Kyle lays it on the line

[nelg]

Go read for yourself.
http://www.hardforum.com/showthread.php?s=&postid=1025103810#post1025103810

Good to see that he is willing to confront his detractors. I found the language used a little imprecise insofar that it still leaves room for speculating on his true feelings.

For example is he sorry about his ET remark or just that he made it public ?

 

[Doomtrooper]

What we do have with NVIDIA is a tremendously strong relationship. A relationship beyond what I ever thought would be nurtured. NVIDIA is making it a habit to talk with us. NVIDIA listens to us. They use HardOCP as a voice of the community that I still know in my heart and soul that we represent. New solutions in terms of press/community/manufacture relationships are being tried at NVIDIA and seem to be paying off in spades. We do not have that same relationship with ATI and the only time they contact HardOCP is when they have something to tell us.

That is all you need to read , considering Nvidia uses [H] in their PR docs they send to reviewers. .
Kyle thinks he is so big that ATI needs to contact him for approval

1. I have never personally taken payments in any capacity from NVIDIA. KB Networks, Inc. and KB Accessories, Inc. have never accepted payments from NVIDIA. (NVIDIA has never advertised with us.) Technically, I am CEO/President of the company that owns this site and have final edit on all content. HardOCP editors are paid by KB Networks, Inc.

Of course not, but he has from BFG a strategic Nvidia partner and others DUH !!

 

[Althornin]

Wow, he still comes off as biased and unfair.
If i could still post there, i'd ask how he had time to ask ATI about hte quak issue in THREE days, among other things.

Eh, i simply disagree 100% WRT to his comments about #dMark03 and synthetic benchmarks, and now he is backpedaling on the prior knowledge of the aniso issue.

My error in the entire issue and one that probably polarized a lot of readers is that I questioned ET’s motives publicly. If I could go back and undo anything in reference to the situation, that would be it. That statement upset a lot of people and angered them when I think that they otherwise would not have. My apologies to ET and our readers.

about time.
Unfortunately, this is the FIRST time you have actually come out and said, ET was right.
EVERYTHING else i have read that you posted on the issue boils down to "its not a cheat, its an optimization. If you cant see the cheating, its not cheating. And i told you so that sythetic becnhmarks were bad and useless". Your biggest mistake, kyle, was in not even bothering to make a news post to the effect that you disapproved. Everything you said about it was colored with your "so what its just 3dmark" rhetoric, and THATS what pissed alot of people off.
Sure, you didnt want to "undermine your stance on 3dmark" - good for you. you ended up sounding like you didnt care about cheating. great move. I also dont see why saying "see, we told you synthetics were bad, but nVidia, how wrong and unethical etc..." would be so terrible either. you basically soldout your readers to maintain your stance on an entirely seperate issue.

At HardOCP it is our duty to our readers to call it like we see it.

too bad on the whole cheating issue you let your agenda get in the way, and its only months later, in a forum post, that you finally make the statements people were looking for when the shit hit the fan.

 

[Ichneumon]

The main thing I seem to pick up from the post is that Kyle is finally actually saying flat out that all the hypocrisy that they've been spewing lately is because they wouldn't dare change their view on 3dmark03 even slightly no matter what it meant they had to say...

 

[K.I.L.E.R]

Why is it that when one thread about Kyle becomes closed that another thread immediately pops up?

If this keeps up I think Rev should consider opening a forum dedicated to Kyle. :?
Or we can have 1 thread dedicated to him rather than mods closing them all up and having others make new ones.

 

[Wolf]

Probably because ever day there is a new sick twist to this never ending soap opera.

He is the laughing stock of the hardware community. His empire is slipping time for a little damage control. His writing looks like it was looked over by 50 lawyers before it was posted.

 

[Bouncing Zabaglione Bros.]

When's it going to be a front page editorial? Otherwise tucking it away in the forum is like a newspaper putting an apology of a front-page screwup into the corner of page 19 in the hope that no one notices they have been grudgingly forced to eat their words.

AFAICT, there's a lot of back-pedalling and contractictory statements here. There's no apples-to-apples possible, he still talks about Quack in the same breath, 3Dmark2003 is a useless benchmark for being cheatable when UT2K is a good benchmark, [H] knew about the NV triliner issues months ago, but not in the reviews, they are big friends with Nvidia, but too small a company to do in-depth work and report on issues like benchmark cheating, they don't do things that other sites have raised (like benchmark cheating) etc.

