Your Thoughts on 3DMark threats to HardOCP.

[hordaktheman]

But the whole point is you CAN play those games, 3d mark you can't play so as a "gamers" benchmark its kinda pointless.

... thereby rendering [H]'s benchmarking pointless to 90% of the gaming population.

If it were true, that is.

But thankfully it isn't, as one benchmark is generally consistent with most others, so using 5 benchmarks generally gives a similar outcome as using 50. And in the same sense, since 3dmark generally gives an outcome similar to ones given with the other benchmarks, using it is no worse than using those, even though you can't play 3dmark.

 

[Ice Nine]

Can we quit with all the infantile taunting here? It takes a big man to knock somebody down, it takes an even bigger man to offer him a hand up when he's on the ground.

Yeah, but that isn't my decision to make. That's Kyle's.

I'm not taunting anyone at Futuremark. They don't care about my opinion. Those of us who hang out on efnet's #3dfx have been calling 3dmark "Uselessmark" for years now. Instead of making legal threats against a website that expresses an editorial opinion, they should be reading the writing on the wall.

Benchmarks serve a purpose. But all Futuremark's 3dmark series benchmarks do is serve as a marketing tool for ATI, NVIDIA and Alienware. If what they did was more relevant to the enthusiast community, maybe they'd get more sales from enthusiasts instead of marketing dollars from the above companies.

Prior to this 3dmark05 garbage being released, people speculated it would be "Pixel Shader 3.0, FP32 and 3Dc ready", but NONE of that appeared to happen. Gee, why is that?

Is it because ATI gives money to them and didn't want to look bad for not supporting SM3.0 or FP32? Or is it because Nvidia gave money to them and didn't want to look bad for not supporting 3Dc? Honestly, what did 3dmark05 prove?

I remember in previous 3dmark benchmarks, they'd have tests for Bump Mapping, Pixel Shaders, Transformation & Lighting, etc etc etc. What do we have now with 3dmark 05? Oh gee. We have another Doom3 mockup. We have a blimp & a large snake. And a firefly. Followed by a slide show!

Yeah. That's just great. Welcome to the world of the obscure, Futuremark.

 

[hordaktheman]

3dMark2001 was never AS video dependent. It scaled with CPU's much better even when it first came out than 2003 or 2005 have.

Obviously a "video/game" benchmark is going to be very videocard dependent, but there should b/c some CPU scaling too. Because in most real world games, besides Commanche (which doesnt scale for anything) the CPU does a lot of work (AI, physics,etc.) and can improve gameplay considerably.

And a 3Ghz P4 isn't 3x faster than a 1Ghz P3, b/c of IPC, etc. SO bad comparison. I wish the damn CPU/GPU comparison thing worked so I could quote some #'s for average CPU's/ GPU's

Absolutely correct, but the problem comes with finding an adequate middle-ground. The last game I played that was more cpu limited than gpu limited was... C&C: Generals (I think).

As far as I know, tomorrow's games are far more shader limited than cpu limited (or so I've read), so basing benchmarks more on video performance than cpu performance makes sense. And in FM's defense 3dmark05 is a lot more cpu limited than 03, even if it's less so than 3dmark2001. I'm willing to bet that with next-gen vidcards, we'll see a much wider cpu-scaling in 05.

 

[UberL33tjarad]

As far as I know, tomorrow's games are far more shader limited than cpu limited (or so I've read), so basing benchmarks more on video performance than cpu performance makes sense. And in FM's defense 3dmark05 is a lot more cpu limited than 03, even if it's less so than 3dmark2001. I'm willing to bet that with next-gen vidcards, we'll a much wider cpu-scaling.

2100mhz isn't a large enough difference to see a change in framerate? i had my 2.8 at 1.4 (14 x 100) and got 2100, then raised to 3.5ghz and got 2250. in 3dmark03 that would be at LEAST 1000 points.

as for games, playing far cry at highest detail w/ 1.4ghz and 3.5ghz doubles my framerate from 25fps on dam level (on cliff looking down) to 45fps. haven't tried d3 b/c it runs like shit on anything other than medium mode anyway.

 

[eastvillager]

eh, futuremark is just pissing into the wind.

If, IF they actually followed this site on a regular basis, they would've known to formulate a more effective attack(in other words, they could've seen what the phantom pissing and moaning has done for that company, lol).

Not like anybody uses their crap for anything serious anyways. Download new version, check out demos, erase it. Repeat for each new version. Uhoh, maybe I'll get a letter now.

