Kyle's new thread @[H]

[demalion]

Rev,
I think it is a nice way to split off side discussions that are closely related to technology and hardware discussions, and I think that is what this thread does...its title is clearly stating what it is about, and if it doesn't interest it would be easy to skip based on that title.

And I disagree with your premise about game benchmarking...I think it held some merit before shaders, but I think standardized instructions for graphics output make 3dmark 03 much more directly useful...the shader execution in 3dmark 03 is much more directly pertinent to other games using shaders than what have been termed "synthetic" benchmarks before. I've given my reasons in prior discussions about 3dmark here.

 

[dan2097]

if nobody had used 3DMark or Shadermark etc. would anyone have realised that the actual DX9 abilities of ATI's hardware is better than that of CineFX's?

Shame nvidias trying to cheat on both benchmarks to make up for this deficit

That thread on hardocp is laughable, it makes nvnews seem unbiased. There is absolutly no way Nvidia cant be cheating on 3d mark 03 and they cheated the scores by over 20%!

Comparing to Atis quake thing people like hellbinder have feasible reasons for it (r100 optimizations not working on r200) which do make sense meaning it could jsut have been a bug, and besides it was only like a 10% increase max and the next drivers gave that perforomance and then some meaning the cheating drivers were even representing the true potential of the cards.

Nvidias 50% deficit in mother nature is a hardware issue which while they try to hide with drivers will likely come back to bite consumers when dx9 games actually become available. Based on 3d mark 03 results youd have thought the card would be MORE future proof than the r9800 pro

 

[Rodéric]

Another "BTW".

I wish that you guys would stop posting a thread like this in this particular forum. In fact, I wish you guys would simply stop posting threads like this here per se. At the risk of sounding "elitist", I'd really appreciate it if we could return to the old, old Beyond3D Forum, where almost all threads are about 3D technology and that if you want to sound off about other website opinions, do it at other forums. Again, sorry if this sounds kinda "elitist" or pompous.

I'm not going to decide what to do with this thread... John can have the pleasure

Agreed.

 

[Matt]

...
So to directly answer your question, I am not sure how one can cheat at altering the score of a benchmark that obviously has no value to begin with.
...

I think he's said he had a laugh at some of the comments in this forum. I really think he should pay more attention and go back and read some of those comments again. Hey, but I'm not a webmaster, what do I know? If you run a website, maybe you stop needing things like reasoning to support your opinions.

As a webmaster, if you hold Opinion #1, then people who have Opinion #2 and Opinion #3 and Opinion #4 will say "maybe you stop needing things like reasoning to support your opinion" and attack said webmaster for holding an opinion they don't agree with. That's what it's like being a webmaster - holding an opinion means pleasing 1/4th of your readers, while the rest either shrug it off (those who don't have an agenda) or attack him for the opinion (those who do have an agenda).

There are so many times that I don't agree with someone, especially webmasters, but I don't attack them unless they make a mistake over and over again (like saying that refresh rates suck on LCD monitors) and don't respond to my email correcting them.

3DMark03 has become exactly what I feared, a scapegoat within an industry that is having a hernia. Both ATI and NVIDIA are having a tug-of-war, and they've placed 3DMark03 directly in the middle, and I don't know if I can trust a benchmark that has THAT much importance placed upon its shoulders. Especially when you have to pay $5,000+ to become involved with it. Do you?

 

[RussSchultz]

elitist diatribe

Agreed.

Agreed.

Now if we could just get posters like Dan2097 to shorten their "feelings on NVIDIA" to a link to some canned response. I can't count the number of people who joined in the last week or so who posted a nearly identical statement.

 

[demalion]

Matt,
You didn't address the central point of my comment, namely that the reasons for an opinion are important, no matter who you are. It isn't his being a web master that is an issue for my criticism, but having that fact as a substitute for presenting valid reasoning for your opinions.

Please note the last sentence of what you quoted again, and perhaps recognize that I've provided reasons for it besides him being a "webmaster".

 

[WaltC]

Agreed. I'll let this one run its course but in the future any such threads will probably be moved to the General Discussion forum.

