Benchmarking the Benchmarks

[leon4112]

"It is time to put your money where your mouth is, or maybe where your keyboard is. HardOCP sets out to prove that real world video card testing is where it’s at. Beware, we may make you feel dirty every time your run a benchmark from here out!"

Discuss

http://enthusiast.hardocp.com/article.html?art=MTQ2MSwxLCxoZW50aHVzaWFzdA==

 

[schumi84]

I say good on them. Yes synthetic benchmarks may gave a rough indication of one gfx cards performance compared to another, but they say nothing as to how the cards will perform in real world situations, which i, as a consumer, would be much more interested in.

Synthetic benchmarks are good fun, and nice to compare scores for e-peen, but at the end of the day if it isnt going to run the games that you want to play at the resolution you want to play them at then you are spending 100-300 quid on something to play benhcmarks.

 

[Tute]

I'd still take Anandtech over [H] as I prefer their testing. The way [H] does it - one card against another, really unclear graphs, etc. is just daft.

And what was the point of the article? The whole thing seems to be one big stab at Anandtech.

 

[Shocky-FM]

I'd like to see anandtech do something simalar to back up there testing methods, but it would probably backfire.

 

[Chronictank]

I say good on them. Yes synthetic benchmarks may gave a rough indication of one gfx cards performance compared to another, but they say nothing as to how the cards will perform in real world situations, which i, as a consumer, would be much more interested in.

Synthetic benchmarks are good fun, and nice to compare scores for e-peen, but at the end of the day if it isnt going to run the games that you want to play at the resolution you want to play them at then you are spending 100-300 quid on something to play benhcmarks.

Synthetic Benchmars = comaprable results
Hard OCP benchmarks cannot be used to compare cards, ever
because the tests by default are not the same
all it does is show the performance of one card, in one instance of a game
Thats why every comparison on that website is worthless

for years now, and there is no doubt that we have taken a lot of heat from various sources. Most of it is generated through ignorance and the human quality of being resistant to change.

That did make me laugh though, the irony

 

[Squink]

[H]ardOCP used to be my premier hardware site and forum. I don't touch them now, since Kyle's hubris is absolutely monumental. "Yeah guys we look after our own [while taking vast advertising revenues from BFG, eVGA etc and taking random childish pops at our competitors]".

Subjective settings as to what's the 'best' a card can run don't help me much. I'll choose how much AA I want, what fps I like and what IQ settings I'll use thank you very much. Still, it's the blinkered consumer who doesn't read as many reviews as possible on a product he plans to buy.

 

[queamin]

Synthetic Benchmars = comaprable results
Hard OCP benchmarks cannot be used to compare cards, ever
because the tests by default are not the same
all it does is show the performance of one card, in one instance of a game
Thats why every comparison on that website is worthless


That did make me laugh though, the irony

At least they playing the game/s with the cards and not just using in game/canned benchmarks you might aswell run 3dmark 06 and pick a card on the score, i get no where near the frames in crysis as in the ingame benchmark and same with in game benchmark in Fear and IF Anandtech use the intro as a benchmark how can the relate to the game?.
We had this before where both ati and nvidia have been court fixing benchmarks for certain big games.

 

[queamin]

Sites to me are a very rough guild i wait until people here get a card to see what they think as they play the games soon you get a picture of what the card can do.

 

[LoadsaMoney]

i get no where near the frames in crysis as in the ingame benchmark

Well you won't do, as thats not how you play Crysis is it.

 

[Chronictank]

At least they playing the game/s with the cards and not just using in game/canned benchmarks you might aswell run 3dmark 06 and pick a card on the score, i get no where near the frames in crysis as in the ingame benchmark and same with in game benchmark in Fear and IF Anandtech use the intro as a benchmark how can the relate to the game?.
We had this before where both ati and nvidia have been court fixing benchmarks for certain big games.

At least read my post if you are going to quote me,
as a comparison their review is not worth the internetspace it is taking up,
as a idea of what a card in its own right is going to perform the test is fine
(although [H] seem to almost completely at random change settings to get the result they want)

edit: came across a bit harsh sorry