Faking Benchmarks, The Good Benchmark List, How To Spot Cheats and More...

**Rippon** · 04-14-2006, 02:52 PM

I unintentionally discovered a way to over double a score in science mark.

I did this twice to verify. I had it run all of the tests, I got impatient because I was running F@H in the background and knew it would screw up my test results, so I clicked skip, or stop, on the primordia test, it said are you sure, I clicked yes, then I let the rest of the tests finish. In the end, It said I sucessfully completed all tests (even though I clicked stop/skip half way through primordia). My score for primordia was 2500 or so, something really ridiculous to get on my A64 3000+ OCd to 2.5GHz.

I did this again after I turned off F@H to make sure it wasn't an error involved with that. It wasn't I did it again and I got the same result. I am not sure if it will do the same with the other tests, but it definitely messed up (doubled) the one test score.

Someone please verify this for me.

Also, sorry if this has been posted before, I did a quick search and didnt find anything so...

Thanks for verifing.

**Rippon** · 04-16-2006, 09:01 PM

Originally Posted by Rippon

I unintentionally discovered a way to over double a score in science mark.

I did this twice to verify. I had it run all of the tests, I got impatient because I was running F@H in the background and knew it would screw up my test results, so I clicked skip, or stop, on the primordia test, it said are you sure, I clicked yes, then I let the rest of the tests finish. In the end, It said I sucessfully completed all tests (even though I clicked stop/skip half way through primordia). My score for primordia was 2500 or so, something really ridiculous to get on my A64 3000+ OCd to 2.5GHz.

I did this again after I turned off F@H to make sure it wasn't an error involved with that. It wasn't I did it again and I got the same result. I am not sure if it will do the same with the other tests, but it definitely messed up (doubled) the one test score.

Someone please verify this for me.

Also, sorry if this has been posted before, I did a quick search and didnt find anything so...

Thanks for verifing.

Can anyone verify this?

**Joe Camel** · 04-25-2006, 05:21 AM

i think this belongs here...

555.22 FSB with Yonah?

and 925MHz on the memory.

or did someone figure out the golden key to OCing Yonah

http://valid.x86-secret.com/show_oc.php?id=89136

http://valid.x86-secret.com/records.php

**hixie** · 04-27-2006, 09:46 AM

Can someone verify if those two are real or fakes please.
These were posted in on a overclocking competition in Hong Kong. The scores seem pretty high for a X1800GTO (with unlocked 16pipelines).

http://www.geocities.com/darkknighta.../Cheat_002.jpg
http://www.geocities.com/darkknighta.../Cheat_001.jpg

Thanks.

**guess2098** · 05-16-2006, 08:15 PM

Originally Posted by Joe Camel

i think this belongs here...

555.22 FSB with Yonah?

and 925MHz on the memory.

or did someone figure out the golden key to OCing Yonah

http://valid.x86-secret.com/show_oc.php?id=89136

http://valid.x86-secret.com/records.php

that is the bug for CPUz, not cheat

**beast-usa** · 05-30-2006, 04:43 AM

Hey Joe

Wow I never thought people would do so much to cheat.
I have never even tried what would be the point.

**Jose** · 06-02-2006, 11:49 AM

Ihave a question: If I have a relative long list of computer configurations and their claimed benchmarks. Id there a place I could compare them with what is expected? I am doing a study of inflated claims in a project and I need all the help I could get. You can PM or contact me at peterdehoundt@yahoo.com

TIA

Jose

**Gebba** · 06-06-2006, 02:19 PM

can i beleve when somone says that hi can use higher then org multipel on the cpu?
i have alredu sean pictures of unlockt amd64 and some opterons.
of courese i can beleve it when i seas a unlockat socket a but i dont know what to think about amd64.

cheating is no fun,you can never get e better computer becouse of cheating,just tweaking can gives you better score and better fps in games

**tom12519** · 06-21-2006, 09:49 PM

Originally Posted by perkam

Thx for the more understandable syntax Kunaak, it helps

BTW, your faked SuperPi shot is funny--> An 18s SuperPi WOW, man, what kind of knocked out system would you need to get THAT in real life.

