamark score with SLI 8800GTX at ONLY 4832MHz with a Quad, is 261895 marks.....
We ALL now that the VGA's gives you about 2k - 3k MAX on this bench....
QUESTION: How come and he get's THIS score at ONLY 4832MHz?....You'll tell me it's a QUAD......Welllll I've ALSO tested it with a QUAD at 4852MHz.....NO GO....:confused:

THIRD:
His new 2k5 score with SLI : http://service.futuremark.com/compare?3dm05=2654983
k|ngp|ng's score with SLI : http://service.futuremark.com/compare?3dm05=2408750

2000 marks over k|ng|ng?....Hello!....
k|ngp|ng benches TOO MUCH time 3D and OPB JUST STARTED.....Tweaked OS ONCE AGAIN OR tweaked VIRTUAL OS?.....You tell me....

AND REMEMBER: We are talking about 2k5 bench.....You don't get THAT EASY marks from it.....It's not 2k1 with a difference of 2k marks +/- .....It's the 2k5 bench.... :stick:

Take a GOOD look at their ram speed....OPB seems to be 1094MHz and k|ngp|in's seems to be 1188MHz.....

8800GTX's ram frequency can be calculated as: 716MHz (STANDARD - MAIN) PLUS what you see on the compare link......
So k|ngp|ng's is: 716MHz + 472MHz = 1188MHz REAL
and OPB's is: 716MHz + 378MHz = 1094MHz REAL

I don't bite it either.....:stick:


FINAL:

At least as for the 2k5 with the SINGLE VGA OPB's score, I'm 100% positive that it wasn't done at 653MHz VGA's Core OR it is FAKE.....
You try it and tell me.....151fps with the VGA at ONLY 653MHz Core?...NO WAY.....
Take a look at Overklokk's SINGLE VGA clocked high: http://service.futuremark.com/compare?3dm05=2565590.....How about his NATURE score?....That bad too that is beaten with ONLY 653MHz core by OPB?......
Take a look at Oppainter's new score: http://service.futuremark.com/compare?3dm05=2646810
How about his NATURE score too?....Beaten by OPB with ONLY 653MHz Core?..... ;) :stick:

It was done at higher VGA's clocks and he doesn't want to say so?....
That's a lie as for me if he did so......Excuse me for this post of mine but that's how I feal.....:(


.

[oohms]

Sure, it does look a bit suspicious, but didnt this happen before a long time ago in 2k1 with the dual processor tweak or something similar?

My point is that there is a possibility that he may have found a tweak that he might not want everyone else to know, and especially since the 8800 drivers at this point are a bit flaky, who knows? :confused:

[dinos22]

oh boy another :banana::banana::banana::banana:fight is brewing but there are some incredible scores there from OPB :confused: but somehow i don't think there will be answers in thisthread if people start attacking OPBs credibility.......as it was the case with Pi

i'm having a tough time thinking he would cheat :confused: :confused: :confused: :confused: hopefully he does join in to share his views about this

[hipro5]

Sure, it does look a bit suspicious, but didnt this happen before a long time ago in 2k1 with the dual processor tweak or something similar?

My point is that there is a possibility that he may have found a tweak that he might not want everyone else to know, and especially since the 8800 drivers at this point are a bit flaky, who knows? :confused:

Another tweak again?....:rolleyes:

Nature on 2k5 at 653MHz core WHAT tweak makes it CLIMB at 151fps?....

[K404]

Hey! I saw the score on HWBot. I assumed he hit 6 in stead of 9 by mistake when he entered his score, unless theres a sign of 6xxMHz core elsewhere?

I do think the score is obtainable with a single card

:toast:

[zbogorgon]

i hope this time we will see some resolution

[dinos22]

[B]Nature on 2k5 at 653MHz core WHAT tweak makes it CLIMB at 151fps?....
wouldn't you like to know :p: j/k

[Astri]

Take a look at his drivers.They are not FM Approved.

Btw, Hello XS :)

[Entsafter]

Kevin runs his card/cards at or above 880/1150 on ln2

[dinos22]

george did you bother asking Kevin about this first before going at it like this......maybe it would have been smarter before people start taking sides

[massman]

Another tweak again?....:rolleyes:

Nature on 2k5 at 653MHz core WHAT tweak makes it CLIMB at 151fps?....

I guess he just pressed the '6' instead of the '9'.

Look at the frequencies he normaly benches at ... most of them above 800MHz.

[TiduZ]

Well, OPB is one of the top-3 benchers in the world, and i am sure he knows how to tweek his settings realy extreme..

[eva2000]

just guessing

1. different drivers ?
2. clocking gpu/ram too high can throttle with lower end results as well
3. LOD tweak differences
4. 680i with G80s seem to provide higher performance clock for clock than 975x/965p from my 8800GTS/SLI tests
5. any bios flash/shader tweaks ?
6. PCI-E frequency tweaks ? seems may have more gains on 680i than 975x/965p from what i tested with Asus 680i striker versus D975XBX/P5B Deluxe.

I believe also Kev got advice and tweaks from mickeymouse as well for 3d stuff especially for sub zero GPU cooled runs. Don't forget Kev is part of TeamOCX and they have a few more than capable 3dmark benchers as members http://www.hwbot.org/listMembers.do?teamId=132 who can all give advice to each other

[dinos22]

Well, OPB is one of the top-3 benchers in the world, and i am sure he knows how to tweek his settings realy extreme..
dude he cannot outtweak hipro or KP by that margin unless there is something thatseparates them (unequal setups.....orsome kind)

[[cTx]Philosophy]

Will be interesting to see whats up with his tweaks if this is the case, I want em :)

