DLP BFI for Consoles

Ask about motion blur reduction in gaming monitors. Includes ULMB (Ultra Low Motion Blur), NVIDIA LightBoost, ASUS ELMB, BenQ/Zowie DyAc, Turbo240, ToastyX Strobelight, etc.
BFI
Posts: 19
Joined: 23 Nov 2021, 01:25

DLP BFI for Consoles

Post by BFI » 23 Nov 2021, 01:51

Does anyone know of a DLP projector that the PS5 and Series recognise as 120Hz in 3D mode?

I can't even get side-by-side working with PC on my W1210ST (it's an HTPC with a cheap Nvidia card that won't offer anything above 60Hz), and I believe the consoles require HDMI 2.1 for 120Hz?

So I'm after a newer 240Hz DLP with 2.1 ports, but are any confirmed to support both consoles in 3D mode, and do their 3D modes offer 8.3ms persistence at 120Hz (good enough for me), or even better at 4.2?

Failing that, are there any non-OLED TVs with adjustable strobe duty? I almost bought a QN95A, but then discovered its BFI only functions aggressively at 60Hz, which I know I dislike from single strobe on my BenQ XL. I don't want OLED because I leave static images on mine a lot, but I'd consider anything else with adjustable BFI.

Although curiously, I find BFI ineffective at 120Hz on my XL, with lots of blur remaining at strobe duty 001. Unfortunately, I also find strobe duty 30 too flickery on single strobe, so overall I'm unsatisfied. Whereas the smoothness of motion on my 1210 feels fantastic, I just want to be able to halve its persistence.

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

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 23 Nov 2021, 15:09

BFI wrote:
23 Nov 2021, 01:51
Although curiously, I find BFI ineffective at 120Hz on my XL, with lots of blur remaining at strobe duty 001.
Are you doing framerate=Hz?
VSYNC ON or RTSS Scanline Sync or framerate cap?
Lower your Hz and raise your framerate to perfectly match each other.
Otherwise you get the same CRT 30fps at 60Hz problem.
This issue affects all impulsed displays and all strobe backlights.

Image

Make sure to not confuse blur with duplicate images. Also strobe crosstalk can be an issue (especially if you're not doing Large Vertical Total QFT modes). Make sure you describe your artifact more accurately, based on things like LCD Motion Artifacts as well as Strobe Crosstalk.

Warming up your LCD will reduce the strobe crosstalk (60 minutes power up in a cold winter in room), combined with Vertical Total 1350, combined with Strobe Utility adjustment, should help reduce crosstalk-related duplicate images, leaving only framerate-related duplicate images (fixable by framerate=Hz). If not good enough, you need Overdrive Gain tuning such as XG2431 explained at www.blurbusters.com/xg2431 -- this will allow it to outperform 120Hz motion clarity on BenQs.
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: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 23 Nov 2021, 15:12

Part 2 of 2
BFI wrote:
23 Nov 2021, 01:51
Does anyone know of a DLP projector that the PS5 and Series recognise as 120Hz in 3D mode?
It is sometimes tricky. Ideally, one can force your projector into 3D mode by going into the menus, but sometimes you need to switch to a different resolution to more easily do that. You need to first switch to a 120Hz mode, then force frame-sequential 3D on via the projector menus. Then it effectively behaves as a motion blur reducing mode for 120Hz signal sources.
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!

BFI
Posts: 19
Joined: 23 Nov 2021, 01:25

Re: DLP BFI for Consoles

Post by BFI » 23 Nov 2021, 19:02

Thanks for the replies Mark. I only game on consoles, so mostly at 60fps, which unfortunately means I can't take advantage of the 8.3ms frame persistence of 120Hz displays. I don't like the flicker on my XL2720 with single strobe at its minimum of duty 30, but find the image duplication from 60fps games with 120Hz strobing as bad as no BFI at all.

