I Know you dont want to see this

[aNabolic]

Cat V6.14.1.6307. Same as OPP's

http://www.dyonisis.pwp.blueyonder.co.uk/26k.jpg

http://service.futuremark.com/compare?2k1=6729581

Im honestly not cheating. I have given you a desktop pic of my config after the run.

Brett

[aNabolic]

Just to add. I have published the score so you can see it. I will take it down straight after. You were right those Cats are faster :( didnt lower my score by installing them though.

This is annoying. How can I Bench myself If I keep getting scores like this. I really dont want to reformat and start again. I also re-installed 3dM2k1 to make sure. No joy :(

Brett

[GreenBeret]

I'm sorry for being picky, but that tells us nothing more than what we have already known about your system.

Can we see the RivaTuner's card overclock, or better yet, can you run 3DMark03 ?

[charlie]

Brett,
Well let them try to explain THIS one away...hehe
I still believe it's probably a TWEAK not a cheat. There IS "something" that you're doing, as this is not duplicatable by anyone else. And if indeed it's a masterful tweak, I congratulate you. Although, if this hobby has become a contest of who can write their own code and change 3D properties, I'm out.
I think you should just be honest. Say it's some RADICAL windows optimizations...or something like that. You don't have to give up the secret, but as long as it seems like it's so easy for you people are gonna' be doubters. And truly...running a quick 3dmk03 will give you credibility, too. I'm not really giving you a hard time...I just think there's SOMETHING more to the story.
What we really need is a FM engineer to come here and explain to all of us how this score is legit...
C

[macci]

I dont think anyone has called you a cheater? All that has been said is that your system is producing bugged scores. I can't tell whats up with it but I hope you understand this yourself too?

HEre take a look and tell me if you can't find anything funny about it? :)
http://www.solidhardware.com/macci/bug_vs_dj.gif
The 4373MHz rig (aka D's rig) running with 540/414 card... your rig has over 400MHz slower CPU and lower FSB and 30MHz lower video mem..and higher score :P Sounds kinda silly dont you think?

reformat, reinstall windows, reflash BIOS, reinstall 3dmark etc etc

[aNabolic]

Just tried an 03 Run it locked on Nature. The 9800 doesnt like 03 at that GPU clock. But the tests I have published on ORB for 03 are normal !!!

I say again - I have no Windows optimisations or no secret. I have tweaked XP for overall system performance the same way everyone does. Disabled unwanted services etc... I do have the system set optimised for Background processes as opposed to the default of Appication performance.

Nothing out of the ordinary at all. Im as stumped as you are m8.

Brett

[GreenBeret]

reformat, reinstall windows, reflash BIOS, reinstall 3dmark etc etc

He should keep that Windows installation for "academic purposes" or whatever. Install Windows and everything on another hard drive and run the test again maybe ?

[charlie]

Or another idea, Brett
You could shut up EVERYBODY with a well done video showing everyhting from boot to final score...leaving the monitor ON of course.
BTW, what DO you see if you leave hte monitor on?
C

[aNabolic]

I do see it Macci - But Ive only just got my system set up how I like. Im going to stop Benching. Its less hassle and isnt of major importance to me. I know my system is fast and will play any game I throw at it. Its a Gaming rig ot a banchmark tool.

The fun has gone out of this for me now anyway. Fun while it lasted though.

>>Edit>>Charlie, I would hate to think. This sounds real dumn but I have never seen it run through at these clocks. It hurts me to see all the Artifacts. I go by the temp on the front of the Vapo. When that starts dropping its done.

Probably should look as the secret prolly lies ther :eek:

Cya,

Brett

[macci]

The 9800 doesnt like 03 at that GPU clock.
could you try it at lower GPU clock then? :)

[aNabolic]

My Sig contains 03 tests at a lower clock. It doesnt matter. These test are nerfed. Ill take them all down.

Sorry for the hassle.