It's the same old nonsense just regurgitated again. The tone is reasonable, but the logic is as faulty as ever. Nothing new here.

 

[K.I.L.E.R]

When's it going to be a front page editorial? Otherwise tucking it away in the forum is like a newspaper putting an apology of a front-page screwup into the corner of page 19 in the hope that no one notices they have been grudgingly forced to eat their words.

You would be surprised at how much that has happened. I know I was.

 

[StealthHawk]

I think this says it all:

If game specific optimizations were grounds for dismissing benchmarks in actual games, I would think that we might not have any benchmarks left at all. In the case of UT2K3, I think the optimizations were valid and therefore deliver a better experience to the gamer playing the game.

 

[Archaeolept]

reads like he got religion,

after a whiff of grand jury.


Saying that "Our benchmarks will focus on IQ that we deem to [be] comparable" instead of "purist settings" effectively means that we are to take his judgements at his word; as such tests are personal opinion, and, for the most part, not reproducible.

Its an argument for faith instead of science. and the perfect refuge for subterfuge.

NVIDIA listens to us.

my precious...

 

[Evildeus]

I thought it was the rules to not comment on B3D fora other thread put in other fora, seems to be the contrary now...

 

[Dave Baumann]

I would post this at [H] but, well, I can't...

We discussed how the drivers handled the Bi/Tri settings and asked for specific availability to get around pre-set or driver determined Trilinear algorithms. We also asked that "pure" AF be used as well. So all in all, we were asking for the user to have the ability to decide their own level of Trilinear Filtering and that has been granted, as you will see in the next drivers released. Quite frankly I see the problem as being solved before it ever became an issue.

I fail to understand this. Pre 44.03 the "Application" setting had full Trilinear, and with 44.03 the "Quality" mode is supposed to have full Trilinear, so if they didn't know about the UT2003 thing essentially they were asking for something they were told was already there...?

It could easily be argued (and has been repeatedly) that every benchmark using Trilinear Filtering in an environment with surfaces other than 0 and 90 degrees over the last year have given ATI an edge that NVIDIA did not have.

Only if you are talking about rendering with AF enabled - under deafult rendering options you should be calling for full Trilinear - there is little difference in the numbers of samples taken with most boards when doing straight, full Trilinear filtering.

What we do have with NVIDIA is a tremendously strong relationship. A relationship beyond what I ever thought would be nurtured. NVIDIA is making it a habit to talk with us. NVIDIA listens to us. They use HardOCP as a voice of the community that I still know in my heart and soul that we represent. New solutions in terms of press/community/manufacture relationships are being tried at NVIDIA and seem to be paying off in spades. We do not have that same relationship with ATI and the only time they contact HardOCP is when they have something to tell us.

I might suggest that the burden for improving the relationship lies not with ATI, but with [H]? Perhaps if some of the issues that have cropped up over the past few months had been handled differently then ATI would have been more willing to have a more open dialogue. For instance - my banning from [H]'s forums has not going unnoticed by ATI, including their PR, and from my conversations with them they think that was pretty unreasonable. Its actions like these and inactions in other areas that probably doesn't make ATI want [H] to be their confidant.

I may respectfully suggest that the onus there lies with [H] to improve the relationship, afterall if they don't then they are doing a diservice to their readers. While NVIDIA was the undisputed king of the hill over the two years prior to R300 its a good thing to insuate yourself into their talks, but ATI have proved that there are now two 3D vendors who can claim top class hardware - if this turns out to be a sustained effort by ATI then then by not doing the same with them as you have with NVIDIA you'll only be giving your readers half the story...

For those they say we are NVIDIA biased go back and read where we have repeatedly recommended ATI Radeon cards for nearly a year now above all others. Read the hundreds of favorable news plugs and links we have extended to their product. I really feel as though we have gone out the way sometimes to let our readers know how great a product the R3XX line is.

Thats because, until NV35 ATI have been peerless in many ways - you should be praising good hardware for being good, you'd look foolish if you didn't.