It's kind of like being stuck in some bizarre Twilight Zone where illogic is logic, inaccurate is accurate, PR spins become acceptable technical data, and 2 + 2 = 5. Or like a scraped patch of skin which itches so much you feel compelled to scratch it and then the wound gets reopened and the cycle repeats...

nVidia could easily defuse the situation and apply some balm by admitting to a couple of things and pledging to rid its drivers of their ability to recognize specific benchmarks and invoke certain behaviors which only occur when those benchmarks are encountered. So far, though, nVidia hasn't even begun to admit any culpability whatever. I mean, can we assume that when nVidia "looks into it" and "fixes" their "driver bugs" that this behavior in its drivers will cease? Who knows? It's maddening.

I'm quite sick of the whole topic and would like nothing better than to see it resolved. But with the poverty of nVidia's responses thus far, I don't think nVidia's interested in much except trying to convince me that 2 + 2 = 5, and that's not going to happen.

 

[Dean]

Is Kyle an Nvidiot? NO
Is Kyle an ATI FanATIc? NO
Is Kyle stubborn to the point of never accepting another viewpoint? Possibly YES

You will not see Kyle express any negative views regarding Nvidia's alleged cheating in 3dMark simply because he knows if he does so, it would somehow validate 3dmark as a benchmark(which he refuses to recognize).

He has his opinion and may alienate many of his followers for doing so. That is the way he has always been and we are not going to change it.

Dean

 

[Doomtrooper]

3DMark03 has become exactly what I feared, a scapegoat within an industry that is having a hernia. Both ATI and NVIDIA are having a tug-of-war, and they've placed 3DMark03 directly in the middle, and I don't know if I can trust a benchmark that has THAT much importance placed upon its shoulders. Especially when you have to pay $5,000+ to become involved with it. Do you?

Really, and what has changed sinced 3Dmark 01..Nvidia hardware is having major problems with the new benchmark following WHQL standards. There is no favortism of code here, AMD, Intel and Microsoft are not ganging up on Nvidia.

3Dmark 2001 was ok because Nvidia hardware excelled in it for a year, and that is the only reason why 3Dmark 03 is causing issues, because Nvidias product line is not competetive..period end of story.

How many cards has Nvidia sold during the last couple of years on 3Dmark 2000 and 2001, the hypocracy from Fan sites and Webmasters is rediculous.

It's 'ok' when we are winning, but when we lose on the benchmark 'it is not fair'

Anyone who doesn't see that is blind...I especially like reading the old reviews at [H] and Fan Sites when the Geforce 3 was dominating with PS 1.1 support and how it was the cats ass in benchmarks.

This entire fiasco sickens me, and the only review site in my favorites is this one, the rest are a bunch of puppets.

 

[dan2097]

Now if we could just get posters like Dan2097 to shorten their "feelings on NVIDIA" to a link to some canned response. I can't count the number of people who joined in the last week or so who posted a nearly identical statement.

But I want everyone to have my opinion so we can all be happy

 

[rwolf]

The funny thing is that even as Kyle condems 3DMark03 it is the only benchmark we can trust.

No other benchmark has checks and balances to detect cheating. Do you think Carmack is going to issue an audit saying Nvida is running Doom 3 24% faster because they are altering the benchmark?

I would trust the results of 3DMark03 anyday over any other timedemo.

Even if you are given tools to record your own demos you are not save because they can detect and replace shaders with their own code.

Kyle is dead wrong, if anything 3DMark03 has proved that we do need synthetic benchmarks.

 

[dan2097]

Even if you are given tools to record your own demos you are not save because they can detect and replace shaders with their own code.

Whats wrong with replacing shaders if it yields the same result and can be used anywhere in that game? I suppose you have to hope that the driver people are honest and replace the shaders that have EXACTLY the same result though...

EDIT: the reason you shouldnt do it in synthetics is your not going to optimize in the same way for EVERY game released, well atleast the type of optimzation Ati was doing for 3d mark 03. Nvidias is not even optimizations as it reduces IQ as well.

