• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

New 3Dmark interview, and pic

F

fallguy

Guest
B3D has an interview with the 3Dmark team. A pretty lengthy one, and some very interesting info in it. The thing that caught my eye the most was this line,

The next 3DMark will require a fully PS2.0 capable graphics card for all game tests. None of the game tests will use shaders below version 2.0, and the benchmark is really mainly targeted at the second generation of DX9 compliant cards. Of course users with older/first generation DX9 hardware can run the benchmark and get reliable results, but the tests loads aren’t really designed for that generation.
Click to expand...

I think thats great, no more DX7/8 tests. They also say you will have the option to run in PS2.0, 2.0b, and 3.0 depending in your hardware.

http://www.beyond3d.com/interviews/fm04/

sneak.jpg
 
as a demoscene fan i like 3dmark in that regard, its fun to watch the demos run

but there is still nothing better than using games to evaluate the gaming experience, it just makes sense
 
GraphixViolence here makes a good point:

http://www.beyond3d.com/forum/viewt...start=20&sid=7aace7854cda83263a8f500f6a552534

So it looks like 3Dc is out because it's only supported by one IHV, but partial precision shaders are in. How many IHV's support that optimization again? It could be argued that each of these features can provide significant performance improvements for one vendor's hardware only, while having negligible impact on image quality.

If Futuremark is going to take a position on this issue, should'nt they at least try to be consistent? I personally agree with their policy of only using features supported by multiple vendors, but I don't see how _pp hints fall into that category.
Click to expand...

And lets not forget the main problem here. This isn't a game, or even uses a gaming engine for any known game.
 
If Futuremark is going to take a position on this issue, should'nt they at least try to be consistent? I personally agree with their policy of only using features supported by multiple vendors, but I don't see how _pp hints fall into that category
Click to expand...

Because _pp hints is in the DX9 and OpenGL spec.

3dc ain't. It's strictly a proprietary format of texture compression, nobody CAN use it but ATI (hell, I think they even have a patent on it).

Partial precision is in the spec, and anybody who wants to *can* use it if they want. 3dc isn't, and can't be.
 
dderidex said:
Because _pp hints is in the DX9 and OpenGL spec.

3dc ain't. It's strictly a proprietary format of texture compression, nobody CAN use it but ATI (hell, I think they even have a patent on it).

Partial precision is in the spec, and anybody who wants to *can* use it if they want. 3dc isn't, and can't be.
Click to expand...

This is taken directly from the 3Dc whitepaper:

The new 3Dc texture format is supported in DirectX 9.0 using a FourCC code, and in OpenGL using a
vendor-specific extension. 3Dc is expected to be incorporated in some form into future versions of
these programming interfaces. To assist in this process, ATI is making all the details of this format
openly available.
Click to expand...
 
3Dc is expected to be incorporated in some form into future versions of
these programming interfaces
Click to expand...

Indeed, well, we'll see if it happens.

As of right now, partial precision is *ALREADY IN* the spec both APIs. 3dc isn't. So, maybe someday it will be....I guess we'll see what Futuremark's reaction would be when it is.
 
it would be cool if futuremark would make games that you can actually play instead of obnoxious controversial benchmarks that always seem to screw over one ihv or the other...
 
theelviscerator said:
Looks cool...does the flaming ever stop?
Click to expand...

such sensitivity...I'm beginning to wonder if anyone on this board has ever seen a real flame...
 
it would be cool if futuremark would make games that you can actually play instead of obnoxious controversial benchmarks that always seem to screw over one ihv or the other...
Click to expand...

Well, 3dMark03 DID accurately predict the performance of this (and last, and the one before) gen's hardware in today's D3D games.
 
I could hurl epithets and curses that would make a hellknight blush but is that really the point? I remember basic training being cussed at by the DI thinking,, what a puss, my old man is twice as tough as this guy....

He didnt like it when I laughed at him too much....but I made it through even though I had instantly made a spot on his sh list...
 
Bump for those who are interested. Going to be very interesting when this new version comes out, and that interview is great.
 
Brent_Justice said:
that picture looks like something that could be done in UT2K4 IMO

FarCry looks better than that IMO
Click to expand...

I agree that FarCry looks better than that hehe...but it still looks really cool :p
 
I like the 3dmark series for the demo but have never liked the way they give the results of the benchmark. Instead of baseing the score entirely off the number of frames per second a system can do, they counter in stupid cpu scores, sound scores etc etc to give the final mark. If they was just to keep it to the frames per second a machine can obtain and base the mark the machine gets entirely off the frames per second then it would be a valid system performance score.
 
Shane said:
Instead of baseing the score entirely off the number of frames per second a system can do, they counter in stupid cpu scores, sound scores etc etc to give the final mark.
Click to expand...

Sound scores? Seriously?
 
dderidex said:
Well, 3dMark03 DID accurately predict the performance of this (and last, and the one before) gen's hardware in today's D3D games.
Click to expand...

Fair enough.

