Samsung 81 Series anticipation thread

[vtms]

Now that I think about it, dynamic contrast is a misnomer.
It should be called dynamic range.
It is not really contrast at all. You cannot see it simultaneously and contiguously.

Now the dynamic dimming works within a frame, so it really does increase static contrast.
Right. Dynamic contrast for 81s is infinite but so what. The only thing that counts is static contrast/dynamic range within a single frame.

[vtms]

Remember to at least double those numbers for 81 series. The thing is incredible :D
I was just about to edit my last comment and say the same thing about the numbers. :D

[vtms]

48x27 backlight resolution:

http://i13.tinypic.com/4xnn50y.jpg

...and 12x7 resolution:

http://i15.tinypic.com/5z6g93s.jpg

First picture - BrightSide backlight resolution (more or less)
Second picture - 81 Series backlight resolution (more or less)
Cool stuff. :)

I'm beginning to see how even the low 12x7 resolution would have value if it's properly diffused...

...but 48x27 (1300 individually modulated LEDs) would be ultimately cool. :cool:
In that great bit-tech article on BrightSide technology I recall reading something about maximum backlight resolution beyond which there can be no more discernible improvements in PQ (so that LED cluster per pixel scenario would be totally unnecessary). Who knows how far 81 series is from reaching that maximum. Perhaps not as far as we think? ;)

[conan48]

This thread is making me think I should go back to school :D This is getting way to technical. The thing is, we can speculate till the cows come home, but we wont know ANYTHING until someone sees one of these bad boys with lots of different content to see how this low rez LED display will work.

[go81]

Who knows how far 81 series is from reaching that maximum. Perhaps not as far as we think? ;)

Here's simulated brightness difference between brightside and samsung 81 series displaying that Agatha Peak picture:

http://i13.tinypic.com/53z8ggp.jpg

White color means that there's no difference. Black would mean 100% difference. As good as brightside? Almost. :D

[rbarg]

Posted today by "alleged" BB rep - I stopped by a Magnolia last night and they said they would definately carry the 81 series as soon as it ships.

Comments?

http://crave.cnet.com/8301-1_105-9743599-1.html

TalkBack
1 message - 1 new!
Post a comment


120Hz briefly explained and 100,000 : 1 contrast!? New!
I currently work for the largest "brick and mortar" electronics chain on the planet, Best Buy. For those of you that done know there are almost 1200 stores in the U.S., Canada, and China, and the sales people at their stores have not been on commission for nearly 20 years. I have worked there for a little over four years, and have seen a lot of technological innovation in this little amount of time. I remember when the only HD content available was Monday night football, and I am very excited to see how the HD world is expanding.

If you do not read about HDTV’s everyday or do not work in the industry, all of the technical specs and various anagrams and other 4 letter words of the HDTV world can be hard to digest. I am excited to see that Samsung is finally catching up to Sharp, Sony, and Toshiba and is launching a set that is capable of 120Hz refresh rates.

Normal televisions have a refresh rate at 60Hz which is alright for SDTV, but in the world of 1080p and beyond it is somewhat behind the times. The televisions that have a 120Hz frame rate conversion are able to show 120 frames per second by quickly repeating each frame by a multiple of 120 to reduce pixilation and motion blur AKA “judder” that is normally seen by video enthusiasts, with the most discriminating sets of eyes.

A normal TV broadcast is 30 frames per second, which breaks down to 30x4=120, or repeating each frame from four times. Movies are slightly different, and have a frame rate of 24 frames per second which breaks down to 24x5=120, or showing each frame five times.


Moving onto the contrast ratio of 100,000:1, I will believe it when I see it. From all of the brand reps, and websites and other various sources from which I get my information, I have come to find that there is really no bench mark for contrast ratios. I do enjoy the vivid colors, and accurate blacks that the current high end Samsungs have at a contrast ratio of 15,000:1. However when looking at the Samsung LN-Txx65F series I am unimpressed with the amount of pixilation that I see compared to any of Sony’s current offerings, even their entry level KDL-xxS3000 series, or any of the Sharp Aquos televisions. Hopefully the 120Hz frame rate conversion fixes this problem.

The LED technology in the Samsung DLP televisions does offer a much richer, deeper, fuller black than normal micro-display television technologies, and I assume the spread of the LED into a flat panel display will also yield excellent results. The ability to turn of each pixel individually as a plasma television can do will yield much higher range of black tones, and of course make many more shades and tints of each color.

I feel that the picture is promising, but hope that the horrible pixilation that plagues many of the Samsung LCD’s now. However, many consumers, especially those that wear glasses and or contacts, may have a harder time discerning the difference between a televisions with a 120Hz refresh rate to that with one of a 60Hz. Leaving the overall color reproduction and brightness more desirable than the smoothness of the picture.


by steve7410 (See profile) - July 15, 2007 10:01 PM PDT

[Andrew67]

Posted today by "alleged" BB rep - I stopped by a Magnolia last night and they said they would definately carry the 81 series as soon as it ships.