 

[Typedef Enum]

I believe the following is true of most Beyond3D visitors/regulars...

What a fuc*ing idiot. I can't believe _anybody_ gives a rats ass _what_ he thinks.

 

[Matt]

3DMark03 has become exactly what I feared, a scapegoat within an industry that is having a hernia. Both ATI and NVIDIA are having a tug-of-war, and they've placed 3DMark03 directly in the middle, and I don't know if I can trust a benchmark that has THAT much importance placed upon its shoulders. Especially when you have to pay $5,000+ to become involved with it. Do you?

Really, and what has changed sinced 3Dmark 01..Nvidia hardware is having major problems with the new benchmark following WHQL standards. There is no favortism of code here, AMD, Intel and Microsoft are not ganging up on Nvidia.

3Dmark 2001 was ok because Nvidia hardware excelled in it for a year, and that is the only reason why 3Dmark 03 is causing issues, because Nvidias product line is not competetive..period end of story.

How many cards has Nvidia sold during the last couple of years on 3Dmark 2000 and 2001, the hypocracy from Fan sites and Webmasters is rediculous.

It's 'ok' when we are winning, but when we lose on the benchmark 'it is not fair'

Anyone who doesn't see that is blind...I especially like reading the old reviews at [H] and Fan Sites when the Geforce 3 was dominating with PS 1.1 support and how it was the cats ass in benchmarks.

This entire fiasco sickens me, and the only review site in my favorites is this one, the rest are a bunch of puppets.

3DMark 2001 didn't have the amount of spotlight it has on it in 3DMark03. There wasn't hundreds of websites posting about 3DMark 2001 and cheating and drivers and patches that circumvented cheating in specific drivers. There wasn't a company attacking 3DMark 2001, saying it wasn't a valid benchmark. Also, 3DMark 2001 used a real game engine, which made it a more viable benchmark, since the whole point of benchmarking video cards is to evaluate performance in games. One of the reasons why we (3DGPU) stopped using Quake 3 as a benchmark, besides that it was getting old, was that cheats were coming out for it to get better scores. If companies like NVIDIA and ATI are going to cheat in 3DMark03 to get better scores, I don't see any point in using it.

As for the whole "winning" thing you keep bringing up, I don't see the scoreboard. I don't see facts on whether cards are being sold because said cards are getting higher scores in 3DMark. There is no winning and losing, the only winners and losers are those that buys the card, and either get what they want, or don't.

 

[RussSchultz]

Do you think Carmack is going to issue an audit say Nvida is running Doom 3 24% faster because they are altering the benchmark?

Yes. Why not?

Ignoring the obvious retort of "its not his job", he's very outspoken and has no financial interest in either ATI or NVIDIA liking him.

 

[Doomtrooper]

There wasn't a company attacking 3DMark 2001, saying it wasn't a valid benchmark. Also, 3DMark 2001 used a real game engine, which made it a more viable benchmark, since the whole point of benchmarking video cards is to evaluate performance in games. One of the reasons why we (3DGPU) stopped using Quake 3 as a benchmark, besides that it was getting old, was that cheats were coming out for it to get better scores. If companies like NVIDIA and ATI are going to cheat in 3DMark03 to get better scores, I don't see any point in using it.

Hardly, 3Dmark 2001 showed a Kyro based card would be slow in Max Payne, yet they performed in the game better than a Geforce 2 MX
It was not accurate..period..and I remember reading your posts on the old 3DGPU forums where you were pissed you could not break 10,000 marks (so I don't want to hear the score is not important).
How many games use the Max Payne engine ??

http://www.rivastation.com/go_e.htm?http://www.rivastation.com/review/max_payne/max_payne_01_e.htm

since the whole point of benchmarking video cards is to evaluate performance in games.

3Dmark was meant to show what 'future' titiles may look like, using the advanced features of video cards using standard DX code. Was there PS 1.1 games when 2001 was released..NO, was there PS 2.0 games released when 2003 released..No..Hypocrasy Matt.

