mdrejhon

Fantastic stuff. Founder of Blur Busters here. All 240 Hz mythbusting is super welcome in the refresh rate race to future retina refresh rates! The Slo Mo Guys should do some high speed video of refresh cycles in real time. There's the TestUFO test, as well as the article with IPS + TN + OLED High Speed Video Comparision High Speed Video of IPS LCD High Speed Video of TN LCD High Speed Video of OLED More high speed videos can be found at www.blurbusters.com/scanout

Chief Blur Buster here. Consider other important monitor features of a high-Hz monitor. -- Are you sensitive to stutters or tearing? Consider variable refresh rate tech, such as G-SYNC or FreeSync. Note: If you have an NVIDIA GPU, you may favour G-SYNC. If you have an AMD GPU, you may favour FreeSync. Unless you're getting a new GPU anyway. G-SYNC and FreeSync is very forgiving of low and fluctuating frame rates, and will fix a lot of stutters/tearing but won't fix motion blur. -- Are you sensitive to motion blur? And want CRT motion clarity? Consider motion blur reduction tech, such as ULMB, ELMB, LightBoost, DyAc Note: Motion blur reduction uses a strobe backlight to make motion clearer like a CRT. One minor disadvantage is amplifying microstutters (normally hidden by LCD motion blur). It runs runs smoothest when running framerate = refreshrate = stroberate (triple locked produces perfect stutterfree motion like 60fps arcade games on a 60fps CRT). Make sure you upgrade your GPU so you can run at very high steady framerate for beautiful CRT-clarity zero motion blur (Ability to do CRT-era clarity; Nintendo/arcade butter-smooth panning with zero motion blur). New 240Hz monitors include much brighter and colorful strobe backlight features than older LightBoost monitors (often dim), though some 144Hz monitors have gotten much brighter and better than old LightBoost (alas, not always). Some high-Hz monitors provide you with both of the above, as a toggleable "either-or" option (not simultaneously without a hack). Some of us get eyestrain from motion blur and aren't sensitive to TN-vs-IPS color, while for others, color quality is important (ViewSonic has been good at out-of-box color, for example). There's so many different user preferences and sensitivities out there.

Linus, if BENQ sends you a Z-series monitor, make sure it supports V2 firmware (BENQ announcement of V2 firmware). There were BENQ Blur Reduction artifacts with V1 firmware for BENQ Blur Reduction. Only the V2 firmware can allow BENQ Blur Reduction to turn from "worse than LightBoost" into "better than LightBoost" via adjustment from third party utilities such as Blur Busters Strobe Utility. (This app can make BENQ Blur Reduction better than LightBoost, on average, if adjusted properly).

It makes strobe-free CRT motion clarity possible. e.g. CRT clarity without the CRT flicker. 1ms persistence translates to 1ms of motion blur during 1000 pixels/second (e.g. panning/strafing/turning). Frames are static, while eyes are continuously moving. As you track moving objects on a screen, your eyes are in a different position at the beginning of a visible refresh, than at the end of a visible refresh. That creates motion blur as the static frame gets smeared across your retinas. The shorter the persistence, the less motion blur, as persistence-based motion blur is like a camera shutter (60Hz sample-and-hold = 1/60sec of motion blurring = like panning a camera at 1/60sec shutter speed). Sports photographs at 1/250sec, 1/500sec, and 1/1000sec camera shutter speeds are noticeable apart if the motionspeed is fast enough, and likewise, the same also applies to displays as well -- 120Hz is not the final frontier and not even 500Hz is, either. Good educational animations include www.testufo.com/eyetracking and www.testufo.com/blackframes for those who want to understand persistence better. You can either flash for 1ms (120Hz flashed 1ms each, with black gaps in between). CRT phosphor, strobe backlight, light modulation. Or you can have 1ms frames with no gaps between frames. Completely strobe free, flicker free. That automatically requires 1000fps (either real or via interpolation) Both methods would have the same amount of motion blur. It makes it easier to pass a theoretical holodeck turing test, "Wow I didn't know I was standing in Holodeck", because finite frame rates have side effects (stroboscopic effect, mousedropping effect, wagonwheel effect, motion blur, flicker versus motionblur tradeoff, etc) and going to true 1000fps@1000Hz would pretty much make low persistence possible without needing strobing. Currently, all low-persistence displays all require light modulation (phosphor, light modulation, strobe backlight, etc), because there's no way to get 1ms persistence without light modulation at the time (unless you fill all 1ms timeslots -- aka 1000fps@1000Hz). Some useful reading: - Michael Abrash of Valve Software: Down the VR Rabbit Hole (he comments on 1000fps) - Why We Need 1000fps @ 1000Hz This Century - Understanding Persistence: Strobed & non-strobed, CRT vs LCD - Educational Animations: www.testufo.com/eyetracking and www.testufo.com/blackframes For example, popular strobe backlights (e.g. LightBoost, ULMB, BENQ Blur Reduction, Turbo240 etc) flash the backlight for as little as 1-2ms, once per refresh cycle. The only way to match that low amount of motion blur is to fill all timeslots (2ms persistence would require 500fps@500Hz to be completely flickerfree/strobefree, and 1ms persistence would require 1000fps@1000Hz to be completely flickerfree/strobefree). Conclusion: Stop spreading the myth that there is no benefit beyond 120Hz. People like Oculus, John Carmack, Michael Abrash, Valve Softare, and myself of Blur Busters, all unamiously agree persistence is important, and we all recognize that the engineering challenge of 100% strobe-free low-persistence unavoidably requires ultrahigh frame rates. So that's why all current low-persistence displays are light-modulated in some way (e.g. phosphor, flicker, black frames, strobing, etc). NOTE: GtG (transition/movement state) is different from persistence (static/visible state)

