Dave Baumann of Beyond3D joins ATI

[secretanchitman]

Originally posted by: josh6079
It's good to see that this is still about Dave's promotion......

lol i was thinking of writing the same thing...

 

[RobertR1]

Originally posted by: josh6079
It's good to see that this is still about Dave's promotion......

haha yeah! i was the about to start a new thread: "The REAL Dave Baumann joins Ati thread."

 

[redbox]

So....let me get this both Nvidia and Ati had a chip code that would use optomised drivers for the 3dmark test? How is this different then using driver optomization to get better performance in FEAR and Oblivion and any other game. It's not like the cards didn't produce those frames is it? The cards still had to produce those results on their own right? Or did it produce fake FPS, where it said it got 50fps but only got 40fps?

 

[kylebisme]

The difference is optimizations for a game, as long as they don't reduce image quality, only serve to benefit the end users. On the other hand, optimizations for a benchmark only serves to invalidate the results of the tests and mislead potential customers into thinking the hardware is capable of doing more than it actually can.

A short explanation of how they each cheated:

Nvidia re-wrote shader code to make the benchmark run faster on their hardware, doing less math and visibly reducing image quality. And they also took advantage of the fixed viewable distance in the benchmark fly-through by forcing their hardware to not draw polygons beyond a certain distance, that didn't effect image quality in the benchmark fly though; but when going off the rails; stuff that should have been drawn came up missing and hence that "optimzation" was by no means anything that could be used in an acutal game.

In response to what Nvidia was doing, ATI started to re-arrange shader code to make the benchmark run faster on their hardware, but while providing mathematically equivalent results with no loss in image quality. This much the same as what is still done in games by both companies today; but still poor form in a benchmark which is designed to compare products under the same workload.

Beyond that, Nvidia first denied their cheating and claimed bugs for as long as they could before simply refusing to stop composing the integrity of the benchmark and only put a stop to their cheating after releaseing updated hardware. When ATI got caught they admitted their fault and removed their cheats from their drivers. Hence Munky's statement that Nvidia's cheating was blatant, and hence the reason I took issue with Nelsieus's attempt to downplay Munky's statement with the insinuation that ATI had committed equivalent offences. Yeah, they both cheated back in the days of 3dmark03, but any reasonable man can see that one did so far more blatantly than the other.

 

[redbox]

Snowman thanks for sorting that out. While I don't think that what each company did was a golden standard, I don't think it was that wrong either. In my eyes 3dmark isn't a very good cross platform benchmark. A game would be a much better benchmark because you are actually going to be playing it and wanting it to run nice. No one buys a $500 card and only wants to run 3dmark, well anyone with a brain doesn't.

I do use 3dmark however when looking to see how much my card has gained when overclocking it is a much more stable result then playing through a game where fps jump all over the place.

 

[kylebisme]

Yeah, the 3dmark suites can be very handy for that. However, Futuremark's intent with their graphics benchmarks has always been target the latest version of DirectX and implement it's features into game-like situations so that we can get an idea of how our hardware handles such things well before anyone has had enough time to actually build a full game based on that new version of DirectX. Some of their efforts have been arguably off the mark (to pun the term ); but their intent is noble and their benchmarks, at least withstanding those effected by driver cheats, have generally provided results which fall in line with the games that came out latter down the line.

 

[BFG10K]

Cheating is cheating. The fact that you feel compelled to defend one "cheater" over the other is disturbing, but expected from someone like you, nonetheless.

Nope, sorry, the two aren't even close. ATi had generic shader re-ordering but unfortunately they relied on application detection of 3DMark for them to work which is where the cheating came from. When called out they apologized and removed the detection in the next driver. As for Quack, that was a bug and when they fixed it performance and IQ went up.

When nVidia was caught out in 3DMark first they denied it, then they claimed it was a bug, then they claimed 3DMark was invalid, then they badgered FutrureMark to change their stance. In the end nVidia put out some ridiculous "optimization guidelines" which they violated even before the ink had dried.