As for the whole "winning" thing you keep bringing up, I don't see the scoreboard. I don't see facts on whether cards are being sold because said cards are getting higher scores in 3DMark. There is no winning and losing, the only winners and losers are those that buys the card, and either get what they want, or don't.

Misleading consumers is wrong, if you have to 'cheat' in Synthetic benchmarks, you have to 'cheat' in games.

 

[Matt]

There wasn't a company attacking 3DMark 2001, saying it wasn't a valid benchmark. Also, 3DMark 2001 used a real game engine, which made it a more viable benchmark, since the whole point of benchmarking video cards is to evaluate performance in games. One of the reasons why we (3DGPU) stopped using Quake 3 as a benchmark, besides that it was getting old, was that cheats were coming out for it to get better scores. If companies like NVIDIA and ATI are going to cheat in 3DMark03 to get better scores, I don't see any point in using it.

Hardly, 3Dmark 2001 showed a Kyro based card would be slow in Max Payne, yet they performed in the game better than a Geforce 2 MX
It was not accurate..period..and I remember reading your posts on the old 3DGPU forums where you were pissed you could not break 10,000 marks (so I don't want to hear the score is not important).
How many games use the Max Payne engine ??

since the whole point of benchmarking video cards is to evaluate performance in games.

3Dmark was meant to show what 'future' titiles may look like, using the advanced features of video cards using standard DX code. Was there PS 1.1 games when 2001 was released..NO, was there PS 2.0 games released when 2003 released..No..Hypocrasy Matt.

As for the whole "winning" thing you keep bringing up, I don't see the scoreboard. I don't see facts on whether cards are being sold because said cards are getting higher scores in 3DMark. There is no winning and losing, the only winners and losers are those that buys the card, and either get what they want, or don't.

Misleading consumers is wrong, if you have to 'cheat' in Synthetic benchmarks, you have to 'cheat' in games.

Pfft, the Kyro2 is a TBR, you know that. It always gave odd results. It did exceptionally well in some games, while it did horrid in others. That's common knowledge.

You must have me confused with someone else, I never got pissed at my score in 3DMark, and I never tried to break 10,000. The only thing I got miffed about was that I was told by a friend that my processor could go higher than 2.2GHz, but it couldn't. It had nothing to do with 3DMark.

I said the whole point of benchmarking is to evaluate performance in games, I didn't say the whole point of 3DMark. Read a bit slowly, please.

I agree with your last statement, and it's unfortunate. That's why all consumers who want to buy a video card (or any product for that matter), should ask in forums and/or email people in the know. The internet is a fantastic tool to become an informed buyer. I answer emails everyday from people needing advice on what to buy, and it feels good when they email back saying they like what they bought based on my advice.

 

[dksuiko]

It's 'ok' when we are winning, but when we lose on the benchmark 'it is not fair'

I agree. But with 3DMark01, everyone who wasnt nVidia blasted the benchmark (only need to scrollback into the archives of these forums to see that, I even remember a thread where you argued with worm telling him that nVidia was paying him off) while nVidia loved it. Now that 3DMark03 is out and ATI is winning, everyone but nVidia loved it while nVidia hates it. The flip-flopping goes both ways.

 

[Randell]

The flip-flopping goes both ways.

ohh yes

 

[micron]

Another "BTW".

I wish that you guys would stop posting a thread like this in this particular forum. In fact, I wish you guys would simply stop posting threads like this here per se. At the risk of sounding "elitist", I'd really appreciate it if we could return to the old, old Beyond3D Forum, where almost all threads are about 3D technology and that if you want to sound off about other website opinions, do it at other forums. Again, sorry if this sounds kinda "elitist" or pompous.

I'm not going to decide what to do with this thread... John can have the pleasure

Hey...you know what? I would rather have the post deleted than have you lay that "your tainting the purity of B3D" guilt trip on me....
Why do you long for the good old days so badly Reverend?
This website has become "the trusted authority" when it comes to 3d technology, and you played a part in making it so. Be proud of the way that it is today.....and yesterday.....


Also.....you really dont need to be such an elitist snob all the time....
I cant believe I'm saying that, but you are always making me feel like shit as a new member here..... Thanks buddy.