so intense, im loving it! :D congrats macci, on taking the lead... CMON OPP MAKE AMD FIRST TO HIT THE 10K BARRIER.
Printable View
so intense, im loving it! :D congrats macci, on taking the lead... CMON OPP MAKE AMD FIRST TO HIT THE 10K BARRIER.
Why is none of them using a 5900/5950u with Vmods and some exotic cooling?
I thought these cards did even better than the 9800XT in '03?
Or am i mistaken?
What's the drivers Futuremark approve for the 5950 ultra's?Quote:
Originally posted by Zeus
Why is none of them using a 5900/5950u with Vmods and some exotic cooling?
I thought these cards did even better than the 9800XT in '03?
Or am i mistaken?
I think they're still 52.16 iirc.
/me thinks you're mistaken:D
Sorry, i'm not really into 2003, but what's wrong with the 52.16 drivers then?
Damn school, i wanna stay up and see what happens.
sms me on 04230.... naaaah.
Off Topic: The 52.16 drivers have had their 3DMark03 enhancements taken from them in the updated version 3.40.Quote:
Originally posted by Zeus
Sorry, i'm not really into 2003, but what's wrong with the 52.16 drivers then?
The drivers after them have those enhancements re-instated which Futuremark are calling "cheats".
In other words, Futuremark just couldn't be stuffed releasing an update everytime nVidia release a driver update.
Whereas ATI have followed Futuremarks request of not "cheating", that's why ATI's drivers from 3.9-4.2 are support/approved.
Not really into 3DMark03 either, just remember seeing some websites posting results for 2003 with 53.-- forceware being asked to take them down or re-test them with Futuremark approved drivers....
Seems like nVidia can't help themselves, gotta be no#1 no matter what.
On Topic: I can see 10K being taken down within the next week.
/ponders why Oppainter hasn't used Cat 4.2's yet.
/thinks that 4.2 is Opps's secret weapon to break 10k!
/thinks Macci can get those CPU clocks up to take the title for "first to 10K!"
/enough of thinking!!
Macci just needs a few hundred mhz's and hes there. But to be hoenst i'd rather see an amd cpu get there first.
AMD wont beat INTEL in 2k3 IMO. Ive watched the futuremark ORB for months and INTEL has always had a slight advantage there..
I look at my 3700 with a 9500 and cry, lol.
Come on macci you can do it bro :toast:
just a little bit more :)
it's not about Intel or AMD, it's about WinXP and Win2k. 3D'03 runs better WinXP and so does Intel.Quote:
Originally posted by kommando
Macci just needs a few hundred mhz's and hes there. But to be hoenst i'd rather see an amd cpu get there first.
Great scores guys:)
Sorry for going off topic but thanks for the info... :am:Quote:
Originally posted by Minnyboy
Off Topic: The 52.16 drivers have had their 3DMark03 enhancements taken from them in the updated version 3.40.
The drivers after them have those enhancements re-instated which Futuremark are calling "cheats".
In other words, Futuremark just couldn't be stuffed releasing an update everytime nVidia release a driver update.
Whereas ATI have followed Futuremarks request of not "cheating", that's why ATI's drivers from 3.9-4.2 are support/approved.
Not really into 3DMark03 either, just remember seeing some websites posting results for 2003 with 53.-- forceware being asked to take them down or re-test them with Futuremark approved drivers....
Seems like nVidia can't help themselves, gotta be no#1 no matter what.
On Topic: I can see 10K being taken down within the next week.
/ponders why Oppainter hasn't used Cat 4.2's yet.
/thinks that 4.2 is Opps's secret weapon to break 10k!
/thinks Macci can get those CPU clocks up to take the title for "first to 10K!"
/enough of thinking!!
IIRC macci was also the first to 10k in 01.....lets see if history can repeat itself. I think it can :D.
As he is making headway in 03, he should be doing 01 aswell. Improving his 01 score that is, or he could be just letting us all wait then out of the blue and good score comes. I doubt that it would beat the FX though, thats one hard ass s.o.b. :eek:
How come everyone isn't running WinXP then ?Quote:
Originally posted by QuadDamage
it's not about Intel or AMD, it's about WinXP and Win2k. 3D'03 runs better WinXP and so does Intel.
