not gonna lie, I can't see the difference with ULMB on

[.kernel 0x000]

My monitor is the ROG Swift PG278Q, my graphics card is Nvidia GTX 980, my OS is Windows 8.1 (mouse fix applied to explorer.exe). Yes I disabled G-Sync first (as needed).

The only game I tested it on is CSGO, and it just made my screen darker (I also tried switching between the various ULMB values, making it darker each time) and I really don't see a motion difference compared to ULMB off (I usually use G Sync ON and 144Hz when ULMB if off). I don't really see much of a difference with Windows either, just the darkened screen that kinda hurts my eyes. Why do I not see it?

Also as for CSGO, is it a bad idea to play it with g sync on? CSGO's fps_max is set to 144, and I notice cl_showfps 1 reports my fps as 142-143 (flickering between those two values). I don't notice any input lag, is there any reason for me to change my settings?

[Falkentyne]

If you are NOT using internet explorer, or if you ARE, and can use ULMB at 100hz refresh rate (should work down to 85 hz), go to testufo.com, moving picture/alien invasion tests, and test with and without ULMB.

I guarantee you will see a difference.
Once you do that then you go back to the CSGO game, you should notice it more.

Also experiment with various strobe lengths. I don't know what it's called in ULMB, but using 100% (highest persistence) isn't much different than 144hz without ULMB. Try the lowest or second lowest setting and make sure the brightness is set to 100%.

[Chief Blur Buster]

ULMB works best; if:
1. Framerate matches refresh rate (easier with VSYNC ON)
2. You keep your microstutters under control (1000Hz mouse, no mouse microstotter, no graphics microstutter)
3. Your motionspeed is very fast
4. You are tracking your eyes on fast-moving objects while you are turning/strafing/panning/scrolling.

There are many use cases where you won't see a difference. And other times the difference is super-dramatic. It depends.

If you only stare at crosshairs, you won't see ULMB difference. It helps when trying to identify fine details while moving eyes (e.g. scrolling text, high-speed flybys over camoflaged ground, sniping while moving fast such as Scout in Team Fortress 2, etc). A good test is http://www.testufo.com/photo and the TestUFO Moving Map Test. At motionspeeds such as 960pix/sec or 1920pix/sec, it becomes impossible to read the street name labels without ULMB. Only with a strobe backlight, such as ULMB, it becomes to read small fast-moving text (like on a CRT).

Example: It's not possible to read the street name labels without low-persistence (ULMB/strobing):



Turn on ULMB, and you will easily read the street name labels in this panning-map animation. Works best with a full screen browser (click on animation and view maximized).

[ciopenhauer]

Make sure your second monitor is disabled if you have one. That was my problem when I didn't notice any effects from ULMB. (though I also had to reduce the strobe phase and duty).

[Chief Blur Buster]

Make sure your second monitor is disabled if you have one. That was my problem when I didn't notice any effects from ULMB. (though I also had to reduce the strobe phase and duty).

Strobe phase?

As far as I know, NVIDIA hasn't enabled strobe phase adjustability in ULMB -- just strobe duty ("ULMB Pulse Width" in OSD). I presume you're using the BENQ equivalent of ULMB, known as BENQ Blur Reduction?

[ericl]

The Asus ROG Swift is excellent for CS:GO and I can help you optimize for it.

If you're seeing the screen 'dim' down a little, then it is likely working just fine.

Here's what you'll want to set:

- For the best performance, I found the pulse width to be set at anywhere between 60-80.
- Then I increase the dynamic colors in the Nvidia control panel (this makes the players 'pop out' a bit more in CS. This is especially important if you're using ULMB)
- Obviously, you'll want to disable Vsync and make sure Gsync is disabled (it should be, otherwise you wouldn't be able to use ULMB)

The difference isn't as huge on the Asus ROG Swift as it is on other monitors BECAUSE the 144mhz mode is so damn good on it. Obviously, for fast motion games ULMB is better but even without it, the Asus ROG Swift performs like a champ in 'standard' 144mhz mode.

On other monitors (such as the BenQ 2411z), it is MUCH more apparent when you enable strobing.

So in a way, it's not that the strobing isn't working well... it's just that the monitor does both Gsync, regular and strobing really well. This makes distinguishing the differences between them more difficult.

Hope this helps.

[.kernel 0x000]

The Asus ROG Swift is excellent for CS:GO and I can help you optimize for it.

If you're seeing the screen 'dim' down a little, then it is likely working just fine.

Here's what you'll want to set:

- For the best performance, I found the pulse width to be set at anywhere between 60-80.
- Then I increase the dynamic colors in the Nvidia control panel (this makes the players 'pop out' a bit more in CS. This is especially important if you're using ULMB)
- Obviously, you'll want to disable Vsync and make sure Gsync is disabled (it should be, otherwise you wouldn't be able to use ULMB)

The difference isn't as huge on the Asus ROG Swift as it is on other monitors BECAUSE the 144mhz mode is so damn good on it. Obviously, for fast motion games ULMB is better but even without it, the Asus ROG Swift performs like a champ in 'standard' 144mhz mode.

