Will 3DMark03 ever become popular?

[Cotita]

I think the reason why 3dmark 2001 became so popular is because of its MAX-FX engine. People tought there were going to be several games based ot that engine, unfortunately only Max-Payne used it.

Several review sites have reported that they won't be using 3dmark 03 as a benchmark, while other will just post them for reference but rely more on 3dmark 2k1.

Wether 3dmark 03 represents the performance of future games is questionable. We'll just have to wait and see.

How ever, I think that the majority of games, specially 1st/3rd person shooters, racing games, sports games and the like will probably use one of the already popular 3d engines, like Lithtech, Unreal and Unreal2 engines, the quake engine, AMP II, EAGL and last, but certainly not least, the Doom3 engine.

So to better reflect the performance of future games, I think that reviewers should use benchmarks that use those engines.

I know that if my pc can handle UT, then it should have no problems running Undying, or if my vidcard runs Alien vs predator2 then it should run No one lives forever2 quite well, and if my pc runs NFL2003, then it can run Fifa2003 or NFS6 or NBA 2003 without a hicup.

These are games people actually play, not a sinthetic 3d benchmark.

So in the end it doesn't really matter if the Radeon9999 Plus Ultra Hyper is faster than the geforceFX Warp25 Omega Turbo in 3dmark03, but which one is faster in real games.

[kyleb]

your just lissening to the vocal minority Cotita. 3dmark03 is just as good if not beter at its inteneded purpose than the last one; look at 3dmarks backers for a moment and ask yourself why they would support a bad benchmark:


Strategic BETA Members:
AMD, ATI, Intel, Microsoft


Active BETA Members:
Creative, Matrox, S3Graphics, SiS


BETA Members:
ALi, Dell, CNET, Gateway, Imagination Technologies/PowerVR, InnoVISION Multimedia Trident Microsystems


found here (http://www.futuremark.com/betaprogram/).

[gokickrocks]

Originally posted by Cotita

Several review sites have reported that they won't be using 3dmark 03 as a benchmark, while other will just post them for reference but rely more on 3dmark 2k1.


review sites stated that they WILL use 3dmark03, just not the FINAL scoring

[jbirney]

I think the reason why 3dmark 2001 became so popular is because of its MAX-FX engine. People tought there were going to be several games based ot that engine, unfortunately only Max-Payne used it.

Do you remember who was hyping 3dmark2k1 the most? nV was at the launch of the original GF3 card. It was the only "tool" at the time that showed off what a programmable TnL (Ie VS/PS) can do for a card/game. NV at that time was the King for the last few years. They had their PR machine in full view and a lot of people said ahhhh nice benchmark. Since NV said its good, then a lot of people did not bother to think. I do agree a real engine was a plus. But we all know that MAX-FX engine performance != Unreal Tournament performance != Serious engine performance, != Volition (red faction) performance , pretty much it does not equal any of the game engines you listed so what good is it then? Then how many real games were based off that engine vrs all the others? Right a couple at best (I can only think of Max Payne) vrs a whole bunch. So I don't buy a real engine as the contributating factor.

Several review sites have reported that they won't be using 3dmark 03 as a benchmark, while other will just post them for reference but rely more on 3dmark 2k1.

Yeap its too bad these sites can not think for themselves and have to rely on NV PR. Did you read TR or B3D 3Dmark stuff? If so you can see that most of NV claims do not hold that much water. And the fact that it does not represent real games? WTF? IT has NEVER EVER represented real games. Not once. If they were really upset about this then they should have wrote the PDF back when 3dmark1999 came out.

Wether 3dmark 03 represents the performance of future games is questionable. We'll just have to wait and see.

Well we are still waiting to see if 3dmark2001 was representative of games :) But yea jury is out on 3dmark2k3....


So to better reflect the performance of future games, I think that reviewers should use benchmarks that use those engines.

Can not happen. I agree real games is the best but forward looking? What happens when the game engine gets complete overhauled? Based on your statement I could look at how UT preformed and make a prediction based off that of how a card will work in UT2k3? Will a video card today runs JK2, SOF2, RtCW will give me an idea how it will perform in DOOM3? Heck you can not even look at the same engine and predict how different titles will react. Case in point. Frame rates for cards in Q3 do not carry over to RtCW, JK2, SOF2. For example the 8500 looses big time in standard Q3 benches to the GF4 ti4200. But its right on the heels if not ahead of a GF4 ti4200 in the other Q3 based games (RtCW, JK2, SOF2, ect). So even from one game to the next you can not really predict how things are going to run.