 

[hordaktheman]

2100mhz isn't a large enough difference to see a change in framerate? i had my 2.8 at 1.4 (14 x 100) and got 2100, then raised to 3.5ghz and got 2250. in 3dmark03 that would be at LEAST 1000 points.

as for games, playing far cry at highest detail w/ 1.4ghz and 3.5ghz doubles my framerate from 25fps on dam level (on cliff looking down) to 45fps. haven't tried d3 b/c it runs like shit on anything other than medium mode anyway.

I was referring to the really, really high-end cards like x800XT's.

 

[TeroS]

Yeah, but that isn't my decision to make. That's Kyle's.

Prior to this 3dmark05 garbage being released, people speculated it would be "Pixel Shader 3.0, FP32 and 3Dc ready", but NONE of that appeared to happen. Gee, why is that?

3DMark05 supports sm3 and all variants of sm2, you can even force it to run on any of those for HW that supports it. And yes, it does support FP32. 3Dc is not there, at least yet, since it is one vendor specific. If that changes, we'll probably add it there too.

Tero

 

[UberL33tjarad]

 

[Flipin]

To me, it would seem that Kyle is more in the line of entertainment than unbiased news and reviews. Shades of grey turn into black and white and so on. I have yet to find the review (or bash, however you take it) of the new 3dmark, so I really can't say for sure, but it would seem that Kyle managed to piss Tero off pretty good. I certainly wouldn't be too happy if my work that I'd been doing for years is suddenly bashed to the ground (without giving a proper explanation it would seem). I certainly don't think it's useless for what it's meant for: comparing hw & results. It sells, doesn't it?

As for the lawyers, that was a low blow from Tero. However, in this part of the world, a threat of lawyers is hardly taken serious, so it would seem that he misjudged the weight of his statement. Unlike in America, if the case really does go to the court, it's out of the lawyers' hands so no number of expensive lawyers will help.

I think of reading hardocp mainly as entertainment, but many seem to think it of as an unbiased source of information. This however does not always seem to be true, rather, it would seem that Kyle is provocating more companies to take part of this farce and exaggerate stories in hopes of more readership.

This is obviously my opinion & point of view, so it may, or may not mean anything to you. And just to make it clear, I think Tero's first response was very immature, but that has been covered pretty well already.

----
Jamie

 

[UberL33tjarad]

I was referring to the really, really high-end cards like x800XT's.

would it make a difference? at least with slower cards, like my 9700pro, a faster cpu would be more evident than with a x800xt or 6800u, but that's not the case with 3dmark05. people throw the "cpu-limited" term around too much without even understanding what it means.

 

[bigstusexy]

It must truly be humbling for someone at Futuremark to have to apologize to a guy who runs a site called "HARD OverClocked Penis".

Kudos, Kyle

I think you have been miss-lead about what this site's full name is
its: Hard Overclocker's Comparison Page. Its one of those little known facts thats rarely said around here.

 

[[Ch]amsalot]

I would like to emphasize that we really encourage everyone to send and publish any and all feedback, good or bad

It bothers me that ATI is allowed to publish things like this:
http://www.ati.com/companyinfo/press/2004/4787.html

In which they tout their product is best because some insanely talented overclocker can push his X800XT PE to a core of 806MHz (up from factory 520MHz). This gives consumers a false impression of the capabilities of the card. Moreover, since only fps is recorded, there is no way for us to know if there weren't major visual abnormalities such as artifacting that occurred during the benchmark of this and other overclocked cards.

NVIDIA is little better in this regard, having touted a SLI result of 7299 here. SLI is not available to 99.9% of the public at this time, again making the press release misleading and useless to consumers. What was worse with NVIDIA's test was the fact that the results weren't even registered on the database, making it impossible for us to verify the score. In any event, in what must be embarrassing to NVIDIA, overclockers have already beat the SLI score with non-SLI'd, overclocked 6800 Ultras.

I would like to see limits on what Futuremark's partners can publish in terms of results. Specifically, they shouldn't be allowed to toot their own horn based upon the overclocking efforts of others. In fact, I'd like to see all partners be required to use an independent tester using factory-default clocked cards with approved drivers to gain the 3DMark2005 score. That would be most beneficial to consumers looking to evaluate products based upon 3DMark scores.

As the publisher of the product, it is in your ability to control this type of information. I eagerly await your and everyone else's commentary.

