![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
I still am not sure why anyone would buy an Asus videocard. they are always overpriced
Let me be the first one to say here to [H]ard|OCP, I have been around for a little while and I've seen the benchmark of the HD2900XT 512Megs but never the HD2900XT 1GB (Unless I've failed the search function) and I will personally be willing to lend it to [H]ard|OCP for some good testing on this card against the higher end cards such as the GTX and the likes. I would also point out to AtkinS that I do whole heartly agree with you with what you had to say about the card themselves.DEAR hardocp, im a huge user of ati graphics cards, cause, very easy, gives me more performance for the money in norway, mb not in us though, after watching prices etc.
Okey well, i can say overclocking a 3870 doesnt give any benefit at all(almost i say nothing) my benches on my comp was up 1 FPS WOOHO, while my 2900 XT was pure ownage compared, hitting well over 14k's 3dmark while 3870's overclocked max does, guess what? 12k, i got sucky cpu so take a rethought on the score a amd X2
well, what i wanna say, take 3870 and take a 2900 XT, and simply, clock them
I think its strange that the 2900 XT gave me much improvement while the 3870 doesnt give all to much diffrence in performance i dont know why, but i wanna know why, i have sold all my 3870's except one i gave to my gf and i have sold my 2900 XT's aswell, which i ran at 1100 mhz core vs org 743.
Facts and specs about 2900 XT vs 3870, the 3870 is a faster but still slower 2900 XT, 512 bit was decreased to 256 bit, lost 30 gb/sec bandwidth there, core got up from 740 mhz to 777 mhz to compensate, and the powerconsumtion got lower, but the clocking ability went down aswell, making the 2900 XT still the bad ass gphx cards to clock, and i can asure you guys, they clock well, 1100//1200, i dont see many with 1200 mhz memory, but i ran it, it gave me improvement over 1000 which i see lots of people run vs 828.
This is something I would like to see as well. On my own note: I know [H]ard|OCP does an interesting way of doing benchmark, I've nothing against the highest playable settings. I'm sure some of us probably would like to see an apple to apple test (uncanned benchmark) so some may accurately compare one against one other on the exactly same settings. Just a thought.Kyle, interesting question you ask at the end of the article.
I wonder if your time could have been better utilized doing a different kind of review. WAIT, I am not talking about benchmarking approach here
I would really love to see you guys do a comprehensive review of the features of some "key" GPUs as far as features other than raw performance of games are concerned.
I have searched far and wide, and I have yet to find such a review.
Things I would love to see reviewed\discussed:
- Image quality (games\video\pictures)
- Multimonitor support (multi head support for games and others), performance\compatibility issues.
- Resolution\monitor\TV support and orientation (landscape\portrait), monitor scaling quality
- Driver features (OC\Utilities\Diagnostics\OS support)
- Other functionality (Physics\Computation, etc...)
- Warranties
- Support History
Of course, those are just some thoughts, in addition to the great stuff you usually include (performance\power usage\temps).
I guess, in essense I am looking for a kind of state of Video Card technology besides the usual key metrics.
These expensive GPUs we have are remarkable pieces of technology, let's see how they compare.
I hope you guys can do something like this, because I am sure it will required a bit of subjective comparison, and [H] is a site with the cred to do that.
What value would that serve? The point of their testing shows whether or not the card has the ability to make the next step.
This would be no different than testing breaks on a car where the test is to stop before running off a cliff. If one car fails by 10% and the next better car fails by 5%, they both fail.
And please...the 3870 doesn't run close to the 8800GT.
If you don't know what AA and AF do at certain settings, or have a good idea, you have no business buying that expensive of a card. PLUS< they SHOW you in the IQ samples, below the article.It does in HardOcp testing
How am I supposed to know how much impact those two or three changes in IQ settings have ?
For me it can just as well be set randomly on the whim of testers.
Yeah ... 15% overclock for 3/5 fps .. sucks