Wow, thanks for the mention of Blur Busters! (I've now posted this video on Blur Busters) I have not mentioned this publicly until now: Palmer Luckey gained exclusive access to a pre-beta version of TestUFO motion tests more than 1 year ago, long before it launched! This was long before Oculus snagged John Carmack as CEO. I'm one of the parties that convinced Oculus to eventually go down the "low-persistence" path during December 2012 / January 2013 conversations. Mark Rejhon (aka Chief Blur Buster)

G-SYNC locks your framerate to the refresh rate of the monitor. What's different is that the refresh rate isn't locked anymore

Sent reply. Blur Busters was a hobby (unfunded originally, now paid only a bit in advertising fees), and focussed a lot on the technology aspects rather than the gaming aspects. There will slowly be more and more gaming coverage on Blur Busters, but remember Blur Busters started as an electronics hobby (scanningbacklight.com) for my Arduino scanning backlight project before LightBoost was discovered. Yes, my unfunded, unpaid electronics hobbyist site (Arduino Scanning Backlight) slowly became commercial (renamed to Blur Busters), covered by Amazon advertising, so I do have to slowly withdraw from posting links since the universe from free personal site to commercial site became blurred. That said, I keep good relationships to most sites; several are looking forward to my Input Lag Tester that I'm currently developing for 2014 (including Linus, who's also interested). Once I crossed the "hobbyist-versus-commercial" boundary, that's when HardOCP banned me (they have also banned other reviewers, including pcmonitors and others). I have even taught other reviewers about LightBoost, nobody else understood what the heck it was, and how it exactly worked. As flawed as all websites (including Blur Busters), most people say Blur Busters explain LightBoost the best of all. I plan to be one of the better G-SYNC sites too, as I will also continuously expand the G-SYNC stuff (based on reader suggestions including yours too -- I'm accepting suggestions). I tirelessly spent literally hundreds of hours in the last 12 months explaining to so many people, including people at Oculus, people at NVIDIA, people in many commercial companies. You don't know how many big emails I have written behind the scenes. (I can only educate a few people at a time, a few at a time -- and there are 7 billion people in the world). Slowly, it became effective. Which is a small miracle: EIZO, NVIDIA and BENQ have mailed me their monitors to test out (and I've given all of them private feedback based on my sage knowledge). If the hours were divided by the advertising revenue, it's all well below minimum wage -- but it's a hobby that I love dear (at least the electronics, science & tech parts). You know; I'm more of a technology/electronics/geek guy that loves to explain the technology stuff (e.g. how LightBoost works), even to the point of making monitor manufacturers realize that LightBoost is something worth paying attention to! But yes, I know several want me to start talking more often about games! I also agree that we're on the cusp of a gaming monitor revolution (G-SYNC & multiple strobing technologies) and I'm ideally positioned with my engineering knowledge since I know the technology aspects of a lot of the monitor technologies long before most. Suggestions on emulator testing is accepted; send these direct to mark[at]blurbusters.com and feel free to also send the same emulator suggestions to Linus too... P.S. Out of respect for Linus's in-progress G-SYNC work, I did not post a link to my GSYNC review in THIS thread, because I knew I was posting the link a little bit much. And I had already posted a few links to other Blur Busters stuff already here, so I have to throttle back, you know. P.P.S. Humblest apologies in advance. I do get TOO ENTHUSIAC about Blur Busters sometimes. Guilty as charged.

