SuicideUZI
Member
Well I have no doubt it will have better image quality / black levels than LCD when the final product comes out, but how will it handle motion and input lag compared to OLED and highend plasma's?
-
Hey Guest. Check out your NeoGAF Wrapped 2025 results here!
Sony Crystal LED - And a New Display Technology Was Born
- Thread starter Raistlin
- Start date
- Status
Bitmap Frogs
Mr. Community
http://www.sony.co.uk/biz/product/camcorders
Out of these 8 professional ecosystems, only three are tapeless.
Tape is not dead (yet).
Flying_Phoenix
Banned
People who play last generation consoles and prior?
This too.
It still looks like shit.
I'm not sure about the DigitalBeta and DV's but yes Betamax had a monopoly on the professional market. Especially TV market.
Though his "prices don't lower with the professional market", is wrong though. Companies discontinue stuff and its sold to other vendors or at the used market with Steam sale like discounts. Though you're talking about something going from $8,000 to $2,000.
I'm not sure why you're making assumptions about OLED vs plasma - at least it sounds like you're implying they're similar.
I'd suspect in terms of motion handling, OLED and XLED will be similar. They have really fast response times, but likely both use sample-and-hold ... so temporal resolution should be behind Plasma without processing. What processing they can use at this time is unknown.
As for lag, that depends on the model and the specific tech used. LG's W-RGBW apparently uses at least some processing for color reportedly ... and I suspect all will use processing to improve motion resolution. How that impacts lag depends on how they go about it. XLED likely requires less processing for color and the like, but it will still have processing for other stuff. So does Plasma. We'll have to wait and see once there's products available for measurement of everything.
Really, this is very similar to OLED from a high level. I'd imagine the fabrication process may be quite different (or maybe not) ... but basically it's replacing organic diodes with inorganic diodes in a conventional RGB matrix. Though this seems to bring the diodes much closer to the surface (it's a bit unclear ... but the surface may literally be the tops of the diodes) which improves efficiency an viewing angles.
Oh shi----
Speculation from some AVSers is that the LCD TV being paired against the XLED is the 55 XBR-HX929. If so ... DAMN. It's one of the highest rated LCD's yet released.
Can any owners here confirm or speculate whether that the case?
Another interesting piece of speculation ... this may actually be a Quantum Dot display (QLED or QD LED as they've been referred to). I would think they'd be more than happy to mention QD or nano crystals, but maybe since people just aren't all that aware of its significance there's no reason to mention it as yet? Sony actually is known to be R&Ding quantum dot displays.
http://en.wikipedia.org/wiki/Quantum_dot_display
The rep does mention it's semiconductor based ... and the fact they're calling it 'Crystal' LED (QD displays use nano-scale crystals within the diode) ... it's interesting. Hmmmmmmm ....
Speculation from some AVSers is that the LCD TV being paired against the XLED is the 55 XBR-HX929. If so ... DAMN. It's one of the highest rated LCD's yet released.
Can any owners here confirm or speculate whether that the case?
Another interesting piece of speculation ... this may actually be a Quantum Dot display (QLED or QD LED as they've been referred to). I would think they'd be more than happy to mention QD or nano crystals, but maybe since people just aren't all that aware of its significance there's no reason to mention it as yet? Sony actually is known to be R&Ding quantum dot displays.
http://en.wikipedia.org/wiki/Quantum_dot_display
The rep does mention it's semiconductor based ... and the fact they're calling it 'Crystal' LED (QD displays use nano-scale crystals within the diode) ... it's interesting. Hmmmmmmm ....
GTP_Daverytimes
Member
Wow, that is amazing if true. It clearly looks better and with more development time it might look even more awesome, Sony has a real winner in their hands if they play the right cards.
Yeah, if it's sitting next to Sony's top panel - that impressive (and ballsy
)I'm still not convinced it's QD though. It sounds like the fab process is behind OLED in terms of costs (and coincidentally, it's main issue is with blue - lol). Then again, I suppose it's possible Sony thinks they've found a more cost-efficient process?
It would be crazy if Sony has already made the jump to what some argue is the next generation of display tech beyond OLED.