I think I'm asking for the impossible: an impulse-like display I can still buy new. Or maybe there's a used plasma I can look out for? I only need 1080p at 60Hz in SDR, and the BenQ TK700STi might be ideal for me if it had some very mild BFI to make your street map readable at 60Hz.

Are there any options for 8.3ms frame persistence (or even 4.2) without image duplication for 60Hz content? This is my ultimate question if you've only time to answer one.

Maybe there's a 32" monitor or a laser projector. And if you reply again, do you think I'd be any happier with the XG270 over my XL2720, given my dislike of BFI flicker at 60Hz? Even if I bought a nice 240Hz TK700STi and a powerful PC, the games I play are typically locked to 60fps anyway.

If anyone has any thoughts on the Sony A80J, I'd also appreciate those. Particularly the results of smoothness 3 with clarity 1 or 2. Cheers! :)

[edit]

Another quick question for Mark: have you ever approached ViewSonic about adding BFI to their PX701? If it had a low, medium and high option I'd order one straight away!

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

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 23 Nov 2021, 22:33

BFI wrote:
23 Nov 2021, 19:02
Another quick question for Mark: have you ever approached ViewSonic about adding BFI to their PX701? If it had a low, medium and high option I'd order one straight away!
Not currently.
BFI wrote:
23 Nov 2021, 19:02
I think I'm asking for the impossible: an impulse-like display I can still buy new. Or maybe there's a used plasma I can look out for? I only need 1080p at 60Hz in SDR, and the BenQ TK700STi might be ideal for me if it had some very mild BFI to make your street map readable at 60Hz.
Are you aware of these existing large-format impulsed displays with low persistence? Some which supports consoles.

Optoma UHD50X 240Hz DLP projector: 4.2ms persistence (at 240fps)
You don't need BFI to get 4.2ms persistence, you simply need to run 240fps 240Hz for strobeless blur reduction, with 1/4th the motion blur of 60Hz. That does require a powerful GPU and you can't run console games at 4.2ms persistence non-strobed (since consoles only go up to 120Hz = 1/120sec persistence = 8.3ms)

LG CX OLED Televisions From 48 Inches to 75 Inches: 4.2ms persistence
These supports consoles too! They can do 4K 60Hz (12.9ms persistence and 4.2ms persistence options) and 4K 120Hz BFI (4.2ms persistence). For the best large-format impulsed display I recommend the LG CX OLED. These specific OLEDs have a black frame insertion feature that reduces motion blur by 75% -- they have only 4.2ms frame persistence for both 60Hz and 120Hz content at the maximum BFI settings. The C9 and C1 OLEDs cannot do this.

Oculus Quest 2 VR Headset and Valve Index VR Headset: 0.3ms persistence!!
This is more PC only, but so superlative that this must be mentioned here. The other option is an Oculus Quest 2 VR headset -- er, Meta Quest 2, but I have a fondness of "Oculus"... -- as it is one of the best impulsed LCDs I have ever seen in my lifetime, with less motion blur than a CRT tube! You can load BigScreen app, which puts an IMAX screen right in front of you, you even can look down and see VR cupholders, and then be staring into an impulsed display that's more than 10 feet in diagonal (in a Holodeck like space). I give a total standing ovation at one of the most perfect zero-blur LCD ever invented, the Valve Index VR LCD, and the Oculus Quest 2 VR LCD.

The Quest 2 is higher resolution which makes it much higher quality for streaming a 1080p computer desktop screen within a 4K virtual computer-room environment (Holodeck style rendering of a computer room or a home theater room in the "Virtual Desktop" app).

I realize if you want this LCD, you must accept the Facebook ball-and-chain attached to your leg, and pull it along with its privacy considerations. But, John Carmack did a superlative job with a perfect 256x256 GtG100% heatmap, something I almost never see on any LCDs, for perfect zero crosstalk -- Quest 2 has none, zero, nada, zilch, and achieves true real world 0.3ms MPRT persistence with zero crosstalk and zero duplicate images, top/center/bottom is perfect when viewing www.testufo.com/crosstalk in the in-VR Chromium-compatible Oculus web browser.