It should, with this command line: mame -throttle -nosyncrefresh -nowaitvsync -notriplebuffer (That's a friendly tip for you, Linus ) Tech Note for emulator programmers: you just need Direct3D with the D3DPRESENT_INTERVAL_IMMEDIATE for the best, low-lag, G-SYNC synchronized emulator performance, calling Direct3D Present() executes immediate monitor refresh (if you just only need to blit an emulated frame buffer to a Direct3D frame buffer); Technically, the monitor no longer has a concept of a refresh rate; it's waiting for the GPU now. The programmer now controls the timing of the monitor's refresh by deciding when to call Direct3D Present() which does the render time (which is pratically instantaneous when blitting a single emulator frame buffer) and immediately output the refresh without waiting for old-fashioned refresh cycles, Now, if Linus doesn't do it first, I will test this for Part #2 of the Blur Busters GSYNC review (my part #1 is up already). Yes. You want 77.316Hz? It'll run at exactly that, too. (G-SYNC monitor sitting in front of me, too)

For those wondering about G-SYNC's ability to do stutter-free variable frame rates, there's a good animation of G-SYNC behavior on the TestUFO animations site. (Do make sure you run a stutter-free VSYNC-supported web browser, for an accurate simulation!)

In addition to several overclockable IPS 120Hz panels, don't forget there's also 120Hz VA monitors now -- the two models by Eizo, including the consumer FG2421 (from 120Hz monitor list). The "Z" series BENQ monitors include an official strobe backlight (for CRT-quality motion on LCD) that's easily enabled via a monitor button. So you don't need a utility to activate LightBoost. All IPS monitors become very blurry during motion -- www.testufo.com/photo -- you only get crystal clarity on CRT, plasma, LightBoost, Turbo240, and BENQ Blur Reduction. And some of these new strobe backlights don't degrade colors anymore, unlike some older LightBoost models!! 60Hz = baseline 120Hz = 50% less motion blur strobed = 80-95% less motion blur

It's very important to note that there's many different developments that makes things interesting for 60Hz vs 120Hz vs 144hz: - Strobe backlights allow LCD's to have the motion clarity of a CRT. They include LightBoost, Turbo240, BENQ Blur Reduction (XL2720Z), etc. - Strobed 100Hz and 120Hz has less motion blur than non-strobed 144Hz. - There's also non-TN 120Hz now available (EIZO's 120Hz VA panel, and Overlord/QNIX/etc 120Hz IPS) - Overclocked monitors (e.g. IPS 1440p), more resolution and better colors, but much more motion blur There's a big list of 120Hz monitors that lists all of these, for those wondering.

A great way to figure out why it's is not feasiable without hardware modifications: 1. Watch high speed videos of traditional display refreshing: - - - 2. Observe all the above refresh at an exact interval (synchronous), refreshing like a clock, tick-tock, exact schedule. That's the way displays have functioned for more than 75 years, since the invention of the first televisions. Synchronous refreshing. 3. G-SYNC eliminates the exact schedule of a display refresh. A display can refresh at random moments. e.g. one refresh, then another refresh 7.689 milliseconds later, then the next refresh 35.32189 milliseconds later, and so on. The schedule is eliminated. The exact interval is eliminated. G-SYNC no longer refreshes on clockwork. 4. And G-SYNC can do it without flicker or color modulations (traditionally, different refresh rates can have different color quality: e.g. 60Hz vs 120Hz often needs different color calibration). G-SYNC can essentially change refresh rates over 100 times a second (intervals between all refreshes can all be completely different!), without any noticeable side effects. That's an additional, challenging engineering challenge.

Since I'm hearing impaired (deaf), keyboards have been my telephone. So here goes, after three passes: Probably could get to 115 after a few more runs. EDIT: After sixth run, I did:

I now have the EIZO FG2421. (I'm going to be posting a detailed review of this monitor later this month) Preliminary impression: Generally great 120Hz VA panel. I'm getting the benefits of LightBoost, without the color quality degradation. 5000:1 contrast ratio and the ability to read the map labels in the TestUFO Panning Map Test. Viewing angles are roughly 80-90% as good as IPS.

Yes, it's a TN panel. So you need to decide whether having great colors is important (go IPS) or having less motion blur (go LightBoost). One idea -- At the $300 price point, a compromise is the QNIX QX2710 Evolution 2 -- it's an overclockable IPS. It does not have a strobe backlight, however, but will have less motion blur than the average IPS.