Below is the FutureMark audit report from 3DMark that shows just how badly nVidia cheated, and this is to say nothing of the thousands of shaders nVidia subsituted in games.

Wat Are The Identified Cheats?
Futuremark?s audit revealed cheats in NVIDIA Detonator FX 44.03 and 43.51 WHQL drivers. Earlier GeForceFX drivers include only some of the cheats listed below.

1. The loading screen of the 3DMark03 test is detected by the driver. This is used by the driver to disregard the back buffer clear command that 3DMark03 gives. This incorrectly reduces the workload. However, if the loading screen is rendered in a different manner, the driver seems to fail to detect 3DMark03, and performs the back buffer clear command as instructed.

2. A vertex shader used in game test 2 (P_Pointsprite.vsh) is detected by the driver. In this case the driver uses instructions contained in the driver to determine when to obey the back buffer clear command and when not to. If the back buffer would not be cleared at all in game test 2, the stars in the view of outer space in some cameras would appear smeared as have been reported in the articles mentioned earlier. Back buffer clearing is turned off and on again so that the back buffer is cleared only when the default benchmark cameras show outer space. In free camera mode one can keep the camera outside the spaceship through the entire test, and see how the sky smearing is turned on and off.

3. A vertex shader used in game test 4 (M_HDRsky.vsh) is detected. In this case the driver adds two static clipping planes to reduce the workload. The clipping planes are placed so that the sky is cut out just beyond what is visible in the default camera angles. Again, using the free camera one can look at the sky to see it abruptly cut off. Screenshot of this view was also reported in the ExtremeTech and Beyond3D articles. This cheat was introduced in the 43.51 drivers as far as we know.

4. In game test 4, the water pixel shader (M_Water.psh) is detected. The driver uses this detection to artificially achieve a large performance boost - more than doubling the early frame rate on some systems. In our inspection we noticed a difference in the rendering when compared either to the DirectX reference rasterizer or to those of other hardware. It appears the water shader is being totally discarded and replaced with an alternative more efficient shader implemented in the drivers themselves. The drivers produce a similar looking rendering, but not an identical one.

5. In game test 4 there is detection of a pixel shader (m_HDRSky.psh). Again it appears the shader is being totally discarded and replaced with an alternative more efficient shader in a similar fashion to the water pixel shader above. The rendering looks similar, but it is not identical.

6. A vertex shader (G_MetalCubeLit.vsh) is detected in game test 1. Preventing this detection proved to reduce the frame rate with these drivers, but we have not yet determined the cause. Page 4 of 7

7. A vertex shader in game test 3 (G_PaintBaked.vsh) is detected, and preventing this detection drops the scores with these drivers. This cheat causes the back buffer clearing to be disregarded; we are not yet aware of any other cheats.

8. The vertex and pixel shaders used in the 3DMark03 feature tests are also detected by the driver. When we prevented this detection, the performance dropped by more than a factor of two in the 2.0 pixel shader test.

 

[Ackmed]

Originally posted by: keysplayr2003

Originally posted by: Ackmed
Why should someone who is right, back down. And let more ignorance and misinformation be spread around?

Why dont you "back down"?

Can I ask if any of you are getting through to the FUD spreader in question, Ackmed?
What exactly should I back down from? Or is this just something you're saying back to me because I said it? You know, like nursery school.

I find it hypocrtical that you tell someone else to back down, when you dont do it yourself. You're posting as much as anyone, and replying to people who arent even talking to you. So in short, take your own advice.

 

[redbox]

Originally posted by: TheSnowman
Yeah, the 3dmark suites can be very handy for that. However, Futuremark's intent with their graphics benchmarks has always been target the latest version of DirectX and implement it's features into game-like situations so that we can get an idea of how our hardware handles such things well before anyone has had enough time to actually build a full game based on that new version of DirectX. Some of their efforts have been arguably off the mark (to pun the term ); but their intent is noble and their benchmarks, at least withstanding those effected by driver cheats, have generally provided results which fall in line with the games that came out latter down the line.