Brett

[charlie]

Brett,
If it's a Windows tweak, dont' give in to these guys...
If you happened to be the first guy who put together a specific configuration, then defend it.
If I discovered that setting my desktop to PeeWee Herman gave me 27K 3dMarks, HELL I wouldn't tell anyone...I'd enjoy the dominance!
C

[aNabolic]

LOL !!!!! Youll love this :)

You were right I was wrong........

http://www.dyonisis.pwp.blueyonder.co.uk/lol.jpg

http://www.dyonisis.pwp.blueyonder.co.uk/lol2.jpg

Now what the hells that about !!!!!!!!!!!

Weird §§§§ going on for sure. Now even I believe it :eek:

Was fun while it lasted though.

Brett

[charlie]

Interesting bug though...
Should mail the HDD to Macci or somebody for a postmortem
;)
C

[ctgilles]

Publish that one :D just for laughs :D
;)

[Jeff]

:lol: That's pretty good. :lol:

;) Oh well... at least you(and everyone else) knows something's up.

[aNabolic]

Originally posted by charlie
Interesting bug though...
Should mail the HDD to Macci or somebody for a postmortem
;)
C

Nice thought but these two Raptors are staying put :)

And CT, It let me store it on the ORB but wont let me publish. Damn spoilsports :)

Brett

[charlie]

Well ya' know Brett, you've still got a first page rig that you obviously use as a 24/7 machine...which is more than most can say. Do you think Opp plays Medal of Honor at 4.2ghz and 578/422 for two or three straight hours?? Not...
So looking back in retrospect, can you think of anything you've done that might have been whee the bug was intro'ed? And yeah, I wouldn't give up a pair of Raptors either.
And FWIW I've been benching 3dmk for like 8 or 9 months an dhave always set perf to "application performance" optimization.
I just set it to "background processes" and my default rig
3.2ghz
r9700 @ 325/310
oem window install
went from 3dmk2001 16,544 to 16,632
I'll have to try it on the big rig...
thanks Brett!!!
C

[aNabolic]

Charlie,
I did mess around with the Cat3.5's and various versions of the ATI Control Panel. I went back to the 3.4's and tried to downgrade the CPL. Didnt seem to work properly so I installed the 3.1 CPL. Wich still gave me more options but installed OK.

Thats it really. Nothing special. Ill try and think. Im constantly tweaking this and that so it isnt easy to narrow it down to specifics. Ill see what I can come up with though.

Must admit its a great "TWEAK" whatever it is. ;)

Brett

[Kunaak]

just an idea.

you can run a S-video cable from your videocard, to any VCR, hit record, boot up your PC, and record everything, Bios Settings, scroll around in windows to show your windows settings, run the tests too.
then run the tape back to your PC and put it on your website or something for all to download.


this way, theres no doubting you.
no need to edit it, or cut it down for time.
anyone that wants the real details will deal with downloading a big file (after all how many of us downloaded 177 megs for 2003?)

this way, it gives us a chance to really see whats there, and squashes all the people who wanna say your running wire frame tests, and lowering the 9800 to the lowest speed possible, and whatever other theory there is about.

can you do that?

;)

[aNabolic]

Kunaak - Did you see my 03 result just up a bit. I agree there is something seriously weird about my setup. I have taken all scores down from the ORB.

Mind you Id love to publish that 16k 03 run ;)

Im going to try and figure out what I have done to cause this very very stranage abnormality.

Brett

[charlie]

MAcci,
I think that guy's score with nearly 4400mHz is underachieving, big time.
C

[ctgilles]

actually, some time ago I was thinking too...
why the hell invest like 2 to 3 grand in a rig just to get some elite numbers ? :rolleyes: bought me a shuttle and then I realised I missed it :)
but I have to say Brett, that chip is excellent under a Vapo :) care to trade? :D

[SpicyHuevos]

Damn we were right :slobber:
Most of you should realize we never called him a cheater.
We just stated that something was bugged in his score nothing more.

[Hardass]

It has already been established that no cheating was done, lets not even go there again. I want this thread to stay open.

Hardass.
Xtreme Systems Staff.