Great scores guys:)
I'm puzzled !
Regards
Andy
maybe cause the AMD does better with 2000.Quote:
Originally posted by zakelwe
How come everyone isn't running WinXP then ?
I'm puzzled !
Regards
Andy
i really think OPP should try the AMD a64 with XP on 3d03.
Quote:
Originally posted by Stang_Man
maybe cause the AMD does better with 2000.
i really think OPP should try the AMD a64 with XP on 3d03.
IF AMD does better on 2000 then it is about the cpu and not just the OS ......
maybe Quad is right and WINXP is faster for everything. Though most people run who run AMD also run Win 2k still on the front page.
Regards
Andy
The hunt is on again :D
34 to go at the moment..
:slobber: :eek:
Keep goin :toast:
:toast:Quote:
Originally posted by macci
The hunt is on again :D
34 to go at the moment..
You're not going stop just under again like your AMD64 3200 29999.999 score are you , just for Opp to nip in and bag it !.. that would be too much ..you'd explode! :D
Regards
Andy
33 to go :D
lol, poiint by point.Quote:
Originally posted by macci
33 to go :D
Currently with 2 laps to go, Macci is in a lead by 1min, Oppainter is 2nd place, with Fugger lurking close behind in 3rd. Who will win the XtremeSystem race! Macci & Fugger are using 32-bit engine, while Oppainter's machine is 32/64bit power! It will be a close finish. Macci & Fugger both using Cat 4.2s, while Oppainter's is running with Cat 4.1s.
Outstanding Macci:toast:
And they said that 10k couldn't be done with our current hardware.
Take down 10k my man,,, before I do it:D
OPP
:toast:Quote:
Originally posted by macci
33 to go :D
Ive been checking ORB all day, no sign of the 10K mark yet but its just around the corner.
Thats intresting OPP is'nt it. People were saying that 10K couldnt be broken with current hardware. They were presuming that it couldnt be done because the scores were at 9.7K for long enough but now its only 33 marks to go.
:toast:
a bridge on the card exploded (!) during a 3dmark run (it was 20secs from the end of nature at 4505/657/471) :D
fixed the card and the hunt continues :D
Thats the spirit. Cards blowing up but he continues.Quote:
Originally posted by macci
fixed the card and the hunt continues :D
That effort deserves a 10k.
PTK
my hats off to you macci, you are truely a master of your trade. Unless op can do 10k with -40c, he wont get there.
As i was walking into my shop i noticed a strange shadow on the ground, so i looked up. And there was my three stage, up on the roof. I knew what it wa doing, it talked about it all the time. See, things just havent been the same for it since i added a compressor. The new compressor just didnt fit in, and well, the three of them just drove eachother mad....
oh yes, i knew exactly what it was doing....
but there wasnt enough time... I yelled "NOO please not yet".. right as i said that, she looked at me. Closed her eyes, and just let go. i ran to her but.....She was pronounced a peice of crap that didnt work at aproxomatly 6pm 2-17-04
maybe if id spent more time on her design....
more time
i just wanted more time
You sure u was'nt going to break 10.5K with that run :D Now that is what you call extreme... ;)Quote:
Originally posted by macci
a bridge on the card exploded (!) during a 3dmark run (it was 20secs from the end of nature at 4505/657/471) :D
I told you people would start blowing things up, such is the competition, though Europe and USA are still safe one video card and one extreme cascade seem to have had " unhappy " endings as Copyman would say.
Looks like macci managed to make an unhappy ending happy again and so has a clear run now if his soldered together card can stand the pace. Macci, that is a nice card, be careful, don't go beyond manufacturers recommended voltage :D
CC, now for something really cool in time for r420 ? Opps will put it to the best of use I am sure, he is Mr Predictable, Mr Predicatably Good ;)
I was going to test 4 sticks of bh-5 tonight to se how they would go , but this is more exciting, might just browse the forum.
When 3dmark03 was released Worm said that people would be lucky to get 5000. 1 Year later we have this. Excellent progress.