Finally, what possessed you guys to make the default resolution 1024x768? Does anyone with high-end systems really use such a low resolution anymore?

Sincerely,
[Ch]amsalot

 

[Ice Nine]

Futuremark should care wether we uninstall there product.. thats the whole point.. thats how they make money,.. people installing the program and then buying and upgrading to get more options.. they should care ALOT wether we uninstall the program..

Actually no, They don't care about end-users. Remember 3dmark is only a marketing tool for:

1. ATI
2. Nvidia (actually i'm not even sure if Nvidia pays 'em)
3. Falcon Northwest
4. Alienware

They're the ones who pay Futuremark the bucks. Enthusiasts don't care about purchasing this product. And why would they? All it gives you is the ability to post more ORB projects (and just how many of THOSE do you need?).

Everyone else just grabs the keygen if they want to see the rest of the 3dmark "benchmark" suite.

 

[Tekara]

Something I have interest in is that online gaming is becoming more popular. Both massive, like FFXI, and minor, like counterstirke, multiplayer games. These games tend to be more cpu intensive due to processing locational data of other players, 3Dmark does nothing to gauge my overall performance in these games.

Along those lines I would also love to see 3dmark gauge my internet connection. Many people have noticed that ping latency is not enough to gauge perfance due to different packet sizes resulting in different latencies.

There are my suggestions.

 

[Ravenrex]

Everyone else just grabs the keygen if they want to see the rest of the 3dmark "benchmark" suite.

Great, thanks...So in addition to being rude we're all just theives as well. Thanks...please stop speaking on behalf of the enthusiast community. Any more help and we'll all be in court.

 

[Impulse]

Something I have interest in is that online gaming is becoming more popular. Both massive, like FFXI, and minor, like counterstirke, multiplayer games. These games tend to be more cpu intensive due to processing locational data of other players, 3Dmark does nothing to gauge my overall performance in these games.

Along those lines I would also love to see 3dmark gauge my internet connection. Many people have noticed that ping latency is not enough to gauge perfance due to different packet sizes resulting in different latencies.

There are my suggestions.

Testing with a regular game engine is by no means easy... These days picture quality's gotta be taken into account as well as the average fps and how low/high it goes, then there's optimizations NVidia and ATI tend to build in... Doing similar testing with a MMORPG and having it be anywhere near reliable is next to impossible.

Not only is it near-impossible to reproduce the ongoing events of the world from test to test, but it's even harder to reproduce said results on your end. You may not play in the same world/server/etc. as the reviewer. Even if you do maybe the day you do your testing there's 1,000 more odd players online. Heck even as the reviewer goes to switch cards the movement on the server may have changed dramatically enough to impact the results.

It's just not very feasible... As for connection testing, there's a lot of very good sites that do this already on the net, I think 3dMark is trying to stay focused on games and 3d performance. Try DSLreports.com

 

[ccbutler]

Actually no, They don't care about end-users. Remember 3dmark is only a marketing tool for:

1. ATI
2. Nvidia (actually i'm not even sure if Nvidia pays 'em)
3. Falcon Northwest
4. Alienware

They're the ones who pay Futuremark the bucks. Enthusiasts don't care about purchasing this product. And why would they? All it gives you is the ability to post more ORB projects (and just how many of THOSE do you need?).

Everyone else just grabs the keygen if they want to see the rest of the 3dmark "benchmark" suite.

AMEN!!!!! Bout time someone hit the nail on the head... marketing tool indeed. ATI/NV is a marketing tool for FM too... they've all been caught in the same bed together.

Is it just me? Or do these publicly traded corporations all need to be put on a tighter leesh. The word "litigation" has almost completely lost all meaning when you "voice an opinion" about a company and they want to sue your ass off.

 

[Ice Nine]

Great, thanks...So in addition to being rude we're all just theives as well. Thanks...please stop speaking on behalf of the enthusiast community. Any more help and we'll all be in court.

Not thieves. Just people who aren't going to shell out money for a useless product that tells us nothing. Yes, I installed 3dmark05. Yes, I used the keygen to see the "novel" extras. Yes, I uninstalled the whole ball of wax when I quickly realized the product is *useless*.

And I think I *DO* speak for the enthusiast when I say we would rather put our hard-earned dollars towards a new piece of hardware, fast cable connection or new game than waste $20 or $30 on a product that has no relevance to what we do as a "community".