[dinos22]

just guessing

1. different drivers ?
2. clocking gpu/ram too high can throttle with lower clocks as well
3. LOD tweak differences
4. 680i with G80s seem to provide higher performance clock for clock than 975x/965p from my 8800GTS/SLI tests
5. any bios flash/shader tweaks ?
6. PCI-E frequency tweaks ? seems may have more gains on 680i than 975x/965p from what i tested with Asus 680i striker versus D975XBX/P5B Deluxe.
george KP benches nvidia....he would know betterhow to hit sweat spots than OPB don't you think....just a wild guess but the amount of experience thetwo guys have my money would be on KP in 3D

the only thing that stands out are the drivers really

[eva2000]

who knows .. i'm not taking sides just stating possible factors as to the discrepancy between the two

I haven't used a 8800GTX before so not sure how they exactly fair on either platform... i do know i got slightly higher scores with my 8800GTS on Asus 680i striker than my P5B Deluxe at same clocks and PCI-E frequency bump on 680i striker had an effect on scores where as P5B Deluxe was negligible

[dinos22]

who knows .. i'm not taking sides just stating possible factors as to the discrepancy between the two

I haven't used a 8800GTX before so not sure how they exactly fair on either platform... i do know i got slightly higher scores with my 8800GTS on Asus 680i striker than my P5B Deluxe at same clocks and PCI-E frequency bump on 680i striker had an effect on scores where as P5B Deluxe was negligible
hey don't get me wrong i'm not saying you are but most of those are not discrepancies as KP benches E68 and hipro 975x

btw what is the difference percentage-wise

[massman]

hipro, I guess he ran it with SLI @ 653MHz cores
IMO - it was just hwbot that gone mad and got that as single card result (there were such bugs recently)

Nope, the score was submitted on the site itself, so it has nothing to do with the forumbots

[ojdr2001]

May be the reason is the same oppainter wrote in his 3D scores cards where at stock speeds and they weren't... a simple mistake

I think it would be more correct from your side to ask Kevin first privately before you come to public... but why Am I not surprised with your attitude?

[eva2000]

hey don't get me wrong i'm not saying you are but most of those are not discrepancies as KP benches E68 and hipro 975x

btw what is the difference percentage-wise
i've had as much as 150-200pts diff in 3dmark06 between platforms.

Don't forget shader tweaks in bios as well.. whether Kev uses them or not who knows

[Pedro Rocha]

Another tweak again?....:rolleyes:

Nature on 2k5 at 653MHz core WHAT tweak makes it CLIMB at 151fps?....

I already see that score and seems really suspicious.

No is NOT a Tweak I'm almost sure, i'm doing tweaking for 3dmark for many years and I know how far tweaks can go.

Maybe Onepagebook is usinig a cheat software named !@#$% that have that effect.. I think that cheat works in all 3Dmarks except 2001..

Well these guys never stop, they think they can delude all the world - that's really sad :(

BTW : George in Cebit next month I will personally explain you some good old scholl 3D tweaking :)

Edited.

Program removed, keeping cheating from happening, starts by keeping the tools outta peoples hands.

-Kunaak-

[Pedro Rocha]

george did you bother asking Kevin about this first before going at it like this......maybe it would have been smarter before people start taking sides

Kevin (or one of the members of this vast fan club) is completely free to came here and give some clear explanations about this AND about their SuperPI "tweaks"

I did that when I discover the ACPI tweak that allows me and Ricky to break the 40K barrier in 3Dmark01 few years ago..

[massman]

I already see that score and seems really suspicious.

No is NOT a Tweak I'm almost sure, i'm doing tweaking for 3dmark for many years and I know how far tweaks can go.

Maybe Onepagebook is usinig a cheat software named !@#$% that have that effect.. I think that cheat works in all 3Dmarks except 2001..

Well these guys never stop, they think they can delude all the world - that's really sad :(

I always respected you and your fabulous results Pedro, but this is really sad. Just talked to MM, apparently kevin runs his sli run at 880 and above. As you can see, his sli 05 run he benched at 906/1215 ... I don't think it's weird he could have run his card at 953/1145, so maybe he just mistypted ...

[boblemagnifique]

May be the reason is the same oppainter wrote in his 3D scores cards where at stock speeds and they weren't... a simple mistake

I think it would be more correct from your side to ask Kevin first privately before you come to public... but why Am I not surprised with your attitude?

Because, That make nearly 2 years that it makes mysterious scores :eek:

Because, they is explanations dont prove anything each time (vidéos either, I see) :rolleyes:

Because, that do much of because :D

[dinos22]

Kevin (or one of the members of this vast fan club) is completely free to came here and give some clear explanations about this AND about their SuperPI "tweaks"

I did that when I discover the ACPI tweak that allows me and Ricky to break the 40K barrier in 3Dmark01 few years ago..
yeah that's true but what if george just simply compared apples and oranges.....we know Kevin is a primadonna and will not join this thread if his life depended on it because he;s too proud

i donno it doesn't look like all options are exhausted yet to claim this is a cheat as you and hipro are insinuating i donno.....excuse my ignorance if i'm missing something here :confused:

i don't think you guys are sure either as he would have been banned or stripped of his status here for SuperPI thread "tweaks"

[dinos22]

Because, That make nearly 2 years that it makes mysterious scores :eek:

Because, they is explanations dont prove anything each time (vidéos either, I see) :rolleyes:

Because, that do much of because :D
and the thread derailment begins :D hahahahahah :p:

[massman]

(and, hwbot IS buggy. Also check out this score
http://www.hwbot.org/result.do?resultId=559573 and many others...)
(i didn't state Kinc as a faker - used that to prove that hwbot has errors)

Not a bug, Kinc didn't enter the gpu clocks ;)

[TiduZ]

george KP benches nvidia....he would know betterhow to hit sweat spots than OPB don't you think....just a wild guess but the amount of experience thetwo guys have my money would be on KP in 3D

the only thing that stands out are the drivers really

I guess you are right, kinkpin is the true master of nvidia gear..

[massman]

Same chance of OPB not putting a tick in "SLI/Crossfire" ;)

Correct

[dinos22]

I don't believe the guy is capable of controlling LN 2 with 3 pots at -90 for the vga's ,knowing that hes only begon to start and beat the master himself just like that, that just ain't going to happen.