H_Prestige
Banned
Agreed. I remember when I first saw the Sony XEL-1 a few years ago, it just didn't have the silky smooth motion of a plasma.
Pristine_Condition
Member
I got some info from my buddy at CES today who saw the demo and talked to some guys.
Replicant: to somewhat answer your question from last night, I PMed him for an answer and he said the rep said those comparison numbers are based on the current top-of-the-line 55" Sony LED TV with local dimming. (So I'm assuming that's the XBR HX929!)
Apparently the big takeaway beyond the performance numbers is this: The goal of the tech is indeed to build the front surface of the display entirely out of the front-facing elements of the LED array, so there is no front glass, no actual "screen," ect. This will also make huge displays far easier to make in the future.
This is apparently the big deal here. The size issue. This was emphasized to my friend very clearly. Of course he is a buyer for an exclusive super luxe home theater company that specializes in the types of installs where things like 150" plasma tech is actually relevant, and a sister company that does AV for commercial applications and venues, so they might have driven this point home for him more than with more normal consumer-oriented folks.
EDIT: Raistlin: I just saw your post before I clicked "submit." As you can see, my friend at CES is reporting the same thing. He did not get this info from AVS, but from the show itself. While he did not mention a specific model number, it's pretty easy to connect those dots.
Like all the current self-emitting tech, here is potential for image retention. But with LED, less so than plasma and OLED. So I'd suspect after two to four hours of static display, (whatever the engineers come up with after testing) they'll kick on some sort of management routine, I'm sure. They've gotten very good at this.
...
Friend said he'll try to PM me some more specific info and stuff later. He said it's pretty exciting though. He went into this show not expecting much more than the curiosity of seeing the OLED stuff, but he's been surprised and impressed by both the new Panny line which he loves right now, and all this future stuff, far more than he expected.
Replicant: to somewhat answer your question from last night, I PMed him for an answer and he said the rep said those comparison numbers are based on the current top-of-the-line 55" Sony LED TV with local dimming. (So I'm assuming that's the XBR HX929!)
Apparently the big takeaway beyond the performance numbers is this: The goal of the tech is indeed to build the front surface of the display entirely out of the front-facing elements of the LED array, so there is no front glass, no actual "screen," ect. This will also make huge displays far easier to make in the future.
This is apparently the big deal here. The size issue. This was emphasized to my friend very clearly. Of course he is a buyer for an exclusive super luxe home theater company that specializes in the types of installs where things like 150" plasma tech is actually relevant, and a sister company that does AV for commercial applications and venues, so they might have driven this point home for him more than with more normal consumer-oriented folks.
EDIT: Raistlin: I just saw your post before I clicked "submit." As you can see, my friend at CES is reporting the same thing. He did not get this info from AVS, but from the show itself. While he did not mention a specific model number, it's pretty easy to connect those dots.
Like all the current self-emitting tech, here is potential for image retention. But with LED, less so than plasma and OLED. So I'd suspect after two to four hours of static display, (whatever the engineers come up with after testing) they'll kick on some sort of management routine, I'm sure. They've gotten very good at this.
...
Friend said he'll try to PM me some more specific info and stuff later. He said it's pretty exciting though. He went into this show not expecting much more than the curiosity of seeing the OLED stuff, but he's been surprised and impressed by both the new Panny line which he loves right now, and all this future stuff, far more than he expected.
So apparently it is the 929 being compared to it.
Thanks to MaXPL. Apparently the video I posted previously states it at around 1:05 (that's what I get for watching without sound). Unfortunately though, the vid has been moved to private :\
That reminds me. Assuming this is QD (which I'm really not), remember when some of us postulated this may be a short/mid-term solution ... but OLED makes more sense long-term? Forget that. QD actually has the long-term potential to cost even less than OLED amongst other potential advantages. It can also be used in all sorts of products like OLED ... phones, lighting, can be flexible. It even has the prospect of eventually being printable on paper.
While they both have very promising long-term fabrication efficiencies, QD actually has lower component costs. While the core emissive components are in part organic, OLED actually uses rare earth elements ... which only continue to escalate in price. QD apparently doesn't.
