Blur Busters Forums

.

sssouthpaw wrote: ↑

14 Jan 2022, 00:39

The idea of blur reduction + vrr is to remove the strict strobing rule of limiting frame rate to fixed value of monitor refresh rate (or vsync). But using it at lower frame rates is also very important. TestUFO refresh and frame rate is always locked to your currently set Hz setting, meaning any type of real vrr testing is not supported in TestUFO yet. Fortunately we have a solution - Smooth Frog is absolutely necessary to be used for correct vrr pursuit camera testing.

TestUFO will test only:
monitor set to 240Hz: 240fps, vrr 240Hz
monitor set to 120Hz: 120fps, vrr 120Hz
monitor set to 60Hz: 60fps, vrr 60Hz
...

Smooth Frog will test any combination:
monitor set to 240Hz: 240fps, vrr 240Hz
monitor set to 240Hz: 120fps, vrr 120Hz
monitor set to 240Hz: 60fps, vrr 60Hz
or whatever you need
monitor set to 120Hz: 120fps, vrr 120Hz
monitor set to 120Hz: 60fps, vrr 60Hz
...

So what you see in TestUFO will not reflect what you see in game if test conditions are not exactly the same. Because for each set refresh rate monitor applies different setting/timings (moves the single strobe to currently set refresh rate): viewtopic.php?p=73457#p73457 (scroll down to check the pulses at different refresh settings).

If you want to re-test MPRT + Adaptive-Sync I suggest using Smooth Frog with monitor set to 240Hz and doing pursuits for three different fps limits: 239, 120 and 60 fps. That way you cover full range and not just 240fps. Make sure to set correct camera shutter. For each refresh rate it needs to be readjusted. But until Smooth Frog gets a proper sync track for validating camera tracking accuracy (like in TestUFO) I suggest to use s1/60 for 239 and 120 fps, for 60fps s1/30. There is no need for slow motion recording. It should end up looking something like this. Read more here, but for this case skip steps 2 and 5: Simple smartphone pursuit camera guide. If you do it, it would be first proper demonstration on the Internet of MSI's vrr strobing implementation.

Btw, does enabling MPRT really lock out the overdrive and brightness controls like user manual says?

liquidshadowfox wrote: ↑

13 Jan 2022, 15:42

Did you at least tested the VRR + strobing? even with the defects? I want to know if there's any heavy crosstalk if not, it's all good.

Looks like MSI pushed the crosstalk just a little more off the screen (up) compared to XG. Also ufo ghosting test outside fullscreen mode at 1440p is not centered to middle screen. Real center line is just bellow most bottom light cyan track. Here are both vrr strobing at 240Hz and 240fps: MAG274QRX, XG279QM modified OD 13

I noticed some red ghosting. Interestingly only when camera is panning in different direction (3:25). Also you can clearly see the red frame when MPRT is enabled/toggled (2:41). That is KSF phosphor indication, but it seems to be present in much smaller amounts?

Yes, MSI is using the Rapid IPS aka Innolux panel that gigabyte uses on their monitors which also seems to have the KSF phosphor. I heard the higher the refresh rate, the less noticeable the KSF will be (with proper tuning of course)

To my eye it looks like the XG27AQM you can see the eyes on the UFO better (is that image ELMB or ELMB sync? from the URL it seems like it's ELMB sync) and I think it looks better on the XG27AQM with higher refresh (265 fps) but something about the MSI looks like it has a less "softening" effect of the edges vs the XG27AQM that I just can't put my finger on. If the OD is locked during backlight strobing on the MSI and it has little to no overshoot across the refresh range, I'm 100% switching from my XG until I get my eve spectrum I'm excited to see the smooth frog results assuming @sssouthpaw performs the test on it.

You've probably read that in M32Q review. It probably is because of high 240hz yes. This is actually first 240hz ksf ufo I've seen.

I wouldn't recommend directly comparing bad quality ufos shots (out of focus, low resolution, AI upscaling...). You can't get a proper impression of used strobe duration (short - sharp ufo, long - blurrish ufo). That was one of your ELMB-SYNC pursuits.

Optional overdrive is a must for any vrr, both strobed or non-strobed. Especially for such wide vrr range. At least MSI didn't overdo it. It reminds of MAG251RX overall crosstalk.

.

sssouthpaw wrote: ↑

14 Jan 2022, 13:40

We can try it, but only if you want to. It might take some time to do it correctly.

.

sssouthpaw wrote: ↑

14 Jan 2022, 13:54

If monitor is set to 240Hz there should be no tearing bellow 240 fps. OSD fps should fluctuate around set limit, but osd usually doesn't show very accurate results. Before we continue try making vrr work without tearing. One advice would be to enable Vsync with "V" key or in menu (in Smooth Frog v0.92b) and leave it like that. Before we start check if you have an option to change camera shutter speed and ISO.

.

sssouthpaw wrote: ↑

14 Jan 2022, 14:06

Yes, use only 'Frog Pursuit Pattern'. Make sure you're in Pixels per Frame mode. We'll try to do it without sync track this time. For 239 fps use 9ppf 1483pps speed with s1/60 shutter, for 120 use 12ppf 1440pps speed with same s1/60, for 60fps use 24ppf, 1440pps speed with s1/30 but 1/60 should work fine too.
Now just do the same as on test ufo - follow/track/capture the froggies. Try make camera focus nice and crispy.