EDIT: Though Kunaak, I appreciate, as we all do, stuff on programs that CAN be cheated, are there any 3rd party programs that are more reliable/credible then those listed above (I mean as an alternative to CPUZ, etc).

Perkam

Ahhh, how silly that seems now...

Also, is cheating still so easily passed off by others?

**Asazman** · 06-25-2006, 11:04 AM

good read, you learn something new every day ...

**mursaat** · 06-26-2006, 08:40 AM

Originally Posted by tom12519

Ahhh, how silly that seems now...

Also, is cheating still so easily passed off by others?

Well this quote was from 01/12/2004

Originally Posted by perkam

Thx for the more understandable syntax Kunaak, it helps

BTW, your faked SuperPi shot is funny--> An 18s SuperPi WOW, man, what kind of knocked out system would you need to get THAT in real life.

EDIT: Though Kunaak, I appreciate, as we all do, stuff on programs that CAN be cheated, are there any 3rd party programs that are more reliable/credible then those listed above (I mean as an alternative to CPUZ, etc).

Perkam

This is an answer to that, this time on 19/01/2006

Originally Posted by Serge84

We'll see those kind of speeds when they go to diminds to replace silicon.

Speeds like that are impossible even military wise. 32ghz was it. lol On one of those faked cpuz shots.

Only diminds can make those kind of cpu speeds realisticly it would be easaly 3 times that.

To the future!!!

**The_NamelessOne** · 07-02-2006, 02:45 AM

i was trying to demonstrate that a score is a fake in an italian forum...and i unintentionally found a way to cheat cpu-z very easily with an external program ( can i post the program name ? )...i was able to fake cpu multiplier and frequency...validation also works

i think it is a completely new way to fake cpu-z, the type of cheating i used is not listed in the firts post...

http://valid.x86-secret.com/show_oc?id=103363

p.s. bad news...it is possible to modify also fsb/htt.....the only thing which cannot be changed is fsb/dram ratio...but it is not a problem if you want to fake an async score or only increasing fsb....

this is a perfect fake score...also concerning ram speed...the validation i posted before reported cpu/9 as dram ratio, instead of cpu/25....

http://valid.x86-secret.com/show_oc?id=103572

**peteypete271** · 07-10-2006, 02:02 AM

Okies follow Benchmarkian's do I have something to show youse!

I recently sent my 2x 256MB 7800GTX's to a mate in Sydney who volt modded them to 540/1350 (stock 430/1200). Whilst in the "testing for stability phase" he did a 3DMark06 run and achieved the WORLD RECORD 3DMARK06 SCORE WITH MY 7800GTX's!

According to XtremeSystems.org, the World Record is currently 14022 in 3DMark06; www.xtremesystems.org/fo...read.php?t=59753

THIS IS THE 3DMARK06 WORLD RECORD RUN OF 17361 FEATURING MY VOLT MODDED 7800GTX'S IN SLI AND HIS YONAH INTEL CPU; http://img207.imageshack.us/img207/3...6bugrun2ii.jpg

He tried to validate the score with Futuremark but it wouldn't let him. This is what we call a BUG RUN as he wasn't able to mimic this incredible score ever again as ~9.5k was more the average.

But who cares, I have a legit screenshot proving my 7800GTX's helped in obtaining a World Record 3DMark06 score!

A BIG thanks goes out to my amazing GPU Tech in Sydney, Victor, for his continuous online support! Love your work mate!

**heikis** · 07-15-2006, 06:42 AM

Originally Posted by hixie

Can someone verify if those two are real or fakes please.
These were posted in on a overclocking competition in Hong Kong. The scores seem pretty high for a X1800GTO (with unlocked 16pipelines).

http://www.geocities.com/darkknighta.../Cheat_002.jpg
http://www.geocities.com/darkknighta.../Cheat_001.jpg

Thanks.

im sure these are real. x1800gto @ 16 pipes is an X1800XL. x1800xl at those clocks does over 10k for sure. i think even 12K is no problem with the right equipment

**gianni-gt** · 07-17-2006, 05:18 AM

hmmm... isn't the multi of a opteron165 locked at 9x???

http://valid.x86-secret.com/show_oc.php?id=78027

fake?