All 3dmark does is show you some pretty graphics and spit out a number at the end. That number has no correlation to anything realistic or meaningful. All it does is let you argue with other people who have run the same benchmark. Now its scope is even MORE narrow (really just a video card test).

It doesn't take an enthusiast to figure out it's simply a marketing tool that generates a meaningless number.

 

[FrgMstr]

So you got over an unfair banning? Tell me about it. Let's just say I lost some respect for Kyle now since I read that crap of people getting banned over something right. But of course, it's Kyle's website and I can STFU but not when this is going on. Very unfair. Kyle *should* respond to this question of why he banned people for being right. Maybe they insulted or degraded him afterwards but over something right? That's egoism explodism I'm seeing.

-J.

Yes we screwed up 2 years ago. And we publicly apologized for our actions, more than once. A lot of things have changed around here since then. Dunno what else to tell you. You can measure us by our actions today or two years ago, it is your call.

 

[FrgMstr]

While I acknowledge [H]ardOCP and Kyle's right to their opinion and the right to publish their opinions, I must say that simply stating that any product "sucks" without elaboration (as was the case in this instance) is rather poor journalism. And I'm not simply referring to the choise of words (although that could have been done with more tact and maturity), but rather the general tenor of the statement.

If Mr. Bennett wishes to be taken seriously as a computer hardware journalist, he should pay closer attention to the editorials he publishes, regardless of how short or long they may be.

That being said, his subsequent responses to the Futuremark's accusations have been far more in keeping with what I would expect of a professional journalist.

Here are a few you can read. I should have reposted the links along with my suck comment.

Originally Posted by Tim
While we are on this subject and to make sure everyone knows this has been going on for some time, check out the articles on the subject:

Benchmarking Right
Benchmarking Wrong
[H]ard|OCP 11/13/03 Editorial
Cheating the Cheaters

 

[Mister Natural]

Great to see TeroS posting. I think a great thing could result in all this provided the two parties involved would like to see a positive result.
There are a lot of knowledgable folks here and there are some who can manage a productive conversation. If the people at Futuremark would like some input to their product I would think a thread on these forums would be a great place for some feedback.
I also think a visible dialogue between Futuremark and enthusiasts would go a long way in improving the situation as a whole.

 

[Grinder66]

If the people at Futuremark would like some input to their product I would think a thread on these forums would be a great place for some feedback.
I also think a visible dialogue between Futuremark and enthusiasts would go a long way in improving the situation as a whole.

Agreed.

 

[hordaktheman]

would it make a difference? at least with slower cards, like my 9700pro, a faster cpu would be more evident than with a x800xt or 6800u, but that's not the case with 3dmark05. people throw the "cpu-limited" term around too much without even understanding what it means.

At least according to this: http://www.bjorn3d.com/read.php?cID=702&pageID=1051

On another site (that I can't remember) a gain was had just by increasing the fsb (while lowering the multiplier so that cpu speed was the same) on an XT.

 

[Ice Nine]

Great to see TeroS posting. I think a great thing could result in all this provided the two parties involved would like to see a positive result.
There are a lot of knowledgable folks here and there are some who can manage a productive conversation. If the people at Futuremark would like some input to their product I would think a thread on these forums would be a great place for some feedback.
I also think a visible dialogue between Futuremark and enthusiasts would go a long way in improving the situation as a whole.

I personally don't think futuremark cares. And why would they?

There was a time where 3dmark versions prior to 03 were relevant. It drove innovation. At the time these versions were around, not all "top end" cards on the market were capable of running all the tests in the suite. When these versions were released, the highest-end card on the market might have been missing bump mapping (remember Matrox?), hardware T&L (a-la GeForce), or the like. Back then, it was *RELEVANT* because it drive card manufacturers to support new features that developers could leverage.

Back then, your score varied wildly based on what features you card supported.

Since 3dmark03, it seems they've taken the approach of "make something incredibly hard to render that will humble everyone in the industry". While they added these "CPU tests", they BARELY factor into the final score. At the time of its release, pretty much all the top end cards supported 3dmark03's feature tests. But what does it mean now? My Athlon64 3800+ & 6800U combo pounds out 13,000 3dmarks, but STILL drops to near-single-digit framerates in the "Mother Nature" test. Does that make my machine a clunker? Is there some mystery feature my 6800U is missing that would make that test run more quickly? Does this mean i'm going to score single digit framerates in FarCry, Doom3 or Half Life 2 at 1024x768?

No, it doesn't. It's a POOR BENCHMARK that does NOT reflect the real world.

