• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

X800 XT PE...or 6800 ultra who has the better 3dmark 2003 score report here please.

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Dispite my tendancy to degrade those that base their systems performance on 3dmark, i will instead offer some helpful advice sense it's for purely bragging rights and not out of ignorance. First off, why don't you ask "which card overclocks the best to get the highest synthetic test score on applications such as 3dmark". Why? because you are looking to overclock the card higher then your friends'. Actually, maybe in this case, you should build a pci express system and aim for getting an sli setup when they come out. That would be your best bet down the road.
 
Thanks for the advice about SLI, but I don't think that's I'll be waiting for that to come out so I can buy two video cards for 500 bucks a piece.

So getting back to the orginal question.


Please post your 3dmark 2003 scores and 2001 score if you have an unltra or X800 XT !
 
Originally posted by: SilverBack
If you really have to base your decision on a benchmark
here you go........

19k 3DMark 2003
That's a serious score 😎

OP is a poster child for why nV cheating in 3DMock was such a big deal, chumps...I mean users 😛 like him are the reason it was done to begin with 😉
 
Originally posted by: DAPUNISHER
Originally posted by: SilverBack
If you really have to base your decision on a benchmark
here you go........

19k 3DMark 2003
That's a serious score 😎

OP is a poster child for why nV cheating in 3DMock was such a big deal, chumps...I mean users 😛 like him are the reason it was done to begin with 😉


Punisher, 425 Bungholio marks don't lie......you don't want technical obsolescence to emasculate his surging electronic man-box, do you?!

Why the He!! would he care about value, or being able to actually notice a difference in his gaming experience?!

In the world of Futuremark, there can be only one......

 
Originally posted by: Rollo
Originally posted by: DAPUNISHER
Originally posted by: SilverBack
If you really have to base your decision on a benchmark
here you go........

19k 3DMark 2003
That's a serious score 😎

OP is a poster child for why nV cheating in 3DMock was such a big deal, chumps...I mean users 😛 like him are the reason it was done to begin with 😉


Punisher, 425 Bungholio marks don't lie......you don't want technical obsolescence to emasculate his surging electronic man-box, do you?!

Why the He!! would he care about value, or being able to actually notice a difference in his gaming experience?!

In the world of Futuremark, there can be only one......
:laugh: I don't understand why people still proudly display synthetic benchmark results in their sig? 😕 A top notch score requires running@settings below what most of us game with. B3D does a great job IMO of showing where the program can be useful and even at times illuminating, but I just don't think the scores from it are sig worthy I guess.
 
Originally posted by: DAPUNISHER
Originally posted by: Rollo
Originally posted by: DAPUNISHER
Originally posted by: SilverBack
If you really have to base your decision on a benchmark
here you go........

19k 3DMark 2003
That's a serious score 😎

OP is a poster child for why nV cheating in 3DMock was such a big deal, chumps...I mean users 😛 like him are the reason it was done to begin with 😉


Punisher, 425 Bungholio marks don't lie......you don't want technical obsolescence to emasculate his surging electronic man-box, do you?!

Why the He!! would he care about value, or being able to actually notice a difference in his gaming experience?!

In the world of Futuremark, there can be only one......
:laugh: I don't understand why people still proudly display synthetic benchmark results in their sig? 😕 A top notch score requires running@settings below what most of us game with. B3D does a great job IMO of showing where the program can be useful and even at times illuminating, but I just don't think the scores from it are sig worthy I guess.

No, they do. For example, if you can score 7000 with a 9800 Pro you are a 1337 overclocker.
 
Originally posted by: SilverBack
I agree.
Synthetic benchmarks don't mean a thing.
And that's pretty much exactly what I just showed him . eh? 😛
Well, as I stated B3D does a great job of showing where it can be of use, my problem with it has been it's unduely great influence on vid card purchasing. The OP is a perfect example of how influential they can be to where some spend their money. I don't know if it still goes on, but I remember when 3Dmock scores with demographics would be right on the box. I believe many people bought a retail off the shelf card based on almost no other data beyond that, and that's kind of 🙁 IMHO.
 
No, they do. For example, if you can score 7000 with a 9800 Pro you are a 1337 overclocker.
The actual clockspeeds, timings, hard&soft mods, cooling for everything in the system are more eloquent testimony to overclocking prowess than any 3DMock score IMO. Even the best overclocked system won't yield as high a result in 3Dm without all the tweaks that go with it. Tweaks that make my games look like crap, tweaks that have no place in a serious gaming system.

Besides, some drivers that address gaming issues suck for 3Dmock which requires using drivers that despite being great for 3DM do squat for current games. In between which is uninstalling, cleaning, and installing a different set, all to put an extra couple hundred points on to the bench :roll: The score doesn't necessarily reflect a stable overclock beyond ability to get through one run of the benchie either. I am among the group who believe a overclock has to survive hours of grueling stress testing like Gromacs, DIVX encoding, memtest, and P95TT before saying Dayum! sweet overclock.
 