I don't know maybe if I knew what the heck the futuremark's numbers where supposted to mean, then maybe I could use it as a base for next gen game performance, but as it stands I can relate a heck of alot better to fps then I can 5000 futuremarks. In the end it's just a ruler for a person's e-penis with two companies saying 'nope... no... now 4inches is a big futuremark.... really it is! ....'

 

[geo1]

Originally posted by: Nelsieus

Again, reitterating my previous statements, as far as Uttar shifts green, I think you shift red. Neither of you are unreasonable or unfair, and quite frankly, are prominent members of the B3D community. So I too am not trying to go on an "anti-Geo" rant, just balancing the statement of Uttar being "pro-greenie" with the same standards that you're using (which I dub fair).

Have we lost sight of the original context of the comment? The point is, unlike the characterization that Greg likes to flog here every chance he gets, there are several prominent B3D members with a green lean and the technical chops to argue with anyone --including Wavey-- and that continue to thrive there. There's also an NV engineer posting on a regular basis, and several other NV employees have dropped in from time to time on issues of interest. I don't know what other boards you can say that of.

Edit: Oh, and to finally tie this back to the thread topic (what!?), I don't think there's any real doubt that NV was one of the suitors who came calling at Baumann's door with career options.

 

[ginfest]

Originally posted by: josh6079
It's good to see that this is still about Dave's payoff......

Fixed

 

[Genx87]

Really no surprise at all, whilke the reviews are usually pretty good, Daves posts on the forum always had a tint of Red.

Good for him though, maybe he can fix ATI's approach to doing business.

 

[DaveBaumann]

Originally posted by: RobertR1
haha yeah! i was the about to start a new thread: "The REAL Dave Baumann joins Ati thread."

Why, has a fake one joined?

 

[Creig]

Congratulations on your new position, Dave.

 

[erikistired]

Originally posted by: 5150Joker

Originally posted by: Gstanfor

Originally posted by: 5150Joker

Originally posted by: Gstanfor
Not always he hasn't. You need to carefully read what he posts in the forums to see his ATi bias, he hides it well in reviews. Nonetheless he should be much happier at ATi, and maybe, just maybe, he will drag their marketing out of the sewer.

Weren't you banned at B3D for being a troll?

No, banned for proving Baumann wrong on technical issues pertaining to DX9 / Geforce FX too many times. Baumann has a thin skin...

In another words, you were banned for being a troll.

how have you managed to not get banned from here 5150joker?

 

[Gstanfor]

I'm not exactly a fan of Martrox over at b3d, but I must say, i found his post below to be "interesting" too, especially in light of Geo's efforts here...

link

While Dave deservers our thanks and the best for his and his families future, I can't help but wonder where those members of B3d that lean to the green are......

Chalnoth?......Last Activity: 26-Jun-2006 06:31
Ailuros?.....Last Post:25-Jun-2006, 07:15
Uttar?.....Last Activity: 26-Jun-2006 08:12

Not trying single out anyone in particular. I just find it "interesting"!

 

[ronnn]

Originally posted by: DaveBaumann

Originally posted by: RobertR1
haha yeah! i was the about to start a new thread: "The REAL Dave Baumann joins Ati thread."

Why, has a fake one joined?

Super congrats!!!

 

[fierydemise]

Originally posted by: DaveBaumann

Originally posted by: RobertR1
haha yeah! i was the about to start a new thread: "The REAL Dave Baumann joins Ati thread."

Why, has a fake one joined?

That made me laugh, thanks dave, and congrats on your new position.