€: o i see it could be real: http://www.xtremesystems.org/forums/...ad.php?t=88876

**Boble76** · 08-10-2006, 03:05 AM

3DMark03 26579

A64 3700+ @ 2.95GHz
single X1800XT 772/913MHz

*muhahaha*

**GOATSLAYER** · 08-22-2006, 08:54 PM

INTEL AINT GOT

ON THIS!
http://www.ocforums.com/showthread.php?t=473684

**Trouffman** · 08-22-2006, 11:18 PM

http://valid.x86-secret.com/show_oc.php?id=112636

6575.98 MHz (692.21 * 9.5)

What's the hell ?

AM2 3500+ 6.5 gigs LOLOL LOOK!
I know this is a continuation of my other thread, but the subject has sort of changed in that I just wanted people to see and laugh at this.

My AM2 and Abit board have been screwing with me lately. I'll be playing CS:S and it will just begin to slow down gradually until I am moving about half as fast as everyone else (which really ****es the other players off). At that point I will alt/tab out to check cpu-z and clockgen and see something like this:

**syne_24** · 08-23-2006, 12:04 AM

Originally Posted by peteypete271

Okies follow Benchmarkian's do I have something to show youse!

I recently sent my 2x 256MB 7800GTX's to a mate in Sydney who volt modded them to 540/1350 (stock 430/1200). Whilst in the "testing for stability phase" he did a 3DMark06 run and achieved the WORLD RECORD 3DMARK06 SCORE WITH MY 7800GTX's!

According to XtremeSystems.org, the World Record is currently 14022 in 3DMark06; www.xtremesystems.org/fo...read.php?t=59753

THIS IS THE 3DMARK06 WORLD RECORD RUN OF 17361 FEATURING MY VOLT MODDED 7800GTX'S IN SLI AND HIS YONAH INTEL CPU; http://img207.imageshack.us/img207/3...6bugrun2ii.jpg

He tried to validate the score with Futuremark but it wouldn't let him. This is what we call a BUG RUN as he wasn't able to mimic this incredible score ever again as ~9.5k was more the average.

But who cares, I have a legit screenshot proving my 7800GTX's helped in obtaining a World Record 3DMark06 score!

A BIG thanks goes out to my amazing GPU Tech in Sydney, Victor, for his continuous online support! Love your work mate!

I highly doubt that is legit. k|ngp|n broke 17K and he was on a pair of 7900GTXs with LN2 cooling. A pair of 7800 and yonah? Yea right!

http://www.xtremesystems.org/forums/...=110956&page=5

**kyosen** · 09-30-2006, 03:26 AM

I noticed a cheat method today.
It's strong...SuperPI-mod and 3DMark2001SE are affected, sadly

But recent CPU-Z is not affected by this method, fortunately

Now, still-image is not enough for evidence of any OC results, I think...
...at least, CPU-Z is must.

Edited: IP address maintenance.

**zakelwe** · 09-30-2006, 03:58 AM

Originally Posted by kyosen

I noticed a cheat method today.
It's strong...SuperPI-mod and 3DMark2001SE are affected, sadly

But recent CPU-Z is not affected by this method, fortunately

Now, still-image is not enough for evidence of any OC results, I think...
...at least, CPU-Z is must.

Hi Kyosen,

How badly does it effect 2001? Is it easy to spot ?

Regards

Andy

**ojdr2001** · 09-30-2006, 05:11 AM

I hope this is not the end of 2001...

**CAL930** · 09-30-2006, 06:38 AM

hi kyosen-san
Very sad event......

**kyosen** · 09-30-2006, 10:32 AM

Yeah, very sad affair...

I could reach 3DMark2001SE - 130k and 3DMark05 - 40k at fake 8GHz.
It seems that ORB rejects publishing suspicious score/clock for 3DMark05,
but I can publish 3DMark05 - 29873 score at fake 5.6G

Edited: IP address maintenance.

**ojdr2001** · 09-30-2006, 10:46 AM

Sad day for benchers

Thanks for sharing kyosen-san

Thread: Faking Benchmarks, The Good Benchmark List, How To Spot Cheats and More...

Thread Tools

Search Thread

Rate This Thread

Display

a sample of cheat

Bookmarks

Bookmarks

Posting Permissions