Originally posted by: DAPUNISHER
No, they do. For example, if you can score 7000 with a 9800 Pro you are a 1337 overclocker.
The actual clockspeeds, timings, hard&soft mods, cooling for everything in the system are more eloquent testimony to overclocking prowess than any 3DMock score IMO.
That's certainly true but listing all that is cumbersome. Like it or not, 3dmark03 is the accepted e-penis around. And in a somewhat general trend, the more 3dmarks you have, the faster your card is.
 
Originally posted by: lordtyranus
Originally posted by: DAPUNISHER
No, they do. For example, if you can score 7000 with a 9800 Pro you are a 1337 overclocker.
The actual clockspeeds, timings, hard&soft mods, cooling for everything in the system are more eloquent testimony to overclocking prowess than any 3DMock score IMO.
That's certainly true but listing all that is cumbersome. Like it or not, 3dmark03 is the accepted e-penis around. And in a somewhat general trend, the more 3dmarks you have, the faster your card is.
I'll agree with you that it has a basic capability to indicate the level of performance a card is capable of, and that a unusually high score, when not the results of a bug/glitch indicates that the system is heavily overclocked/high performing. However, I'm not a lemming so don't follow the "popular e-penis" standard, most reputable review sites are able to show the performance of a card without the need for 3DMock, and only noobs are likely to rely on the assistance of the benchie to help them determine what card to get. A decision that@one point could have led them to make a poor choice 😉

In the final analysis, it will make a difference/have meaning to some, but not for others=celebrate diversity? 😕
 
Not everybody knows (or cares) about the difference between DirectX and OpenGL, between GPU core clock speed and memory clock speed, between AMD Mhz and Intel Mhz, etc. 3DMark makes for a good general rule-of-thumb for the less informed consumer.

Heck, having a person base a video card purchase on 3DMark scores is at least better than those who purchase based on memory size (256mb must be TWICE as good as 128mb!)
 
Originally posted by: Creig
Not everybody knows (or cares) about the difference between DirectX and OpenGL, between GPU core clock speed and memory clock speed, between AMD Mhz and Intel Mhz, etc. 3DMark makes for a good general rule-of-thumb for the less informed consumer.
Does it? 3Dm2k1se is very dependent on CPU speed, while '03 scores does not reflect the ability of say a ti4200 to run pre DX9 games or to run the DX9 games on the older path it uses. Or even what the differences in those paths mean to the gaming experience for that matter. Therefore the consumer has to know enough about the bench to understand this, along with how drivers and other system specs effect scores. Otherwise they are just going off the score the card gets which is going to be a bad decision as often as not. Based on the fudged drivers that artificially inflated scores, using it as the basis of a purchase could have/most likely did, leave some with a poor overall choice in the price range. The potential for such abuse, even now, makes it a very sketchy basis for purchases IMO.

Heck, having a person base a video card purchase on 3DMark scores is at least better than those who purchase based on memory size (256mb must be TWICE as good as 128mb!)
Very good point, and readily conceded 🙂

 
Okay, I'll revise my wording slightly. 🙂 3DMark makes for an iffy rule-of-thumb, but is better than nothing.
 
See this whole dicusssion is based on the fact whether I would buy a card based on a 3dmark score, That is not going to be the decision maker for me, I would just like to know what people out there are getting for those high end cards. If you have either one of the high end cards please post ! ! ! !


At this point I think I would probably go for the ultra due to the fact that it's on shelves and you can actually buy it online for a decent price.
 
3Dmark01 is based mainly on processor perfromance , whereas 3DMark03 is video card based.. Make your decisions on research not Benchmarks.. To many variables will effect the overall benchmark, so the more you tweak the higher you score.
 
OK this form is not baes off 3dmark scores just what people benched with high end video card....why can't you understand that.....damn
 
Originally posted by: Ademi99
OK this form is not baes off 3dmark scores just what people benched with high end video card....why can't you understand that.....damn

I believe I already answered your question in my first post.


If you REALLY insist on basing your purchase on 3DMark03 scores then you should know that out of the top 20 scores, 19 used an ATI X800XT PE (or an X800 Pro VIVO flashed to PE speeds). The lone Nvidia card was a 6800 Ultra at 13th place.

FutureMark Hall of Fame


If you're looking to outscore your buddies, then it's pretty obvious you want either an X800XT PE or an X800Pro VIVO w/1.6 ns memory and re-flash it into an X800XT PE. You can do a search of the Video forums to find out which brands carry the 1.6ns memory.


Link to thread with more info
 
my dad gets computer magazines where they test pre built computers and laptops from companies like dell, and systemax and places like that.....and when they test for grpahics performance they always use, comment and base 3D performance off the 3Dmark 03 score, or somethimes the 3d mark 01 score.....either way they are using a single synthetic bench mark to help readers decide on whats best......Geforce FX 5950 can rack up a great 3d mark 03 score.....but we all know its SM2.0 sucks dog nuts.
 
Geforce FX 5950 can rack up a great 3d mark 03 score.....but we all know its SM2.0 sucks dog nuts.

And thank god we know that! Somebody might be tempted to spend $338+ at one of the four vendors on Pricewatch still selling this discontinued card and not be able to play those 5 SM2 games very well!

(which doesn't invalidate your point about 3dMARK's utility)
 
Back
Top