[FUGGER]

Just for laughs,

I have a few hundred less Mhz than DJ and less GPU clock yet I outscore him...

I know the 1:1 bandwidth owns his 5:4 reduction to keep Mhz high. Did Brett ever mention if he was using 1:1?

I was reaching 175FPS in 2K1 nature, didnt get a full run to post.

[Hobocrow]

Hey Brett...Nice '03 score :D

I really think that since it's an obvious anomaly with your system, maybe you should just change jobs and build systems...I can see a new mathematical formula in the future: 2K1@26K+=0 :)

Best to you for being and honest man ....rare! :toast:

[FUGGER]

Brett has shown that he can duplicate the scores and improve upon them. He has done the right thing in removing the bugged scores in question. In no way is Brett a cheater, and he will remain as one of the top benchmarkers on the team.

Thanks for being up front, sorry for the grief in all the confusion.

[charlie]

Brett,
Reinstall your o/s and start benching again...
C

[luihed]

Now there's an honest man..... hats off to you Brett and maybe the time will come that youd be #1 again..... without bugs..

[Iolao]

I re-state what I said in the other thread. I'm really sorry for you because I can imagine what should mean having so high scores and realize they are bugged. Don't mind any longer, format, reinstall OS and keep benching, your rig is very nice nonetheless.

[iolair]

Hmmm.....just wondering....if Brett can achieve these scores at lowered operating speeds due to some kind of setup anomaly, how can we be 100% sure that *all* scores are not affected by some lucky setup anomaly in one degree or another, regardless of operating speeds? Could there be a flaw in the basic premise of this particular benchmarking program?

http://home.pacbell.net/iolair/index.html

[Iolao]

well, Brett anomalies are pretty evident. In the two scores of 2k1 some test vary even of 20fps! This is not realistic if you use the same settings.

[SpicyHuevos]

I have a few hundred less Mhz than DJ and less GPU clock yet I outscore him...
well yes you are also using different drivers:)
And test 1 is the only test where cpu mhz even matter a little:)
In 2001 where system speed actually matters youre nowhere near me.
So it aint the comp its the drivers thesystem mem speed that catapult you almost 20 more marks ahead of me;)

[aNabolic]

Originally posted by Iolao
I re-state what I said in the other thread. I'm really sorry for you because I can imagine what should mean having so high scores and realize they are bugged. Don't mind any longer, format, reinstall OS and keep benching, your rig is very nice nonetheless.

Cheers, but no need to feel sorry for me. I have now reverted to all the scores I had before my system seemed to go into overdrive. Im installing XP on another drive as I type so I will be interested to see how I do with a fresh install and driver set.

Not sure what I will do If I get the same results. Will prove its not software at least. Oh and, incase anyone is interested Im running the ABIT 1.5BETA Bios.

I actually think this is pretty cool. And does show a serious issue with the Futuremark tests. Its also nice to have a "unique" system :)

I know I have a front page rig so its not such a major loss. And I had my 5 minutes of fame :banana:

Ill be back benching as soon as I slow my system down - Jeez thats a weird statement when it comes to benching :eek:

Cheers,

Brett:D

PS. All the results are saved on the ORB so if I ever get mad I know how to piss people off :stick:;)

[sandman]

So, your still going to bench for XS, right?

[aNabolic]

Originally posted by sandman
So, your still going to bench for XS, right?

If the other members dont mind. Then of course :)

Just thinking. Has anyone overclocked there 9800's so far that the screen is more artifacts than the test itself. I just wondered if because I probably pushed my core way past its unmodded state could take. Maybe it just stops rendering properly and zooms through the artifacts. Just a thought, although Im sure you have done this.

Brett

[Schneider]

You should try running the test with rivatuners anti-detector scripts on.
You can download the scripts from Guru3d (http://www.guru3d.com)
I'm not sure if they affect ati drivers..but its worth a shot

[makatee]

Of course we dont mind you benching for xtreme...keep up the work and know that we are all behind you