Thanks to MaXPL. Apparently the video I posted previously states it at around 1:05 (that's what I get for watching without sound). Unfortunately though, the vid has been moved to private :\
Unknown. If it's similar to traditional LED, they do age ... though I believe more linearly than OLED ... so I wouldn't be too concerned. If it's QD, then there's even less to worry about.
That reminds me. Assuming this is QD (which I'm really not
), remember when some of us postulated this may be a short/mid-term solution ... but OLED makes more sense long-term? Forget that. QD actually has the long-term potential to cost even less than OLED amongst other potential advantages. It can also be used in all sorts of products like OLED ... phones, lighting, can be flexible. It even has the prospect of eventually being printable on paper.While they both have very promising long-term fabrication efficiencies, QD actually has lower component costs. While the core emissive components are in part organic, OLED actually uses rare earth elements ... which only continue to escalate in price. QD apparently doesn't.
Cool beans.
Now that we know the original 3.5x contrast thing was in daylight viewing, and that night viewing is essentially off the charts ... any chance you can ask your buddy if he heard anything further regarding response times?
All we've heard is 10x faster ... but 10x what exactly? I'm not sure if they even disclose the pixel response measurements for their LCDs?
I hope you're right. Would be awesome if they were this far along.
Pristine_Condition
Member
I've already asked him about response times. I'll bug him some more later.
Thanks. Regardless of whether he has or can get the specifics ... it would be good to hear his description of the general motion characteristics. Both in general, and compared to the 929.
Oh ... and I suck at typing lol
look at them even the ones that offer tape offer ssd
thats from the srw-9000
The SR master uses flash
HDcam uses flash
XDcam uses optical /ssd
nxcam is cc , ssd and memory stick
hdv is tape , cc ,hdd
there are only two tape only ecosystems on that page and they seem to just be older models that are still sold
You're so right. Companies should never try anything new and just go with the flow. That definitely will result in the best advancement in technology over time.....
Not sure how I didn't notice this previously.
See those bands there? Journalists noted they didn't see anything like that while viewing ... but it was showing up in some photos. Looks like they're using some type of simulated backlight scanning to improve temporal resolution?
See those bands there? Journalists noted they didn't see anything like that while viewing ... but it was showing up in some photos. Looks like they're using some type of simulated backlight scanning to improve temporal resolution?
I don't know what that means!
But I've some questions I'm itching for someone to ask Sony. Principally timescales for release, manufacturing costs (surely they must be there seeing as Sony have seemingly ditched OLED for this tech) and some realworld comparisons to OLED on contrast and response times (unless these have already been released??).
Ahhh ... here's a copy of the video mentioning the LCD - http://www.youtube.com/watch?v=5dxXUtx2RN0&feature=player_embedded#!
Around 1:05, he states it's their top-of-the line Bravia (which is the 929). Now the reason I missed it before is the original upload (now gone) had comments stating the sound stopped halfway through so I didn't bother listening at all. Now that this is fixed, some other interesting tid-bits are available.
The claim is that the response time is practically instantaneous - better than other display techs. Now I don't know if that's hyperbole (or possibly taking advantage of the fact large-size OLED's aren't commercially available) ... but the fact this appears to be using scanning means it must be pretty damn fast This may already have full temporal resolution.
http://www.neogaf.com/forum/showpost.php?p=33910019&postcount=116
Response times are unknown other than being a magnitude better than top-of-the line LCD. How that compares to OLED we're not sure ... but it's obviously fast. Since this appears to be using simulated backlight scaling, it may also have better (perfect?) temporal resolution. We don't know what if any processing the upcoming OLED's are using to improve temporal resolution.
Around 1:05, he states it's their top-of-the line Bravia (which is the 929). Now the reason I missed it before is the original upload (now gone) had comments stating the sound stopped halfway through so I didn't bother listening at all. Now that this is fixed, some other interesting tid-bits are available.
