Separate names with a comma.
Discussion in 'Displays' started by Mark Rejhon, Jun 11, 2013.
1080P and burn in issues, no thanks!
Plasma TVs like the Panasonic VT50 are not fit to display even 60 FPS of fast motion, let alone higher. You get all sorts of weird stuff, like warping effects due to the way the subfields work. They're designed and optimized for 25-30fps Television/Cinema where panning speeds are strictly controlled. Even consoles are generally running at lower framerates and there's no rapid (twitch) panning thanks to the controller handicap. You feed these TVs 60fps of ultra smooth mouse-driven PC motion, and they're well out of their depth.
Add to that the retention, dithering, etc. etc. - I tried really hard to make Plasmas work as PC gaming monitors, but it didn't take me long to realise I was pissing in the wind. Increasing the refresh rate wouldn't have made it any more viable.
Skakruk:You're still spewing that BS I see. It's is not that noticeable in-game unless you're one those people who are more interested in image artifacts than the game. Not to mention, LCDs have far worse problems (as most LCD TVs are PWM based) and OLED isn't affordable yet.
For the record, Samsung plasmas have better motion than Panasonic PDPs (but with elevated black levels).
Vega: Modern plasmas are extremely difficult to burn-in. I leave static pictures on all the time. It's only an issue on LG plasmas.
1. Viewing distance. Sitting computer-monitor-proportional distances (3 feet from 42" plasma) can cause artifacts to be easier to see.
2. Plus (see below), sensitivity to the specific types of artifacts that plasma has, versus sensitivity to the specific types of artifacts that LCD has.
Some people are color blind. (reportedly ~8% of population!). Some people are not.
Some people have better hearing. Some people have poorer hearing.
Some people are more sensitive to motion blur. Some people are not.
Some people get headaches by looking at a display. Some do not.
Some people are sensitive to intra-frame temporal effects (rainbow artifacts on DLP, subfields artifacts on plasma). Some are not.
Some people are more uncontrollably distracted by certain display artifats. Some people are not.
So, if you're all for equality in diversity of humanity, and your sentence is clearly an assumption someone is "interested" in artifacts; some people uncontrollably see them as a distraction (not too unlike a high-pitched noise screaming annoyingly into your ears, while others can't hear the noise at all). It's THAT annoying to a small proportion of people. Some artifacts are THAT uncontrollably annoying to some human eyes, just like a high-pitched coliwhine. Or a symptom such as a PWM-dimming headache you get that others cannot get, etc. Vision is NOT THE SAME among all humans.
For my eyes (when excluding colors), the LCD look is much, much more pure/natural than plasmas. I am sensitive to high-frequency elements of a DLP and plasma (I can see DLP rainbow artifacts easily & I can see plasma subfield artifacts easily). However, accounting for colors, plasmas are certainly better. This is not a lie for my own specific set of human eyes (and many others), even if not your set if eyes. This is a lot of human eye's preference.
I don't quite remember, but I think this was the muppet who claims to be a reviewer, yet has all the discernment of my grandmother when it comes to identifying flaws in the displays allegedly being reviewed. What next, shall we have the hearing impaired reviewing Hi-Fi gear?
Newsflash: This is Hardforum, an enthusiast website. Generally, you can expect the people who post here have a keen interest in things like image artefacts.
Well I have testufo working on the latest Opera and I have some crazy discovery to share
First test was tracking the blurtrail line by eye (Running 1440x1080p @ 100hz/100fps on TH-42PF30U)
Here is the interesting bit
If your eye isn't moving there is ZERO motion blur, the blurtrail, UFOs and every other graphic remains as sharp and clear as any static image at any speed including 6400pps!
I can reproduce this even at 60hz, the only purpose higher refresh rates serve is reducing eye-tracking based blur
Sample and hold blur exists and is a bane of displays. It is related to your eyes and movement, that is why you have to use a pursuit camera to get an approximation of what the blur looks like to your eyes. In real world screen usage, you are going to get sample and hold blur constantly in FoV movement based games (1st/3rd person cgi world perspectives) unless you stare unblinkingly at one point in the middle of the screen and never even judder your eyes slightly at any angle, which is impractical if even possible(I doubt it).
