Vertical 1350 and Crosstalk Question

Adjusting BENQ Blur Reduction and DyAc (Dynamic Acceleration) including Blur Busters Strobe Utility. Supports most BenQ/Zowie Z-Series monitors (XL2411, XL2420, XL2720, XL2735, XL2540, XL2546)
jazzyy
Posts: 2
Joined: 02 Jun 2015, 02:47

Vertical 1350 and Crosstalk Question

Post by jazzyy » 02 Jun 2015, 02:52

The guide says that for the Vertical Total 1350 trick we should set our Crosstalk to zero. Does that mean that I should set it to the farthest left setting? There's no numbering on the bottom panel as seen in my attachment and the guide doesn't seem to specify other than a number.

Sorry for the dumb question.

Image for clarity: https://i.imgur.com/a138niD.png

Falkentyne
Posts: 2793
Joined: 26 Mar 2014, 07:23

Re: Vertical 1350 and Crosstalk Question

Post by Falkentyne » 02 Jun 2015, 04:03

I prefer using vertical total 1500 or VT 1502 (whichever one works and doesn't cause strange pulse errors or stuttering) instead of VT 1350.

But to answer your question, setting the slider to the *far left* is the most compatible setting and will work with all refresh rates and custom resolutions. This will leave some crosstalk around the last 1/10th of the screen. Setting the crosstalk slider to the left corresponds to "strobe phase 000" in the service menu.

There are _some_ issues with setting the crosstalk slider to the far left that are not that critical but can be noticeable at low refresh rates.

The first is, the crosstalk at the very top of the screen is only completely 100% clear if the strobe duty is set to 001 (lowest persistence; dimmest screen) and is only unnoticeable (meaning theres no real ability to notice it outside of a testUFO full screen alien invasion test) if the strobe duty is no higher than 003. At strobe duty 003, you will see the very top of the strobe pulse start to show errors along the top few pixels of the screen, which have no effect on any real world gaming usage whatsoever, but as you raise the persistence (increases the brightness and lowers the amount of blur reduction, since strobe duty 003 is a rather dim screen), the strobe crosstalk at the top of the screen moves downwards more, and once you go past strobe duty 006 (1.0ms persistence), it actually gets large enough to be visually distracting in gameplay. There's no way to stop this from happening: raising the strobe duty lowers the crosstalk at the top of the screen and moves it downwards, but to a FAR less degree than changing the strobe phase. With the crosstalk at the far left ,I find strobe duty 006 (1.0ms persistence with a VT tweak) to be the absolute limit to what is acceptable crosstalk at the top of the screen. Any higher (even though I get a brighter picture), and it becomes annoying.

The second is that having the strobe phase at 0 ADDS one frame of input lag compared to having the strobe phase set to the maximum value (or the highest value before the backlight shuts off due to the persistence dropping to 0--this only happens if you are using a VT tweak and this changes based on the refresh rate). Masterotaku found this out a long time ago but no one believed him then, but he was right all along. This is because past a certain strobe phase, and you can test that by raising and lowering the strobe phase, on the TestUFO, alien invasion test, you will notice that the UFO's move either "ahead" or "back" one entire frame. This is the reason that the default strobe phase for a full factory reset or fresh firmware flash is 100--Benq's presets are optimized for lowest input lag, and the least crosstalk at the top of the screen (similar to Lightboost mode).

http://forums.blurbusters.com/viewtopic.php?f=13&t=560

While having a strobe phase of 000 and losing 1 frame of input lag (how much ms you lose depends on the refresh rate, of course) is probably insignificant at 120hz refresh rate, but at 50hz (using a VT of 1360 single strobe; do NOT try 50hz WITHOUT a VT tweak, ever!), a strobe phase of 000 has 20 ms (!) more input lag than strobe phase of 100. It's measurable and noticeable in call of duty advanced warfare, with vsync at 50hz.

Now, VT tweaks lower the position of the crosstalk at the -bottom- of the screen (basically, pushing it downwards; VT 1500 pushes it downwards more than VT 1350) while not affecting the position of the crosstalk at the -top-. If there is too much strobe crosstalk at the TOP of the screen (e.g. if you are using a high persistence value, for example), then you can gain 1 frame of improved input lag and make the top of the screen perfect, by raising the strobe phase. Without a VT tweak, you can go to phase 100 at all refresh rates,

With a VT tweak, the maximum strobe phase is limited by the refresh rate. Going past the maximum value will shut off the backlight until blur reduction is toggled off (basically the persistence decreases as phase increases until persistence drops to 0). Some values for maximum strobe phase (the persistence will be 0.167ms at the maximum strobe phase, or the equivalent of strobe duty 001--you will notice that if you try to raise strobe duty at the maximum strobe phase, nothing happens, right?)