These are games people actually play, not a sinthetic 3d benchmark.

Synthetic benches give us a guess how cards might perform in future games. A good review will have real life bench and a few synthetic thrown in. You should never base a review off one or the other.

[creedamd]

it would be dumb for Hardware reviewers not to show the 3dmark2003 score, if a graphics card runs 2003 smooth, the chances are exceptional that it will run current games like butter, 2001 is less reliable, it is cpu intensive.

To not use 3dmark03 is ignorant, and nvidia biased.

[Cotita]

I wonder what will happen if the nv30 outperforms the r300 significantly in doom3.

Not likely, but several people are claiming that the true performance of the nv30 will be seen in doom3.

If I remember correctly, hard ocp won't be using 3dmark03 at all, beyond3d won't post final scores (like they do with 3dmark01), a few others are "evaluating" it and some others like tech report will include it in their set of benchmarks.

There has been rumors that one of the main reasons why nvidia stepped out from 3dmark03 is because futuremark refused to use CG for its pixel shader tests and used assembly instead.

[Shinri Hikari]

The debate will not end here, as some will agree and some will not.:cool: And who said their a minority?:D More like 50%, I will guess.:cool: And some just do not care!;) :D

[abb]

Will 3DMark 03 ever become popular?--Yes, it already is. The majority of reviewers have already stated that they will implement this benchmark in their testing. The only ones that are opposed to this benchmark is Nvidia. The only reason tor that is because they do not have the technology yet (and probably will not for some time) to take full advantage of this benchmark. The Ti4600's score of 1688 (avg), really makes Nvidia look like crap. Especially that being their cadillac right now. Then again, remember when 3DMark 2K1 came out, and the only card to score a high score was the GeForce 3? Well, same story except that now it is not Nvidia and they do not like it.
Abb

[Cotita]

Originally posted by jbirney


Can not happen. I agree real games is the best but forward looking? What happens when the game engine gets complete overhauled? Based on your statement I could look at how UT preformed and make a prediction based off that of how a card will work in UT2k3? Will a video card today runs JK2, SOF2, RtCW will give me an idea how it will perform in DOOM3? Heck you can not even look at the same engine and predict how different titles will react. Case in point. Frame rates for cards in Q3 do not carry over to RtCW, JK2, SOF2. For example the 8500 looses big time in standard Q3 benches to the GF4 ti4200. But its right on the heels if not ahead of a GF4 ti4200 in the other Q3 based games (RtCW, JK2, SOF2, ect). So even from one game to the next you can not really predict how things are going to run.



Synthetic benches give us a guess how cards might perform in future games. A good review will have real life bench and a few synthetic thrown in. You should never base a review off one or the other.

That analogy applies exactly the same with 3dmark03, if the radeond9700 beats the geforceFX in 3dmark03 (or even if it was the other way around) means nothing to doom3 because their 3d engine is completely different.

Even games that share the same engine have different results. As you mentioned the radeon 8500 was slower than the ti4200 (and even the geforce3) in quake 3, but almost meet the ti4200 when running MOF and in sometimes even beat it in RTCW.

I think that benchmarking has become much more complicated, since its becoming much more difficult to have apples to apples comparison when you have so many different variables.

[tamattack]

Exactly. That's why you want to use more tests, not less. That's another reason to add 3dMark2003 to your benchmark test suite.

[Admiral Horror]

When does the geforceFX Warp25 Omega Turbo come out? Do u have any info on the specs?

[DaveBaumann]

Originally posted by Cotita
If I remember correctly, hard ocp won't be using 3dmark03 at all, beyond3d won't post final scores (like they do with 3dmark01), a few others are "evaluating" it and some others like tech report will include it in their set of benchmarks.

There has been rumors that one of the main reasons why nvidia stepped out from 3dmark03 is because futuremark refused to use CG for its pixel shader tests and used assembly instead.

A.) Beyond3D, as a general rule, does not use the final score from any version. You may find it in a few reviews (dunno, haven't looked at all of the other reviewers reviews) but it certianly won't have been used in mine or Rev's in the past year or so.

B.) Why would Futuremark use Cg in the first place? 3DMark is a DX benchmark and considering DX has its own HLSL it would frankly look very odd if they even thought about using Cg!