I completely disagree that the only purpose of high hz (input+output) combined with high fps is reducing eye tracking based blur.
While i haven't tested the VT50 for pc use, almost none of that is true at all for the VT60.
It displays flawless blur-free 120hz if you put it in 3D mode (2D-3D conversion, or feed it 60hz side-by-side which turns into 120hz).
You can count the little windows on the castle at the fastest speed ( Mark's moving photo http://www.testufo.com/#test=photo&photo=quebec.jpg&pps=3840&pursuit=0 ) , and the picture looks about as crisp as lightboost. It is also possible at 3000pixels/s at 60hz, although it's certainly not perfectly crisp.
It is, however, far better at motion @60hz than any 60hz IPS/VA out there.
Seeing dithering simply means you are sitting too close. Minimum distance is ~2m for a 50".
The only issue is the little rainbow artifacts (which i perfectly understand could be a dealbreaker for some), but again, they are very rarely visible unless far too close to the screen.
That, and the 60hz flicker because of short persistence/60hz general choppiness issue until/if the refresh can be raised.
Burn in seems about as likely as a CRT these days, as in if you really really mistreat it, it can happen.
I haven't even seen any retention yet on mine.
Real deal breaker on those plasmas is that they don't make 30" variants
Its not a deal breaker if you can get a room that has enough space to sit further away.
Its more comfortable and better for your eyes too.
oops I meant it's the only solution not purpose
How did you go with gaming so far? fyi there should be no dithering on Cinema/True Cinema/Pro1/Pro2 picture modes
Gaming has been fine on it, but i have a hard time standing 60fps for fast motion.
Haven't had much time to experiment more with timings since I have been very busy lately.
I've mainly watched movies/series on it, the 120hz 3d is excellent (better than 3d lightboost, digital cinema, or anything else i've seen), and movies and series are great in 60fps 2d as well.
I have zero issues except the 60hz choppiness for gaming and the 60hz flicker for whites/bright colors. The quality in all other regards is, as most reviewers also seem to agree, reference quality.
Even if I can't get the refreshrate up, I'm still very happy with it for movies/series. We will see.
I know what you mean, I simply run 1440x1080 for certain games because 100hz is a big improvement for shooters and the like. I'm holding out for a VT70 with display port but the future of Panasonic plasma is uncertain with only 2% market share or something
I think you misread my post. I never even mentioned blur.
No, I'm afraid this is bullshit in my experience. Maybe some people are less sensitive to it, but I tried a range of viewing distances and if I was to increase my distance from the TV far enough to make the dithering unnoticeable, the effective size of the screen would end up being 2/3 the size of my 23" monitor. The distance required to eliminate dithering was just too great.
And that was 'best case scenario' dithering. In 'gaming mode' the dithering was more pronounced, so you were forced to choose between lower processing latency and less dither.
That's television view distance. Not suitable for some of us.
Common proportional view distance is approximately 1:1 for a primary computer monitor.
24" monitor = common desktop view distance of ~24" (e.g. 24" monitors)
50" monitor = common desktop view distance of ~50" (e.g. HDTV mounted on wall at back of deep desk)
In fact, many people sit closer to the SEIKI 50" (wall mounted at the back of their desk, 3 feet away from their head), because of the pixel density of the SEIKI. From a computer monitor usage perspective where you want to use it as a primary monitor for everything, I don't necessarily buy the "too close" stuff on every use case; this is why plasma becomes flawed as a computer monitor. You're forced to sit further back just to gain the full plasma benefits, which is unacceptable for many if they like to use Visual Studio, PhotoShop, and videogames, on the same TV panel as a computer monitor. As a result, people who are sensitive to plasma artifacts, means plasma won't let you sit close enough comfortably without seeing plasma artifacts. It automatically disqualifies plasma for many people who want to scale the angular coverage proportionally to a computer monitor. Plasma is great when it fits your use case. For desktop usage, I find close viewing of LCD motion blur (of non-LightBoost) is more comfortable than close viewing of plasma artifacts. Your mileage will vary based on your use case.
