Very peculiar response times in Cooler Master GM27-FQS

[EsaT]

I think this could use own thread.
There's really something new/unusual going in response times:

Cooler Master GM27-FQS ARGB review - TFTCentral

The latest 27" gaming screen from Cooler Master with a 27" 1440p Fast IPS panel, 165Hz refresh rate. FreeSync Premium and USB-C connectivity

tftcentral.co.uk

Response times and overshoots being pretty much constant at different refresh rates just don't fit into variable overdrive.
That should show as clear slowing down of response time from decrease in overdrive pulse strength when refresh rate decreases.
Like shown in Asus PG32UQX review.

So I wonder if overdrive has been "decoupled" from frame refresh and overdrive is applied for say 5ms duration...
Followed by pixel's control signal value being changed to correspond to what input signal asked for rest of the frame's duration.
Applying fixed level of overdrive for fixed duration would result in such constant behaviour.
And would solve all too familiar "overdrive good for low Hz is hopelessly slow for fast Hz/overdrive good for high Hz is unusable overshooting at low Hz" problem at once.

Certainly additional processing power requirements for panel controller shouldn't be slightest issue with today's IC manufacturing.
And with 300+ Hz LCD monitors I guess communication speed/bandwidth need between controller and LCD matrix is solvable.

Scope shots would have told far more about this peculiar behaviour.

 

[Baddass]

Not been deliberately ignoring your PM mate, just on holiday at the moment and trying to have a break (clearly I’m now failing! Lol)

Will do some further investigation on this when I’m back and have chance

 

[EsaT]

Let's hope AUO has designed such "decoupled" overdrive.
Because who knows how many more years we'll be stuck on LCD before self emissive pixel techs are mature.

Besides solving bad/non-optimal overdrive for the fps problem, it would also improve experience at lower fps:
Response times there are limited by need to keep overdrive signal strength low.

It would be also simpler than variable overdrive in sense of one LUT being enough for whole refresh rate range and having no need to guess frame time of starting frame.


In case someone needs primer on how overdrive, or really Response Time Compensation, works, here's good document made by National Semiconductor:

https://www.ti.com/lit/an/snla185/snla185.pdf

just on holiday at the moment and trying to have a break (clearly I’m now failing! Lol)

It's not holiday if you aren't working on something in most of the days...

Myself keeping last week of "summer" holiday now, so with pause today it's likely back to using 8½ ton "garden hoe" tomorrow.
Might otherwise even have more than partial day to take photos of autumn's colours.