On other monitors (such as the BenQ 2411z), it is MUCH more apparent when you enable strobing.

So in a way, it's not that the strobing isn't working well... it's just that the monitor does both Gsync, regular and strobing really well. This makes distinguishing the differences between them more difficult.

Hope this helps.

hmm thanks it does, but it seems the browser often gets stuttering errors according to the site.

Also what would you say would be better for playing Arkham Asylum on the ROG Swift, v-sync ON, MaxSmoothedFrameRate=122.000000 with 144Hz mode or with 120Hz mode? I know one might ask "why are you not using G Sync?!" but it seems my framerate is way smoother with v sync ON than g sync ON in that game (g sync ON I can get fps drops to 70fps, v sync ON the lowest I ever get is 110fps, this is all tested in the same room with no npc's). Also am I correct in having maxsmoothed be 122 and not 120?

[Chief Blur Buster]

This is quite normal. As a rule of thumb:

...If your framerate is low or variable, GSYNC often looks much better.
...If your framerate is steady and very high, ULMB can look better.

VSYNC ON is problematic for input lag, but if you're playing solo or the lag is low enough, the motion quality can sometimes be more important. It all depends on type of game & gameplay, and personal preference. Now if you're a professional competitive that hates tearing/stutter, then aiming for lowest input lag at high framerates, GSYNC 144Hz + fps_max 140 is often a very good combination (game throttling of framerate is lower lag than external throttling such as by drivers or monitor). But it will still have more motion blurring during eye-tracking situations, than ULMB. So it's all about very subtle tradeoffs.

As for browser microstutter, I have seen certain versions of Chrome (38 and 39) get more microstutter with TestUFO; I consider this a browser bug. So try a different browser (e.g. Internet Explorer 100Hz is often stutter-free in TestUFO), make sure you check http://www.testufo.com/browser.html to improve TestUFO performance. And, if you are using ULMB, you want to lower Strobe Duty Cycle a little bit. Try adjusting Strobe Duty Cycle while watching the TestUFO Panning Map Test, you'll see street name labels become clearer/blurrier as you adjust Strobe Duty Cycle. It's a brightness-versus-clarity tradeoff, especially at 1920 pixels/second.

[.kernel 0x000]

Now if you're a professional competitive that hates tearing/stutter, then aiming for lowest input lag at high framerates, GSYNC 144Hz + fps_max 140 is often a very good combination (game throttling of framerate is lower lag than external throttling such as by drivers or monitor)

interesting, I've noticed with fps_max 144 in CSGO and G Sync on, I get 142-143 (constant fluctuations) which I am told is to prevent v-sync style input lag. So this is the driver doing this? And it would be better if I chose fps_max 140 so that then the game is doing the limit? Also what is the magic number to choose? Like how did you know 140 and not 139 or 142?

And on another note for a single player game (Batman) is switching the monitor from 144Hz to 120Hz when a game can only get 120Hz a good idea (assuming v-sync is used)?

[Falkentyne]

if that test weren't so seizure inducing at 1900+ pix/second (I'm still crosseyed now), that would probably be the best tests in the history of mankind (it's still a GOOD awesome real world test of how beneficial strobing is and what we missed when those old CRT's went byebye). At 1920 pix/second, you can actually see visible blurring at 1.0ms persistence that becomes nice and clear at 0.5ms persistence. Now think of moving objects or fast turning in FPS games, and you can see there is a benefit of having a 0.5 ms persistence, but brightness becomes a definite tradeoff now. The image would be too fast (you'd need 3800+pix/sec) and the screen too dim, to be able to perceive and test 0.167 vs 0.250ms persistence (the two lowest possible values on a Z series benq, only available in the service menu).

BTW here's something you can do to find a true real world test of what persistence would benefit you the most in gaming.

Download the Pixperan persistence test. This test was created back when CRT's were still in use, and was used to test persistence and ghosting on LCD's. But I think people FORGOT just how useful the 'readability test' really is.

Set your desired refresh rate, turn on your motion blur reduction/strobe setting,
Go to the readability test and set tempo to 30.
Then set your persistence (strobe pulse width/strobe duty) to where you can get the correct answers at tempo 30, at least a decent amount of the time (Benq users should make sure that the strobe crosstalk field is at the top and bottom of the screen so adjust strobe phase/use VT tweaks as needed)

The point where you can see the text clearly at tempo 30 in a dimly lit room is probably the ideal highest strobe width/persistence you can use for games, and then you can go lower as needed for more comfort or clarity. I find 1.0ms persistence to be a nice tradeoff from clarity to brightness. I can see a difference in word clarity going from 1.0ms to 0.5ms, although some of the awkward letter pairings make it harder for me to get some answers, than just the clearer scrolling.