When people sit that close to a 50" 4k monitor, it is a completely different scenario.
They are effectively using that 50" 4k in the same way and the same PPI/viewing distance as you would use 4x 25" monitors mounted together in a stand, it can't be compared.
When comparing a 50" 1080p to a 24" 1080p, it's a different thing.
The 50" is more than four times the size.
I am currently using only the 50" for a while to see how it works, and sitting 50" away is way to close.
Around 2m for a 50" seems to give the same experience as using a 24" at normal viewing distance.
Of course it entirely depends on the user. I had to wait for the phosphor trails and flicker to be reduced before I could own a plasma. I was entirely happy with LCD until PWM trailing became an issue (along with uniformity, elevated black level and viewing angle issues due to LED backlighting). Based on my reviews and my colleagues, I feel that LCD TVs haven't improved since 2010 (and in some cases, they worsened). Back in 2010, I reviewed a 40C580 CCFL-LCD by Samsung. It retailed for ~ £380 (a low end model) and had ANSI black level of 0.03 cd/m2 (peak brightness was set to 120 cd/m2). To this day, we have yet to review a LCD TV with such black level. In fact, some have elevated to high as 0.07/0.08 cd/m2. Viewing angles have also worsened (possibly due to LED backlighting) and motions hasn't improved at all. Without MCFI, LCDs are only capable of 300 lines of motion. It seems manufacturers have ceased improving LCD TVs and are focusing on design, cost and features. Considering money is being diverted to OLED and low-cost, high-efficient LCD, I guess this shouldn't be a surprise.
Panasonic plasma is definitely not ideal for PC gaming. Aside from the obvious issues, there's also the 1.5-2 frame lag. My statement was mainly aimed at Skakruk, who seem to think warping is a massive issue.
PS: Samsung plasma could be ideal since they have very clear motion and very little flicker. But it's processor is questionable though.
I never claimed to be a reviewer. I do actually review for HDTVtest. Here's a little teaser:
Every display has it's faults. Some directly impact the quality of viewing, some are minor. Warping falls under minor. The effect has to be observed in order to be seen. In other words, you have to know when and where it would occur and track its movement.
I've played countless games on my 42G30 (which has far more flaws than 50 and 60 series) and I haven't seen warping during game-play (60p) or when viewing high-motion video. There are occasions where distortions can occur, but its not something I would personally fuss over (especially knowing that the alternative have far more issues).
Even at TV viewing distances the dithering was easily visible on the 50 series plasmas.
Bollocks. How do you think I discovered the bloody issue in the first place? Did I go scanning the set for a phenomenon I didn't even know existed? No. I was playing a game and noticed the vertical boards on a picket fence were curved during pans. Later, I saw straight pillars warping in the same way in another game. Then l noticed that a bright slab of rock on a dark background appeared to be floating when the scene was in motion. Then in F.E.A.R, a game full of sharp contrasts and rich shadows, I saw whole scenes appearing warped/disjointed. THEN, and only then, did I decide to try and figure out the cause. At this point I was ignorant to the mechanism of subfields.
Actually I'm already familiar with it
That's how I created www.testufo.com/eyetracking
It's proof that speed of LCD GtG pixel transitions aren't the limiting factor on modern LCD panels, but the sample-and-hold effect.
Modern LCD's have GtG transitions that are a tiny fraction of a refresh. The remainder of motion blur is forced upon your eyes caused by eye-tracking across high-persistence (scientific references).
Sample-and-hold effect is always crystal clear if you don't move your eyes, but creates motion blur when you move your eyes. Your eyes are continuously moving, even while static frames are displayed during milliseconds. Your eyes are in a different position at the beginning of a refresh than at the end of a refresh, causing the statically-displayed frames.
Long persistence phosphor creates a sample-and-hold effect. A plasma with 5ms of phosphor decay will generate approximately 5 pixels of motion blur during 1000 pixels/second, as an example. The phosphor decay curve does soften this somewhat, as eye-tracking motion blur becomes linear-style (brief strobes) or gaussian-style (gradual strobe with fadein-fadeout). Some CRT tubes have ultrafast short persistence phosphor that mostly decays in only 0.1 seconds. That means ultrafast eye motion is as clear as a static image, with no perceptible motion blur (~0.1 pixel of blurring during 1000 pixels/sec would not be noticeable).