Regards
Andy
i thought that was funny. oh well
not exactly. AMD is faster in Win2k and Intel is faster, way faster in WinXP, but the real problem is 3D'03 is optimised for WinXP, Intel's favourite:)Quote:
Originally posted by zakelwe
IF AMD does better on 2000 then it is about the cpu and not just the OS ......
maybe Quad is right and WINXP is faster for everything. Though most people run who run AMD also run Win 2k still on the front page.
Regards
Andy
I'm not so sure about that QD.Quote:
Originally posted by QuadDamage
not exactly. AMD is faster in Win2k and Intel is faster, way faster in WinXP, but the real problem is 3D'03 is optimised for WinXP, Intel's favourite:)
If you look at the compares, OPP has a higher GT1 score even with the lower overall score:
macci/OPP
341.3/343.5
66.8/66.5
57.5/57
58.1/57.5
Then again macci doesn't have his CPU at max. 3DMark says 4106mhz. But the point remains that OPP has more system power at the same overall score.
"Then again macci doesn't have his CPU at max"
His old p4ee overclocked better too. Maybe if he swaps cpus and puts his old one back in or something ;)
No it didn't. He can bench 3DM01 at 45xx mhz with his current one.
maybe his cooling cant make it cold enough or his cpu is no longer stable at that high clock? Id imagine he would have had it at 4.5GHz by now if he could
346.6 in GT1 on my rig, didnt get any runs in yesterday and shut down early.
Great job macci, GL on 10K Ill be lucky to hit 9.9k
:doh: :doh: :doh:Quote:
Originally posted by macci
a bridge on the card exploded (!) during a 3dmark run (it was 20secs from the end of nature at 4505/657/471) :D
fixed the card and the hunt continues :D
can u post a pic? lol
how many mod have u made?
- 33 wow
Melkizedek, Yea I have a macro shot of the dang part that exploded :D
Still at 9967. that one was at 4434MHz (246x18 1:1) 654/471.
The CPU will do the bench at 4524MHz (251.3x18 1:1) now. Good card clocks at that speed should do the trick.
Good luck macci, I wanna see how the card exploded.
let the ES release the POWER:devil:Quote:
Originally posted by macci
Melkizedek, Yea I have a macro shot of the dang part that exploded :D
Still at 9967. that one was at 4434MHz (246x18 1:1) 654/471.
The CPU will do the bench at 4524MHz (251.3x18 1:1) now. Good card clocks at that speed should do the trick.
one session with another MB 15 x 300mhz (FSB 5:4)is possible?
250x18 1:1 2-2-2-5 has proven to be a little faster than 300x15 5:4 2-2-2-5 I think...
that's a no brainer :)Quote:
Originally posted by Soulburner
250x18 1:1 2-2-2-5 has proven to be a little faster than 300x15 5:4 2-2-2-5 I think...
Yup 250 1:1 is faster than 310 5:4
It's not really, because it looks like the high FSB would be king but in testing it really isn't.Quote:
Originally posted by unrealneo
that's a no brainer :)
Most people don't know this...I wouldn't call it a no-brainer...
every time i see "last post by macci", i anticipate just a few points away from 10k..Quote:
Originally posted by macci
Yup 250 1:1 is faster than 310 5:4
i think you already did 10k... i think you're brewing up 10100:)
heh who knows, but Macci just got his cpu to 4.4GHz so maybe that did it for 10k :D
:rolleyes:Quote:
Originally posted by Soulburner
I'm not so sure about that QD.
If you look at the compares, OPP has a higher GT1 score even with the lower overall score:
macci/OPP
341.3/343.5
66.8/66.5
57.5/57
58.1/57.5
Then again macci doesn't have his CPU at max. 3DMark says 4106mhz. But the point remains that OPP has more system power at the same overall score.
OPP's faster in 3DMark2001 right? why because his chip is more powerfull. So why does OPP lose in '03 even though his chip's faster? WinXP maybe?
I don't know the vid clocks OPP had on his card though. Anyways let's not crap this thread.
Awesome score macci, you know you can't give up now:toast:
But how about this
15 x 250 1:1 = 3750
15 x 300 5:4 = 4500
Is someone trying to say 3750 can be faster than 4500 ?????