I have a theory, i will post it soon when i have teh resulst, it's pretty easy you know, and the guy is using the same crap again for sure :)
actually OPB does have a smart way of benching with LN2 which gives him more control so you should actually look into it before you post something funny :p: i'm serious :)

[Rovtar]

Kink you just have to admit that somebody is better than you, that's all
...

[[OCM] El Polo]

yes, I know I'm not a good bencher at all
yes, I know I'm not a international star of OC
yes, no one knows me here

but I've got a brain too, and I think (I don't affirm!!) that a tweaks/OC/bench guru can type three numbers without typo error....
furthermore when his previous benchies (spi i.e) were not accepted by many of the world's best benchers...

[Magnj]

OPB :bow:

Why do you have to bring hostility around? Cause he can tweak better than you? Cause he typed in a wrong digit?

[lawrywild]

ROFL..

and it starts again.. this thread is so :banana::banana::banana::banana:ing lame

and started by the person who left his own team on hwbot to give xs a lead over ocx :rolleyes:

[s e t h]

the super-pi issue is solved as far as i'm concerned...
all one has to do is like 100 runs and one run will often come out substantially faster and create the differences seen in OPB/coolaler runs.
they are most likely just bugged runs imo
super-pi is far from consistent

[Planet]

I have noticed that if I do a search of dual or single card it keeps showing me dual card results. That was a few days ago maybe its been fixed.

[perry_78]

This is quite sad indeed.

I'll give him the benefit of the doubt and assume he missed the SLI checkbox, but either way, his clean sheet and integrity have been blemished a while ago.

Sad.

[K404]

Well...tweaks can make up those FPS I think, if you really tried.


I`m 90% sure that the score submitted is single card, so only 2 pots to control.

I controlled (without style or grace) 2 pots with my first ever runs so with some practice, sure its possible to run "properly" with 2 pots, single handed.

I think its a mis-type. Anyone here never made a spelling mistake?

Storm in a teacup.

[kromosto]

when i see the title of the thread i think OPB opened this thread as he likes to talk about him like third person :).

i am in xs nearly 4 years and never seen somebody blamed for nothing. there always been a suspicious score for someone but when a thread begins about it he comes and explains how it is done and everything is solved.

now if there is no fire then why it is smoking. and it begins to smoke too often. he is not coming here and defend himself because of his proud i dont think so, bacause he cant defend himself in a place like here he needs to defend himself in a place that he has more supporters. what i believe he is a talented cheater.

he needs to open a thread in here named like onepagebook answering questions and has to give satisfactory answers to questions.

[cronic]

Here we go again. Let me get this straight. All that would have to be done is to mistype an entry and you call him a faker. Why would someone with copper rev.2 vga MousePots bench at stock speeds? Use some common sense guys. Didn't OPP make the same mistake earlier and nobody at TeamOCX was calling him a cheater or fake. When will you guys learn how to properly handle a result in question? You don't start threads and call people a cheater without asking the person first. Its common sense people. Hipro I respect you as a world class voltage guru, but your actions recently have definitely called into question your character. It really is a shame that threads like this repeatability pop up at XS. Once and for all can we not end this BS between certain members here at XS directed towards OCX?

[blind_ripper]

that 05 is a fake or he is still not saying any thing to us .
with thos clocks u should only get about 120fsb in 05 canyon flight , tested it by my self M8 .
now im telling u 05 had some nice little tweaks for ATI X800 but notthing yet for 8800GTX !

but this is gone be a flame war again , so ppl calm down and if u dont have anything good to post like aaarrrggg its a fake again . then plz shut up :)

[eva2000]

There is no way that you can gain 8 FULL FPS in GT1 at same cpu speeds with lesser GPU cards.
There is no tweak that can do such magic :)
And to push it even further, when you tweak the :banana::banana::banana::banana: out of nvidia cards your GT1 wil go down slighty only giving advantage in GT2 at its minimal and GT3 full.
there is if the card and cpu are on completely different chipset/platforms.

[Zeus]

Guys, just don't know, i haven't investigated the scores but i know it ain't easy to top scores coming from k|ngp|n for one.

I mean, how difficult can it be to just step in here and say: "sorry guys, i just mistyped the VGA speeds" ?

I think Hipro has done his homework to come out with these questions.

I'm keep myself from condemnation, i know Kevin is a master in tweaking memory, it's possible he has more tweaks up his sleeve that we can think of.

[K404]

Hey Pallmall...i`d be willing to give what I said a try, but i`m minus a CPU. :( Thats another kettle of worms..

If this is still an issue when i`m back up an running (if I still have the gear) i`ll give it a shot

[massman]

I mean, how difficult can it be to just step in here and say: "sorry guys, i just mistyped the VGA speeds" ?

Very difficult if you're not online.

[Kunaak]

Thread Restored...

[nn_step]

well it could have to do with the beta I sent him

[DDTUNG]

This is a strong reminder to everyone posting in this thread to keep it a civil and constructive discussion. Any flaming will be dealt with swiftly and decisively.

DDTUNG:cool:

[zakelwe]

It's unfortunate that Futuremark still does not put up correct card clocks and people have to write them in the descrription on the ORB. Now the top overclockers are not even putting them there so you have to go to hwbot to find them. I request that people put video card clocks in the ORB like in the past, it makes it easier for viewers to see details.

It could easily have been a typo on hwbot, there is no denying that.

Another discrepency though is the SLI runs in 05 for Kingpin and OPB. They have very similar clocks but OPB is quite a lot higher in both GT, GT2 and GT3. As GT1 and GT2 are both cpu dependant and GT3 is more graphics dependant ( I assume at this level still ) then this means that two tweaks must be applied that Kingpin is not doing, one for cpu and one for gpu. Other possibilities are bug or cheat but again we will never know for sure except that is high and I doubt OPB will tell XS what his tweaks are, nor will he admit to cheat or bug so we are no nearer to pinning it down one way or another.

