HardOCP's idea of playable is laughable at best.
I wouldn't say it's that bad. But the bottom line with their review method is - you have to trust the reviewers. And they introduced that method because of rampant driver cheatings from graphics chip vendors. So they're basically saying "Trust us instead of trusting nVIDIA/ATi"
Considering their GPU reviews are pretty much done by 2 persons (Kyle and Brent), I could hardly say their test results/suggestions are reliable - 1. Accuracy (dozens of tests done by 2 people?) 2. Bias (I read their personal preferances between the lines, all the time) 3. Questionable results and comparisons (I think this issue is closely related to the issue 1 and issue 2, their numbers are often "off" and the numbers they acutally are "selecting and showing" are, to say the best, not self-proving - if not "biased")
On top of that, their attitude is like "You don't like the way we do things, then go away" Well, that actually made me keep some distance from them. Don't get me wrong, HardOCP is still one of my favorites and where else could I possibly get a hands-on reviews on Doom 3 like that? But certainly not a good attitude on their side. I would rather aggressively tackle down any issues raised up against their reviews, instead of simply saying "get lost."
All in all, however, if the method becomes more sophisticated and involving more trust-worthy (knowledgeable, less-biased) people, I'd say they have a good potential to grow in graphics reviews area. More than I have a reservation on their method, I totally ignore a test like 3DMark or some other artificial-looking numbers thrown on a graph.
lop