GPU reviews are all flawed

[MajinCry]

Quite a lot o' sarcasm kicking around 'ere. Come on guys.

I'm in agreement with the OP; the data should be pulled from multiple hour-long playing sessions, rather than a couple half-minute beelines.

In reviews, I've yet to see one which accounts for things such as Windows power settings and settings specific to the driver control panels (e.g, Catalyst's Texture Filtering options).

That shiznizzle is important.

But aye, Railven's got the right idea; think of the current slew of stats as a very rough, probably biased guideline of how a card performs.

 

[boozzer]

When you factor in that reviews are primarily subjective (ie what a reviewer wants to focus on) it would be rather hard to regulate it.

Now, if they turned it into a real scientific study (ie not bloggers/guy_A "lets start a website") perhaps you can get some regulations.

As of now, just buy what you want. Use the reviews to give you an idea of what to expect.

opinions, sure they can be as different as they come.

but not testing methodology or bars or graphs.

a performance avg that include games like project cars? seriously?

 

[nitromullet]

If only all hardware sites come together and formulate a mechanism of standardize testing.

When you factor in that reviews are primarily subjective (ie what a reviewer wants to focus on) it would be rather hard to regulate it.

Now, if they turned it into a real scientific study (ie not bloggers/guy_A "lets start a website") perhaps you can get some regulations.

As of now, just buy what you want. Use the reviews to give you an idea of what to expect.

I don't think a standardized testing methodology is a good idea...

If there was some standardized test NV and AMD would design drivers around it to make their cards look the best in those specific situations.

I actually think a difference in methodologies benefits us more. If they use different methods, but come up with roughly the same relative result, it actually tells us more than if everyone does the same test and gets the same result.

Plus, can you imagine what the forums would be like? You'd have the usual suspects nitpicking every single result they didn't agree with trying to find a violation of some minutia of the standards.

 

[railven]

opinions, sure they can be as different as they come.

but not testing methodology or bars or graphs.

a performance avg that include games like project cars? seriously?

And that's the subjective part. I'm pretty sure no one here cares about WoW performance, but I tell you what, 99% of the time that is the first page I turn to just to get it out of the way.

As someone who doesn't play games like Civilizations, those should be completely thrown out too, as it's also an outlier in favor of AMD.

So where does the reader/consumer's desires start/stop and the reviewers conclusion rest?

Just don't read that site. Simple, done.