With some refresh rate overrides found in some apps or via SideQuest -- you can run Android apps, or movies, or non-VR PC games (on a virtual monitor), or Remote Desktop at 60Hz, 72Hz, 80Hz, 90Hz or 120Hz to play your 2D and 3D (NVIDIA 3D Vision, but better, via VorpX) games on a large virtual computer monitor that's more gigantic than your real computer monitor. Streaming PC to VR does add a bit latency, but usually less lag than the average latency of an average TV set, anyway, and the average latency of a DLP projector...

It is also much more eye-friendly than Real3D and Disney3D too (gimmick glasses). Just stick to "Comfortable" rated VR content, and stay away from the VR roller coaster apps, and you'll find it amazingly friendly low persistence that's better than everything you've seen recently. Want a 100 inch 90Hz or 120Hz CRT tube in front of you, less blur than a CRT? You got it with the Quest 2 LCD.

Now, some people I know got the Quest 2 as a stand-in for a 60Hz CRT tube, you do need to download SideQuest, and force-configure 60Hz. You can even sideload Android apps onto Quest 2 -- Android apps show up as a floating virtual 100 inch screen in mid-air.

Being 6dof, you can tilt your head, look under, look behind screen, even stare under the virtual computer desk of your virtual desktop running Windows Desktop (being streamed from your PC to the 3D world within the Quest 2 LCD, and texturemapped onto the virtual monitor at full framerate).

Many VR newbies avoid VR because they think they will get dizzy, or they ran a very bad VR headset with a very bad app. The Quest 2 is highly refined next-generation stuff. It's simply app selection. Moving and tilting your head behaves 1:1 symmetric with real life, so you don't get the dizzying vertigo problem of fake 3D systems that are not modern 6dof VR. You don't have to play VR games to enjoy a VR headset's capabilities. You can always use VR headsets only with those Remote Desktop apps, movie-watching apps, and virtual teleprescence (e.g. sitting on a seat on a virtual beach, or flying a kite, or such) -- those generally never make people dizzy like a VR rollercoaster app you might've seen at a VR arcade.
BFI wrote:
23 Nov 2021, 19:02
Maybe there's a 32" monitor or a laser projector. And if you reply again, do you think I'd be any happier with the XG270 over my XL2720, given my dislike of BFI flicker at 60Hz? Even if I bought a nice 240Hz TK700STi and a powerful PC, the games I play are typically locked to 60fps anyway.
If you want zero blur at 60Hz without double images, you have to tolerate 60Hz flicker.
  • Don't use 60Hz with bright Windows desktop
  • Use 60Hz BFI only for motion content (not static content), where motion blur sometimes create more eyestrain (for some) than flicker
  • Reduce brightness when doing 60Hz BFI
  • Increase viewing distance when doing 60Hz BFI
With proper 60Hz BFI hygeine, it's not as terrible when you're needing to motion-blur-reduce retro content.
However, if you can increase frame rate to 120fps 120Hz, that is even better.
You do desire framerate=Hz for the most perfect possible low-persistence operation without double image effects.

So you are SOL for flicker-free 60Hz low persistence.
Because persistence = frametime on sample-and-hold displays.
60fps = 1/60sec = 16.7ms persistence without BFI (even at 120Hz and 240Hz too)
120fps = 1/120sec = 8.3ms persistence without BFI (even at 240Hz too)
240fps = 1/240sec = 4.2s persistence without BFI

Fast-GtG displays will generally appear to linearly follow persistence=frametime behavior of Blur Busters Law (www.blurbusters.com/1000hz-journey ...) so getting a good 120Hz DLP will automatically give you 8.3ms persistence without BFI for 120fps content. The rule of thumb is frametime generally dictates motion blur on a sample-and-hold display (ignoring stutter effects of framerate/refreshrate non-divisible mismatch on non-VRR displays).

