Ops,"I" did it again. Why does Futuremark still ke

[Bouncing Zabaglione Bros.]

Since when did Futuremark sanction cheating? I'd say they are turning a blind eye to it. I haven't yet seen them say "yes, NVIDIA's optimizations are acceptable." In fact, we've heard the opposite. That cheating or optimization, they are still invalid.

If you know it's happening and you keep quiet about it, then you are sanctioning it. Don't forget, Futuremark are in the position of policing the validity of 3Dmark scores. They have an obligation to keep the scores "clean"

The question is, once NVIDIA launches some new official drivers which bring the scores back up, will Futuremark do anything or stay silent? I'm betting at this point that they will just keep quiet.

Just after the 330 patch came out, Nvidia launched new drivers that seemed to put the cheats back in. Performance went up to pre-330 cheat levels. Neither FM or Nvidia made any comment about this. Given it would have been in FM and Nvidia's best interest to put out a statement saying "all cheats have been removed and the GFFX really does bench this well" (especially given all the furore over Nvidia's initial cheating), I can only surmise that the cheating is still happening, but that FM have been induced (either by lawyers or money) to keep quiet about it. Thus in my opinion, FM are sanctioning these cheats, and are complicit in the deceit of both Futuremark's and Nvidia's customers.

 

[Entropy]

Just after the 330 patch came out, Nvidia launched new drivers that seemed to put the cheats back in. Performance went up to pre-330 cheat levels. Neither FM or Nvidia made any comment about this. Given it would have been in FM and Nvidia's best interest to put out a statement saying "all cheats have been removed and the GFFX really does bench this well" (especially given all the furore over Nvidia's initial cheating), I can only surmise that the cheating is still happening, but that FM have been induced (either by lawyers or money) to keep quiet about it. Thus in my opinion, FM are sanctioning these cheats, and are complicit in the deceit of both Futuremark's and Nvidia's customers.

True words.
I quoted them because they deserved to be read again.

While Futuremark could have weathered the nVidia storm and come out in as strong as before, they have now managed to alienate just about everyone. Even their own people, apparently. It is one thing to see yourself as part of the foremost benchmarking outfit in the business, a driving force for PC and graphics development as well as equipment sales. It's quite another to look in the mirror every morning and see someone who helps egg the kids on to spend their money buying kit from corporations out to decieve them, a deception you're an accomplice to. I'd predict that Futuremark will loose other good people in the relatively close future.

But IMHO too much attention in this affair has been focussed on Futuremark because nVidia tactics became obvious in their high profile benchmark. Those tactics aren't limited to 3DMark, and it is interesting to follow how the industry (manufacturers/vendors/media/consumers) react and deal with that.

And I wonder just how much nVidia has lost by sticking to their deceptive tactics, and will loose in the future. While there are blind fans, most people take a dim view of being consistently misled and aren't necessarily eager to support it financially. Or are they? If this had been about cars, and a manufacturer had been found to consistently lie about the information on engine power, safety equipment et cetera, they would be deservedly dead in the market. nVidia obviously sells huge volumes to the preconfigured part of the market, but what about the part where consumers make active choices? If anyone has solid retail sales data, it would be very interesting to see it.

Entropy

 

[Reverend]

Dave, if you have an accusation to make, just say it. Instead of making veiled inferences about Nvidia and FutureMarks integrity.

Dave is giving you a "scoop". Scoops invariably never tells the whole story. If you do not prefer to have such scoops, don't comment or if you do, try to respect the publicly recognized host of your public forum participation. I do not like the inference about dave's charactyer you're making here.

Why don't you interview this guy, and ask him the reasons why he left. At least thats proffessional.

I gave you guys the reason he left a long time ago. Look for my post about this, I don't have the time to search for you. I didn't name him for very obvious reasons but someone here already mentioned his initials. He didn't agree with the route taken by Futuremark about the whole NV/FM fiasco.... and he doesn't agree with it now according to the latest agreements (not public yet, or may never be) between NV and FM. Interviewing him would be like interviewing Scott Sellers about why he never wanted to joined NVIDIA -- Scott could be sued for revealing his true feelings and thoughts... so what's the point?