Once again you'll just have to make your mind up one way or another unless we can have some more 3d bench stats to make a case.

Personally I'd like to see a video of one run from boot up from all the top benchers nowadays which is in the ballpark top score, I'm still not sure if that would be 100% proof but it would be in the right direction.

Regards

Andy

[zbogorgon]

personally I thing sth should be done, his scores are just unrealistic, he is making fun of all oc community
If he was cheating then, he has no reason to stop, cause nothing was done to prevent then since and it will go on and I am quite sure that many are thinking of cheating and indeed they are already doing it or they are going to start, because it's just too easy, just cheat a couple of %, your excause is tweak.

[hipro5]

I'll place a bet here....:)

I'll bench with single card and get some scores......Next day will be broken!........You want to bet?......:rolleyes:

[BenchZowner]

I'll palce a bet here....:)

I'll bench with single card and get some scores......Next day will be broken!........You want to bet?......:rolleyes:

Good one.
But I think you should bet on something else...that someone won't reply here :D

[dinos22]

lol hipro

[Kinc]

Iv'e benn on vaction from all hardware buisness some time know. Will back at the rigg next week. I have been benching 88gtx sli in 05 well over 5GHz cpu and from my experiance 800/1200MHz in SLI is all ou need to max out GT3. Then its all about hte cpu.

I just saw OPBs score and thought what the heck is up with that score. That is bloody imposible on ~5230MHz even with crazy good memory. Then I saw this thread must agree that his score looks strange. There are good LOD tweaks and some bus retalted tweaks for 05 but nothing close to this. From all the tweaks I know I can gain about 200-300 points and that are no real tweaks mainly overclocking the bi16 bus and pcies freqs.

If you ask me from my experience this score aint posible on the cpu clock showing in the compare link, not at all posible.

[LardArse]

I havent benched 05 in a long time but even 06, GT1 and GT2 is pure cpu bound after a G80 gpu speed attainable by watercooling... so for 05 I'm quite sure GT1 and GT2 you can't do anything to up it other than cpu power, high cpu clocks number one effect, FSB and memory bandwidth- number 2 effect, and nothing else.So that 6 fps jump can never be from GPU, only CPU, and a lot of mhz at that

[zakelwe]

The difference in core clock between OPB and KingPin's SLI runs is 0.3% and the difference in score for GT1 is 8% and for GT2 is 4% for reference.

Regards

Andy

[Johnny Bravo]

Out of interest lets say that OPB is a proud person - now Pedro touched on the subject of the old 2k1 tweak of setting multiprocessor to gain about 2k in the scores. If OPB has discovered something like this, I dunno some sorta software "bug" that allows for greater scores AND he's a proud person - then he's hardly gonna share it straight away is he? From what I gather he has been spending alot of time 3D benching of recent. He's also been in contact with MM and a few other good 3D benchers from OCX so he'll had gotten hold of alot of good info. I don't think the score are impossible but they are very "unique" for the clocks. Unfortunately I don't think XS will be hearing from him anytime soon. I'm not a proud guy at all but if I get scores like that then came here and read this thread I'd probably not say anything either. I understand there's some hostility here but there's a big difference between constructive criticism and blatant accusations.

Just my pov

John

[BenchZowner]

@JB:
Don't expect any gains like these ( real big ) from any software tweak on 2k5 & 2k6 mate.

Unless explained, there's nothing leading to a valid score like this one.

[dinos22]

how can you be absolutely sure :confused: (at least you seem to be)

[BenchZowner]

how can you be absolutely sure :confused: (at least you seem to be)

I'm not ABSOLUTELY sure, but 90-95% sure :D

[zakelwe]

I don't want to see this getting into a " How do you know? " " I do know!" set of posts with nothing new or point of view brought to the table without at least some backup data to show your point, thanks.


Regards

Andy

[k|ngp|n]

Hi :D.
Well, don't know what to say about this one as I don't like to be involved in controversy, but here goes. Not trying to stir the pot, just giving my input from my experience benching the HW. When I saw that 33.9k 05' SLI score, it seemed way off the norm for the clocks he is stating in his description. IMO the single card 05' score in question isn't really that out of the norm for the cpu clocks on 680i platform, but he is obviously mis-stating his gpu clocks in that one, which is a whole other issue in itself. What validity does HWBOT have if guys are going to not put the right card clocks in their submissions?...especially someone at the very top who others will compare their results too. I think I know where he learned that move from :rolleyes:
Hitting the wrong numeric key when typing in a submission is the lamest thing I ever heard (if those are OPB's actual words).


He's also been in contact with MM and a few other good 3D benchers from OCX so he'll had gotten hold of alot of good info.
John

With all due respect to MM who is THE best 3d bencher at ocx period, not quite Jonny.
Without giving names, I have asked some of these people directly about his SLI 05' score and the responses I got was:
Not sure whats up...his card clocks were higher than stated at 650, more like 920-1200...could have been caused by a bug...you know how runs are with the 8800's, one minute the score is high next minute low.
Not the kind of response I was expecting.

As for the comparison with my 32k SLI 05' score from november, keep in mind that was done on a pre-production EVGA mobo that had some issues and was done with relatively low bus speed (401x13). On a stryker or retail EVGA with 43*x12 there is definitiely more cpu power to be had in 3d, but it's not even close to the difference between the two scores with the cpu speeds being so close.
Like shamino and some others said WHO KNOW what the HW is capable of, after a certain point gt2 and 3 become very cpu bound and increases in cpu power be it through better timings or tweaks or whatever, results in increases in all game tests. Whatever is making the score seem "artificially" high, is coming from CPU end no doubt regardless of the fact that he is mis-stating his card clocks.