This is exactly how the graphic works at 60Hz vs 120Hz vs LightBoost
When doing a motion at 1000 pixels per second, the milliseconds translates to pixels of motion blur (1 pixel = 1 millisecond during 1000 pixels/sec). TestUFO uses 960 pixels/sec as the default target motion speed, so this is very close to the easy "1 ms = 1 pixel". Therefore, this graph means approximately (+/- 0.5 pixel):
60Hz LCD = ~16.5 pixels of motion blurring
100Hz LCD = ~10 pixels of motion blurring
120Hz LCD = ~8 pixels of motion blurring
144Hz LCD = ~7 pixels of motion blurring
LightBoost 100Hz @ 100% = ~3 pixels of motion blurring
LightBoost 120Hz @ 100% = ~2.5 pixels of motion blurring
LightBoost 100Hz @ 50% = ~2.5 pixels of motion blurring
LightBoost 120Hz @ 50% = ~2 pixels of motion blurring
LightBoost 100Hz @ 10% = ~2 pixels of motion blurring
LightBoost 120Hz @ 10% = ~1.5 pixels of motion blurring
It's similiar to photography: Longer shutter speeds (longer strobe flashes) will create more motion blur (photograph vs display) in a moving sensor (moving camera versus moving eyeball). The motion blur mathematics is identical.
This is quite obvious, since it is consistent with what is seen at www.testufo.com/ghosting at 960 pixels/sec
-- 60Hz; about 16 pixel-widths of motion blurring
-- 120Hz; about 8 pixel-widths of motion blurring
-- 120Hz LightBoost; 1.5 pixel-widths of motion blurring
The 100Hz LightBoost example would be just very, very slightly blurrier than the last photo above (one added pixel of motion blur, maximum), but much sharper than the 120Hz middle photo. The 144Hz non-LightBoost (7 pixels of blur) would be just very, very slightly sharper than the 120Hz middle photo (8 pixels of blur). Just view www.testufo.com/ghosting -- the motion blurring you will see, will be very similiar to one of the above photos. (Different monitor response curves and overdrive settings will affect things such as ghosting and coronas, so your monitor may not match my monitor and overdrive settings exactly, but the sample-and-hold motion blur is unaffected)
The display is often contributing additional (often unwanted) eye-tracking motion blur that is above-and-beyond natural human limitations. So people like us, who dislike motion blur, want to eliminate display-enforced eye-tracking-based motion blur where possible.
The ideal display has EXACTLY the same motion clarity during fast-moving images, and during static images. With LightBoost=10% and VSYNC ON during framerate=Hz motion, fast motion has darn near ZERO motion blur (almost unnoticeable), like a piece of paper moving past your face and your eyes tracking it. No display-enforced eye-tracking-based motion blur caused by the sample-and-hold effect of finite framerates. Basically, to reach Holodeck perfection, you want a display to have less forced unwanted eye-tracking motion blur (above-and-beyond human eye tracking limitations/saccade inaccuracies), so that a panning motion on a display looks exactly the same as physical panning (e.g. piece of paper moving sideways in front of you). Above a certain limit, you can't track fast motion anymore, but a "Holodeck" display should never force unwanted eye-tracking motion blur upon you.
Good Valve Software article discusses eye-tracking motion blur (Michael Abrash), you will now understand this paper better with your new enlightenment.
Michael Abrash: Down the VR rabbit hole: Fixing judder
Imagine wearing VR googles. Imagine turning your head. You get one big display motion blur mess. You can't stare at things while you're turning your head. You can read text on this forum while nodding your head. But you can't read text on a wall in virtual reality while nodding your head, because of the display motion blur on the VR goggle's screen. Can become nauseating and headache-inducing. It's not Holodeck perfection. You need either 1000fps or strobing to fix the VR googles panning motion blur problem. See, you _have_ to fix tracking motion blur for many use cases (FPS gaming, VR goggles, sim racing, and many other situations where motion blur is heavily unwanted).