[StealthHawk]

Originally posted by creedamd
it would be dumb for Hardware reviewers not to show the 3dmark2003 score, if a graphics card runs 2003 smooth, the chances are exceptional that it will run current games like butter, 2001 is less reliable, it is cpu intensive.

yes...and most current games are cpu intensive as well. so what's your point?

[scott123]

3dmark03 is a dismal disaster. Futuremark really missed the mark, and they will probably need to release another one within a year.

If Futuremark cannot get ALL the folks on-board (including Nvidia) for future beta testing, they won't make it in the long run.

Scott

[creedamd]

Originally posted by StealthHawk
yes...and most current games are cpu intensive as well. so what's your point?

point is that it's supposed to be a video card benchmark, not a cpu benchmark, you can take a 9700pro and a 2 ghz system and get a score on 3dmark2001, now take the same 9700pro and put it on a 3 ghz system and what is the score? A substantial improvement. Do the same for 3dmark03, what happens? The score is close to the same. Because it is a benchmark of graphics.

the name of the company is FUTUREmark, one can only hope that games look that way in the future, hopefully they will make games using the same technology that made 3dmark03. Can you imagine the card that will run that thing smooth? It is a good goal for the graphics industry and I believe it will become a standard. I know I will take it into consideration, but not base a purchase on it. Anyone who says " I don't care what it gets in 3dmark03" when purchasing a new card will be a.)lying or b.) a fool.

[creedamd]

Originally posted by scott123
3dmark03 is a dismal disaster. Futuremark really missed the mark, and they will probably need to release another one within a year.

If Futuremark cannot get ALL the folks on-board (including Nvidia) for future beta testing, they won't make it in the long run.

Scott

I don't believe they have to have nvidia on to be sucessful. Ati has the whole graphics market cornered, from value to highend, and probably will for a while. They will win over alot of enthusists in the next few months. 3dmark03 will become a standard.

[Cotita]

Originally posted by DaveBaumann

B.) Why would Futuremark use Cg in the first place? 3DMark is a DX benchmark and considering DX has its own HLSL it would frankly look very odd if they even thought about using Cg!

As far as I know futuremark didn't use HLSL at all but used assembly instead.

Actually the use of Cg might have helped to include more DX9 effects, the downsize is that nvidia cards would be most benefited.

[Kruno]

Originally posted by creedamd
it would be dumb for Hardware reviewers not to show the 3dmark2003 score, if a graphics card runs 2003 smooth, the chances are exceptional that it will run current games like butter, 2001 is less reliable, it is cpu intensive.

To not use 3dmark03 is ignorant, and nvidia biased.

That's what I have been saying. I have used 3dmark to diagnose and test 3dvideo performance. If my 3dmark03 score is great, but an old game is slow. I will know it has nothing to do with my video card. (drivers may have broken the game and iwould ask if that occurs, so I would get further without needing to do much work)

Also if capable of generating a high 3dmark score, games will run good. CPU limited games I have come across are: Commanche 4, DF:BHD and finally DF:LW.

That's it. Every other game takes a large hit with FSAA and AF enabled, even when it's disabled, games like UT03 run so much faster by nudging texture details to medium than their maximum in game options allow. Might I add, disabling sound and all other CPU tasking options don't give me much of a performance increase, if any at all. What's more is that my CPU is out of date. :rolleyes:

If people only knew how to configure things, run games from a CLI with a high priority setting and explorer and just about everything disabled they will find that a vast majority of games are not so "cpu limited" after all.
BTW: This does not include games like Quake 1. ;)

[scott123]

I don't believe they have to have nvidia on to be sucessful. Ati has the whole graphics market cornered, from value to highend, and probably will for a while. They will win over alot of enthusists in the next few months. 3dmark03 will become a standard.
This effort by Futuremark is like amature hour. The graphics look terrible, and the benchmarks are based on conjecture not fact. The 1999/2000/2001 efforts were great, but 2003 IS a disaster.

Futuremark, to be a valid benchmark needs ALL the big players on board. Right now the 3dmark03 benchmark has been deminished, and is not really the benchmark of choise anymore.

I don't know if Futuremark will ever come out with another valid benchmark, but I wish them luck because it's back to the drawing board.