I just picked up another stryker as I blew up the vreg on the one I set 100k with :D. If I had run 05' with the 100k setup, the score would have been closer to 33k, but nowhere near 34k. The 100k clocks were around 950/1220 on cards and 5288mhz(406x13) on cpu.
05 and 03 will get the treatment from me in SLI this week with the stryker using higher bus speed with 12x, better tweaking, and much higher card clocks than early launch runs from Novemeber so I should have more of an apples to apples comparison of whats going on with his 34k SLI score to make a better judgement on it, as I can match the clocks "stated" in that run easily. As for the others like am3, hipro has the peronal experience now to comment on that.
The flaming has to stop though, especially from those who really have no clue what goes on when benching at this level. Anyone who would question Hipro's intentions about bringing this up is out of their :banana::banana::banana::banana:ing mind. Show some respect. The fact is the scores don't add up atm, for me particulary the 05' SLI score. This is the opinion of not just Hipro but MANY top guys. These things need to be discussed to preserve the integrity of what we are all doing here. Nothing personal against opb as I am sure if any of us posted some of the runs that he has recently, they would be in question.

[Johnny Bravo]

With all due respect to MM who is THE best 3d bencher at ocx period, not quite Jonny.
Without giving names, I have asked some of these people directly about his SLI 05' score and the responses I got was:
Not sure whats up...his card clocks were higher than stated at 650, more like 920-1200...could have been caused by a bug...you know how runs are with the 8800's, one minute the score is high next minute low.
Not the kind of response I was expecting.

Fair enough, shows what I know :rolleyes:

[Xtasy]

it sad to see that guys are using tweaks and tweaks for getting more and more points... and if they cant get the results they are begining to cheat. imho extrem tweaking is lame because you have to know the right persons.. so a normal one cant get those scores... even if they run exactly the same clocks..

[massman]

Let's assume that everyone who thinks OPB is a cheater is right. Who would be that stupid to cheat just by entering the gpu clocks wrong? That would be the most idiot way of cheating imho.

The core clock isn't correct, we know, but I don't think this is an example of cheating.

The 'hitting the wrong numbers' was not a statement by OPB, not at all, it was one way to explain the weird core clock.

@Xtacy: Xtreme tweaking is a part of extreme benching.

[K404]

it sad to see that guys are using tweaks and tweaks for getting more and more points... and if they cant get the results they are begining to cheat. imho extrem tweaking is lame because you have to know the right persons.. so a normal one cant get those scores... even if they run exactly the same clocks..


I`ve found plenty of tweaks by myself. I`m unlikely to be the first one to find them, but a lot of people dont need to have everything spoon-fed to them- you get out what you put in, the key to good benching is understanding. Maybe tweaking isnt a 100% fair comparison to an everyday, fully-loaded computer, so with that principle, better stop using Cascade, Dice and LN2.

[eva2000]

The core clock isn't correct, we know, but I don't think this is an example of cheating.


yeah that's true.. look at how many top 50 ranked pi or 3dmark results have their cpu clocks at default by accident while folks immediately above and below them are in 4+ ghz speed arena

[zbogorgon]

@massman
I am going to enter this (http://img217.imageshack.us/img217/7647/fakesg5.jpg) SPi result today on hwbot.org as I figured OPB's secret tweak.
here is my cpu validation (http://valid.x86-secret.com/show_oc.php?id=167209) link at higher freq than SPi was run, so I have the same level od proof as OPB has, so that must be a new SPi 1M world record.

you could expect a couple of WRs from 3DMarks and Aquamark soon

[massman]

@massman
I am going to enter this (http://img217.imageshack.us/img217/7647/fakesg5.jpg) SPi result today on hwbot.org as I figured OPB's secret tweak.
here is my cpu validation (http://valid.x86-secret.com/show_oc.php?id=167209) link at higher freq than SPi was run, so I have the same level od proof as OPB has, so that must be a new SPi 1M world record.

you could expect a couple of WRs from 3DMarks and Aquamark soon

Are you trying to boycot our database?

I'm very disappointed that you would even think of submitting a fake result. I've lost the respect I had for you, mate. I don't mind if you think someone is cheating at all, if you have the proof for it, but I DO mind if you try to cheat yourself. TBH, it wouldn't be strange if you did some great PI times and nobody believes you, just 'cause you once had the intention to sumbit a fake result ...

Sad.

[zbogorgon]

ups, I forgot to wrote "sarcasm on" at the begining of previous thread.

as I said before, If i wanted to cheat, I would already do so when I submited cpuz WR and I am 99% sure that I would be holding SPi 1M WR today, not OPB.

With boycoting I would achieve nothing, I appreciate what you guys have done with hwbot, it is by far the best database of OC achievements on this planet and I am fighting today becasue I don't want everything go to waste.


p.s. but wise man once said, If you can't beat them, join them.

[ahmad]

Thats is the problem: you make a mistake at XS and somebody is bound to notice. Whether the mistake was intentional or accidental, something should be done.

What most of the amature overclockers here don't realize is that running benches is like playing pool. After playing for a long time, you get to see things other people can't. Meaning, a slight variation in scores, especially when running almost identical hardware, needs explanation or something is up.

We got hipro, kinc, kingpin, pedro, zak, eva, shamino all commenting on this. If are not at that level, you need not give your opinion ;)

I would like it if opb said something. Usually remaining silent is a good sign of something..

[zbogorgon]

@ahmad, he commented accusation for his Spi 1M WR all-right. We all where a bunch of idiots that cannot even comprehend the level of his tweaks, sth like that.

[ahmad]

@ahmad, he commented accusation for his Spi 1M WR all-right. We all where a bunch of idiots that cannot even comprehend the level of his tweaks, sth like that.

Seriously? I would have expected more from OPB considering all the work he has done.

[massman]

ups, I forgot to wrote "sarcasm on" at the begining of previous thread.

as I said before, If i wanted to cheat, I would already do so when I submited cpuz WR and I am 99% sure that I would be holding SPi 1M WR today, not OPB.

With boycoting I would achieve nothing, I appreciate what you guys have done with hwbot, it is by far the best database of OC achievements on this planet and I am fighting today becasue I don't want everything go to waste.


p.s. but wise man once said, If you can't beat them, join them.