OK, so Futuremark could have redeemed themselves with 3dmark05. We now have cards with different featuresets. We have the 6800 camp with Shader Model 3.0, FP32, etc. And we have the X800 camp that has 3Dc, Trueform, whatever else. But NONE of these features are tested/implemented/scored or leveraged into that final "3dmark score".

Why? Because futuremark can't afford to lose an advertiser. So they marginalize the "benchmark" by just making the benchmark more generic. Pretty, but still generic. Oh, and ridiculously hard to render, to the point where our cards 2 years from now will probably still stutter on this pig.

I'm actually *impressed* that Futuremark can make a business out of spitting out a meaningless number.

 

[tazzmission]

Tero Sarkkinen is a tool.

That is all

 

[Zebo]

Lawyers suck, there's too many of them suing everyone.

Until *you* need one to defend *your* rights or prosecute a tort then you'll be licking his balls and paying for the privelidge.

Don't hate the game, hate the player who level bullshit to intimidate, cooerse, kill compitition, or just plain want money for doing nothin'...no need to tell you who the "player" is here...dispite appology

 

[corran_horn314]

Tero Sarkkinen is a tool.

That is all

what a great way to elevate the level of discourse, we're really making progress now.

It doesn't seem like anyone has changed their opinion about 3dmark from bickering in this thread

 

[SuperPotato]

Hey hey, classic thread... Enjoyed it.

 

[chrisf6969]

This thread has really outlived it usefullness, someone should close it, and start one, called "Constructive Criticism for Futuremark".

Enough ranting & bitching already. Tero owned up to his mistake. Kyle made mistakes 2 years ago... big whoop. Move ON!!

Ok look for new thread here: Constructive Criticism for Futuremark

 

[Koslov]

Pretty amusing har har
he might a member of the Yasuka family?

 

[Tyklfe]

Way to go Kyle.

I love to see you guys defend your right to speak the truth. Keep up the good work, and I'll keep coming.

 

[slavik]

I think someone overclocking the card and showing the insane score shows DAMN WELL the capability of the card ...

it is "misinformation" to NOT state if there is an overclock or not and if there is what clock rates were "used"

I know your point ... but I think that as long as ATI says that a true hardcore enthusiast was able to do such and such and get this score, then it should be used for marketing to other enthusiasts (do u get my point, though?)

It bothers me that ATI is allowed to publish things like this:
http://www.ati.com/companyinfo/press/2004/4787.html

In which they tout their product is best because some insanely talented overclocker can push his X800XT PE to a core of 806MHz (up from factory 520MHz). This gives consumers a false impression of the capabilities of the card.

 

[slavik]

Kyle, to be honest, it's too bad they didn't sue you ...

the computer news field is rather "dull" at the moment ... I want some REAL news

Kyle, [H] - 1, plaintiffs - 0 (infinium)

was there anyone else who tried to sue you? (digging through years of your news ... would sidetrack me into reviews of old products )

and my "3rd" thought ... I think that Kyle should deserve the Pulitzer prize for great investigative (spellcheck ) reporting

my final thought (tired yet?), Kyle should cover the election ... too much corporations looking out for their own interest ... any chance on an article about the election related happenings in the near future?

 

[Fisofo]

Just saw Tero's apology... I must say it's refreshing to actually get an apology rather than have them push ahead legally (with or without a leg to stand on), hopefully a better understanding of the place for synthetic benchmarks will come out of this. Burn-in tests anyone?

Edit: just saw Tero's other posts, welcome to the forums! It'll be great to work through some of this stuff with you

 

[J4M3S0N79]