The claim is that the response time is practically instantaneous - better than other display techs. Now I don't know if that's hyperbole (or possibly taking advantage of the fact large-size OLED's aren't commercially available) ... but the fact this appears to be using scanning means it must be pretty damn fast
This may already have full temporal resolution.Hopefully this post helps explain it a bit.
http://www.neogaf.com/forum/showpost.php?p=33910019&postcount=116
Contrast seems as good, maybe even better (basically unmeasurable in a dark room - and 3.5x higher than top-of-line LCD in daylight)
Response times are unknown other than being a magnitude better than top-of-the line LCD. How that compares to OLED we're not sure ... but it's obviously fast. Since this appears to be using simulated backlight scaling, it may also have better (perfect?) temporal resolution. We don't know what if any processing the upcoming OLED's are using to improve temporal resolution.
Pristine_Condition
Member
Not sure how I didn't notice this previously.
See those bands there? Journalists noted they didn't see anything like that while viewing ... but it was showing up in some photos. Looks like they're using some type of simulated backlight scanning to improve temporal resolution?
Hmmm. Gonna throw my two cents in here as a pro photographer...
I think it's probably just an effect of Pulse Width Modulation, which is sadly nothing new for us who shoot electronics (and especially DSLR videographers) when shooting LED/OLED screens at certain brightness levels under certain shutter speeds with digital cameras, rearing it's ugly head.
DISAPPOINTMENTON
If true, guess it's hard to speculate on motion handling until some more tech is divulged (or someone squeaks in some sort of testing).
Hopefully we'll at least get some hands-on accounts from people that know what to look for in terms of motion ... or at least a relative comparison to the 929 (though without knowing what processing is turned on for the 929, I suppose relative-only accounts are kind of pointless).
no prob
purnoman3000
Member
I noticed Sony didn't show off a 900-series LED TV, typically their XBR line. Do you guys think they left it out as they may bring the CLED later this year as a replacement for that series?
.
.
Unless everything they've said has been part of an elaborate troll (primarily towards LG and Samsung) ... this ain't coming out this year. If Sony is to be believed, they won't even confirm nor deny whether this will definitely become a product (though obviously unless reliability or long-term costs are expected to be an issue - it's gonna hit eventually).
As for a new 900 series, yes it's kind of weird ... but not without precedence. It's their top-tier display and has received very favorable reviews, so sitting on it for a year or two isn't entirely surprising.
It would have been nice if they at least threw GoogleTV in it and called it a day. Or better yet, added that and 1080p60 3D (though that may be out of their hands).
Actually do you mind going into some detail regarding this topic?
Nice to hear Sony is being very active to turn their TV business around. Would be sad to see all that R&D go to waste.
Seconded. I don't quite understand what relationship PWM might have with that effect.
Pristine_Condition
Member
It's really complicated, but I'll try.
This is how I understand it to work:
A CMOS sensor in a digital camera doesn't actually record the whole frame all at once. It records a sequence of lines/pixels/whatever. It does this very very quickly, so most of the time you don't have a problem, even at fairly high shutter speeds. However, when attempting to record a very fast event. IE a very fast-moving object or say, an electronic flash lighting the scene at a sync/flash duration speed above the recommended speed of the sensor, you can get fuckups.
In the case of shooting an LED display, sometimes, as the CMOS sensor in the camera sequentially records the frame, it picks up the Pulse Width modulation (basically, the light source, in this case the LEDs, flashing on and off, extremely rapidly) which they do to control brightness output of the LEDs. The CMOS sensor records a set of lines while the LED's are on full blast, resulting in a bright "bar" on the screen, and when the CMOS sensor continues on the next set of lines, it sees the dimmed or off LED's, resulting in a dark "bar" for that sequence of lines being recorded. When the next sequence is done, the sensor picks up the LED's turned back on, ect, ect. ect.
If you shoot video through the CMOS sensor, and this occurs at a rate that is captured by the sensor, the effect known as "rolling shutter" will sometimes cause these light and dark "bars" to appear to "roll" through the display, slowly or quickly, or not at all, depending on how the scanning speed of the sensor and the rapid pulsing of the display happens to sync up. (Think of how you can sometimes look at the wheels of a car next to you while you are traveling, and the wheels of the neighboring car seem to turn in slow motion, or sometimes even appear to be turning backwards.) But in the case of a modern LED or OLED screen, it is not the display making these "bars," like you would see in the case of an old-school CRT, where the electron gun is scanning across the tube, it is the scanning of the CMOS sensor in the camera causing them to appear.
