4870X2 3d mark 03 bugged score thread

[Gautam]

Vista SP1 32

128197




Here is my run with 780/900 on cards and 1112fps in GT2, why 1450 is inpossible ? at 870 on GPU and 6200 on CPU

That's still way too low. 4.8Ghz and 790/915 I get 133k with 1180 GT2. And even that is way lower than spl.

SF3D and what about the single card score ,can they be bugged 2 ?

Yes.

[George_o/c]

SF3D and what about the single card score ,can they be bugged 2 ?

According to that :

The black screen bug works even with a single vga. I've proved that with my screen on the first page of this post (143k with 3.6 ghz e8500 and 875/1250 3870 single -.-').

, I assume the bug exists when using a single 4870X2 too

[.p4nzer]

According to that :



, I assume the bug exists when using a single 4870X2 too

Yeah the bug is everywhere!

I have used a 3870... So can be used by any card 'cause zolkorn made it with gtx280.

The 3dmark03 score are totally ed, not only 4870x2.

BTW I don't think someone has used this bug before... Maybe only the newest run must be checked.

Bye

[Achill3uS]

how do you guys make this black screen effect i dont now, i can't make bugged runs with even vista64 and 8.8 cata...
its just completely slow compared to 155k runs at almost the same cpu/gpu clocks, ofc if those are real score

[blind_ripper]

me and massman will see what the bug is on 03 with a single card .
hope we dont need to ban hwbot score's about something stupid like this , although there are no bugs in 01 ,05 and 06.

[AndreYang]

Becuse thats what I have on my table at the moment. Becuase im trying out 06 and Vantage with 4870X2. Thats is the score I get every single time.

It the movie isnt proof enough then what can we do? No more 3dmark 03 I guess.

Please be more calm about the issue Andre, I don't see why you are so agressive with me. I'm just trying to solve the issue so we can continue to use 3dmark 03.

NO..I did NOT ONLY point to you. I didn't want to say who make bug is because everyone is possible. I research this bug for one week and try to find how to control GT1/2/3/4 bugs. Depend on gpu clock, you can easy to handle score you want. I am so disappoint about some guys did them. I have did the same speed/gpu as kinc and control bugs GT2/GT3 close to kinc. YES, the score is the same. I know it's bug because I handle it. If everyone think your score is real, I could say my handle score is real,too.

Forget 03, kinc, until driver fixed them. I didn't respond any 03 messages and pms here until all problem fixed.

[ultimeus]

6 month ago I was doing 123k with a 8800 gts g92 only using bug. Futuremark have to do a patch.

[.p4nzer]

NO..I did NOT ONLY point to you. I didn't want to say who make bug is because everyone is possible. I research this bug for one week and try to find how to control GT1/2/3/4 bugs. Depend on gpu clock, you can easy to handle score you want. I am so disappoint about some guys did them. I have did the same speed/gpu as kinc and control bugs GT2/GT3 close to kinc. YES, the score is the same. I know it's bug because I handle it. If everyone think your score is real, I could say my handle score is real,too.

Forget 03, kinc, until driver fixed them. I didn't respond any 03 messages and pms here until all problem fixed.

But u know how to control the bug even without a black screen?

I wonder how...

I'm curious, i wanna know how it works. Can u tell that?

Bye

[Kinc]

NO..I did NOT ONLY point to you. I didn't want to say who make bug is because everyone is possible. I research this bug for one week and try to find how to control GT1/2/3/4 bugs. Depend on gpu clock, you can easy to handle score you want. I am so disappoint about some guys did them. I have did the same speed/gpu as kinc and control bugs GT2/GT3 close to kinc. YES, the score is the same. I know it's bug because I handle it. If everyone think your score is real, I could say my handle score is real,too.

Forget 03, kinc, until driver fixed them. I didn't respond any 03 messages and pms here until all problem fixed.

But you say you can control it, I only run and this is what I get as shown in the videos. How are you able to control the scores? We need to take this to the right person at FM so they can fix it. Can you share here? Or I can give you contact person at FM that will do the best to remedy this and you can contact directly. They need all the input on this issue as they can get. For now 03 is sadly dead yes

[BenchZowner]

Guys what I think is needed to be done is:

1) Have someone run some normal frequencies on the CPU ( that everyone can reach ) and some "high clocks" on the VGA to "trigger" the blank screen bug or the bug itself as Andre says.
2) Give us the full rig specs ( hardware & software ), and the "extra" software installed and/or used ( example RivaTuner + LoD @ +3.9, etc ).
3) Initiate contact and have someone at FutureMark check this out himself.

And for God's sake don't start flaming each other ( as far as I can see things this haven't been done, it was one person misinterpreting the other person's post )
I might have more to add later on... I'm not 100% concentrated at the moment.

[Gautam]

Yes, FM must fix this ASAP.

[blind_ripper]

Guys what I think is needed to be done is:

1) Have someone run some normal frequencies on the CPU ( that everyone can reach ) and some "high clocks" on the VGA to "trigger" the blank screen bug or the bug itself as Andre says.
2) Give us the full rig specs ( hardware & software ), and the "extra" software installed and/or used ( example RivaTuner + LoD @ +3.9, etc ).
3) Initiate contact and have someone at FutureMark check this out himself.