Hello,
It seems to me that this whole benchmarking fiasco which has been ongoing for years now is getting a little long in the tooth. Many people are very quick to criticize shortcomings of benchmarking software (or any HW or SW for that matter). Usually the claims revolve around the fact that these benchmarking utilities are not truly indicative of real world performance, performance in future games and performance in past games. When 3dmark 2k3 came out, people complained that there was older DX engines in there but what I saw was FM at least attempting to include a wide range of tests, I was actually happy to see that my 9800pro was great at game 1 and lagged at almost all else...this told me that my card was relatively slow at rendering all the nextgen DX9 features…and guess what…it is slow at rendering these, but when I go back and play DX7 and 8 games my computer is more than sufficient. I really think that the brunt of the issue is the attempt to reduce your computer to a single score when it comes to gaming. How would you as a person like to be branded with your IQ? Unfortunately that seems to be what is happening and the sad fact of the matter is that Joe Schmoe who has dollars, and wants to but a fast computer but doesn’t know the details of PCs, is going to buy a comp based on this ‘industry standard’ benchmark score. More often than not, it will lead him to an alienware, XPS or FNW comp. I hate these systems, but are they bad systems? No, some are hella fast, impractical as they may be. My point is this, we are [H] junkies are kinda above this benchmarking mess, there is no way to run a proggy and have a crystal clear analysis of all game engines, and how they run on all GPUs in combination with all CPUs, taking into account the cpu hit of sound, network traffic, legacy components, background tasks….people…the list of variables goes ON AND ON. I say, Download FM for the eye candy at least so we can get a glimpse of cool graphics our comps will be running in ’06. As for benchmarking, and building a system….AMD-Intel, Nvidia-ATi, Windows-Linux…..there is no clear winner…that’s why they are all still in business, and 3dmark won’t crown the victor…neither will hl2, or doom3 or unreal7 or halo 62. Make up your own damn minds…don’t let a benchmark tell you what’s right.

 

[Mr^Yahtzee]

very well spoken.....

 

[Spare-Flair]

I think there has been a lot of overreaction here. I read the apology, and I accept and agree with it. Even from the tone of the original e-mail, you could tell that it was an emotional outburst. I would be too if I had worked for years on a product, put a lot of blood, sweat, and tears into it, only to have it unilaterally lampooned by a premier, high traffic, website that neatly proclaimed in a short news blurb that "3DMark05 sucks as a benchmark".

I understand that that is how Kyle feels but I would be pretty emotional if a very popular and well-read individual suddenly proclaimed, with no caveats or disclaimers, that my product sucked in so few uncertain terms on the front of a news page. This much indicates to me that TeroS reads [H]ardOCP as well.

It was just an emotional outburst. It was an e-mail from an angry and frustrated guy. This vindictive sue me, sue you atmosphere doesn't do much but pad the pockets of the lawyers (which frighteningly, I hope to become one). Many of us absolutely dislike and would never use 3DMark as a benchmark, but is serves as a good point of reference or parallel comparision. If nothing more, I download it because I love the beautiful demos and you can tell a lot of love went into that work. I personally agree that it sucks as a benchmark, but I would have many nicer words to say for it's other redeeming qualities.

Remember, that Futuremark, then Madonion, then Futuremark again, once started as simply a gathering of great artists, muscians, coders, and demo makers like many of us. It's just a shame it has become so corporate and industry giants have muscled their way in with abuse and scandal.

I know many of us still use 3Dmark to prove our [H]ardness as exemplified by the sigs of many users who proudly display their marks like a badge, proving your virility and computer muscle. It is as much an emotional quotient as the horsepower in a car which doesn't reveal the entire picture, but gives us bragging rights around the BBQ or the water cooler. I don't know what the fuck I just wrote there but I hope you get the point. If you've got it, flaunt it. If you think it's deceptive, avoid it. But save the bickering and litigacious attitude for fighting better enemies...like the RIAA, Spammers, Trojan hackers who zombify our computers, proven conmen like Tim Roberts of Infinium, and other such ilk.

 

[FrgMstr]

I understand that that is how Kyle feels but I would be pretty emotional if a very popular and well-read individual suddenly proclaimed, with no caveats or disclaimers, that my product sucked in so few uncertain terms on the front of a news page. This much indicates to me that TeroS reads [H]ardOCP as well.

Tero knows very well that we have published a bit more than "sucks" on the subject as well. While expressing our opinions on a daily basis it is simply not doable to link everything that backs those opinions.

Originally Posted by Tim
While we are on this subject and to make sure everyone knows this has been going on for some time, check out the articles on the subject:

Benchmarking Right
Benchmarking Wrong
[H]ard|OCP 11/13/03 Editorial
Cheating the Cheaters

 

[Spare-Flair]

Hey of course I agree with you Kyle. I guess I'm too much of a softie.

Heck, one time a girl wrote on public forum that "I sucked" and I sent back a much worse and hate-filled e-mail than this. Damn that hurt Wasn't much opportunity to explain the history of that to our collective friends either haha.

 

[Silverghost]

3Dc is not there, at least yet, since it is one vendor specific. If that changes, we'll probably add it there too.

Tero

Well that is true, but then how come nVidia's DST is in 3DMark05, and further, why is it enabled by default?