125 hz: VT 1498: 047
120 hz: VT 1500= 049
100 hz: VT 1502: 059
91 hz: VT 1502: 065
85 hz: VT 1501: 70 (I think)

So, for a strobe duty of 004 (0.668 ms persistence) and for gaming at the lowest input lag, without having too much crosstalk at the bottom of the screen, and having the top of the screen perfect (this is what masterotaku likes to do), you could use:
125 hz: VT 1498, strobe phase 044 (max persistence is 0.668), strobe duty 004
120 hz: VT 1500, strobe phase 046, strobe duty 004
100hz: VT 1502: strobe phase 056, strobe duty 004,

etc.
If the screen is too dark, drop the strobe phase by 2 so you get 1.0ms persistence (maximum strobe duty will be 006 now), but this may put too much crosstalk at the bottom of the screen.

The entire reason that Chief Blur Buster's strobe utility slider is limited to 047 strobe phase for the right most slider, is to prevent a black screen when enabling blur reduction, from having the strobe phase set too high. The highest refresh rate anyone would really use a VT tweak for is 125 hz (since this matches the synchronization of 1000hz gaming mice divided by 4 evenly), and the absolute highest refresh rate that will accept VT tweaks without out of range errors is 128hz (129 hz is erratic, just like 76 hz).

IF you are gaming at 60hz refresh rate with single strobe enabled (flickery screen), you want the strobe phase at 100 or as high as possible, as strobe phase 0000 will cause an extra frame of input lag. And if you are at 60 fps, losing 1 frame of lag can be noticeable.

jazzyy
Posts: 2
Joined: 02 Jun 2015, 02:47

Re: Vertical 1350 and Crosstalk Question

Post by jazzyy » 11 Jun 2015, 10:01

Sorry about the late response! I did those improvements (1502) and I really am enjoying it. I appreciate you putting time in to describe this stuff to me. Thanks!!

Friard
Posts: 17
Joined: 13 May 2017, 00:03

Re: Vertical 1350 and Crosstalk Question

Post by Friard » 13 May 2017, 00:07

Falkentyne wrote:The highest refresh rate anyone would really use a VT tweak for is 125 hz (since this matches the synchronization of 1000hz gaming mice divided by 4 evenly)
Do you have any forum threads I can read on the benefits of synchronization between polling rate and refresh rate

User avatar
RealNC
Site Admin
Posts: 3737
Joined: 24 Dec 2013, 18:32
Contact:

Re: Vertical 1350 and Crosstalk Question

Post by RealNC » 13 May 2017, 22:38

Friard wrote:Do you have any forum threads I can read on the benefits of synchronization between polling rate and refresh rate
http://www.blurbusters.com/faq/mouse-guide

Note that this really only can be an issue with blur reduction. Without blur reduction, it's virtually impossible to spot this kind of micro-stutter. (Actually, I can't spot it even with blur reduction enabled; my vision isn't that good. But many people can.)
Falkentyne wrote:The highest refresh rate anyone would really use a VT tweak for is 125 hz (since this matches the synchronization of 1000hz gaming mice divided by 4 evenly)
Nitpicking mode: It's divided by 8, not 4 :mrgreen:
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Vertical 1350 and Crosstalk Question

Post by Chief Blur Buster » 14 May 2017, 11:20

RealNC wrote:
Friard wrote:Do you have any forum threads I can read on the benefits of synchronization between polling rate and refresh rate
http://www.blurbusters.com/faq/mouse-guide

Note that this really only can be an issue with blur reduction. Without blur reduction, it's virtually impossible to spot this kind of micro-stutter. (Actually, I can't spot it even with blur reduction enabled; my vision isn't that good. But many people can.)
Actually, I can see this microstutter even with blur reduction turned off. I'm sensitive enough to see it.

That said, it's definitely MUCH easier to see the microstutter when blur reduction is turned on.