I hope that's enough to get the point across...I really don't want to get into the notched waveforms and all that stuff. Actually I don't think I'm qualified to go into that level of detail!
Basically, it's an annoying bug that pops up from time to time for us photographers of this stuff. It sucks, because a lot of the time you don't see it happening until you go to post with your footage (especially in video shoots.) That's why I try hard never shoot anything with a screen like that without having the ability to proof it on the spot. I've found when it does pop up, usually a simple adjustment of either the display brightness (which changes the rate of the PWM) or changing the ISO setting on the camera (which changes how fast the CMOS sequences) is enough to change the timing enough to eliminate the issue. Changing the shutter speed would seem the obvious thing to do, but that doesn't work, like it would if you were shooting film, because even though you are dragging the shutter, the sensor in the camera is still going to be scanning at the same rate...it'l just have more time to do more scans with the shutter open. Often there is cussing and fumbling for ND filter involved too. And of course there's always the option of shooting the display in film and popping that image in in post, if you are allowed to do so. Obviously in this case, where you are trying to show the display of a device in a live setting at a trade show, these poor saps didn't have any of those options at their disposal.
Didn't know they used a 929 as a comparison....ballsy. I wonder if they will even release a XBR this year....hell, the 65in. 929 just started selling back a couple months ago, and folks have been waiting for it since last CES. I love my 929 but these smart tv's are already making my set look outdated. 
Brofist
Member
Is it really taking off in any meaningful way? I mean it won the format war kind of by default, but it seems people are going straight to streaming media rather than blu-ray.
And lol at people hoping Sony price this tech below the competition. Like that ever happens.
Hahah yeah. Anything close to 180 degrees is fine. I haven't seen a recent TV that wasn't viewable from all reasonable angles. My current LCD is 4 years old and has a good 175 degrees.
http://www.youtube.com/watch?v=aTCnhc-JHTo
As has been said ... really no one. The point however is that the image retains nearly perfect chroma and luma at any viewing angle ... which includes viewing angles people actually do use that are problematic in other display types.
It's simply an extreme example to convey that it will have no issues. Plus, it looks cool.
Define 'meaningful way'? It's generating good revenue and margins ... and growth is increasing.
And lol at people hoping Sony price this tech below the competition. Like that ever happens.[/QUOTE]
Sony? Since when do they control pricing on this?
What about vertical viewing angles? Is this expected to perform as well for them?
It's the sort of thing you notice when you stand up to play Guitar Hero or whatever.
Yeah, the visible gamma on my old laptop monitor changes a lot when slanted the wrong way.
Valkyr Junkie
Member
Well they need to. Samsung et al have done a good job of already destroying the "LED TV" branding before we actually got TVs remotely LED-based.
For my arcade cabinet, duh. Fighting games are better on a screen that isn't the size of me.
Barkley's Justice
Member
viewing angles: because no one sits in front of their tvs.
It's for when you are watching with a group of people. It's a luxury we all had back when we were on CRT and it's nice that it's coming back.
Thanks for the information! I wasn't aware digital CMOS sensors work sequentially.It's really complicated, but I'll try.
This is how I understand it to work:
A CMOS sensor in a digital camera doesn't actually record the whole frame all at once. It records a sequence of lines/pixels/whatever. It does this very very quickly, so most of the time you don't have a problem, even at fairly high shutter speeds. However, when attempting to record a very fast event. IE a very fast-moving object or say, an electronic flash lighting the scene at a sync/flash duration speed above the recommended speed of the sensor, you can get fuckups.
In the case of shooting an LED display, sometimes, as the CMOS sensor in the camera sequentially records the frame, it picks up the Pulse Width modulation (basically, the light source, in this case the LEDs, flashing on and off, extremely rapidly) which they do to control brightness output of the LEDs. The CMOS sensor records a set of lines while the LED's are on full blast, resulting in a bright "bar" on the screen, and when the CMOS sensor continues on the next set of lines, it sees the dimmed or off LED's, resulting in a dark "bar" for that sequence of lines being recorded. When the next sequence is done, the sensor picks up the LED's turned back on, ect, ect. ect.