Sarcasm or not, I don't like the attitude.

I really feel a bit lost now. When I first started with benchmarks, I never saw anyone being accused of cheating. Now we're about 4 years later and the word 'cheat' is said 'bout every month. I really like what I do at HWBot, because I have the opportunity to spot cheaters and ban/block them. HWBot is becoming a huge database with, thanks to the contest, even more and more older hardware, but somewhere down the road, this huge database is being used as a way to spot possible cheaters. I will ban every single cheater if there's proof that 100% correct, not if there are just a bunch of insinuations, correct or not.

p.s. your wise man was not at all pugnacious.

[zbogorgon]

@massman
I am sorry, I didn't meant to be disrespectful. The thing with cheating today is, that there really is no way to tell for sure if it is a cheat or not. The cheaters we easily spot are morons, that don't understand a thing about benching. If someone who has a slght clue what is going on it is really not a big problem, after that it is up to us, to tell the difference.

And if most if not all the top benchers on the world, says that there is sth fishy with the result, that result needs an explanation. And not by someone unknown, by everyone, the same goes to hipro, kingpin, kyosen, king, shamino, oppainter...

[zbogorgon]

About OPB, he shown us that he doesn't give a crap about XS, we could be dancing on our eyelashes and he wouldn't mind. But I think he caress about hwbot, and as I see, you give a damn, so it is more or less up to you, if you will certify his results or ask for an explanation

[massman]

@massman
I am sorry, I didn't meant to be disrespectful. The thing with cheating today is, that there really is no way to tell for sure if it is a cheat or not. The cheaters we easily spot are morons, that don't understand a thing about benching. If someone who has a slght clue what is going on it is really not a big problem, after that it is up to us, to tell the difference.

And if most if not all the top benchers on the world, says that there is sth fishy with the result, that result needs an explanation. And not by someone unknown, by everyone, the same goes to hipro, kingpin, kyosen, king, shamino, oppainter...

yes, all top benchers gave their opinions and they all said it isn't possible. I'll contact OPB as soon as possible about the 'core635' issue, but what if it's just a typing error? Will you believe him?

Would someone please PM me the exact accusations with the a short theory? I don't know which result is being challenged and which not.

[zakelwe]

Let's assume that everyone who thinks OPB is a cheater is right. Who would be that stupid to cheat just by entering the gpu clocks wrong? That would be the most idiot way of cheating imho.

The core clock isn't correct, we know, but I don't think this is an example of cheating.

The 'hitting the wrong numbers' was not a statement by OPB, not at all, it was one way to explain the weird core clock.

@Xtacy: Xtreme tweaking is a part of extreme benching.

That does not explain the SLI results though.




I really feel a bit lost now. When I first started with benchmarks, I never saw anyone being accused of cheating. Now we're about 4 years later and the word 'cheat' is said 'bout every month. I really like what I do at HWBot, because I have the opportunity to spot cheaters and ban/block them. HWBot is becoming a huge database with, thanks to the contest, even more and more older hardware, but somewhere down the road, this huge database is being used as a way to spot possible cheaters. I will ban every single cheater if there's proof that 100% correct, not if there are just a bunch of insinuations, correct or not.

p.s. your wise man was not at all pugnacious.

There have been accusations of cheating and people caught in the last 4 years, Coolaler got caught out by Shamino and Sami in 3dmark a couple ago for instance, since then there have been a few people but it has passed under your radar, the only difference this time is that the only recent accusations concern one person who is right at the top so everyone knows about it.

Unfortunately it takes a big slip up to be able to definitely say cheat, not something that is likely if people are just slipping a few percent in here and there. Conversly this means it is just as likely to be a tweak as it is only a few percent.

The problem with the tweak though is the longer it goes on without the tweak being shown then the more doubtful it becomes that there is such a tweak.

How many people on this thread would not show a tweak if it was causing such concern to all fellow benchers ? I would, but then again I am more interested in being respected by fellow benchers than being right at the top. It does seem to me that either $$ or points has the potential to warp what people want out of hardware benching.

But no one has proved anything yet and this small percentage gain is equally as likely at this point to be through an unknown tweak than bug or cheating.Unless of course there is more evidence one way or another out there?

It will only take one thing to quash this completely, full details of the settings run and tweaks done that can be replicated by others. Then a lot of people can eat humble pie, surely the prospect of that is worth giving out the tweak all on it's own ;)

Regards

Andy

[zbogorgon]

I am not that much into 3D benches, but as I figured out, his 8800GTX SLI result is more problematic, cause many said it cannot be done with stated CPU speed, and that is not just a typo.
But what I am saying is that his SPi 1M SuperPi really needs a lot of explanations

[kromosto]

i am with zbogorgon from the beginnig.
what he says and what he shows is explaning everything for me.

we need an explanation.

and as zbogorgon says hwbot now has the most important duty for future of the overclocking. if scores are ceritified before getting a full explanation of what is done then overcloking is over. this is that simple.

[massman]

I am not that much into 3D benches, but as I figured out, his 8800GTX SLI result is more problematic, cause many said it cannot be done with stated CPU speed, and that is not just a typo.
But what I am saying is that his SPi 1M SuperPi really needs a lot of explanations

What kind of explanation?

He made a 1h13min video of him benching, from the boot-up till the pifast and (3x) the superpi 2M record. The Pifast WR was recently broken by roro.

[massman]

and as zbogorgon says hwbot now has the most important duty for future of the overclocking. if scores are ceritified before getting a full explanation of what is done then overcloking is over. this is that simple.

If you're willing to pay us so we can live from this job and can spend 10 hours a day checking out every result, then I would be able to do that.

BUT, we have that much faith in the benchmark community that we won't question results if we have enough verification, like futuremark link, screenshot, and the score isn't suspicious. If it turns out to be suspicious, we have the 'report to moderator' button ;).