We, as a beta member, means we cannot disclose non-public information. Asking us to do so would mean possibly paying legal fees and fines. Would you be willing to foot such possible bills?

Beyond3D will be watching how things turn out at FM. We will have no hesitation in leaving the beta program if we feel that the way things are going at FM doesn't agree with our own principles. We don't know all the details at this point. When we do, we can decide what to do but we not be able to reveal it due to NDAs. We may still be tied to NDAs for a certain period if we decide to leave the beta program, so anything we may have to say about why we may be leaving the beta program will have to come upon NDA expiry. This should not infer that we have already decided what we're going to do based on what we know because, as mentioned, we don't have all the details (FM has not emailed us)... I'm only saying all of this because I'm annoyed by the apparent fact that you don't seem to know what it means to be a FM beta member and all the legal tie-ins this means.

 

[Ailuros]

While this is true in general, can you look at the facts of this whole nVidia/3DMark03-thing and still believe valid optimizations to be the cause for the recent re-alignment of 3DMark-scores?

Clipping planes are like Partisans. They randomly appear in your driver code, when you expect them the least

 

[WaltC]

Re: Ops,"I" did it again. Why does Futuremark stil

Ops,Nvidia did it again. New drivers Dets 44.67 optimized for 3dmark03 patch 3.30.FX made much better scores again.
Why does Futuremark still keep silence?
3dmark03 now is meaningless, isnt it?

44.03 is the latest official nVidia Detonator package officially released by the company from www.nvidia.com . Everything else is either leaked or an OEM driver (which may use different, and misleading, numbering schemes.)

I think all that FM can reasonably be asked to do is to issue a recompile patch along the lines of 330 every time nVidia officially releases a driver package. I think FM is entirely right to ignore all other "releases."

I saw a blurb somewhere yesterday which stated nVidia would officially release a new Detonator set on Monday, Aug 11. Have no idea whether it's true, of course. But if it is then that's when FM should do another patch. If and when nVidia releases another official Det package, if FM at that time (within a few days) refuses to issue another recompile patch then I'll agree with you.

 

[Fred]

Reverend, theres ways of putting things that are more apt to be dissapproved of. Like it or not, Dave is an integral part of the 3d graphics community and his word (and yours) carry more weight and therefore more responsibility than mine or any others. Responsibility of the press, and all that silly stuff.

His comments could be taken like 20 different ways, some of which are accusatory in nature.

For instance, I could infer (as I initially did) that he was making aspersions about the integrity of 3dmarks future programs, because as he put it, a 'stand up' guy left the very same day that Nvidia rejoined. Inference: all 'stand up' guys should feel the very same way. Thats one way of interpreting it, wrong as it may be.

It would be unfair to 3dmark to do just that, despite what others on this forum posted. No information has been given as to the details and the 'whys' to this mans departure, and for all we know they still have the consumers best interests at heart. Unless you wish to show evidence to the contrary?

I don't like it when Kyle mudslings publically, and veiled mudslinging is just as bad.

(My excuses go to Dave of course, as I doubt that was the intent. Just a quibble with the wording is all)

 

[Mark]

So... now that nVidia apparently "endorses" 3DMark03 once again, any takers on how long it'll be before [H] reintroduces it back into their reviews? How about the "Oops, sorry we were wrong. 3DMark03 is a great benchmark" editorial by Kyle?

 

[nyt]

So... now that nVidia apparently "endorses" 3DMark03 once again, any takers on how long it'll be before [H] reintroduces it back into their reviews? How about the "Oops, sorry we were wrong. 3DMark03 is a great benchmark" editorial by Kyle?

Why don't you wait instead of speculating about some more potential bashing? Ain't you tired of it? AFAIK they made some efforts with their latest review, you could acknowledge that. IMHO the discussed move from 3DMark03 will further prove [H]'s point regarding this very benchmark. Now if they do consider using it again, it's another story.

 

[thatdude90210]

So... now that nVidia apparently "endorses" 3DMark03 once again, any takers on how long it'll be before [H] reintroduces it back into their reviews? How about the "Oops, sorry we were wrong. 3DMark03 is a great benchmark" editorial by Kyle?

I predict FM will introduce a new patch 340, it will do absolutely nothing but will allow Nv and [K] to save face, "patch 340 has address our previous concerns so 3Dmark03 is now useful."