It's a law of physics like speed of light, so you're SOL for 60Hz content.
Strobed: You gotta tolerate 60Hz flicker
Nonstrobed: You gotta tolerate 16.7ms of blurring

Future displays may be able to emulate phosphor fade (e.g. 1000Hz displays using 16 refresh cycles to simulate phosphor fade effects per 60Hz refresh cycles), to soften 60Hz flicker, so there will be some goldilocks compromises coming in the future, but for now, it's beyond the scope of displays on the market at the moment.
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: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 23 Nov 2021, 22:51

TL;DR: OLED Burn-In Hype Is Overblown After A Few Easy Tips And A Colorimeter
BFI wrote:
23 Nov 2021, 01:51
I don't want OLED because I leave static images on mine a lot, but I'd consider anything else with adjustable BFI.
OLED burn-in is way overblown. OLEDs stop burning-in fast when BFI is permanently enabled. The LG CX OLED burns-in much slower with the dimness of perpetual BFI operation.

Moreover,
  1. LCDs degrade too. The backlight of an LCD becomes 50% dimmer after about 3-5 years of heavy use. In addition, due to liquid settling effects on large-format LCD panels, some screens also get very splotchy (inconsistent grays), like splotches that grow bigger on an old VA LCD in a 10% or 20% gray field.
  2. Dark Mode + taskbar autohide + orbit operation can turn annoying burn in into much hazier wear-and tear that's slightly less than LCD wear-and-tear. Use a very animated screensaver, to better wear-level the burn-in. So when burn-in finally appears years later, it just looks like faint hazy splotches.
  3. Enabling BFI darkens the OLED to the point where burn in is much slower. There's a geometric slowdown in burnin involved. OLEDs running at half brightness tend to burn in more than 4x slower, and OLEDs running at quarter brightness tend to burn in more than ~16x slower. It's not exactly that perfectly inverse square relationship, but it's very sharply inversely exponential kind of like that - the burn in curve sharply falls off beyond a human gadget generation (e.g. 5year to 10year ownership). At some point, the OLED lasts more than an order of magnitude longer without burn in. At some point, OLED burnin becomes slower than LCD backlight wear-and-tear (especially with strobe-brightness-enhancing voltage-boosted strobe backlights that may accelerate brightness wear-and-tear; a forum user reported that a 7 year old VG248QE had fallen very dim after long-term LightBoost use)
  4. I've seen airport LCD displays with burn in too (permanent image retention from displaying static images for years nonstop), so LCDs are not completely immune to burn in effects. Most of the time it's far more temporary on LCDs, but LCDs are not completely immune. It's much easier to erase burn-in on LCDs, but sometimes they even go permanent in extreme cases (airport arrival/departure LCDs displaying high contrast static stuff in the same locations for multiple years).
  5. Do a few minutes of googling and you see zero burn-in OLED threads like this (by the thousands of threads in hundreds of different forums and websites) throughout the whole Internet, from both amateurs and professionals. I literally copy and pasted that link in just 15 seconds flat of kindergarten-level Google-Fu.
RTINGS OLED burn in contest is definitely useful science, but it's like showing off a scary bonfire demo to scare people away from ever using a small cigarette lighter or matchbook to light a candle.

Lots of people just want to drive OLEDs very brightly, which speeds up burn in geometrically. Reducing OLED brightness by 50% can turn 1 year burn in into 4 year burn in for example. The curve varies a lot, but it's quite sharply geometric for some of them. So I've seen 1 year OLEDs with burn in and 7 year old OLEDs with no burn in. I've seen OLEDs with ZERO burn in after 3 years, and I've seen LCDs with splotching appearing after just 1 year. The venn diagram overlaps. Yes, it's easier to abuse, overdrive and destroy an OLED, but you don't have to.

As long as you're calibrating your OLED to average ~100nits (strobed or nonstrobed) then you can practically run it tens of thousands of hours and have zero visible burn-in on the OLED. Even with image orbit turned off!! If you don't know the brightness, just spend $100 extra on a color calibration sensor and adjust your OLED to practically burnin-proof levels, knowing your USB-connected computer sensor told you reliably that your bright whites were only 100 nits in an easy-to-launch app... The price of safeguarding your $2000 OLED, pony up that insurance fee for low persistence nirvana...