C
nope. 4500 will kill 3750.Quote:
Originally posted by charlie
But how about this
15 x 250 1:1 = 3750
15 x 300 5:4 = 4500
Is someone trying to say 3750 can be faster than 4500 ?????
C
of course. it's 750mhz we're talking about.
just hit 10k someone please:D
Nope...Quote:
Originally posted by charlie
But how about this
15 x 250 1:1 = 3750
15 x 300 5:4 = 4500
Is someone trying to say 3750 can be faster than 4500 ?????
C
18x250 1:1 is faster than 15x300 5:4 ;)
Well, of course :rolleyes:Quote:
Originally posted by Bennah
Nope...
18x250 1:1 is faster than 15x300 5:4 ;)
But since most peeps have P4 CPU with 1 multi....
let's say a 2.8... if you have very good BH-5 and can run 270 1:1 for 3780mHz. Now the CPU is not at it's limit. Let's push the CPU farther to 295 or 4130mHz 5:4 or 236mem speed, of course in practice the 5:4 is faster. P4 1:1 benching is totally irrelevant (except if you have 18X ES like Macci) because the CPU's are capable of more raw speed.
The 4130 5:4 will be speedier than 3780 1:1 all day long.
C
we know, what you said has no relevance to what anyone said, however. :) (although true)Quote:
Originally posted by charlie
Well, of course :rolleyes:
But since most peeps have P4 CPU with 1 multi....
let's say a 2.8... if you have very good BH-5 and can run 270 1:1 for 3780mHz. Now the CPU is not at it's limit. Let's push the CPU farther to 295 or 4130mHz 5:4 or 236mem speed, of course in practice the 5:4 is faster. P4 1:1 benching is totally irrelevant (except if you have 18X ES like Macci) because the CPU's are capable of more raw speed.
The 4130 5:4 will be speedier than 3780 1:1 all day long.
C
Who's crapping this thread, and why are you rolling your eyes?Quote:
Originally posted by QuadDamage
:rolleyes:
OPP's faster in 3DMark2001 right? why because his chip is more powerfull. So why does OPP lose in '03 even though his chip's faster? WinXP maybe?
I don't know the vid clocks OPP had on his card though. Anyways let's not crap this thread.
We are having a discussion...what's wrong with that?
You realize what i'm saying is, I don't think OPP has a disadvantage because of running Win2k. His GT1 score is higher even though the rest are lower (because of lower card clocks).
GT1 is much more CPU dependant. That shows its chugging along just fine...IMO if he had macci's clocks he would be pulling at LEAST the same or higher score.
yes it did....one guy said one thing, another said something different... and blah.Quote:
Originally posted by unrealneo
we know, what you said has no relevance to what anyone said, however. :) (although true)
in orian's computer, running 3.9Ghz 3:2, 325FSB was faster in 3d01 than 3.5GHz 5:4 (not sure on the fsb on that one...)
Is it only me or is 3dmark03 search facility really really slow at the moment ? I've done a search on XT9800 and macci's score is 9923 ..is this the latest, i thought it was closer to 10k than that ?
If I do a search on all it just times out.
I'm surprised that 5:4 is so much of a hindrence that 50MHz FSb still does not get it above 1:1, good to know though.
I know what Charlie is saying, with a locked multiplier you just have to bite the bullet and 1:1 will be the first thing to go so that the cpu is not speed capped artificially.
I've been thinking the same thing about AMD64, at present I have a VIA Abit board that does not like clockgen so it runs at 240x10 or so 1:1. If I moved over to the nforce3 Gigabyte K8N that is still sitting on the shelf and run it 300x8 6:5 would the 6:5 and slightly slower nforce3 chipset mean it is not much quicker ?
Regards
Andy
Soulburner, let me explain it.
XP gives better GPU scores in 03' than 2K, even at the same clocks. GT1 will favor 2K, as its mostly all system.
So basically the question is do you run XP on the AMD and lose system power for better clocks, or get the GT1 boost with 2K and suffer a few tenths of a frame in the latter tests? Intel works best on XP so it has that first advantage.
Sucks about OPP's lack of cooling. Its a 2-man race now. :(
zakelwe, I havent uploaded any new results to the ORB yet. the 9967 run is still on my bench HDD.