 

[redbox]

Originally posted by: Gstanfor
I'm not exactly a fan of Martrox over at b3d, but I must say, i found his post below to be "interesting" too, especially in light of Geo's efforts here...

link

While Dave deservers our thanks and the best for his and his families future, I can't help but wonder where those members of B3d that lean to the green are......

Chalnoth?......Last Activity: 26-Jun-2006 06:31
Ailuros?.....Last Post:25-Jun-2006, 07:15
Uttar?.....Last Activity: 26-Jun-2006 08:12

Not trying single out anyone in particular. I just find it "interesting"!

I don't see anything 'interesting' in that link. Looks like they have all recently visited the forum ....soooo?

 

[Keysplayr]

Originally posted by: Ackmed

Originally posted by: keysplayr2003

Originally posted by: Ackmed
Why should someone who is right, back down. And let more ignorance and misinformation be spread around?

Why dont you "back down"?

Can I ask if any of you are getting through to the FUD spreader in question, Ackmed?
What exactly should I back down from? Or is this just something you're saying back to me because I said it? You know, like nursery school.

I find it hypocrtical that you tell someone else to back down, when you dont do it yourself. You're posting as much as anyone, and replying to people who arent even talking to you. So in short, take your own advice.

My own advice eh? Ok. Let me advise myself to do some digging. What kind of digging? Oh, I don't know. Be back later.

 

[Nelsieus]

Originally posted by: BFG10K

Cheating is cheating. The fact that you feel compelled to defend one "cheater" over the other is disturbing, but expected from someone like you, nonetheless.

Nope, sorry, the two aren't even close. ATi had generic shader re-ordering but unfortunately they relied on application detection of 3DMark for them to work which is where the cheating came from. When called out they apologized and removed the detection in the next driver. As for Quack, that was a bug and when they fixed it performance and IQ went up.

When nVidia was caught out in 3DMark first they denied it, then they claimed it was a bug, then they claimed 3DMark was invalid, then they badgered FutrureMark to change their stance. In the end nVidia put out some ridiculous "optimization guidelines" which they violated even before the ink had dried.

Below is the FutureMark audit report from 3DMark that shows just how badly nVidia cheated, and this is to say nothing of the thousands of shaders nVidia subsituted in games.

Wat Are The Identified Cheats?
Futuremark?s audit revealed cheats in NVIDIA Detonator FX 44.03 and 43.51 WHQL drivers. Earlier GeForceFX drivers include only some of the cheats listed below.

1. The loading screen of the 3DMark03 test is detected by the driver. This is used by the driver to disregard the back buffer clear command that 3DMark03 gives. This incorrectly reduces the workload. However, if the loading screen is rendered in a different manner, the driver seems to fail to detect 3DMark03, and performs the back buffer clear command as instructed.

2. A vertex shader used in game test 2 (P_Pointsprite.vsh) is detected by the driver. In this case the driver uses instructions contained in the driver to determine when to obey the back buffer clear command and when not to. If the back buffer would not be cleared at all in game test 2, the stars in the view of outer space in some cameras would appear smeared as have been reported in the articles mentioned earlier. Back buffer clearing is turned off and on again so that the back buffer is cleared only when the default benchmark cameras show outer space. In free camera mode one can keep the camera outside the spaceship through the entire test, and see how the sky smearing is turned on and off.

3. A vertex shader used in game test 4 (M_HDRsky.vsh) is detected. In this case the driver adds two static clipping planes to reduce the workload. The clipping planes are placed so that the sky is cut out just beyond what is visible in the default camera angles. Again, using the free camera one can look at the sky to see it abruptly cut off. Screenshot of this view was also reported in the ExtremeTech and Beyond3D articles. This cheat was introduced in the 43.51 drivers as far as we know.

4. In game test 4, the water pixel shader (M_Water.psh) is detected. The driver uses this detection to artificially achieve a large performance boost - more than doubling the early frame rate on some systems. In our inspection we noticed a difference in the rendering when compared either to the DirectX reference rasterizer or to those of other hardware. It appears the water shader is being totally discarded and replaced with an alternative more efficient shader implemented in the drivers themselves. The drivers produce a similar looking rendering, but not an identical one.

