it still doesn't make sense. You can't say that two cards get more or less the same framerates, and then follow that up with "but at different resolutions". That's a completely useless statement.
I did suggest you read the
review, and not just that one page, but it's lengthy and I can see where you're missing my point..
The review at HardOCP attempts to discover the best image quality settings (resolution+AA+AF) that can be used in a given game, while still allowing you to max-out all its visual settings and achieve playable framerates (they use 30fps as a minimum desireable framerate, and consider anything 60+ as essentially gravy). Oddly enough, that's exactly what I attempt to discover when I sit down with a new game, so I find that approach extremely helpful. My initial comment about "more or less the same framerates" was a poorly worded and very loose interpretation of the wiggly-line graphs which show exactly how well each card performed this task -- the lines tend to overlap more often than not = "more or less the same framerates". OR, the very slight variation in average FPS scores shown = "more or less the same framerates".
That said, take a look at the table on this page of the review, which shows a breakdown of all the tests run in this manner, and the IQ settings used on each card:
http://www.hardocp.com/article.html?art=NjExLDEwYou will see that all four cards tested are capable of running those games at playable framerates in high resolutions, and with varying degrees of AF and AA turned on. You will also see that the two new ATI cards were able to use higher-quality settings than the 6800. In the case of the X800Pro, the diffence is minor, but X800XT vs. 6800U (flagship to flagship), there is a considerable difference. Hope that helps clear up what I meant there..
Unless you're trying to gloss over the misgivings of something, it's generally best, when comparing two things, to compare them on as close to equal footing as you can. It tends to be more useful for generating a valid comparison between things. Unless of course, one is trying to hide something.
Again, I can see where you're coming from there.. people have all kinds of opinions about HardOCP and their methods, and that is - of course - the way it should be. As I said, I tend to find it usefull to look at how well a card can supply a combination of playable framerates+max detail+best IQ (and if HardOCP is part of some sort of hidden-agenda black-ops video card propaganda machine, then I remain grateful that they aren't the only source of info on the planet).
For an equal-footing comparison, you can skip to
this page (but
read the part about "Apples to Apples"):
http://www.hardocp.com/article.html?art=NjExLDg=I agree with the author that the other method is far more usefull
to me. Even still, out of 6 tests the 6800 takes top position in only 2, and by a very small amount.
Anyway, performance be damned -- they're all fucking insane!! What's far more interesting to me (and only barely, becsuse I ain't buying
any of this tech) is that the 6800 is still producing shittier rendering on some of the shaders in Far Cry than the 97/9800's!!
----
The dactylic pentameter consists of too [sic] parts separated by a diaeresis. Each part consists of to [sic] dactyls and a long syllable. The spondee may take the place of the dactyl in the first part, but not in tEh [sic] second... -Harkness.
This comment was edited on May 5, 15:49.
-----
I'm not even angry. I'm being so sincere right now, even though you broke my heart and killed me.