Have to try the 10k again later today.
Ok, starting to lose track of what is going on , so much is happening.Quote:
Originally posted by macci
zakelwe, I havent uploaded any new results to the ORB yet. the 9967 run is still on my bench HDD.
Have to try the 10k again later today.
have to agree with Hallowed, shame Opp is seemingly out, must be frustrating if you can't bench with a chance of getting the title back. Maybe CC and Opp have a good plan B. FUGGER being quiet again maybe a new run soon.
Good luck macci.
Regards
Andy
It has nothing to do with card clocks? :DQuote:
OPP's faster in 3DMark2001 right? why because his chip is more powerfull.
If you take the total FPS number from car hi and lobbys you'll see that P4 EE ES and FX are pretty much equal on that one too.
A64 at 11x275 1:1 2-2-2-5-8-16 (it runs SuperPi 8M at that speed) should come quite close to EE and FX thou.. ;)
Some just pop the cap allready :)
I wouldn't discount Oppainter just yet. He still might have a few more pts up his sleeve. I think Macci is wise to wait and see what his challengers are going to do next. 10K will be broken, but Fugger and Oppainter will have to push Macci to get it.
I didn't know anybody was? I'd be glad to see any of the three get it.Quote:
I wouldn't discount Oppainter just yet.
What a nonsensical statementthat is. He should wait for the other guys, to see what they do, and possibly pop 10k? ;) He should just go for it, they all should.Quote:
I think Macci is wise to wait and see what his challengers are going to do next.
Sheesh man.....errr...ummmm.....jah ;)
"You're almost at the top of Mt. Everest, but I think you should wait to see what the other guys are gonna do"
;)
His Cascade is dead. Unless he gets new one built, like today, he isn't going to be in the race for 10k im afraid :(.Quote:
Originally posted by Kanavit
I wouldn't discount Oppainter just yet. He still might have a few more pts up his sleeve.
Macci was first to 10k in 3DMark2001...and its looking like he will be in 2003 as well.
Heh, I can imagine CC working doubletime trying to get a new cascade rolling...
Its nearly an endurance race, first to 10k, and times running out for two teams here. :stick:
If the 4.2's are 150 points over the 4.1's then OPP would have it, too. :(
i didn't say card clocks are not important, i said i didn't know what clocks OPP runs his at.Quote:
Originally posted by macci
It has nothing to do with card clocks? :D
If you take the total FPS number from car hi and lobbys you'll see that P4 EE ES and FX are pretty much equal on that one too.
A64 at 11x275 1:1 2-2-2-5-8-16 (it runs SuperPi 8M at that speed) should come quite close to EE and FX thou.. ;)
Also macci, i wasn't talking about A64, timings and superpi, i was stating FX-51 is simply the most powerful 3D chip and EE wins in '03 only because '03 is optimised for WinXP that's it.
I have to agree:toast:Quote:
Originally posted by QuadDamage
i was stating FX-51 is simply the most powerful 3D chip and EE wins in '03 only because '03 is optimised for WinXP that's it.
OPP
Here it is:Quote:
can u post a pic? lol
http://www.akiba-pc.com/broken.jpg
It was not designed for 657MHz 2.28V :D
damn, this 10k race seemed a lot longer than most had expected :p:
damn macci, saw that and almost had a heart attack. (before pic came up)Quote:
Originally posted by macci
Here it is:
I thought you'd done it! :D
This hardware doesn't want to go to 10k very easily it seems. I've got 9962, 9965, 9966 and 9967 now - all with different GPU/RAM/CPU clocks... :brick:
The card has taken insane amount of punishment and it still keeps on going :D I'm sure she'll get 10k in the end...if not.. :bsod:
I'm sure the magic number will appear when you're least expecting it. :)
you mean you cant hit 10k even with your cpu higher then 4.2ghz?
i mean sure... its not gonna effect it much... but even if you added "just" 200mhz, shouldnt that give you the 30-35 more 3dmark2k3 points you need?
Maybe you'll need CAT 4.3 ;)
I got up to 9945 at 4305MHz (287x15 5:4)/654/474, at 4435 (246x18 1:1)/655/473 it is 9966. 4505/657/471 should be quite close to 10k.