Scott

[StealthHawk]

Originally posted by creedamd
point is that it's supposed to be a video card benchmark, not a cpu benchmark, you can take a 9700pro and a 2 ghz system and get a score on 3dmark2001, now take the same 9700pro and put it on a 3 ghz system and what is the score? A substantial improvement. Do the same for 3dmark03, what happens? The score is close to the same. Because it is a benchmark of graphics.

i fully realize that 3dmark03 is a video card benchmark. the problem, as has been stated more times than i can count, is that it is called a gamer's benchmark.

it is not called a video card enthusiast benchmark. now that says to me that 3dmark03 should be indicative of future performance. like i said somewhere, we will see how Doom3 runs on all these cards. right now there is no proof one way or another that says 3dmark03 will or will not be indicative of games in the future.

Anyone who says " I don't care what it gets in 3dmark03" when purchasing a new card will be a.)lying or b.) a fool.

well like i said, we will see how Doom3 runs. if we don't see the performance pattern in Doom3 that 3dmark03 has, then i wouldn't care about the 3dmark03 score at all.

i never cared about my 3dmark2001 score. great, my friends with gf2mxes had pitiful scores compared to mine because they couldn't run the Nature test. and their cards choked in 32bit color. but i know that the gf2mx performed poorly compared to my card based on in-game benchmarks already. 3dmark2001 never meant anything compared to real games either. and how many games do i own that have special DX8 effects....hmm, zero.

[digitalwanderer]

Originally posted by scott123
3dmark03 is a dismal disaster. Futuremark really missed the mark, and they will probably need to release another one within a year.

If Futuremark cannot get ALL the folks on-board (including Nvidia) for future beta testing, they won't make it in the long run.

I dunno, I sort of think you got that backwards. If nVidia doesn't get on-board with Futuremark soon I think it's gonna just keep being another PR black-eye for 'em.

They're looking like sore losers because their shiny new cards ain't all they were cracked up to be, plain and simple.

I'm sort of shocked that 3dm2k3 is already as popular as it is...I really didn't expect it to be adopted as quickly as it has been. I still pretty much use 3dm2k1se to bench D3D as I think it's more accurate and representative of most of the games out today...but 3dm2k3 is gonna be added to me benching regime after my next upgrade I'm pretty sure. (Ok, I just can't STAND scoring under 2000! ;) )

Right or wrong, 3dm2k3 is going to be one of the new standards...regardless of what nVidia says about it. :)

[Kruno]

Originally posted by digitalwanderer
I dunno, I sort of think you got that backwards. If nVidia doesn't get on-board with Futuremark soon I think it's gonna just keep being another PR black-eye for 'em.

They're looking like sore losers because their shiny new cards ain't all they were cracked up to be, plain and simple.

I'm sort of shocked that 3dm2k3 is already as popular as it is...I really didn't expect it to be adopted as quickly as it has been. I still pretty much use 3dm2k1se to bench D3D as I think it's more accurate and representative of most of the games out today...but 3dm2k3 is gonna be added to me benching regime after my next upgrade I'm pretty sure. (Ok, I just can't STAND scoring under 2000! ;) )

Right or wrong, 3dm2k3 is going to be one of the new standards...regardless of what nVidia says about it. :)

Unreal Tournament 2300? WTF??? :confused:
3dmark 2300?
:confused:

Pass around what you are smoking plz. :D

[digitalwanderer]

Originally posted by K.I.L.E.R
Unreal Tournament 2300? WTF??? :confused:
3dmark 2300?
:confused:

Pass around what you are smoking plz. :D

I always share, but I don't get where I messed up. :confused:

Please share the funny as I must be extra thick-headed, I love laughing at meself. :)

[DaveBaumann]

Originally posted by Cotita
As far as I know futuremark didn't use HLSL at all but used assembly instead.

Actually the use of Cg might have helped to include more DX9 effects, the downsize is that nvidia cards would be most benefited.

Thats exactly as I wrote in our Postmotem article. However, under no circumstances does it make sense for them to use Cg when there is a perfectly good, native DirectX, vendor unspecific, HLSL in DirectX9.

Really, the only time a DX developer would use Cg is if they are porting across to OpenGL as well, if you are doing DirectX only development, why would you use Cg?

[Kruno]

Originally posted by digitalwanderer
I always share, but I don't get where I messed up. :confused:

Please share the funny as I must be extra thick-headed, I love laughing at meself. :)

2k3 is 2300. :)

:)