 

[just me]

Not surprised at any of these events. FM has changed since their 'original' Marks with the most significant changes after 2001. Look who came 'onboard' since then & look how MO changed & what it has become. $$$ is the only reason FM is in business anymore & I wouldn't doubt if SW & a few others follow AJ's lead.

Seems the NV NDA muzzle is on here too. How does it feel to have to watch every word for fear of legal repercussions?

"Keep your friends close & your enemies closer" & under an NDA.

.02,

 

[WaltC]

As of Monday NVIDIA rejoined the Beta program. As of Monday one of the most stand up guys I know left Futuremark.

Well, just like me to reply to the first post without reading the thread...and miss the most important post in it....Tells me pretty much all I need to know. I am really bothered by the fact that there's been no press release from nVidia reversing the press release it made when it quit last year.

 

[WaltC]

Beyond3D will be watching how things turn out at FM. We will have no hesitation in leaving the beta program if we feel that the way things are going at FM doesn't agree with our own principles. We don't know all the details at this point. When we do, we can decide what to do but we not be able to reveal it due to NDAs. We may still be tied to NDAs for a certain period if we decide to leave the beta program, so anything we may have to say about why we may be leaving the beta program will have to come upon NDA expiry. This should not infer that we have already decided what we're going to do based on what we know because, as mentioned, we don't have all the details (FM has not emailed us)... I'm only saying all of this because I'm annoyed by the apparent fact that you don't seem to know what it means to be a FM beta member and all the legal tie-ins this means.

Rev, nVidia quit the program last year and made plenty of public statements--negative ones--about FM and their benchmarks. I can't see at all how nVidia would be less NDA-bound than B3d, and apparently nVidia wasn't bound at all. So I'd say you have at least as much freedom to say what you like as nVidia has done for the past 8-9 months. I'd say you have nothing to worry about. I can't see how FM could hold you to any higher standard than it held nVidia. (I would think an NDA going beyond the code you've seen and used would be very excessive.) If you are concerned for some reason about this then I would suggest you do what nVidia has done--find a convenient proxy to release the info for you, if you are concerned about releasing it directly.

It also strikes me as very odd that as a partner in the program you have not been informed of these developments, if they have actually occurred, and there is nothing stopping you as a partner from emailing FM to inquire (no need to "wait" on them necessarily.) Perfectly legitimate, straightforward, question which all the partners have a right to know about. I mean, it's not as if FM keeps its partner list a secret...

 

[WaltC]

...
For instance, I could infer (as I initially did) that he was making aspersions about the integrity of 3dmarks future programs, because as he put it, a 'stand up' guy left the very same day that Nvidia rejoined. Inference: all 'stand up' guys should feel the very same way. Thats one way of interpreting it, wrong as it may be.

...

Fred, I don't think it's possible for anyone, anywhere, to cast more aspersion on FM than FM has thus far cast upon itself. I have been trying to give them the benefit of the doubt, however, and was hoping they'd be quick to release another recompile patch shortly after the next official, bona-fide Detonator release. I wanted to give them that much room, at least.

But if Dave has received the straight skinny on this, and FM and nVidia are conspiring to keep nVidia's renewed participation a secret, that will tear it for me. Without a doubt.

Frankly, if this situation is the correct one as described, I am amazed--amazed--that ATi or any other reputable partner would continue to pay to be a member of this program. I hope we'll get some quick confirmation of these events, one way or the other.

 

[StealthHawk]

Re: Ops,"I" did it again. Why does Futuremark stil

Ops,Nvidia did it again. New drivers Dets 44.67 optimized for 3dmark03 patch 3.30.FX made much better scores again.
Why does Futuremark still keep silence?
3dmark03 now is meaningless, isnt it?

44.03 is the latest official nVidia Detonator package officially released by the company from www.nvidia.com . Everything else is either leaked or an OEM driver (which may use different, and misleading, numbering schemes.)

I think all that FM can reasonably be asked to do is to issue a recompile patch along the lines of 330 every time nVidia officially releases a driver package. I think FM is entirely right to ignore all other "releases."

