ATI Marketing

[Quitch]

There's an interview with Adam Kozak from ATI marketing over at Rage3D.gr

http://www.rage3d.gr/board/showthread.php?s=&threadid=247

While the depth of the interview leaves a lot to be desired, the answers go to show that, while quiet, ATI haven't been asleep. It does seem that 3dMark might be in trouble though, since not only are nVidia attacking them on one side, but the actions of nVidia towards 3DMark appear to be influencing ATI.

Being on top also means you have to continue to be one step ahead of your competitor. In this respect, it has also forced us to be more creative in thwarting our competitors expected moves. This has been demonstrated in canned benchmarks, especially within the last year. We've been trying to shift the attention away from these benchmarks towards ones of real world experiences that customers are most likely to experience. Since May, we've been playing every game that has come out and have created repeatable benchmarks using FRAPS to help demonstrate the lead our products enjoy. Getting this information out into the public as well as deflecting our competitor's natural resistance to this method has been important as of late.

The wording would suggest though that he isn't drawing a line between synthetic and game benchmarks, but rather between ones created by the developer, and custom ones. An interesting divide, and something that shows they really have been paying attention over the course of the year.

 

[Dave Baumann]

ATI has long since been telling people to take the top ten best selling games at any one month (of which, invariably won't be being benchmarked) and compare them.

 

[Quitch]

I guess you're right there, I remember that reviewers guide of their's.

 

[cthellis42]

Certainly it'll get a lot less goofy when games figure they should really all be bundling the ability to record personal demos for FRAPSing (and provide their own analysis tools that help reviewers and the public see what they feel is important.

 

[Mark]

You don't really need recordable demos either (though it makes it a hell of a lot easier). By doing multiple runs through a game sequence, then averageing the numbers reported by FRAPs, you can get fairly accurate results as well.

 

[digitalwanderer]

You don't really need recordable demos either (though it makes it a hell of a lot easier). By doing multiple runs through a game sequence, then averageing the numbers reported by FRAPs, you can get fairly accurate results as well.

True, but it is a pain in the ass to graph the results of it!

 

[Geo]

". . .created repeatable benchmarks using FRAPS to help demonstrate the lead our products enjoy. Getting this information out into the public. . ."

What exactly does this mean, and how are they trying to do the latter? Are they sending these "repeatable benchmarks" out to website editors? Anyone seen any evidence of this making it into reviews anywhere?

 

[Hellbinder]

". . .created repeatable benchmarks using FRAPS to help demonstrate the lead our products enjoy. Getting this information out into the public. . ."

What exactly does this mean, and how are they trying to do the latter? Are they sending these "repeatable benchmarks" out to website editors? Anyone seen any evidence of this making it into reviews anywhere?

Thats exactly what it means... They send "packages" and other things to web editors and magazines pointing out the Results, asking them to verify for themselves, and Urging them to use MODERN Games to test with in this manner.

The bottom line is that it is hard to compete when Nvidia Flatly chooses to and DOES Falsly infate or Outrigth CHEAT in virtually every canned benchmark and demo level known to man.

And they openly intend to keep doing it.

 

[Geo]

They send "packages" and other things to web editors and magazines pointing out the Results, asking them to verify for themselves, and Urging them to use MODERN Games to test with in this manner.

Okay, but has anyone seen this stuff end up in a review somewhere? If not, why hasn't it?

 

[cthellis42]

Probably not really an important thing to mention, unless one is writing an essay on benchmarking itself, and the actions and reactions of the big IHV's over the past year. They're just offering suggestions--and probably the same ones I myself would offer. Hehe... Not much to write home about.

 

[jb]

They send "packages" and other things to web editors and magazines pointing out the Results, asking them to verify for themselves, and Urging them to use MODERN Games to test with in this manner.

Okay, but has anyone seen this stuff end up in a review somewhere? If not, why hasn't it?

FS posted something about this in one for the FX reviews. They noticed that when they ran the stock timedemos and compared them to custome time demos the results were...interesting

 

[Hanners]

Okay, but has anyone seen this stuff end up in a review somewhere? If not, why hasn't it?

If you take a look at the way [H] test video cards now, it seems to be largely FRAPS-based testing. Anandtech do it a fair bit now too. I'm sure there are others, but those are the main two that spring to mind.

 

[parhelia]

Being on top also means you have to continue to be one step ahead of your competitor. In this respect, it has also forced us to be more creative in thwarting our competitors expected moves. This has been demonstrated in canned benchmarks, especially within the last year. We've been trying to shift the attention away from these benchmarks towards ones of real world experiences that customers are most likely to experience. Since May, we've been playing every game that has come out and have created repeatable benchmarks using FRAPS to help demonstrate the lead our products enjoy. Getting this information out into the public as well as deflecting our competitor's natural resistance to this method has been important as of late.

I'm surprised why ATI feels the need to test each and every game which comes out with FRAPS, but well, if that gives them moral satisfaction like they seem to indicate, why not.
I'm not saying that it's bad or anything, I'm merely surprised

 

[digitalwanderer]

I'm surprised why ATI feels the need to test each and every game which comes out with FRAPS, but well, if that gives them moral satisfaction like they seem to indicate, why not.
I'm not saying that it's bad or anything, I'm merely surprised

Why? It seems like a good idea to me to compare how all the games run on your hardware, and fraps is the best way I've found to consistantly measure performance.

 

[John Reynolds]

I'm surprised why ATI feels the need to test each and every game which comes out with FRAPS, but well, if that gives them moral satisfaction like they seem to indicate, why not.
I'm not saying that it's bad or anything, I'm merely surprised

Because a lot of games now being released were developed on NV hardware, and if there's going to be a solvable problem with ATI hardware the company obviously needs to know about it ASAP so they can work on a fix (CoD, KotOR, etc.).

 

[xGL]

Good point.