Zowie XL2720 no inverse ghosting while MBR: off, AMA: high

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

Re: Zowie XL2720 no inverse ghosting while MBR: off, AMA: hi

Post by Falkentyne » 02 May 2017, 15:03

Ok I flashed it. No idea what was changed besides the logos and ID string. I remember someone having a "Zowie V5" also. I can't tell a difference in anything. Probably some stuff with displaypilot or something.

I'm more than 200% sure that 50hz single strobe still causes the monitor to hard reset itself because there is no single strobe data for 50hz so the LED current gets increased by 1.8x by the MBR flag anyway, triggering OCP.

(remember these monitors were reverse engineered from Lightboost. That's why V1 version of the monitors double strobed from 50hz to 85hz, which is exactly what Lightboost monitors did if you checked the "Keep monitor in 3d mode: always" in the 3d vision tab, and that's why both MBR and Lightboost increase backlight current by 1.8x when the strobe signal is set). While 50hz had double strobe data, Benq neglected to add any single strobe data for 50hz, yet backlight current still gets increased, triggering a hard reset (if the strobe duty is higher than 006 and brightness is set to 100%).

Chief, I know you probably don't believe me, but I still personally think the fact that VT tweaks work is an 'Unintended bug" with how Benq designed Lightboost to work with the monitor.

I believe I can prove this :

If you still have your old XL2420TE/TX or XL2720T (make sure its NOT the 120hz panel but the 144hz panel; the XL2720T used a CMI panel (same as VG278H), so it's unknown whether you would get a scrambled screen or not.
However someone here DID test a VT 1500 tweak on an XL2411T monitor. It worked but the screen was scrambled.

Try these timings (do NOT use the NVCP for this! ToastyX CRU only):

H:2560
48
32
2720

V: 1440
3
5
1500
Refresh rate: 100hz.
(this will require the DVI pixel clock patcher. Displayport will cause link errors unless you have updated firmware by Benq; this is what XL2420Z and XL2720Z V4 firmware fixed (I found this information on Benq's support page replies)

This works on the XL2420Z, XL2411Z, XL2720Z, XL2430T, and SHOULD work on the Lightboost 144hz panels.

VT tweaks above 1185 will cause an "Out of Range" error on Asus screens because their firmware blocks all custom VT totals.

tl;dr: Lightboost sends a "VT 1500" signal to the firmware in a different way than a custom 1500 VT resolution.
(you can check this in the factory menu; VT 1497-VT 1502 report 1280x1440@60hz as the resolution while VT 1481-VT 1485 report "Out of Range", with a resolution of 1920x1440 (which is why the OOR error appears on the OSD).

User avatar
Chief Blur Buster
Site Admin
Posts: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: Zowie XL2720 no inverse ghosting while MBR: off, AMA: hi

Post by Chief Blur Buster » 03 May 2017, 11:48

Falkentyne wrote:tl;dr: Lightboost sends a "VT 1500" signal to the firmware in a different way than a custom 1500 VT resolution.
In more proper and clearer terms, it could be considered "It creates its own equivalent of a large vertical total internally".

But, more accurately:

It does an accelerated scanout internally. The only purpose of large vertical totals at the signal level is to force "instant-mode" panels to accelerate their scanout. But monitors can do the accelerated scanout internally without needing large vertical totals. (VT1149 is simply like an "ON" switch for LightBoost, and is unrelated to scanout velocity in the particular case of LightBoost). Essentially, the LightBoost monitor is creating a defacto equivalent of a large vertical total by partially buffering a refresh cycle then doing an accelerated scanout (reportedly approximately ~1/170sec). Marc Repnow (StrobeMaster) of Display Corner figured that out a few years ago.
StrobeMaster wrote:The updating of the screen (i.e., writing new pixel values) is accelerated so as to leave the pixels more time to settle before the backlight is flashed.
To scanout onto the panel faster than the display signal is delivering to the monitor (one pixel at a time at the dotclock rate) -- it becomes necessary to buffer (metaphorically, like filling an empty a water tank with a slow pipe) before scanning out faster (metaphorically, to pour water faster than the incoming pipe could fill it).

Also, one advantage of doing Large Vertical Totals at the signal level, is that it avoids buffer input lag. Because of this, XL2720Z strobing can have 1-2ms less lag than LIghtBoost (at VT1147 or VT1149), while having similar strobe timing/phase (very clear screen centre) -- this is because VT1350 is a higher dotclock than VT1149, and delivers the refresh cycle faster to the monitor (about (1350/1149)ths faster) -- a little more than 1ms faster -- and avoiding the need for any partial pre-buffering which LightBoost/ULMB does. Yes, it is a tradeoff --
easy LightBoost/ULMB -versus- calibrateable strobe backlights (that can potentially have a smidgen less input lag).

For other readers not aware, see Engineering for High Quality Display Motion Blur Reduction -- to better understand why monitors need accelerated scanout & longer pauses between refresh cycles, for blur-reduction strobe backlights to work successfully.
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