I saw a blurb somewhere yesterday which stated nVidia would officially release a new Detonator set on Monday, Aug 11. Have no idea whether it's true, of course. But if it is then that's when FM should do another patch. If and when nVidia releases another official Det package, if FM at that time (within a few days) refuses to issue another recompile patch then I'll agree with you.

These actions actually bring up an interesting dilemma though. How exactly did NVIDIA claim the number one spot in 3dmark03. What driver were the cards using? Certainly it wasn't 44.03...

So Futuremark is willing to accept possibly tainted drivers but is not willing to release a new patch. That is very irresponsible. Of course, there may be something bigger here going on.

If NVIDIA releases a new driver on Monday with the same ridiculously high scores, and Futuremark does nothing, then it will be obvious to everyone that they have sold out their integrity.

 

[Dave Baumann]

But if Dave has received the straight skinny on this, and FM and nVidia are conspiring to keep nVidia's renewed participation a secret, that will tear it for me. Without a doubt.

I don't think there is any secret here, its just that the deal is rather fresh. I've probably just done a rather poor job of preannouncing things!

 

[Rookie]

Dave wrote:

As of Monday NVIDIA rejoined the Beta program. As of Monday one of the most stand up guys I know left Futuremark.

Rev wrote:

Beyond3D will be watching how things turn out at FM. We will have no hesitation in leaving the beta program if we feel that the way things are going at FM doesn't agree with our own principles. We don't know all the details at this point. When we do, we can decide what to do but we not be able to reveal it due to NDAs. We may still be tied to NDAs for a certain period if we decide to leave the beta program, so anything we may have to say about why we may be leaving the beta program will have to come upon NDA expiry. This should not infer that we have already decided what we're going to do based on what we know because, as mentioned, we don't have all the details (FM has not emailed us)... I'm only saying all of this because I'm annoyed by the apparent fact that you don't seem to know what it means to be a FM beta member and all the legal tie-ins this means.

Does your b3d guys imply when next monday NVIDIA back to beta program,B3D will quit ?

 

[Fred da Roza]

Re: Ops,"I" did it again. Why does Futuremark stil

So Futuremark is willing to accept possibly tainted drivers but is not willing to release a new patch. That is very irresponsible. Of course, there may be something bigger here going on.

If NVIDIA releases a new driver on Monday with the same ridiculously high scores, and Futuremark does nothing, then it will be obvious to everyone that they have sold out their integrity.

How do you fight a company that has no ethics and another that is willing to sell theirs? This is a serious dilemma for ATI. If Futuremark won’t enforce the validity of their benchmark, does ATI have any other choice but to cheat as well? If it destroys the usefulness of Futuremark's benchmark so be it. They made their bed.

 

[Sazar]

So... now that nVidia apparently "endorses" 3DMark03 once again, any takers on how long it'll be before [H] reintroduces it back into their reviews? How about the "Oops, sorry we were wrong. 3DMark03 is a great benchmark" editorial by Kyle?

as has been stated already... this move.. if [H] keeps their stated stance with regards to 3dmark03 == they were right... hard as it may be to believe

ergo... [H] MAY yet be vindicated for SOME of their statements made over the past several months...

has yet to be seen but I for one am not going to jump the gun and start another KB bashing thread... at least not this time...

until such a time as an editorial or other is released by [H]... any comment on their exected position in light of the events that have apparently taken place is pure speculation... which is all well and good... but bashing == non-constructive...

btw... cheers for the news and info Dave/Rev... appreciated...

 

[digitalwanderer]

Re: Ops,"I" did it again. Why does Futuremark stil

How do you fight a company that has no ethics and another that is willing to sell theirs?

Well generally with sarcasm, although I tend to rant and stomp about a bit in outrage and disgust too.

 

[Arun]

Re: Ops,"I" did it again. Why does Futuremark stil

How do you fight a company that has no ethics and another that is willing to sell theirs? This is a serious dilemma for ATI. If Futuremark won’t enforce the validity of their benchmark, does ATI have any other choice but to cheat as well? If it destroys the usefulness of Futuremark's benchmark so be it. They made their bed.

Well, nVidia tried to quit 3DMark with 0 community support.
This time, ATI would have nearly full community support ( certainly including mine ) - so that's an option.
And then there's what Digitalwanderer said, hehe


Uttar