You can still occasionally special-treat yourself by boosting to true HDR operation (nonstrobed, though) when watching movies or playing certain games with HUD disabled (like Final Fantasy), that won't be a problem to boost the gamut occasionally to OLED deliciousness. Then go back to safely reduced brightness SDR mode with strobing for burnin-proof operations of an OLED showing Windows.

I know a few Visual Studio developers use OLEDs (48" and 55" 4K OLEDs on the wall at the back of a deep 4 foot computer desk) and report no burn in if adjusting brightness down. It's like a matrix of 4 1080p PC monitors in multimonitor mode in a 2x2 grid, except it's one uniform panel. So it's not too big nor text too small, if you optimize for an approx 1-meter / 3-to-4-foot view distance from LG's smaller OLEDs, and use approximately 150% DPI zoom (give or take, choose your preferred DPI). Everyone has their preference, but judicious setup of LG OLED makes it a wonderful giant computer monitor too. You may want to enable minor image orbit though, but it's a treat to the eyes.

Other forums are full of advice, and thousands of people are reporting zero burn in on their OLEDs because they followed these instructions.

WIth due respect, are you serious about avoiding OLED? Tips on avoiding burn in on OLED is easier to learn than riding a bicycle. You *don't* have to steer your bicycle intentionally into a car or a curb. Learning how to configure the OLED for avoiding burn in isn't that hard, and if you're scared, get $100 equivalent of an insurance policy (aka "Colorimeter") to literally wear-speed-limit your OLED pixels with a nit cap, and call it a day, no?

Also, an OLED is cheaper than a car, but easier to avoid burn-in for a few years on than crashing a car into a wall once they purchase a colorimeter (your equivalent of a driving license in terms of staying out of burn-in jail). Yet people are more scared to buy an OLED for computer use, than purchase a death-machine-on-wheels? Take a driving course if you want to drive a car. Or buy a colorimeter and limit your OLED to 100 nits if you want to avoid OLED burn in. Duhblankstare.

(Most CRTs back in the day didn't go above 100 nits anyway...)

Winky wink ;)

TL;DR: OLED Burn-In Hype Is Overblown After A Few Easy Tips And A Colorimeter
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!

BFI
Posts: 19
Joined: 23 Nov 2021, 01:25

Re: DLP BFI for Consoles

Post by BFI » 24 Nov 2021, 01:52

This is all great info Mark - thanks for the effort. :D

I realise I have to choose between flicker or 16.7ms persistence with 60fps games, but on console I don't expect 120fps games to arrive for years, if not decades. The nice thing about the response of my DLP is that motion feels much smoother than anything other than a CRT, but with its sample-and-hold display I still see motion blur with saccadic eye movements.

I'd love to see ViewSonic or any other DLP manufacturer add a BFI option, ideally one where you can set how long and often you want the mirrors to hit the beam stop for. I think that seems a better method than blocking a segment of the wheel, but I could be wrong. Blocking the wheel might be easier.

I have just tried my XL2720 on single strobe with brightness at 100 and strobe duty at 010 (quite dark), and do find that easier on my eyes than duty 30 with brightness at 0. Phase is 100 as per Falkentyne's recommendation.

BFI implementations on TVs seem undocumented and lacking in granularity, where if 60Hz is too flickery for my liking there's typically no adjustment. I've read that Sony's A80J outputs 120Hz on settings 1 and 2, and 60Hz on 3. And you have to have interpolation on 1 (of 3) to use this feature. I believe 60Hz BFI has to be enabled for 1080+ lines of motion resolution (I know you prefer MPRT but I can't find a figure), else it drops to an unacceptable 650, and as low as the base 300.

