View Full Version : Anyone read this ATI interview?
ChrisRay
04-02-05, 09:28 PM
http://3dcenter.org/artikel/2005/03-31_a_english.php
It's an interesting article on how ATI percieves benchmarks. Also doesnt seem to be as "overloaded" with PR as some interviews.
3DCenter: What do you think about the obsession with numbers of parts of the hardware scene with regards to benchmark results of new products? At what level are measurable differences really relevant for you for the evaluation of a new graphic adapter?
Greg Ellis: Performance is just one element, among many, that can be used to differentiate products. Certainly things like driver support, compatibility, price, ease of use, etc. must be considered as well.
As to what constitutes a "relevant" performance gap - there are many factors to consider, most of them subjective.
I think we can agree that very low framerates (e.g. 6 fps vs. 5) fail to discern a realistic difference because both products are obviously inappropriate to the task and settings under test.
The same is true of very high framerates (e.g. 275 fps vs. 240), where neither choice is a bad one - both products are obviously quite good at whatever it is that is being tested.
In the middle ground, subjectivity becomes important. One user might find 50 fps to be infinitely preferable to 40 fps in one particular game, while at the same time noticing no difference whatsoever between 38 fps and 32 in another title. Another user might have a different opinion.
I've noticed a number of reviewers who make a point of including their subjective impressions along with the scores they quote. I think that's a good thing.
At the same time, I would not wish to discourage the community of enthusiasts who spend time and money tuning up their machines to get the best possible number in a particular test. This can be fun and exciting too.
Got to say I entirely agree with this answer.
Yeah, I read it in German and I also have to agree. Mr. Ellis seems to be a cool guy :)
i have to agree also...i think what he said was totally reasonable and very level headed.
jbirney
04-03-05, 12:18 PM
Thanks Chris, always love it when there are few PR filters on the interviews!!!
saturnotaku
04-03-05, 12:37 PM
From a PR perspective, it was an absolutely perfect answer to that question. Very well done whoever coached him - even more props if he wasn't. :)
It looks like he spent some time with the guys at hardocp.
The only problem with benchmarks in reviews is the wording they use for different leads.
Especially when card x leads 33 to 30 frames on card y and they call it overwhelming.
But when card y leads 66 frames to 60 they call it a small victory.
Where do you draw the line between noticeable and useless performance increases? it's still 10% at the end of the day
ChrisRay
04-03-05, 10:51 PM
It looks like he spent some time with the guys at hardocp.
The only problem with benchmarks in reviews is the wording they use for different leads.
Especially when card x leads 33 to 30 frames on card y and they call it overwhelming.
But when card y leads 66 frames to 60 they call it a small victory.
Where do you draw the line between noticeable and useless performance increases? it's still 10% at the end of the day
I myself usually draw the line between playability and unplayability. I think its reasonable to say if one card is getting 250 FPS and the other is getting 230. Both cards are completely playable in that game. However I think its up to the person "viewing" the article in question to determine the level of playability.
When I do my investigations or reviews. I try to cover as many resolutions with Anti Aliasing amounts as possible given some time constraints with fraps. Dependent on the what I am testing will usually determine if I have the time to cut back certain resolutions. I mean for instance. No point in benching an FX 5200 at 1600x1200. But I have always tried to let the people reading decide what was playable/unplayable.
I myself usually draw the line between playability and unplayability. I think its reasonable to say if one card is getting 250 FPS and the other is getting 230. Both cards are completely playable in that game. However I think its up to the person "viewing" the article in question to determine the level of playability.
When I do my investigations or reviews. I try to cover as many resolutions with Anti Aliasing amounts as possible given some time constraints with fraps. Dependent on the what I am testing will usually determine if I have the time to cut back certain resolutions. I mean for instance. No point in benching an FX 5200 at 1600x1200. But I have always tried to let the people reading decide what was playable/unplayable.
Far Cry on my old P4 was a stutterfest, in some points (outdoor) It'd just slow down to a crawl or pause, at 1024x768, indoors however it runs smooth as a .. whatever, still I'm thinking that the game was playable even with the few hickups, sure it sucks when you're firing at something, but you can allways reload a checkpoint.
