Blur Busters Forums

Hello,

talking about displays with G-Sync module (i.e. with variable overdrive).

How much in general is the impact on blur, ghosting and latency when
a 240 Hz display only receives 144 fps? Could it perform WORSE than a
similar quality 144 Hz display?

I know there are many factors involved but maybe you understand my point.

Imagine you consider to replace your 144 Hz monitor with a 240 Hz monitor
since that will clearly give you some benefit in a game that you can run at
240 fps. But then you realize that you also want to play another game which
you know only runs at 144 fps. You clearly don't want a regression in terms of
blur, ghosting and latency with the new 240 Hz display when running at lower
144 fps.

Thanks

lemmeblur wrote: ↑

08 Sep 2021, 07:23

talking about displays with G-Sync module (i.e. with variable overdrive).

How much in general is the impact on blur, ghosting and latency when
a 240 Hz display only receives 144 fps? Could it perform WORSE than a
similar quality 144 Hz display?

Variable overdrive on a display containing a G-SYNC module should ensure that the selected overdrive preset looks consistent across the VRR range, regardless of framerate.

That said, a higher native refresh rate does not necessarily equal better overdrive performance, which can depend on the given panel's native GtG characteristics, as well as the individual monitor model's overdrive tuning implemented by the manufacturer, G-SYNC module or no. I.E. 144 FPS on a native 144Hz monitor, 144 FPS VRR at 240Hz on a native 240Hz monitor, and 144 FPS on a native 240Hz monitor set to a physical refresh rate of 144Hz could all have differing overdrive performance. There's no guarantees.

You'll need to read reviews (rtings, TFT Central, etc) between the models you're considering, along with owner feedback (where possible) to get a baseline comparison where GtG and overdrive performance is concerned.

Thanks for the reply!

That's what I find quite difficult because many tests/reviews for 240 Hz monitors seem
to test only at 240 Hz (+60Hz and 120Hz in some cases) but not at "<= 144 FPS VRR".
At least I assume when they show numbers for 240 Hz then they tested with 240 FPS "input".

So I feel unable to actually make a decision. In my particular case I am thinking about
going from Asus ROG Strix XG279Q (144Hz) to Asus ROG Gaming PG279QM (240Hz).

Found a very good review for the PG279QM on "Hardware Unboxed" and on prad.
Also bought the XG279Q some years ago based on a prad review.
Still with all the reviews I don't know if the PG279QM will perform at least as
good as the XG279Q at around 144 FPS VRR.

Is there some easy to use comparison table somewhere that makes it easier to
see how 240Hz displays actually perform when fed with only 144 fps?

Thanks again for help!

lemmeblur wrote: ↑

08 Sep 2021, 10:41

That's what I find quite difficult because many tests/reviews for 240 Hz monitors seem
to test only at 240 Hz (+60Hz and 120Hz in some cases) but not at "<= 144 FPS VRR".

Again, with variable overdrive on native G-SYNC monitors, anything between 0 and 240 FPS at 240Hz with VRR enabled should maintain the same overdrive appearance the panel does at 240Hz non-VRR, so there's no need for them to test it across the VRR range in that case.

You only have to worry about this scenario with FreeSync/G-SYNC Compatible monitors, as 99% of them don't have any form of variable overdrive, and instead use the static overdrive parameters of the max refresh rate at all times, regardless of current VRR refresh/framerate. This creates more overdrive artifacts because different refresh/framerates require different overdrive intensities to look the same during VRR operation.

Alright this is why I started with "talking about displays with G-Sync module (i.e. with variable overdrive)".
I just wasn't sure if variable overdrive avoids all kinds of possible regressions. If it does then everything should
be perfect You know I was a bit confused by one thing that the guy from Hardware Unboxed said in his review
at time 6:06.

He said:
"[...] delivers great performance at all refresh rates, around 200 to 165 Hz I do think the overshoot level is get
a bit high but below that variable overdrive kicks in to balance out performance all the way down to the lower
refresh rates where you still get a 5 to 6 ms experience.".

I am not 100% sure what he means. To me it sounds that with 200 to 165 fps VRR (though he says Hz not fps)
the overshoot was not perfect and that variable overdrive only kicked in below that range?
That confused me or at least raised some doubts that we can fully trust variable overdrive when
there are ranges where it does not "kick in".

lemmeblur wrote: ↑

08 Sep 2021, 16:12

I am not 100% sure what he means. To me it sounds that with 200 to 165 fps VRR (though he says Hz not fps)

I believe he's referring to the overdrive performance when the monitor is forced to a physical 165 or 200Hz refresh rate, not when you're playing at 165 or 200 FPS at the native physical 240Hz refresh rate with VRR enabled.