Because I'd be happy with 1080p SDR with impulse-like 60Hz, I feel it shouldn't be this hard to find a screen! The TK700STi and PX701 have everything except BFI features, and I'm sure I can't just set them to frame-sequential 3D and output my PS5 at 60Hz? I'd jump at one if I could. The Samsung QN95A is also reported to be too flickery for most people at 60Hz, again with no adjustment.

Do you know anything about the rolling scan implementations in newer OLEDs? If it only increases input lag in the single figures I'd hope to see it become common. I don't understand this race for brightness! I don't even like HDR, but that's me.

So it's maybe looking like a Sony A80J as my best bet, since OLED comes closest to matching DLP's responsiveness for smooth motion. I'd really like to know its MPRT figures at each adjustment level under Smoothness and Clearness, but I can't even confirm it's using a rolling scan method, nor its input lag figures under each setting. It's so much money for something I'd use for years when its motion performance isn't well documented. For gaming, at least.

I'd like everything but the persistence on a newer PJ, but 240Hz is wasted on me in terms of framerate output. Why can't console games just be 120fps?!

BFI
Posts: 19
Joined: 23 Nov 2021, 01:25

Re: DLP BFI for Consoles

Post by BFI » 24 Nov 2021, 02:41

Quick follow-up question (and then I'll try and stop bothering you!): would an HDMI switch with selectable 60Hz BFI output be feasible? I don't know how sensitive displays might be to tripping with momentary blanking applied to the video wires.

I think such a device at 100 USD would sell really well, giving all console gamers the option of a few strobe duties. Press a button to cycle between 5, say.

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

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 24 Nov 2021, 03:05

Yes, BFI implementations on TVs have been crappy. RTINGS does among the best job figuring this out.
BFI wrote:
24 Nov 2021, 02:41
Quick follow-up question (and then I'll try and stop bothering you!): would an HDMI switch with selectable 60Hz BFI output be feasible? I don't know how sensitive displays might be to tripping with momentary blanking applied to the video wires.

I think such a device at 100 USD would sell really well, giving all console gamers the option of a few strobe duties. Press a button to cycle between 5, say.
Theoretically a box doing software-based black frame insertion is possible but persistence would only be adjustable in refresh cycle granularitty, rather than sub-refresh MPRTs.

Like www.testufo.com/blackframes -- Software BFI emulating 60Hz on a 120Hz LCD will only reduce motion blur by up to 50%. There are some issues with this interacting with LCD voltage inversion algorithms (image retention) but there are also workarounds for that.

Now it does scale far beyond -- a box can kind of emulate a CRT using a 1000Hz sample-and-hold display. (Much like the reverse version of playing a 1000fps high speed video of a 60Hz CRT). This would be essentially a video processor box equivalent of this RetroArch feature request that I added.

It will be a long-term progression, but I wouldn't expect a box like this anytime soon. Such a box in small-batch hobby manufacturing run will cost well over 100 dollars per unit.
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: 11648
Joined: 05 Dec 2013, 15:44
Location: Toronto / Hamilton, Ontario, Canada
Contact:

Re: DLP BFI for Consoles

Post by Chief Blur Buster » 24 Nov 2021, 03:18

BFI wrote:
24 Nov 2021, 01:52
I'd love to see ViewSonic or any other DLP manufacturer add a BFI option, ideally one where you can set how long and often you want the mirrors to hit the beam stop for. I think that seems a better method than blocking a segment of the wheel, but I could be wrong. Blocking the wheel might be easier.
BFI on DLP is achieved simply by turning the DLP pixels turned off for. It's not done via the colorwheel.

Since DLP generates color depth temporally, you do have a color depth tradeoff, where 50% BFI will halve color depth of DLP, creating more contouring artifacts.
BFI wrote:
24 Nov 2021, 01:52
Phase is 100 as per Falkentyne's recommendation.
It depends on
BFI wrote:
24 Nov 2021, 01:52
I believe 60Hz BFI has to be enabled for 1080+ lines of motion resolution (I know you prefer MPRT but I can't find a figure), else it drops to an unacceptable 650, and as low as the base 300.
1080+ lines of motion resolution (it's vertical lines on a horizontally scrolling test pattern) is now almost meaningless to my brain; I don't even remember how much blur that is. It scales really wonky at different motion speeds and different resolutions, while MPRT is motion resolution independent.