5. In game test 4 there is detection of a pixel shader (m_HDRSky.psh). Again it appears the shader is being totally discarded and replaced with an alternative more efficient shader in a similar fashion to the water pixel shader above. The rendering looks similar, but it is not identical.

6. A vertex shader (G_MetalCubeLit.vsh) is detected in game test 1. Preventing this detection proved to reduce the frame rate with these drivers, but we have not yet determined the cause. Page 4 of 7

7. A vertex shader in game test 3 (G_PaintBaked.vsh) is detected, and preventing this detection drops the scores with these drivers. This cheat causes the back buffer clearing to be disregarded; we are not yet aware of any other cheats.

8. The vertex and pixel shaders used in the 3DMark03 feature tests are also detected by the driver. When we prevented this detection, the performance dropped by more than a factor of two in the 2.0 pixel shader test.

It's very telling how you, Munky, and Snowman rush in to defend ATI when all I stated was how "both nVidia and ATI had their own driver cheats." (which I cited by 4 sources, including references from FutureMark, themself).

Kind of like when a principal calls in a student who immediatly screams "I didn't do it!" before the principal even says anything... :roll:

(Awaits Josh to bawl "But BFG has a Geforce card! Ha! You can't reprehend him!")

 

[kylebisme]

I did not rush to defend anything, I patiently gave you plenty of time to clarify your attempt to deprecate Munky's comment on Nvidia's blatent cheating.

It is very telling how you continue to ignore reality, kinda like the ex-husband who says "who cares if I screwed prostitutes regularly, my wife confessed to making out with her old boyfriend once."

 

[Nelsieus]

Originally posted by: TheSnowman
I did not rush to defend anything, I patiently gave you plenty of time to clarify your attempt to deprecate Munky's comment on Nvidia's blatent cheating.

It is very telling how you continue to ignore reality, kinda like the ex-husband who says "who cares if I screwed prostitutes regularly, my wife confessed to making out with her old boyfriend once."

Who's trying to deprecate nVidia's blatant cheating?
I believe my words were:

Originally posted by: Nelsieus
To be fair, both teams had their own "cheating methods."
But I agree both can have very low tactics at time. The recent (well, occuring in the last year) slideshows that were distributed via TechPowerUp had shown just how low these two companies could stoop.
But saying one is "clearly dirtier" than the other is quite a biased statement, imho.

You're the one who's ignoring reality, all because I associate ATI with the driver cheat incidents (omg!) which did take place, and apparently, you must not like when someone casts ATI in a poor light (yet attack nVidia's blatant cheating!)

Not only are you ignoring reality, but you're delusional. Atleast Munky and BFG are actually making points. You're just trying to get attention... *sigh*

EDIT: Atleast Munky and BFG are atleast attempting to making points...

 

[kylebisme]

So who do you think was on my account last night typing out those paragraphs explaining the details of the issue at the request of Redbox?

 

[redbox]

Originally posted by: Nelsieus

Originally posted by: BFG10K

Cheating is cheating. The fact that you feel compelled to defend one "cheater" over the other is disturbing, but expected from someone like you, nonetheless.

Nope, sorry, the two aren't even close. ATi had generic shader re-ordering but unfortunately they relied on application detection of 3DMark for them to work which is where the cheating came from. When called out they apologized and removed the detection in the next driver. As for Quack, that was a bug and when they fixed it performance and IQ went up.

When nVidia was caught out in 3DMark first they denied it, then they claimed it was a bug, then they claimed 3DMark was invalid, then they badgered FutrureMark to change their stance. In the end nVidia put out some ridiculous "optimization guidelines" which they violated even before the ink had dried.

Below is the FutureMark audit report from 3DMark that shows just how badly nVidia cheated, and this is to say nothing of the thousands of shaders nVidia subsituted in games.