Most game engines do not synchronize mouse poll rate to gametime updates (although I think Quake can do it). Now, 1000Hz poll rate perfectly synchronized to a 250Hz display (I wonder if any one of the new 240Hz displays can be overclocked to 250Hz... hmm!) Without synchronized poll rates, overkill poll rates is our option to reduce/solve microstutters.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Vertical 1350 and Crosstalk Question

Post by Chief Blur Buster » 14 May 2017, 14:17

BTW, there's a new TestUFO Strobe Crosstalk test at www.testufo.com/crosstalk

This is the same as the old TestUFO Alien Invasion but is designed specifically for quicker testing of strobe crosstalk -- you simply click to go to full screen mode -- and the URL is much easier to remember.

You adjust Strobe Phase/Area while viewing this test -- the double-image zone will shift around while you're adusting the setting.
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Falkentyne
Posts: 2793
Joined: 26 Mar 2014, 07:23

Re: Vertical 1350 and Crosstalk Question

Post by Falkentyne » 14 May 2017, 15:52

Chief, is there a way to remove the orange message once you go fullscreen? It remains there and is a little jarring for those with OCD.

User avatar
RealNC
Site Admin
Posts: 3737
Joined: 24 Dec 2013, 18:32
Contact:

Re: Vertical 1350 and Crosstalk Question

Post by RealNC » 14 May 2017, 20:09

Chief Blur Buster wrote:Actually, I can see this microstutter even with blur reduction turned off. I'm sensitive enough to see it.
I hate to be "that guy" (I really do), but have you ABX'ed it? I was convinced that I'm sensitive enough to detect a LOT of things in the past. As soon as I did my first actual test, I found out I was BS'ing myself :P
SteamGitHubStack Overflow
The views and opinions expressed in my posts are my own and do not necessarily reflect the official policy or position of Blur Busters.

User avatar
Chief Blur Buster
Site Admin
Posts: 11647
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Vertical 1350 and Crosstalk Question

Post by Chief Blur Buster » 14 May 2017, 23:03

The variables must obviously be defined (e.g. Framerate used, framerate capping used or not, which game engine, which refresh, TN is slightly easier to detect mirostutters on than IPS, etc) Microstutter intensity, on average, does goes down at higher framerates. e.g. 300fps versus 1000fps. However, don't forget harmonic microstutters are easier to see at say, near multiples, such as super-accurate 362fps at 120Hz than it is at random framerates (modulus 120 creates a remainder of 2) -- 2 microstutters a second. (A Titan and a fps_max 362 still shows very faint two-skip-a-second microstuttering in Source Engine games). Content also matters. The low contrast edges versus higher contrast edges.

Mathematically: At 2000 pixel per second smoothpan (keyboard strafe, not mouse, to eliminate mouse microstutter error margin), 300fps microstutter can still have an edge back-and-fourth (edge vibrate) amplitude of (2000/300) = ~6 pixels as the 300fps essentially does a beat-frequency microstutter against the 144Hz refresh rate. This is easily hidden in the (2000 / 144) = 14 pixels of motion blurring of 144Hz sample-and-hold (plus 2 extra pixels of motion blurring from 1ms TN GtG at 2000px/sec). Visibility of microstutter is greater when hitting low harmonics -- like 241fps or 361fps at either 120Hz -- and when bypassing mouse microstutters (e.g. Autopanning, or using keyboard strafe, etc). Sometime, I'll have to write about harmonic-frequency microstutters and how they fade in and out, as framrates modulates through the multiples.

The point of diminishing returns apply here based on the variables, and the bar is raised when in strobed modes, but the 300-400fps does not completely eliminate microstutters for me on 144Hz TN-panel LCDs -- VSYNC ON (laggy as it is) is still subtly but noticeably smoother than 300-400fps VSYNC OFF (which is better for competitive) for me. I don't see it if I focus on the game, but it's there if I pay attention to noticing it especially in close-range keyboard strafing.

This is getting off topic though for this thread, but worthy of opening this discussion in a more appropriate area (defining the variables properly, etc.).
Head of Blur Busters - BlurBusters.com | TestUFO.com | Follow @BlurBusters on Twitter

Image
Forum Rules wrote:  1. Rule #1: Be Nice. This is published forum rule #1. Even To Newbies & People You Disagree With!
  2. Please report rule violations If you see a post that violates forum rules, then report the post.
  3. ALWAYS respect indie testers here. See how indies are bootstrapping Blur Busters research!

Post Reply