And for God's sake don't start flaming each other ( as far as I can see things this haven't been done, it was one person misinterpreting the other person's post )
I might have more to add later on... I'm not 100% concentrated at the moment.

me and massman will give u some score's soon .
and like bill say's pleas keep it clean , started this thread becaus it would start in other threads.

so guy's dont blame any one yet plz, its hard to see top guy's freaking about this little bug !

[Kinc]

I have removed my score from the ORB since I want to wait with 03 until this issue is solved. Going to do some other fun stuff instead

[Gautam]

I ran at 5600MHz (multi is actually 10, left EIST on ):



Kinc's run at 5600MHz:



I don't think 30MHz on the GPU and 70MHz on the memory could possibly make up an 8000 point deficit. Bugs aside, something seems to be hampering the CF performance. This time around this was a P5E64 WS Evolution, for me it performs exactly the same as the Rampage Extreme.

I don't care about the bugged runs at all at the moment, I'm more concerned in finding out what this gap is.

[filmbot]

If you took the time to make this run, why not match the clocks exactly?

[massman]

I don't think 30MHz on the GPU and 70MHz on the memory could possibly make up an 8000 point deficit. Bugs aside, something seems to be hampering the CF performance. This time around this was a P5E64 WS Evolution, for me it performs exactly the same as the Rampage Extreme.

I don't care about the bugged runs at all at the moment, I'm more concerned in finding out what this gap is.

Which of the two is a bugged run: Kinc's high score or your low score? What memory freqs are you using?

[CCC]

http://www.hwbot.org/result.do?resultId=766920

this score is bugged ot hell and back

1100fps gt1 at 6ghz
but 800fps nature gt4 at only 820/1000

come on

[Gautam]

If you took the time to make this run, why not match the clocks exactly?

Because I can't. :-/ Furthermore, I don't even have any way to go past 1000 on the memory.

[Gautam]

Which of the two is a bugged run: Kinc's high score or your low score? What memory freqs are you using?

I don't think either are bugged, to be honest. I can do the same runs all day, and I know Marcus can do his all day. Like I said, I don't care about the bugged runs atm. This is a separate performance issue with 4870X2 CF.

Kinc, joe_cool and spl have a lot of runs that are consistent. They are the only three with "high" runs. Mine are more "normal." I'm thinking there is something special going on with the configuration.

The mem's 1:1 at 5-4-4-15 because the board was being an ass, but it doesn't make a difference. Raising GPU clock from 790/915 to 842/1000 helps only by ~1500 pts, so there's no reason that going to 877/1070 will suddenly get much higher.

[ahmad]

6 month ago I was doing 123k with a 8800 gts g92 only using bug. Futuremark have to do a patch.

Not ATI specific issue.. why do people continue to flame ATI for this bug?

This bug affects ALL (yes, even matrox )

[DMH]

Not ATI specific issue.. why do people continue to flame ATI for this bug?

This bug affects ALL (yes, even matrox )

Because it's we're it's more noticeable?

[STEvil]

I don't think either are bugged, to be honest. I can do the same runs all day, and I know Marcus can do his all day. Like I said, I don't care about the bugged runs atm. This is a separate performance issue with 4870X2 CF.

Kinc, joe_cool and spl have a lot of runs that are consistent. They are the only three with "high" runs. Mine are more "normal." I'm thinking there is something special going on with the configuration.

The mem's 1:1 at 5-4-4-15 because the board was being an ass, but it doesn't make a difference. Raising GPU clock from 790/915 to 842/1000 helps only by ~1500 pts, so there's no reason that going to 877/1070 will suddenly get much higher.

Possibly PCI-E timings or frequency?

[hipro5]

I ran at 5600MHz (multi is actually 10, left EIST on ):

Kinc's run at 5600MHz:

I don't think 30MHz on the GPU and 70MHz on the memory could possibly make up an 8000 point deficit. Bugs aside, something seems to be hampering the CF performance. This time around this was a P5E64 WS Evolution, for me it performs exactly the same as the Rampage Extreme.

I don't care about the bugged runs at all at the moment, I'm more concerned in finding out what this gap is.

You see difference EVERY 10MHz core frequency on those new 4870.....They are VERY powerfull....

[Gautam]

You see difference EVERY 10MHz core frequency on those new 4870.....They are VERY powerfull....

I see it, but it will not be 7000 points. I gain 1500 points going from 790/915 to 842/1000.

I won't suddenly gain 7000 points going from 842/1000 to 877/1070.

Right now there are only 3 people alive that I see with "high" scores. And look at everyone else with "normal" scores.

Everyone that's trying to answer either has normal scores or doesn't even have the cards. I want to hear answers from the 3 people with the high scores. I don't care about how to bug the runs right now. We need answers from these 3 people, simple as that.

So where are your scores anyways George? If you can match Kinc, joe_cool, and spl, and have an answer, I'd love to hear it.

Edit: Here is what I get at 842/950 at 4.8GHz.



spl:



Now don't tell me it's the 8MHz extra on the GPU doing it lol. (And no, before anyone brings it up, it's not the FSB either, just look at Kinc's monster score at only 400MHz FSB)

[r1rhyder]

Here is my bugrun to add to the others. Can you say busted.


By r1rhyder at 2008-09-06