Wat Are The Identified Cheats?
Futuremark?s audit revealed cheats in NVIDIA Detonator FX 44.03 and 43.51 WHQL drivers. Earlier GeForceFX drivers include only some of the cheats listed below.

1. The loading screen of the 3DMark03 test is detected by the driver. This is used by the driver to disregard the back buffer clear command that 3DMark03 gives. This incorrectly reduces the workload. However, if the loading screen is rendered in a different manner, the driver seems to fail to detect 3DMark03, and performs the back buffer clear command as instructed.

2. A vertex shader used in game test 2 (P_Pointsprite.vsh) is detected by the driver. In this case the driver uses instructions contained in the driver to determine when to obey the back buffer clear command and when not to. If the back buffer would not be cleared at all in game test 2, the stars in the view of outer space in some cameras would appear smeared as have been reported in the articles mentioned earlier. Back buffer clearing is turned off and on again so that the back buffer is cleared only when the default benchmark cameras show outer space. In free camera mode one can keep the camera outside the spaceship through the entire test, and see how the sky smearing is turned on and off.

3. A vertex shader used in game test 4 (M_HDRsky.vsh) is detected. In this case the driver adds two static clipping planes to reduce the workload. The clipping planes are placed so that the sky is cut out just beyond what is visible in the default camera angles. Again, using the free camera one can look at the sky to see it abruptly cut off. Screenshot of this view was also reported in the ExtremeTech and Beyond3D articles. This cheat was introduced in the 43.51 drivers as far as we know.

4. In game test 4, the water pixel shader (M_Water.psh) is detected. The driver uses this detection to artificially achieve a large performance boost - more than doubling the early frame rate on some systems. In our inspection we noticed a difference in the rendering when compared either to the DirectX reference rasterizer or to those of other hardware. It appears the water shader is being totally discarded and replaced with an alternative more efficient shader implemented in the drivers themselves. The drivers produce a similar looking rendering, but not an identical one.

5. In game test 4 there is detection of a pixel shader (m_HDRSky.psh). Again it appears the shader is being totally discarded and replaced with an alternative more efficient shader in a similar fashion to the water pixel shader above. The rendering looks similar, but it is not identical.

6. A vertex shader (G_MetalCubeLit.vsh) is detected in game test 1. Preventing this detection proved to reduce the frame rate with these drivers, but we have not yet determined the cause. Page 4 of 7

7. A vertex shader in game test 3 (G_PaintBaked.vsh) is detected, and preventing this detection drops the scores with these drivers. This cheat causes the back buffer clearing to be disregarded; we are not yet aware of any other cheats.

8. The vertex and pixel shaders used in the 3DMark03 feature tests are also detected by the driver. When we prevented this detection, the performance dropped by more than a factor of two in the 2.0 pixel shader test.

It's very telling how you, Munky, and Snowman rush in to defend ATI when all I stated was how "both nVidia and ATI had their own driver cheats." (which I cited by 4 sources, including references from FutureMark, themself).

Kind of like when a principal calls in a student who immediatly screams "I didn't do it!" before the principal even says anything... :roll:

(Awaits Josh to bawl "But BFG has a Geforce card! Ha! You can't reprehend him!")

So do you deny that Nvidia cheated more? I will agree with you that any cheating is bad but just like everything there are differing degrees to it. Furthermore the way in which you handle a company after the fact the it is discovered that you did something underhanded says alot about your character. Imagine what would have happend if the enron or worldcom ceo's would have acted better when the news broke about their underhanded deals, or when martha stewart was caught insider trading. The public is more inclined to forgive and forget if you are up front about what went on. If you lie about what you did and then that gets out you are in bigger trouble.

For the most part I just see you descusing the reason someone brought up a fact and not conferming or denying said fact. You don't really need to be pointing fingers unless they are making inaccurate claims.