The perfect display is the infinite-framerate, infinite-resolution display, so there's no strobe effects, no flicker, no sample-and-hold effect (externally-forced eye tracking motion blur), no aliasing effects, no pixellation effects, etc. A perfect zero motion blur display showing fast panning motion (e.g. www.testufo.com/photo ...) should be as perfectly clear as a display being physically pushed smoothly sideways; or turning while holding a tablet/book in your hands. (Like John Carmack demonstrated on YouTube 12min00s). But such displays will probably never happen in our lifetimes, so flicker displays (e.g. CRT, plasma, LightBoost) are a good compromise of motion blur elimination during moving images.
ChadThunder: BlurBusters XP Points +10 -- LEVEL UP!
Enlightened: Now understands the concept of eye tracking motion blur: the sample-and-hold effect
Cool, I didn't realize that.
I've talked to David Mackenzie, Vincent Teoh and and Moderator FoxHounder., who are all fans of Blur Busters now (sent a few PM's), plus a few posts. Here's one of my articles that I posted about my tests of Sony Motionflow Impulse on HDTV Test Forums
BTW, they've (HDTVTest) asked for a unit of my upcoming Blur Busters Input Lag Tester (three-flashing-square method, supports all resolutions and refresh rates; 4K and 120Hz compatible), expected during Winter 2014.
Sony Motionflow Impulse does not use MCFI.
It's 100% pure strobe based on the KDL55W905A, adds less than one additional frame of input lag, and is available in Game Mode!
It returns over 1000 lines of motion resolution on the motion tests pattern on the KDL55W905A.
Pure strobe backlights allow breaking the "300 lines of motion resolution" barrier on LCD, without using MCFI!
You may want to be aware that "lines of motion resolution" isn't future-proof.
Standardizing Motion Resolution: "Milliseconds of motion resolution" (MPRT) better than "Lines of motion resolution"
Most refined post of mine is the following:
And about the arbitrariness of "lines of motion resolution":
Maybe you're more sensitive to subfield rendering. That doesn't mean everyone will experience the same phenomenon. The same applies to phosphor traling and noise.
In any case, you should make it clear that its effects varies from individual to individual instead of stating it as a universal problem.
Blur Busters Master Listing of Known Ultra High Efficency Strobe Backlights
Confirmed 100% MCFI-free low-latency (adds under 1 frame) method of breaking the "300 lines of motion resolution" barrier on LCD's.
1. nVidia LightBoost -- www.blurbusters.com/lightboost/howto
2. Samsung "3D Mode" -- www.blurbusters.com/zero-motion-blur/samsung
3. Eizo's FDF2405W blur reduction mode -- page 15 of manual describes LightBoost-style strobing
4. Sony's Motionflow Impulse -- www.blurbusters.com/sony-motionflow-impulse
Monitor reviewers/television reviewers who are unfamilar with low-latency MCFI-free CRT-clarity LCD's, need to begin studying the science of strobe backlights, as these are gradually becoming more common, with more and more models coming onto the market, including those officially advertised as motion blur reduction backlights. Ultra-High Efficiency Strobe backlights outperform yesterday's scanning backlight which were often gimmicks (inefficient motion blur reduction), TFTCentral explains why strobe backlights outperform scanning backlights, especially for game/computer use.
Correct Neilo -- scientifically, MCFI-free methods of eliminating motion blur, tends to add more visible flicker.
Generally, for MCFI-free operation, motion blur reduction is unfortunately proportional to amount of visible flicker. (and even flicker only seen under highspeed camera)
Variables: Assuming a fixed refresh rate (e.g. 60Hz) without using MCFI, It's an unavoidable flicker-versus-motion-clarity tradeoff. You can balance it pretty well by using a slightly longer flicker that's not human-visible (softer phosphor decay), very good plasmas use approximately about a 5 milliseconds of phosphor decay, which is just about "perfect" for common video motionspeeds, but still limiting for high-end enthusiast gaming.