Overdrive response on the same monitor can change (sometimes even degrade) if you force it to a physical refresh rate below native, as overdrive presets are usually most optimally tuned for the max refresh rate, since beyond 60 or 120Hz console use, manufacturers usually don't see a reason why users would force their physical refresh rate below maximum, especially with G-SYNC available.

lemmeblur wrote: ↑

08 Sep 2021, 07:23

How much in general is the impact on blur, ghosting and latency when
a 240 Hz display only receives 144 fps? Could it perform WORSE than a similar quality 144 Hz display?

Ghosting: Varies. Ideally it's identical (only persistence blur)
Latency: Almost always less

Even 144fps frames gets displayed in 1/240sec at 240Hz, so latency is almost always less, though latency is a very large chain that has both absolute latencies (tapedelay latencies) as well as scanout latencies (not all pixels refresh at same time -- www.blurbusters.com/scanout ...) etc -- latency is not a single number.

You can get better or worse performance when comparing a good 144Hz monitor with a bad 240Hz monitor. But a very good 240Hz monitor will always blow away 144Hz. Sometimes cherrypicking is needed to make sure you don't have any downgrades.

jorimt wrote: ↑

08 Sep 2021, 16:59

I believe he's referring to the overdrive performance when the monitor is forced to a physical 165 or 200Hz refresh rate, not when you're playing at 165 or 200 FPS at the native physical 240Hz refresh rate with VRR enabled.

So you mean he probably set the display to (for example) 200Hz, but since no variable overdrive was active at 200fps the overshoot was a bit high but got better below that 200fps when variable overdrive kicked in? So this observation should not concern anyone who is going to run the display at 240Hz, right.

Chief Blur Buster wrote: ↑

08 Sep 2021, 17:33

lemmeblur wrote: ↑

08 Sep 2021, 07:23

How much in general is the impact on blur, ghosting and latency when
a 240 Hz display only receives 144 fps? Could it perform WORSE than a similar quality 144 Hz display?

Ghosting: Varies. Ideally it's identical (only persistence blur)
Latency: Almost always less

Even 144fps frames gets displayed in 1/240sec at 240Hz, so latency is almost always less, though latency is a very large chain that has both absolute latencies (tapedelay latencies) as well as scanout latencies (not all pixels refresh at same time -- www.blurbusters.com/scanout ...) etc -- latency is not a single number.

You can get better or worse performance when comparing a good 144Hz monitor with a bad 240Hz monitor. But a very good 240Hz monitor will always blow away 144Hz. Sometimes cherrypicking is needed to make sure you don't have any downgrades.

Thank you so much. Now I am confident to purchase the PG279QM as it is supposedly a "very good 240Hz monitor"

lemmeblur wrote: ↑

10 Sep 2021, 13:49

jorimt wrote: ↑

08 Sep 2021, 16:59

I believe he's referring to the overdrive performance when the monitor is forced to a physical 165 or 200Hz refresh rate, not when you're playing at 165 or 200 FPS at the native physical 240Hz refresh rate with VRR enabled.

So you mean he probably set the display to (for example) 200Hz, but since no variable overdrive was active at 200fps the overshoot was a bit high but got better below that 200fps when variable overdrive kicked in? So this observation should not concern anyone who is going to run the display at 240Hz, right.

I mean he went into the Windows or GPU control panel, lowered the monitor's physical refresh rate to 200Hz, and then took the GtG/overdrive readings, which will be different than when that same monitor is set to a physical refresh rate of 240Hz.

In other words, his 200Hz readings do not apply to 240Hz 0-240 FPS VRR at all.

When you play, for instance, at 144 FPS 240Hz with G-SYNC enabled, the monitor is still running at 240Hz, regardless of framerate. G-SYNC only adjusts how many times the scanout cycle repeats per second to match the current framerate. It does not control how fast each of those scanout cycles complete or what the actual max refresh rate is (if you're running 200 FPS 240Hz VRR, the display is still at a physical 240Hz, not 200Hz). This aspect is fixed and is dependent on the currently set maximum physical refresh rate.

At 240Hz, each scanout cycle lasts for 4.2ms. I.E. no matter what your framerate is during G-SYNC operation, if your monitor is set to 240Hz, frames will scan in at 4.2ms per, regardless.

This is why we recommend you always keep your refresh rate at maximum with G-SYNC and use an FPS limit to control frame/refresh "rate" instead, as 144 FPS at 240Hz with VRR, for example, has faster overall frame delivery than 144 FPS at 144Hz with VRR, because the former is scanning in frames at 4.2ms per, whereas the latter is scanning in frames at 6.9ms per.

As for overdrive performance, again, if you have your monitor set to 240Hz on a native G-SYNC monitor, and VRR is engaged, overdrive performance will look consistent across 0-240 FPS due to variable overdrive.