Well either HUBs result are wrong or you and I have have a different definition of fine
.
It’s the former
well, they’re not “wrong” per se, in that they are accurate measurements. BUT they’ve updated their response time testing methodology to something that is, in my opinion, unnecessarily aggressive and no longer representative of real world results. They now capture 3-97% of the response time curve for each measurement , contrary to 20+ years of what the term “response time” represents in the monitor and display market (its always been 10-90%).
That’s why all their results and measurements are now much slower than everywhere else, and much slower than all the results over that time span, so it’s very hard to fairly and accurately compare them.
in my opinion there is no benefit in capturing that much of the curve, it doesn’t translate back in to real world perceived performance any better. it also has significant issues with darker tone measurements which exaggerates numbers further. That’s their choice of course, but it has led to some questionable conclusions in recent reviews.
the Corsair is a good example. They suggest the “fastest” mode at 165Hz which maybe looks ok on paper in their data with that measurement method, but in real usage that’s not practical to use as overshoot is pretty visible. And that overshoot is very bad as soon as refresh rate / frame rate in VRR lowers. In reality the “fast” mode is far better and can be used at all refresh rates comfortably.
if you look at the measurements at TFTcentral (I appreciate you might not have early access yet) and Pcmonitors too you can see response time data that is far more relatable to real performance and makes more sense. Suddenly those figures don’t look bad
That’s using an updated gamma corrected method which provides improved accuracy, but provides figures that actually fit real world perceived motion clarity. That’s then backed by pursuit camera photos from both sites showing that the motion clarity is very good in that mode. Trust me I’ve used it and have the screen
this is the problem with their new method. “Response times” being reported are not representative any more of what “response time” has always meant in the market, or what really translates in to real world usage.
more info on this whole situation, updated gamma corrected measurements and consideration of the 3-97% method in this article if you’re interested
https://tftcentral.co.uk/articles/response_time_testing