 

[hjs]

Well i can, so i quotted you there

 

[5150 Joker]

Go read for yourself.
http://www.hardforum.com/showthread.php?s=&postid=1025103810#post1025103810

Good to see that he is willing to confront his detractors. I found the language used a little imprecise insofar that it still leaves room for speculating on his true feelings.

For example is he sorry about his ET remark or just that he made it public ?

Anyone want to take bets on how soon Kyle will close the thread after he's cornered?

 

[Sharkfood]

I doubt he will get cornered since replies are likely to simply be deleted to keep the thread alive.

What I find of particular interest is how Kyle is running to the "Well, UT2003 is a game" to try and defend the apples->apples comparison while not trusting 3DMark03 to compare videocards.

Was Quake3 not a game? The whole quake/quack debacle became a big issue financed/provided by NVIDIA and reported by [H], but yet we see similar finding with UT2003 now and the same treatment of the situation is not being applied.

It doesn't take a rocket scientist to see the blaring double standards. The only real difference is the IHV involved, and (at a later date) evidence to put the whole issue to bed (original IQ restored without change in performance). Yet this case required an intensive study, fact-finding mission article, and later a mindset of denouncing any scores from ATI running Quake3.

 

[XForce]

New solutions in terms of press/community/manufacture relationships are being tried at NVIDIA..

Interesting..

..and seem to be paying off in spades.

Sure, but for whom?
Kyle, NV or the community?

It's not
"[W]e give you all the information we have and you decide"
but
"[W]e let you know what we consider important for you to know when and if we decide and in case you're to lazy to think, we already have an opinion prepared for you. Would you like ketchup with it?"

Sorry.

 

[demalion]

There seems little wrong with the possibilities allowed by what he outlined would be done in future reviews. What is missing is the idea of accountability to anyone but himself. Everything boils down to "trust him", and I simply ask "why"?

He states regret for his actions about Extremetech, but the damage is done and well established. Where is the effort to undo it? He still used the label "B3DPolice" in the thread without apology, yet the justification for that label was part and parcel with what he now claims to regret. It might be inconvenient and uncomfortable to outright apologize, in public, and to everyone he might have done wrong to, but it was his call to do those things in the first place, wasn't it?

Yes, that's a "lot to ask", but he made a lot of unfounded, ill-considered, and damaging accusations. Is he accepting accountability for that, or just the portion dealing with a bigger media organization than him? Only in a forum, and without apology to anyone else maligned the same way and similarly without provided coherent cause? Such selective action, after such a long time, is giving up self interest by the inch as forced to, not exhibiting fairness.

I see it in essence as being a request to "trust me to handle my relationship with nVidia and the fair evaluation of image quality within this framework that allows fairness", while both ignoring that the framework also allows unfairness and avoiding engaging in the discussion of how he has failed the trust already. To skip from before him being viewed as lacking, to an after he is trying to create, while skipping all the messy, complicated, and "in the past" issue of why he was viewed as lacking in between, except as it suites him "moving on" from it.

One problem I see with that is things like"B3DPolice" and other accusations, as well as bannings and deletions, he made along the way, and whose justification are directly associated with things he said he is changing. By ignoring them, and "gracefully" saying he'll model his reviews like what people want, the issue of how he used such accusations to respond to being told to do this in the first place is just avoided. That's admitting error as it suites your self interest. I'd say that such action indicates that a person can be trusted to look after their self interest and public perception, and that admitting error when it is inconvenient indicates that a person can maybe be trusted to do otherwise.

If this had directly followed the ExtremeTech article commentary on the front page, it would indicate something different. What it follows instead is a long time period of accusations and dedication to a specific anti-Futuremark agenda that presents as its (only, that I've seen) technical argument a literal transcription of nVidia's statements on the matter alone, among other actions. He's covered the Extremetech problem, atleast in part of that one post...what about all the rest?

 

[Doomtrooper]

Beyond3D police is sad considering they are supposed to be publishing the truth, after I saw Dave banned the red flag should have went up, the internet community that knows Dave respects him and his work !. If you look at Dave's post where he was banned, he did not break any rules, just exposed Hardocp and their editors as liars. Who wants to be a part of a forum that bans people for posting the truth, I don't miss being there as his 'readers' are mostly 15 year olds that are looking for 'kickass' blue LED card reviews, with no reagard on how those numbers were reached.