That raises my point, can you judge a game on two or three one minute benchmarks? is it better to play through a game yourself or with "mates" and then pass on your final verdict?
sure my e-pen1s loves my 3dmark05 scores but I am absolutely sure I was more interested in framerates when my hardware wasn't decent then when it was...
ChrisRay
04-03-05, 11:35 PM
Far Cry on my old P4 was a stutterfest, in some points (outdoor) It'd just slow down to a crawl or pause, at 1024x768, indoors however it runs smooth as a .. whatever, still I'm thinking that the game was playable even with the few hickups, sure it sucks when you're firing at something, but you can allways reload a checkpoint.
That raises my point, can you judge a game on two or three one minute benchmarks? is it better to play through a game yourself or with "mates" and then pass on your final verdict?
sure my e-pen1s loves my 3dmark05 scores but I am absolutely sure I was more interested in framerates when my hardware wasn't decent then when it was...
I myself dont find timedemos very reliable. I'd like to point out Doom 3 for instance was very playable on my rig @ 1600x1200 with 4xAA/16xAF in 95% of the game. But the boss fight in the Hell level. Chugged me into the 20 FPS. And it was "Barely" playable.
I myself try to always benchmark in fraps. But thats just my preferred method of doing game benchmarks. I just like putting out results I can trust. What can be more trustworthy to myself than benches I collected manually from in game? ;) I almost always have fraps running in games these days just so I can look for strange performance issues.
I almost always have fraps running in games these days just so I can look for strange performance issues.
Like the first 2, 3 seconds of a level in War of the Ring?
okay... back on topic..
oldsk00l
04-04-05, 02:29 PM
oh man
What a cool answer! I am really glad to see one of "the boyz" from a graphics card company (let alone ATI) frame this issue and put it back into perspective. It's almost like, breaking it down to "Ford vs Dodge" type thing, and that if hardware is already good at something you're really getting down to superficial preferences and one is not "pwning" the other.
Love it, thanks for finding that!!
Treason
04-04-05, 05:27 PM
Quick! One of you big wigs contact your nVidia insider and tell them to hire this guy away from ATi!
:)
Quick! One of you big wigs contact your nVidia insider and tell them to hire this guy away from ATi!
:)
Hmm, I think ati offers better contracts, they just hired some top PR staff and management of nV and it's biggest distributors in eastern europe..
ChrisRay
04-05-05, 06:14 AM
Hmm, I think ati offers better contracts, they just hired some top PR staff and management of nV and it's biggest distributors in eastern europe..
Nvidia guy strolling through Nvidia Europe Headquarters.
Nvidia guy hears some rustling in the bushes and sees a figure.
Ati Guy says, Hey you.. come over here...
Nvidia guy says, Who me?
Ati Guy says, *Nods* You
Nvidia guys comes over.
Ati Guy says, How would you like to work for ATI?
Nvidia guy says, I dunno..
ATI guy says, Awww come on!
Nvidia guy says, Mmm if you put it that way. *jumps over the bush jumping into the ATI convertable throwing his Nvidia badge behind him*
emailthatguy
04-05-05, 06:49 AM
pretty good interview. i wish the catalyst team seemed so human.
Nvidia guy strolling through Nvidia Europe Headquarters.
Nvidia guy hears some rustling in the bushes and sees a figure.
Ati Guy says, Hey you.. come over here...
Nvidia guy says, Who me?
Ati Guy says, *Nods* You
Nvidia guys comes over.
Ati Guy says, How would you like to work for ATI?
Nvidia guy says, I dunno..
ATI guy says, Awww come on!
Nvidia guy says, Mmm if you put it that way. *jumps over the bush jumping into the ATI convertable throwing his Nvidia badge behind him*
:D :D
So, asking nicely gets you this far in the current business eh?
lol
Where is the time that you had to sign contracts that you wouldn't work with a competitor etc..
vBulletin® v3.7.1, Copyright ©2000-2012, Jelsoft Enterprises Ltd.