[zbogorgon]

I didn't see any video, but noone is talking about Spi 2M, we are talking about Spi 1M with ratio below 50000 and CPU speed well over 5.2GHz.

from boot, with stopwatch or sth, or he explains tweaks to sb else that can replicate them.

[cadaveca]

Yeah I gotta agree...without some osrt of standard so that we can decently compare what's going on...some sort of info that explains why one score is better than the other...keeping track of scores is pointless.


I'd really like to see posts with scores like this saying "Look at what I did! Here's how...." and once the score has been duplicated, it can be confirmed as a legit score. Without confirmation, any score is subject to being a cheat, as long as it falls outside the norm.

I'd like to see benching just be benching, and everyone's attitude's left at the door. While I kinda lost a bit of respect for OPB due to these allegations, zbo has lost more from me merely by flogging a dead horse. OPB got kudos from me last time until he posted...

Attitude and personal feelings are getting in the way here. I know issues such as this it's hard to stay impartial, but really...my kids, aged 3 and 5, fight like this.

[massman]

I didn't see any video, but noone is talking about Spi 2M, we are talking about Spi 1M with ratio below 50000 and CPU speed well over 5.2GHz.

from boot, with stopwatch or sth, or, he explains tweaks to sb else that can replicate them.

Some other benchers had a sub50k ratio also and they are not questioned at all. In the same vid he had a run of 9.094 (due to the motherboard, which had problems to run above 1.9v). He explaned everything he single thing he did.

[Messiah]

and where is that video?

[zbogorgon]

who has a ratio below 50k with CPU speeds over 5GHz?

9.094s itself is quite possible, but what CPU speed?

if he did it with CPU speed 5435MHz or less and you personally seen the video?

[kemo]

What kind of explanation?

He made a 1h13min video of him benching, from the boot-up till the pifast and (3x) the superpi 2M record. The Pifast WR was recently broken by roro.

I think know you got it, all PI broken except 1M

Long ago when the record was broken I was not evolved into benching but what I knew at that time that OPB system was faster by all means than anybody else so logically easy WR (check the old thread and you will find that I was defending him) after that you take a deep look you find that the result is odd it can’t be that all OCs noobs and he is the only guy that knows how to bench. So either he knows something big and he should share it or just another fake. And you know faking is easier that OCing

[kromosto]

:) if i have enough money why not.

what i am saying is hwbot is reponsible for opb's recent scores. they can be true or cheat but these scores shows us an important thing. we have a very big hole in our world. this is because our way of thinking becomes old. as the time goes everything changes like crimes as a best example so we made new laws for new crimes. now our overcloking community laws are getting older and older and opb's scores show us that we have to make new laws we have to draw our borders again. as because this is the first time we see our rules and laws that are not written are not enough anymore, we dont know what to do. we are talking figthing about opb some says he is a cheater some says he is the best but we have to see the big picture as soon as possible. this is not about opb only this is about the future of the overcloking so as i said hwbot has the responsibility of opb's recent scores but all the overcloking community has the responsibility to define the new rules make the new laws. if we cant begin this quickly then overcloking is really going to be over. if we cant define our new rules then there will be lots of scores like ops's recent scores everyone gonna believe what they wanted to believe and this is will be the end.

[Gautam]

This deserves emphasis.


READ WELL the whole post before start posting and compare....Use your mind and don't let others to use it for you....

[cadaveca]

Um, those in the right should not care what others think, as they are right. To not even defend that your ARE right, is, in it's ommision, an admission of guilt.

OPB HAS helped lots. Cannot deny that. But now he has stopped. why?

But also, this can be taken as someone trying to force OPB's hand, which, i tihnk may not be a bad idea. He should help the community by simply just squashing all of this crap.:fact: Then there would be no quesiton in anyone's mind. Instead, he says "too bad".

[BenchZowner]

@lawrywild: { friendly as always }

1) OPB is not the only person out there knowing how to tweak RAM good ( sigh )

2) Finding a software tweak has nothing to do with your skills on pure overclocking ( hardware oc & modding skills ) [ knowing how a GPU works, and mostly its drivers helps, but not always ].
For example someone decides to take 2 days of his life in "playing" with Windows Registry, and luckily he finds out a tweak.Does this make him the best overclocker ? Nope.
Would giving the tweak to someone give him the ability & skills to take down a WR ( or even a great score ) ?
And after all, this is why the overclocking community & forums like this one are, for friendly competition & knowledge sharing.

3) Writing guides on Hardware or Software can be done by several people...or should I say numerous.
I've written a guide on kick-boxing techniques in the past, does that make me a kick-boxing champion ? ( or even a "top 20" member ? )

4) Diabolically, someone with a...master plan could've easily taken advantage of being nice at first, sharing information, tips & guides, and then use 'em for "coverage".
for example "I've published some overclocking guides, optimization & tweak guides on overclocking, and helped a lot of people on the net in the past.Now here comes the master plan.I use a "*****" to get some top scores and post 'em.Having published guides, and stuff make my scores valid & 100% trustworthy ? he he, I don't think so.

And after all, we're talking about a more than negligible margin on same/less clocks.

Ending this post...when you've got lots of people accusing you of cheating /or having serious doubts on your scores, what do you do:
a) Remain silent, get lost, and let them doubt your scores & fame
b) Act, show enough proof & go on walking proudly, like you should be ?

[zakelwe]

After two warnings after this thread was re-opened from Kunaak and DDTung and a message from me that I wanted just facts and figures and insights gained from experience or well thought out reasoning we seemingly have people who either cannot read or do not care about those requests.

Feel free lawrywild to open up a thread on Speak freely or Computer related open discussion about this and I will be more than happy in great length to comment on your points as will others, but desist from threadcrapping because quite frankly in this thread you are adding nothing. That goes for everyone else as well.

Thanks

Andy

[lawrywild]

Thread crapping?

Erm what?

This is just ridiculous, someone actually posts something besides OMFG HE'S A BLATANT CHEATER and you delete the posts saying I'm thread crapping?