Liars, there is no other word for it...they dedicate the entire NV31/NV34 review on apples to apples trilinear filtering AF then back pedal and say it isn't possible to do apples to apples
He mentions he talked to another popular review site, guaranteed is was Anand..at that site in question is certainly not famous either for posting accurate information.(see 5900 review and Splinter Cell issues).

The only way it will stop is the IHVs, like what Matrox did.. just stop sending review cards...Kyle has got a fat head, and it needs to be popped like a balloon.

 

[Nazgul]

Kyle still failed to address why he feels it's justifiable to force a "pseudo-Trilinear" filtering only when a specific game very popular for making review benchmarks is run, but said "optimization" does not get applied universally. If the image quality really is comparable, then there's no reason why it should not be applied in all situations. It's not unreasonable to look at the situation and suspect that Nvidia is reducing image quality in a popular benchmarking game in order to gain performance by changing the way one of their own driver settings functions only when a certain filename is executed. I see no justification for this whatsoever, and until Kyle, Nvidia, or somebody provides a convincing answer I see no defense for Nvidia's actions in this UT2K3 issue.

As for Kyle's comments about ExtremeTech, that really deserves to be on the front page. That's where he publicly questioned ExtremeTech's motives and integrity when this whole issue began, the respectful thing to do would be to post a note on the same front page apologizing for making those comments. I think doing so would repair his reputation far more than any forum post would. Of course, I think his language needs to be clearer as well. He did state that ExtremeTech got the facts right, but he didn't actually retract his statement about ET's motives for posting those facts.


Overall, I don't really see Kyle's post as solving much of anything. I still have significant disagreements with [H]'s stance on evaluating image quality & performance. I still have issues with the way they report(or don't report) on Nvidia's blatant manipulation of prgrams widely used to evaluate their video cards. I have a real problem with the heavy-handed treatment people get on his forums. Yet he has no problem taking a swipe at other sites like B3D or Rage3D when they weren't even part of the thread topic("if you want to beat a dead horse, why don't you go over to Rage3D, there's plenty of them over there" is a slight paraphrase of a very recent post of his). I don't think Kyle's going to change anyone's opinion of him with vague forum posts at this point.

Ah well, what do I care if one site's reviews are no longer consistent with my priorities when it comes to 3D video cards. There's plenty other sites out there to get info from.

And K.I.L.E.R, the reason why there's yet another thread is the same reason why there's always another thread, Kyle or [H] have posted something new which folks have questions/concerns/criticisms about, and many folks either don't want to or are unable to talk about it over there.

 

[beyondhelp]

Kyle - "I hope I have answered your three questions sufficiently."

Not in the least, But I do have a much better understanding of your Opinions, Rationalizations, and Justifications.

They are still erroneous because they are based on the false premises that what Nvidia did were Optimizations in 3DMark, not cheats so it was OK; and That IQ manipulation behind your back in UT2K03 using Nvidia drivers is OK, and the IQ "Good enough" for everyone, etc. etc. etc.

Hellooo! Kyle! I want to have the say in what IQ I get onscreen. I paid $XXX's of dollars for this Mega Gaming Wonder and I want to run it full tilt boogy with highest IQ possible. I don't give two shits about what Nvidia thinks it should be run at, nor do I want them hardcoding driver IQ settings based on Game Name, especially if it prevents me from the highest settings. It's purely an attempt to inflate scores in the Benchmark. And it sure as hell does NOT give the best Gaming Experience possible, as you seem to think your reviews represent. Hello! You gettings this Kyle?

Whether IQ is good enough or not, whether called cheats or optimizations, whether in Synthetic or "Game" Benchmarks, Nvidia is dumping workload behind the scenes in certain well known and oft used benchmarks at supposed "equal" IQ settings to it's Competitor to inflate scores and look better than they are. That's fruad, That's False Advertising, That's theft by Deception. That's BOGUS! That deserves a Front Page Editorial, Not lip service in your Forums and being swept under a rug.

I do not care about your subjective opinions of what is "good enough" IQ while playing games. Your not me, nor any other of your readers, and to ASSUME you know what's best for all of us is the height of Egotism.

If the first Law of Journalism is to be Objective, imho You failed miserably.