Very approximate:
300 translates to very roughly 16.7ms MPRT
600 translates to very roughly 8.3ms MPRT
1080+ translates to about anything very roughly 4ms MPRT or less, not possible to know what MPRT number is below approximately 4 milliseconds, because "1080+" caps out.

1ms MPRT is like having 4000+ lines of resolution on a 1080p, which is a non-sequitur because that is beyond resolution of 1080p in both X and Y axis. So, I'm really truly not a fan of lines of motion resolution because it caps out at arbitrarily low MPRTs, because they tuned to more-motion-blurry plasmas so that they "seem" to benchmark as close to a CRT. So, I hate these motion resolution benchmarks that caps out at ~4ms MPRT, when my human eyes can tell apart 0.25ms MPRT and 0.5ms MPRT already (in certain motion tests).

The absurd amount of syntheticness crammed into these old "lines of motion resolution" benchmarks, originally in the analog era, falls to complete failure in this Hz-agnostic, resolution-agnostic era, where "lines of motion resolution" is meaningless if Hz changes or resolution changes, or when it caps out more than an order of magnitude weaker than human vision acuity (when it comes to low-MPRT strobed 4K displays).

Totally, I just put my nose up on these obsolete motion resolution benchmarks that has rotting mold and stink, since it caps out more than an order of magnitude above my motion-clarity sensitivity threshold (strobed 4K at 0.25ms MPRT).

So rheoretically, I ask the forum audience, why quote motion resolution benchmarks that cap out at 4ms MPRT, when my eyes can see 0.25ms MPRT versus 0.5ms MPRT!? I may add a block ban-word filter on those old motion benchmarks just to get those old motion resolution benchmarks to goddamn die faster; that is how much I despise those old obsolete motion resolution benchmarks. I even tell forum members to tell other people in other forums to stop using these old obsolete motion resolution benchmarks, when posting about them in other forums. Just stop. ;)

MPRT(100%) follows Blur Busters Law, so:
0.25ms MPRT is 1 pixel of motion blur at 4000 pixels/sec
0.5ms MPRT is 2 pixel of motion blur at 4000 pixels/sec
1ms MPRT is 4 pixel of motion blur at 4000 pixels/sec
That's one screenwidth per second panning on a 4K display.

Real MPRT is 10%->90% as per www.blurbusters.com/gtg-vs-mprt but I like to use MPRT(100%) because it translates to a very simple linear Blur Busters Law meth (1ms = 1 pixel of motion blur per 1000 pixels/sec).

My works is now cited in more than 20 peer reviewed research papers as already mentioned elsewhere. Even Samsung now cites TestUFO Making Of: Why Are TestUFO 960 Pixels Per Second, in their research paper 57-2: A New Approach to Motion Frequency Metrics Quantifies Motion-induced Blur, recognizing the modernization of motion resolution benchmarks.

Interesting fact: 0.25ms MPRT without strobing requires 4000fps 4000Hz, so the retina refresh rate is really, really high when it comes to wide-FOV high resolution displays. To do motion blur reduction in a completely strobeless manner (since MPRT(100%) is frametime on a sample and hold display). However, for strobed 8K displays, MPRTs as low as 0.1ms becomes human distinguishable from 0.2ms in curated TestUFO test patterns I've created. Now, THAT will require 10,000fps at 10,000Hz to do strobelessly. Great example of the Vicious Cycle Effect where ever-bigger wider-FOV higher resolutions amplify MPRT limitations. The assumption is made, that individual pixels are still visible, but the fact is displays have become bigger and bigger over the years, from 15" CRTs to 21" CRTs to 24" LCDs to 40" TVs to 75" TVs, so that is now propelling more increases in entry-level refresh rates like more widespread use of 120Hz.
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