I think they do what they do well...but games serve as benchmarks just as easily...so why not make games and be known as the company that makes cool games with awesome benchmarking abilities instead of just benchmarks...or better yet...make a system that allows developers to easily integrate complex benchmarking stuff into thier games...

I just think they have a single-minded view of things...all the content they develop could be put to better use...
 
Yes, for example they add certain points to the score based on your cpu speed, what type of card you get etc. There is a article about it on anandtech where they put a 9800pro in a 500mhz p3 and a ti500 in a 3.2 ghz system and it says that the 500mhz system was faster by its 3dmark score. Yet when they observed the frames per second the machines were rendering the 3.2 system blew it away in frames per second rendered.
 
Shane said:
I like the 3dmark series for the demo but have never liked the way they give the results of the benchmark. Instead of baseing the score entirely off the number of frames per second a system can do, they counter in stupid cpu scores, sound scores etc etc to give the final mark. If they was just to keep it to the frames per second a machine can obtain and base the mark the machine gets entirely off the frames per second then it would be a valid system performance score.
Click to expand...


thats is completely erroneous..all you have to do is run the first four tests to get a score..no cpu test no sound test...sheesh..
 
You think? then why is it my friend that runs a stock system with a 3.2 ghz cpu, and a 9800pro with 512mb ram gets a higher score than my system with a 2.4c oc 3.0 with a overclocked 9800pro, 1 gig of ram and when we watch the frames per second my system blows his away yet he obtains a higher score than me. I know what im talking about, and this is the reason people dont trust and base the scores given by 3dmark to be worth crap.
 
Brent_Justice said:
This is taken directly from the 3Dc whitepaper:
Click to expand...

btw FourCC code's are ATI's ways getting around MS not putting official support into DX. Same thing ATI had to do with instancing for their R300+ cards since none of them support SM3 which is what MS has to be there before you can considering instancign being there.

While you can use it in DX doesn't mean its officially supported.
 
Cryect said:
btw FourCC code's are ATI's ways getting around MS not putting official support into DX. Same thing ATI had to do with instancing for their R300+ cards since none of them support SM3 which is what MS has to be there before you can considering instancign being there.

While you can use it in DX doesn't mean its officially supported.
Click to expand...

Yep this is correct, 3dc is not part of the dx9 spec officially.

Partial precision and DTX5 both are part of the spec on the other hand..
 
Shane said:
You think? then why is it my friend that runs a stock system with a 3.2 ghz cpu, and a 9800pro with 512mb ram gets a higher score than my system with a 2.4c oc 3.0 with a overclocked 9800pro, 1 gig of ram and when we watch the frames per second my system blows his away yet he obtains a higher score than me. I know what im talking about, and this is the reason people dont trust and base the scores given by 3dmark to be worth crap.
Click to expand...

Obviously, because he has less junk installed on his system, or it's configured better.

How clueless are you? Futuremark has posted the formula on their website, it ONLY uses game tests 1, 2, 3, and 4 (specifically, game test 1 FPS * 7.3 + game test 2 FPS * 37 + game test 3 FPS * 47.1 + game test 4 FPS * 38.7). Nothing else at all - the rest if just informational. Seriously - run the full set of tests, then run just the 4 'game tests'. Your score will only differ within the margin of error. Check the Futuremark ORB - the top 10 scoring projects of all time only have the first 4 tests run!

Now, that's not to say CPU speed doesn't matter - of course it does - but only because it means the FPS in those 4 game tests are higher. CPU does, after all, effect the FPS in games, too. Not as much as the graphics card (most the time), but it certainly does effect it.
 
Shane said:
You think? then why is it my friend that runs a stock system with a 3.2 ghz cpu, and a 9800pro with 512mb ram gets a higher score than my system with a 2.4c oc 3.0 with a overclocked 9800pro, 1 gig of ram and when we watch the frames per second my system blows his away yet he obtains a higher score than me. I know what im talking about, and this is the reason people dont trust and base the scores given by 3dmark to be worth crap.
Click to expand...


post both links to the scores you are talking about.

or quit making stuff up.
 
Dont take my word for it then becuase I know and I told you I get higher fps than my friend...we watched the fps and in games my system is much faster. Anyways like I said go to anandtech or maybe tomshardware I believe they have a full explanation about this. Im just too lazy to go look it up for you people that think 3dmark scores are a true representation of the frames per second a computer can render.
 
Shane said:
You think? then why is it my friend that runs a stock system with a 3.2 ghz cpu, and a 9800pro with 512mb ram gets a higher score than my system with a 2.4c oc 3.0 with a overclocked 9800pro, 1 gig of ram and when we watch the frames per second my system blows his away yet he obtains a higher score than me. I know what im talking about, and this is the reason people dont trust and base the scores given by 3dmark to be worth crap.
Click to expand...

itS Common knowledge that when you run asychronus you suffer inefficiencies due to latencies. I run stock in dual channel quad pumped 200mhz fsb and when compared to an equivalently say a 2.4 oc'ed to 3.2 I win almost every time
 
You must log in or register to reply here.
Back
Top