If you shoot video through the CMOS sensor, and this occurs at a rate that is captured by the sensor, the effect known as "rolling shutter" will sometimes cause these light and dark "bars" to appear to "roll" through the display, slowly or quickly, or not at all, depending on how the scanning speed of the sensor and the rapid pulsing of the display happens to sync up. (Think of how you can sometimes look at the wheels of a car next to you while you are traveling, and the wheels of the neighboring car seem to turn in slow motion, or sometimes even appear to be turning backwards.) But in the case of a modern LED or OLED screen, it is not the display making these "bars," like you would see in the case of an old-school CRT, where the electron gun is scanning across the tube, it is the scanning of the CMOS sensor in the camera causing them to appear.
I hope that's enough to get the point across...I really don't want to get into the notched waveforms and all that stuff. Actually I don't think I'm qualified to go into that level of detail!
Basically, it's an annoying bug that pops up from time to time for us photographers of this stuff. It sucks, because a lot of the time you don't see it happening until you go to post with your footage (especially in video shoots.) That's why I try hard never shoot anything with a screen like that without having the ability to proof it on the spot. I've found when it does pop up, usually a simple adjustment of either the display brightness (which changes the rate of the PWM) or changing the ISO setting on the camera (which changes how fast the CMOS sequences) is enough to change the timing enough to eliminate the issue. Changing the shutter speed would seem the obvious thing to do, but that doesn't work, like it would if you were shooting film, because even though you are dragging the shutter, the sensor in the camera is still going to be scanning at the same rate...it'l just have more time to do more scans with the shutter open. Often there is cussing and fumbling for ND filter involved too. And of course there's always the option of shooting the display in film and popping that image in in post, if you are allowed to do so. Obviously in this case, where you are trying to show the display of a device in a live setting at a trade show, these poor saps didn't have any of those options at their disposal.
However your contention that modern displays would never show what appears to be CRT scanning is not true. Higher-end LCD's actually intentionally use backlight scanning in order to improve temporal resolution - and that's what is being proposed here. That they are simulating backlight scanning in order to improve motion handling.
As for OLED, I'm not sure if they typically use PWM to drive the pixels, but if they do I suspect it's much like Plasma ... which means per-subpixel, and at thousands of times per second. That would be too fast to record, and if you could ... it wouldn't show banding, it would show a grid effect.
For this TV, while LED backlights may use PWM on the power supply to control brightness, this isn't backlit. It's a self-illuminating display where the LED's are the actual sub-pixels. So much like Plasma (and possibly OLED), if this is using PWM it would be per-subpixel and quite fast. And again, would show up as a grid pattern if it were actually slow enough to be filmed.
Given that they must be able to control luma for each sub-pixel (either with really fast PWM or some other means), I wouldn't think they'd bother controlling overall brightness via a power supply PWM circuit. Seems redundant? Maybe there's a technical reason though that makes that the better solution. The problem here though is that this showing was not a dark-room demo. They were in full light, and one of the intents was to show of the high contrast in a non-light-controlled environment. For that reason, I'm pretty sure they would be set to max panel brightness. So even if these do use PWM on the power supply ... it's pretty suspect they'd be using panel dimming at all for this demo event.
Based on your information regarding how digital CMOS work though, I suspect there may be a different explanation. Assuming this isn't some sort of simulated backlight scanning as originally proposed, it could be the combination of the CMOS with black/contrast-adjusted frame insertion. That technique is used in all of Sony's 120Hz and greater LCD's - and since the assumption is CLED is a sample-and-hold display ... they'd obviously use it here too. Though to make it more fun, in higher-end LCD's both techniques are used
It's already here. There are affordable TVs out now with very wide viewing angles, plasmas in particular.
I know, I own a Panasonic plasma. Though my previous post does make it seem like viewing angles completely disappeared with CRT, whoops.
- Status