This is like some kind of dictatorship, getting rid of anyone that stands in your way.. (and not you personally, this is most of XS)

edit: and no I won't post in speak freely, why didn't you move my posts there instead of deleting what I typed

[zakelwe]

Thread crapping?

Erm what?

This is just ridiculous, someone actually posts something besides OMFG HE'S A BLATANT CHEATER and you delete the posts saying I'm thread crapping?

This is like some kind of dictatorship, getting rid of anyone that stands in your way.. (and not you personally, this is most of XS)

edit: and no I won't post in speak freely, why didn't you move my posts there instead of deleting what I typed

The rules were set out at the start. People have defended OPB and not got deleted, read the thread again if you want proof. You however just went on a rant about "backstabbing". This sort of post does not follow the guidelines laid out.

As I said, your comments are better suited for other parts of the forum and not this thread and I am sure people will still be interested. I did not move them there as that should be your choice, but you have no choice on this thread if you wish to continue doing the same, those are the rules.

Or you can PM me. I love receiving PM's from disgruntled parties as we beat things out and come to some common ground.

Regards

Andy

[[XC]melymel]

In the last thread concerning OPB and SPi tweaks the following idea was proposed but it got lost in flaming:

For OPB to set the record straight he needs to inform an unbiased party of what tweak/s he has used to obtain his scores. This way he is not releasing his tweak to the benching masses so he can still have the upper edge but also can put to rest the cheating alligations and have a great time making everyone eat humble pie. From this thread I think a good person to be told is massman as he is an anti-cheat moderator at hwbot, so clearly doesn't want to see cheating occuring, but at the same time he has not taken sides in this argument.

Does this sound like a reasonable idea? as I think it's clear opb isn't going to speak now no-matter how he got his score/s and people will be free to draw whatever conclusions they want.

:toast:

[Planet]

In the last thread concerning OPB and SPi tweaks the following idea was proposed but it got lost in flaming:

For OPB to set the record straight he needs to inform an unbiased party of what tweak/s he has used to obtain his scores. This way he is not releasing his tweak to the benching masses so he can still have the upper edge but also can put to rest the cheating alligations and have a great time making everyone eat humble pie. From this thread I think a good person to be told is massman as he is an anti-cheat moderator at hwbot, so clearly doesn't want to see cheating occuring, but at the same time he has not taken sides in this argument.

Does this sound like a reasonable idea? as I think it's clear opb isn't going to speak now no-matter how he got his score/s and people will be free to draw whatever conclusions they want.

:toast:

If memory serves me correctly. HwBot crew was given 4+ hours of him benching with ln2 from boot to tweak to runs. If you read back a page or two you would see massman already posted that. I believe he also sent them the tweaks as well. Like mass said there was a video of him with a Pi 1m at 9.084 or something.

[[XC]melymel]

If memory serves me correctly. HwBot crew was given 4+ hours of him benching with ln2 from boot to tweak to runs. If you read back a page or two you would see massman already posted that. I believe he also sent them the tweaks as well. Like mass said there was a video of him with a Pi 1m at 9.084 or something.

I read his posts but I wasn't sure if that was a general video posted or one for verification purposes. If what you have said is true can/has a member of the hwbot crew who was given access to the tweak confirm they do enhance scores legitamately to the level shown by opb? :toast:

[Planet]

I read his posts but I wasn't sure if that was a general video posted or one for verification purposes. If what you have said is true can/has a member of the hwbot crew who was given access to the tweak confirm they do enhance scores legitamately to the level shown by opb? :toast:

I believe he had given massman or fredrick acess to the tweaks. It was only posted for hwbot and a few others to see due to bandwidth.

[Eldonko]

Guys if you have issues with a score click the report link in hwbot. They will investigate and decide whether it is valid or not. Whats the point in beating a dead horse for days and days and days, don’t you guys get tired of this crap? Threads like this just end up in flaming wars and personal attacks and accomplish nothing.

Lets review:
1) Click report bad result
2) Hwbot staff investigates the issue
3) Appropriate action is taken.

[massman]

If memory serves me correctly. HwBot crew was given 4+ hours of him benching with ln2 from boot to tweak to runs. If you read back a page or two you would see massman already posted that. I believe he also sent them the tweaks as well. Like mass said there was a video of him with a Pi 1m at 9.084 or something.

Yes sir, the most interesting 1h13min this year. He explaned every single step he took.

[BenchZowner]

Actually we're going off-topic now.

Here's massman's post:

He made a 1h13min video of him benching, from the boot-up till the pifast and (3x) the superpi 2M record. The Pifast WR was recently broken by roro.

1) SuperPi 2M record video...1M can't be tweaked like 2M ( and even 2M isn't more than a little bit tweak-able )

2) What makes you even think that if he was using a "ch**t" he couldn't have it to load with windows in the background ?

3) Let's say that he has given the tweaks to massman with that video.
Did somebody ( or massman himself ) confirmed the numbers using the tweak ? ...

4) We're talking about 3D Mark05 here, and for his SLI score mostly, which is considered at least "suspicious" by lots of top benchmarkers here and elsewhere.

[dinos22]

@ahmad, he commented accusation for his Spi 1M WR all-right. We all where a bunch of idiots that cannot even comprehend the level of his tweaks, sth like that.
:confused: wtf linkie

[dinos22]

If you ask me, with all the top benchers in the world questioning these results there is a problem. That amount of experience is impossible to ignore.
yeah all the people i would turn to for reference have joined the thread pretty much

[metro.cl]

this threads never end up good.

[zakelwe]

this threads never end up good.

True, they do tend to go downhill but at the start and in the middle they do give food for thought.I am closing this thread now. If another XS mod.supmod.admin.owner wants to open it then no problem.

If anyone wants to post on here a relevant point they feel they missed out on then fine PM me and I will stick it up, either way.

If OPB or anyone wants to send me OPB's tweaks to replicate these results I will also put them up also.

Regards

Andy