What this means is that a finely-tuned strobe backlight (long flashes, capacitor decay to simulate phosphor fadeout), would roughly make an LCD equal plasma in motion resolution. But it's so easy to flash a backlight briefer, to get less motion blur than plasma, too.
No plasma displays can successfully pass the TestUFO Panning Map Test at 1920 pixels/second
Only a few displays can. (e.g. CRT displays, LightBoost strobe-backlight displays, Eizo FDF2405W professional strobe-backlight LCD, etc).
This means there exists a select, choice, few LCD's that actually create less motion blur than a plasma (Without needing to use MCFI).
Statistically at one time in the past, it was measured that the average motion speed of video was approximately 6.5 pixels per frame of motion (this is how FPD benchmark came up with the motion speed of some of their tests). However, videogames don't have builtin motion blur (makes motion blur easier to see), videogames are often viewed closer than television (makes motion blur easier to see) & videogames move much faster (20-30 pixels per frame of motion during FPS, that's a full screen width at 60Hz at 1080p during fast strafing/turning) making motion blur easier to see. So the bar for motion blur sensitivity is much higher. Motion blur stands out far more. That's situations WHEN strobe backlights really sings their praises. Unlike during watching video material.
As you can see from www.blurbusters.com/about/mark -- I used to work in the home theater industry, and create video processor boxes (I invented the world's first open source 3:2 pulldown deinterlacer, found in dScaler, a PC-based video deinterlacing program that was released in the late 1990's, when Faroudja boxes still cost several thousand of dollars! It's also mentioned in "PCinema", page 38 of Stereophile Guide To Home Theater, November 2001 issue), I've got an intimate familiarity with motion blur science, and the uncanny ability to explain motion blur to the masses (e.g. LightBoost). The video processor I worked on a long ago (TAW ROCK) won 2003 Editor's Choice in Stereophile Guide To Home Theater, as it was one of the first video processors to convert 3:2 pulldown to 3:3 pulldown for 72Hz output to a CRT projector. I owned an NEC XG135 CRT projector, and became an expert at convergence, keystone, bow, astig, focus, and was able to get the beam spot size damn tight for 1080p in year 1999, long before 1080p was thought of as a standard.
I've been off for a year. With uni and now the Army, I don't have much time. Besides, I got sick of reviewing the same garbage over and over. Prettier design with performance similar to 2008/9 models. It also got too messy due to panel and processor lottery.
David reviews the all the high-end gear cos he has all the equipment. I'm not sure if I'll be reviewing any more TVs, but may help out Vincent if David were to go. Now with OLED, the differences between various sets are going to be even narrower (maybe nonexistent).
I haven't tested nor seen it, so can't really comment. My biggest issue with any kind of pulsing is maintaining sync, which can lead to multiple ghost images (frame doubling/tripling/quadrupling). I can see it working with TN since TN can respond within 8ms, but not sure about VA and IPS (especially VA since most VAs still struggle to respond within 16ms). Not to mention ambient temperature directly affect pixel response. FoxHounder did point out some downside so I'm guessing Pulse-based Motion-Flow it's still isn't good as Samsung and Panasonic plasmas. At least for now. The other issue is black level. ATM, Plasmas are 6-7 times better than leading LCD. That needs to be addressed as well, but I don't see it happening.
Lines of motion isn't future-proof, but I don't think Vincent will change that any time soon. It still gives pretty good indication tho.
Yes, and maybe you're less sensitive to subfield rendering, but that doesn't mean that no one else will see it or that it's "BS". Better to overstate potential issues than understate them, and I expect supposed 'reviewers' to approach this sort of thing with a more critical eye. This is why I'd take any review from the likes of you with a grain of salt.
Yeah -- the faint crosstalk ghost effect that also happens on several LightBoost displays. Basically, a trailing ghost sharp image (non-blurred) roughly similiar to intensity of 3D crosstalk. I already explain this in the HDTVTest posting about Sony's Motionflow Impulse, and explain exactly why it's a non-issue for gaming. An easy way to test strobe crosstalk (faint sharp trailing ghost double-image) similiar in faintness to 3D crosstalk, is the TestUFO Eiffel Tower Test.