Now I just need to get it thru the dang bench. with lower card clocks I've passed the test at 4524MHz CPU and on the other hand the card has passed the test at 660/474 w/ lower CPU clock.
Yea CAT 4.3 (+50marks) would rock :D
macci, you have pro version, you can't select tests and run them individually?
macci, need more power! :D
hope you get it sorted. I hate it when you get scenarios like that :mad:
I would assume if you can do that, he already does.Quote:
Originally posted by Stang_Man
macci, you have pro version, you can't select tests and run them individually?
But I hear 03 doesn't let you do that.
sounds like all you have to do now is bench with higher clocks on both cpu and VC.
i'd say theres 10k in that system, without any different drivers/tweaks. ;)
Pff, yeah, I mean come on macci, thats so easy. :rolleyes:
what? are you saying its hard to overclock two parts of your system to their full potential at once? :rolleyes:
You dont have the experience to understand the finer things about taking strong hardware to its absolute limits.Quote:
Originally posted by pik-ard v1.1
what? are you saying its hard to overclock two parts of your system to their full potential at once? :rolleyes:
If it were a simple matter of just "overclocking more" it would have been done already. One overclock will determine the stability of the other and consequently you must strike a balance that can live for 20 minutes. Theres more but I don't need to explain it.
but he has done it before...
he has pushed his cpu further then his top scoring 3dmark03 run...
he's pushed it atleast 300mhz more stable enough for benching...
you would have a point if that 4.2ghz is the max he has ever benched it at, but it isnt, so you dont. ;)
"all" he has to do, in my eye, is do the two things at once, and for someone with his skill, i dont see how it could be that hard. ;)
you should note that im not telling him to OC his proccy more then he previously has...
;).Quote:
Originally posted by macci
Yea CAT 4.3 (+50marks) would rock :D
Havent benched them against 4.2's so not sure if they give more in 3DMark2k3... :D
Btw, any idea how many times you have run the bench .. lets say the last 3-4 days ?
PTK
Ps. Keep on running.. its gotta work out soon(tm) :toast:
Clocks up processor, GPU clock will fall.
GPU is 80%+ in 03'. Hence clocks at where they are. I believe this has been mentioned a few times already.
its well known that gpu is by far most important in '03... but, all he is asking for is 34 marks... its not like he needs to up his processor that much for just 34 marks...Quote:
Originally posted by Hallowed
Clocks up processor, GPU clock will fall.
GPU is 80%+ in 03'. Hence clocks at where they are. I believe this has been mentioned a few times already.
well it's just harder the closer you get. kinda like with getting down to 0k (-273°C). the closer you get to that, the smaller the steps forward you make will be. so gaining 34 pts at the level he is at right now is like searching for 1500 pts for the guys around 6000 pts with a 9500 ;)
Yeah. Its obviously (at the moment) the max clocks possible on either component for the best combo or macci would have it already.
This really isn't that hard to understand.
I say the winning formula would be his card @ 660/474. Which macci said that it had passed through the test at that before but his cpu has to be at a lower clock. Therefore run the cpu @ 242x18 (1:1) The good ol' 1:1 action would do the trick ;)
Set the card to its max firstly, then set to the cpu to suit :)
Hallowed is correct.
You may have more CPU speed, and you may have more GPU speed, but running them at the max at the SAME TIME will interfere with each other and sometimes cannot be done.
It isn't like that on my system but obviously is on his.
Yup I know that.
Thats why I said run to the card at its max and run the cpu accordingly ;)
Soul, it seems like the populace here has a hard time understanding the simple law of "dont tell macci how to bench"
We should write this down somewhere.
why not? what we tell macci may be helping him! yes they are right. Macci, leave your gpu at its max oc then get the cpu up one step at a time
:ROTF:
You have seriously GOTTA be kidding me??
Any of US....
telling MACCI how to bench??
:ROTF:
he knows alot, but many of us are also experts at benching. the suggestion macci max his gpu and clock his cpu a notch down was spot on and could be his 10k
Ace you have said some stupid stuff in the past but this takes the cake. :hehe: