hardocp 2900xt xfire review

J.D · 10 Jul 2007 at 14:00

Tom|Nbk said:
Whether they compare with the exact same settings or not, those XFIRE scores are way to low there. The rest of the review may as well be voided to me.

Agree'd.

If it was a fair review then this thread wouldn't be here :rolleyes:

. Looks like some people know what their talking about (The ones with sense) and the others can't see anything wrong with the test even though they pretend to be "in the loop" or so they say. :rolleyes:

Read it, look at REAL test sites with no gay fanboy problems. You "in the know" people might learn something. If a test is flawed then it's flawed. People who defend their purchase are laughable.

Tom|Nbk · 10 Jul 2007 at 15:26

skirkham said:
a manual run-through riding horseback from outside the Imperial City to Chorrol to Bruma ?

skirkham said:
a web site run some benchmarks and all you can say is their biased

if you want to dispute what they have posted used the same system specs they do and benchmark the same way they do

i.e Oblivion testing For testing we have chosen to do a manual run-through riding horseback from outside the Imperial City to Chorrol to Bruma

doing anything else just makes you look like a fanboy

you might also want to look at the article again if you think the gts beat the 2900xt crossfire in Oblivion. You havent grasped that hardocp benchmark to find the best playable experience for graphics cards. And if you dont like this way just look at the apples for apples part

for me hardocp are the first place i go to for graphics card reviews as they are far more realistic as to what you are going to get from your graphics card than running 3dmark

Just done it for you exactly how HardOCP did (a manual run through riding horseback from outside the Imperial City to Chorrol to Bruma).

1680x1050
8xAA (HardOCP used 2xAA)
16x HQAF

Exact same in game settings as HardOCP:

Min Max Avg
30 77 46.925

Proofs in the pudding :rolleyes:

Cyber-Mav · 10 Jul 2007 at 15:30

Tom|Nbk said:
Just done it for you exactly how HardOCP did (a manual run through riding horseback from outside the Imperial City to Chorrol to Bruma).

1680x1050
8xAA (HardOCP used 2xAA)
16x HQAF

Exact same in game settings as HardOCP:

Min Max Avg
30 77 46.925

Proofs in the pudding

had to say, personal scores will always differ from web reviews since thier testing is more extreme. also did you use the same drivers as in the review or some beta drivers?

if you have a problem with hardocp review contact them and confront them. id like to know the truth on their benchmark so ill back you up in asking them for some kind of statement or explanation on this.

Tom|Nbk · 10 Jul 2007 at 15:34

Cyber-Mav said:
had to say, personal scores will always differ from web reviews since thier testing is more extreme. also did you use the same drivers as in the review or some beta drivers?

if you have a problem with hardocp review contact them and confront them. id like to know the truth on their benchmark so ill back you up in asking them for some kind of statement or explanation on this.

Used 7.6's. CBA contacting HardOCP they wouldn't listen never do be a waste of my time. My testing was extreme as theirs Mav I did EVERYTHING they did.

Cyber-Mav · 10 Jul 2007 at 15:45

dunno then, but from other reviews iv seen on the web hardocp review seems to be spot on.

this always happens though, if someone puts up a benchmark and they don;t like it then they just instantly play the biased review card.
can;t see what hardocp or any review site has to gain from putting out false results, the fraps results speak for themselves so can;t really argue with them. most other reviews just show benchmark bars with peak fps results and never take the minimum or average into account. fraps is the best way to do it.

crossfire could have had issues on thier setup, check the mobo or the cpu and system ram they have, is there any know issues with the combination of hardware they use?

KangooVanMan · 10 Jul 2007 at 15:45

Why don't you just post a comment, on the comments page for that review? See what reaction you get.

jaykay · 10 Jul 2007 at 15:47

They were using windows vista which does lower the fps a bit compared to XP.

Tom|Nbk · 10 Jul 2007 at 15:47

Cyber-Mav said:
this always happens though, if someone puts up a benchmark and they don;t like it then they just instantly play the biased review card.

No it dosn't nearly we all know about HardOCP, I doubt they actually have any bias at all I just think there way of testing and the testing it self is most likely utter ***** I would never base a card decision on one of their reviews.

Cyber-Mav · 10 Jul 2007 at 15:53

jaykay said:
They were using windows vista which does lower the fps a bit compared to XP.

they are using XP

Cyber-Mav · 10 Jul 2007 at 15:54

Tom|Nbk said:
No it dosn't nearly we all know about HardOCP, I doubt they actually have any bias at all I just think there way of testing and the testing it self is most likely utter ***** I would never base a card decision on one of their reviews.

whats wrong with thier testing methodology? the apples to apples section is totally fair, puts each card through the same settings, what more can you ask for?

Tom|Nbk · 10 Jul 2007 at 15:57

Looks to me like they stuffed the review up royally

Squink · 10 Jul 2007 at 16:01

Cyber-Mav said:
the apples to apples section is totally fair, puts each card through the same settings

The apples to apples is invariably an afterthought pegged to the end of the review and rarely offers much scope. I think you either buy into their subjective methodology or you don't. Lots of people do.

Cyber-Mav · 10 Jul 2007 at 16:03

Tom|Nbk said:
Looks to me like they stuffed the review up royally

but how? explain....

all i keep hearing is the review is borked because its, a hardocp review, or they are biased, or its the way they test the cards,

give a more detailed explanation on this other than vague one liners. right now it just seems that 2900 owners are clutching at straws here in a desperate attempt to defend thier purchase.

iv looked over thier testing method and its fine, in the apples to apples test all cards go through the same settings and whatever the results are they put them in the table with graph. can;t see any issues with that.

Cyber-Mav · 10 Jul 2007 at 16:06

mulpsmebeauty said:
The apples to apples is invariably an afterthought pegged to the end of the review and rarely offers much scope. I think you either buy into their subjective methodology or you don't. Lots of people do.

best part of thier reviews is they alter the settings in game to whats best for the card. thats what i like most of the review, i can then see that if 2 cards give same sort of fps 1 card may be able to have higher details enabled in the game to do the similar sort of fps as its competitor.

one thing i would like is if they showd what sort of settings can be enabled on each card at different resolutions, e.g 1024x768, 1280x1024, 1600x1200 etc.

would be very interesting.

fornowagain · 10 Jul 2007 at 16:23

KangooVanMan said:
Why don't you just post a comment, on the comments page for that review? See what reaction you get.

Now that's a good idea.

Kyle and Brent are both in there,

http://www.hardforum.com/showthread.php?t=1206447

Tom|Nbk said:
My testing was extreme as theirs Mav I did EVERYTHING they did.

Not quite, you used a lower resolution for a start. Is yours a stock game install, no mods? Same cpu speed? There's a lot of variables.

Tom|Nbk · 10 Jul 2007 at 16:41

fornowagain said:
Now that's a good idea.

Kyle and Brent are both in there,

http://www.hardforum.com/showthread.php?t=1206447

Not quite, you used a lower resolution for a start. Is yours a stock game install, no mods? Same cpu speed? There's a lot of variables.

Totally stock, the resolution is debateable theres barely any difference mate, not enought to affect it as bad as it was in the review anyway.

Schnippzle · 10 Jul 2007 at 16:56

Too bad Hard OCP went the way they did regarding gfx reviews. Always held them pretty high when it came to reviewing hardware....guess this had proven them wrong.

MystaEB · 10 Jul 2007 at 17:00

The results could very well be accurate, unless you run the EXACT same benchmark test they did, with the same hardware, there is no way to say they are incorrect. There are so many factors that affect results, test setup, test settings, resolution, game benchmark run.

Oblivion is a particularly poor example to compare results between tests, one user may run a benchmark run through and another may run what he thinks is the same benchmark run but tiny little things like time of day affect framerate in Oblivion. For all we know there could have been more characters on screen, or more monsters on screen or whatever.

I'm not siding with anyone here as I personally take all benchmark results with a pinch of salt anyway, as most users should. What I find odd however is how whenever a benchmark is released that differs slighty from a users own results, they often go nuts and start claiming ********. This especially happens when the review is nagative. I blame human attachment to be honest, for instance, if someone calls your girlfriend ugly you immediately smack them.

It's also important to remember that users strive for high framerates, and will do everything to get them. Many reviewers don't do this - they take a stock system and use stock configuration to stress a system to produce results.

I had a point when I started this post, I'm not sure if I've covered it but I tried, heh.

knowledge123 · 10 Jul 2007 at 17:26

tbh, at the end of the day, provided the person who brought the card is happy with the card, what does it really matter?
I bought my 2900xt not long after launch and I am 100% happy with it, ATI do an excellent job with the drivers and while it may not be the fastest card on the planet, I don't really care because it gives me the performance I need.
In regards to this review no one site can be the start and end all of all benchmarks, as there are too many variables that can change, best to have a set of at least 3 reputable websites and correlate the results between them all.

queamin · 10 Jul 2007 at 17:59

To be honest i like the way they test as i think it better than canned benchmarks yes they aren't perfect most of us aren't.
Tom if you got a problem with the results post there and see what they say.
Added you never buy anything because of 1 or 2 reviews, i normaly wait until people have brought it and find out what they think.