Fortunately, VA has made some amazing progress in the last 2 years alone. It's finally now possible for strobed VA to have less ghosting than older strobe TN (e.g. XL2420T Revision 1.0 from early 2012). MCFI has forced VA LCD's to refresh faster (240Hz VA LCD's now exist). This gives the perfect opportunity for a creative trick called the double-pass refresh, already documented on Page 15 of the manual of the released FDF2405W LCD (cost several thousands, mainly for what are probably enterprise customers). They use a 240Hz two-pass refresh (one heavily overdriven refresh in total darkness, and one regular cleanup refresh, then followed by a strobe near the end of the second refresh which is overdrive-artifact-free -- as explained here in Eizo's manual), doing 120 strobe flashes per second. The already-released professional-market-only Eizo FDF2405W would have has less sharp-double ghosting than LightBoost displays, because of this very creative VA refreshing technique.
Yep, he didn't test 60fps videogames, which is where Impulse shines massively (IMHO -- it benefits 60fps games massively more than video -- I'd even argue a 10x bigger benefit to games than to video). FoxHounder agreed with me about Motionflow Impulse could be potentially good for videogames, as he hadn't closely considered that use case during his tests. (videogames use sharper graphics, closer view distances, and faster motion, all of which makes motion blur easier to see than with video)
Agreed, it's probably okay for the average home theater audience. That said, I'm trying to contact several test pattern manufacturers to try and consider MPRT as a unified motion blur measurement standard. But I'm just a small drop in a big ocean. That said, if a test pattern disc maker gives you a free sample, please tell them about a future-proof motion blur measurement such as MPRT by providing a link to my articles and the MPRT scientific papers. Thanks
At the end of the day, we need to strike a balance. We don't have a fault free display technology and we don't have an alternative to LCD and Plasma (at least not yet). Overstating a problem that doesn't affect everyone is not a wise move. You can advice them of possible issues that is inherent to various technologies, but that's it.
As things stand, plasma has the upper hand. It's blacks are 6-24 times better than LCDs (no, that's not a typo), it has cleaner and better motion than LCD (generally speaking), has better uniformity and winder viewing angle. Given the pros and cons of both tech, PDP easily wins. But in the end, it all depends on which technology you can tolerate better. Personally, I can tolerate plasma more than LCD. I also don't have to deal with panel and processor lottery.
Agreed - especially in space movies.
(Though for gaming -- black levels are far, far, far less annoying than motion blur for me)
Agreed -- except for Game Mode motion uniformity.
Not for computer use at 1:1 viewing distances.
Not in Game Mode with plasma processing turned off.
Compared to the better strobe-backlight computer displays. I can read the tiny text on the TestUFO Panning Map Test at 1920 pixels/second on my LightBoost display (configured to 10%):
Just try that on any plasma. You can't read any text. Games have fast panning speeds (fast turning, fast strafing, fast panning) and it's lovely to be able to identify fine details without stopping turning/strafing first. And when you use Game Mode on plasmas, you often reduce plasma processing in order to reduce input lag -- so you often get more plasma artifacts in Game Mode (e.g. more banding, because anti-banding algorithms often add input lag).
At 1:1 viewing distance from a plasma, I can see:
-- plasma christmas pixel noise in dark shades of dungeon FPS games (yes, even on Pioneer Kuro, yes even on Elite LCD HDTV)
-- plasma temporal color effects (also seen in high speed camera). (I also see DLP rainbow artifacts even on 6X colorwheels)
-- plasma banding during fast motion
All of them quite annoying during computer gaming at traditional proportional computer monitor viewing distances (1:1 ratio). As a result, even blurry LCD motion becomes much, much cleaner looking (and I still hate the motion blur). The 50" SEIKI 4K is nicer to stare at on a computer desk because of this (e.g. wallmounted at the back of a deep desk), and it even supports true 120Hz from a computer at 1920x1080 to solve a little bit of the LCD motion blur. Mind you, crappy black levels, though. But for gaming, LCD blacks are a lesser of evil for me, at least.
Plasma motion looks rather nice at regular sofa viewing distance, mind you, and they are quite numero uno for a lot of television material, but plasma limitations really stand out when you attempt to use them as 1:1-view-distance computer monitors for gaming, if you have any slight sensitivity to motion-related artifacts.
This is video processing which can be turned off
You are describing the artifacts caused by the interpolation algorithm Panasonic used on the 30 and 50 series, they use subfield and full frame motion prediction which has some really weird affects with CG content (in CounterSrike the stationary text 'shoots' or streaks at random times, as you turn corners you see warping and so on)
I think most people are comparing plasma in its worst modes for CG, 4:2:2 compression (540p chroma) ON, interpolation ON, low gradations ON which are all not representative of the technology
No it isn't. You think we haven't been though all this a thousand times?
Not to mention it's just as present on a V20
Probably the same on V20, I know exactly what warping you are talking about and its not present on my PF30 unless frame creation is on. I don't even know why they would use sub field interpolation it doesn't make any sense unless you were trying to save power
On my UK plasma, interpolation automatically turns on when I use a 50Hz refresh (TH 42PZ80).
Some US TVs might have a similar method when using 60Hz.
Good point Nenu
I have read that NTSC(60hz) uses 10 sub-fields and PAL(50hz) uses 12
I see major flicker at 56hz and under so it makes perfect sense to interpolate up to 100hz
I've only ever used UK spec units at 60Hz and the interpolation was disabled. No one is going to be using 50Hz for PC use.
As much as I hate motion blur, I prefer viewing 1:1 from an LCD, than viewing 1:1 from a plasma or closer. Sitting a mere three to four feet from a 50" television mounted at the rear of a deep desk, for example. Here, plasma motion artifacts are so unnatural looking, that even plain LCD motion blur looks more natural. It's as annoying as coilwhine screaming into my ears; it's not something I'm intentionally paying attention to.
Human vision and real world don't use subfields or temporal methods of color generation. It's totally artificial and unnatural if you have a vision sensitivity to those, especially at 1:1 monitor view distances.
Sofa viewing distance, I love plasma, though. Movies are amazing on plasma! Especially on Kuro and Elite. My two favourite plasmas.
Plasmas will never be good close-view-distance desktop computer monitors for me. I'm amazed how "dirty" looking plasma refreshes look, and high-speed video even agrees too (Have you seen HDTVTest's high speed videos of plasma refreshes? Versus LCD refreshes? Exactly. My eyes notice this, especially when I'm saccading my vision around as I'm a continual moving-objects-tracker. That's not good for a plasma that's sitting right in front of my nose like a computer monitor. LCD refreshes are more natural to humans that are very perceptive of high-speed temporal artifacts). Plasma is beautiful when pixels are too small to be seen, or when your vision system can't see the artifacts.
The perfect Holodeck display is a flicker-free (no subfields, no strobes, no temporal color), infinite-refresh-rate, adjustable-gamut, infinite-resolution display. No motion blur (caused by discrete refresh rates on sample-and-hold displays). No interpolation. No flicker. No strobing. Perfect solid flickerfree even in 1000fps high speed video. A display that looks the same as real life in 1000fps video comparision of the display operating versus real life scenery. LCD and plasma are neither, I know. The perfect Holodeck display won't happen for a long time. Until then, no display is a one-size-fits-all.
The VT60 can not do that perfectly at 1920 pixels @60hz, simply too blurry, but in 2D-3D conversion (120hz refresh, 60hz per eye + brain-postprocessing to meld the images), some of the text is easily readable, some is hard to read but doable, and some is not readable at all.
It only looks a touch worse than on VG248QE.
At 1440 pixels/s 2D-3D everything is readable though some text is still a bit hard.
Still hoping this can in some way achieve 96hz or more, and still pretty sure it will perform as excellently in 2D as it does in 3D.
Is the text readable at 1900 pixels/s even on a crt@60hz ?
When you are talking about 2D-3D conversion does that mean are using 3D goggles?
If you are able to read 1440pps with 60 unique frames and chroma softening (VT60 uses the 4:2:2 color space) image what is technically possible with 125fps+ 4:4:4 2D with full gradations (no noise), makes me sad thinking about what is being thrown away (I probably need another hobbies)