Comments?

Magnolia, yes. Best Buy, no.

A normal TV broadcast is 30 frames per second, which breaks down to 30x4=120, or repeating each frame from four times. Movies are slightly different, and have a frame rate of 24 frames per second which breaks down to 24x5=120, or showing each frame five times.

120hz is all well and good, but it's very important on how that 120hz is arrived at. I have yet to see a single manufacturer of televisions get from 24hz to 120hz the most proper route. Most are taking the 24fps content, doing the tradtional 3:2 pulldown to get to 60fps and multiplying that by 2 to get to 120fps. So it doesn't matter that 120 is divisible by 24 because the 3:2 conversion has still been peformed.

[rkurlander]

Posted today by "alleged" BB rep - I stopped by a Magnolia last night and they said they would definately carry the 81 series as soon as it ships.

Comments?

http://crave.cnet.com/8301-1_105-9743599-1.html

TalkBack
1 message - 1 new!
Post a comment


120Hz briefly explained and 100,000 : 1 contrast!? New!
I currently work for the largest "brick and mortar" electronics chain on the planet, Best Buy. For those of you that done know there are almost 1200 stores in the U.S., Canada, and China, and the sales people at their stores have not been on commission for nearly 20 years. I have worked there for a little over four years, and have seen a lot of technological innovation in this little amount of time. I remember when the only HD content available was Monday night football, and I am very excited to see how the HD world is expanding.

If you do not read about HDTV’s everyday or do not work in the industry, all of the technical specs and various anagrams and other 4 letter words of the HDTV world can be hard to digest. I am excited to see that Samsung is finally catching up to Sharp, Sony, and Toshiba and is launching a set that is capable of 120Hz refresh rates.

Normal televisions have a refresh rate at 60Hz which is alright for SDTV, but in the world of 1080p and beyond it is somewhat behind the times. The televisions that have a 120Hz frame rate conversion are able to show 120 frames per second by quickly repeating each frame by a multiple of 120 to reduce pixilation and motion blur AKA “judder” that is normally seen by video enthusiasts, with the most discriminating sets of eyes.

A normal TV broadcast is 30 frames per second, which breaks down to 30x4=120, or repeating each frame from four times. Movies are slightly different, and have a frame rate of 24 frames per second which breaks down to 24x5=120, or showing each frame five times.


Moving onto the contrast ratio of 100,000:1, I will believe it when I see it. From all of the brand reps, and websites and other various sources from which I get my information, I have come to find that there is really no bench mark for contrast ratios. I do enjoy the vivid colors, and accurate blacks that the current high end Samsungs have at a contrast ratio of 15,000:1. However when looking at the Samsung LN-Txx65F series I am unimpressed with the amount of pixilation that I see compared to any of Sony’s current offerings, even their entry level KDL-xxS3000 series, or any of the Sharp Aquos televisions. Hopefully the 120Hz frame rate conversion fixes this problem.

The LED technology in the Samsung DLP televisions does offer a much richer, deeper, fuller black than normal micro-display television technologies, and I assume the spread of the LED into a flat panel display will also yield excellent results. The ability to turn of each pixel individually as a plasma television can do will yield much higher range of black tones, and of course make many more shades and tints of each color.

I feel that the picture is promising, but hope that the horrible pixilation that plagues many of the Samsung LCD’s now. However, many consumers, especially those that wear glasses and or contacts, may have a harder time discerning the difference between a televisions with a 120Hz refresh rate to that with one of a 60Hz. Leaving the overall color reproduction and brightness more desirable than the smoothness of the picture.


by steve7410 (See profile) - July 15, 2007 10:01 PM PDT

120Hz refresh rate is a scam. There is no way anyone can perceive any difference.

[Karamba676]

120Hz refresh rate is a scam. There is no way anyone can perceive any difference.


You do not know what you are talking about. It is very easy to perceive the 3:2 pulldown judder and it is very easy to percieve fast refresh rates for moving objects.

Digitization noise (spatial and temporal) can be detected even when the digitization bins are much smaller then the absolute resolution for the detector (eye).

[Flash01]

Anyone got some info on who might carry the 81 series in canada?

I really don't have much choices locally anyway so I may have to drive a bit or order online if I feel lucky... Just wondering if anyone asked around at their local dealers up here.

[phigment]

Anyone got some info on who might carry the 81 series in canada?

I really don't have much choices locally anyway so I may have to drive a bit or order online if I feel lucky... Just wondering if anyone asked around at their local dealers up here.

I asked at Gibson Sound and Vision in Waterloo. They called Samsung Canada who didn't even know about the TV yet. Future Shop didn't have it in their system either.

[bmoorhouse]

I have a 4066f that I am having problems with, but I am also dissappointed that it will not pass 5.1 DD received via the HDMI connection to the optical out.

The newest toshiba (177 series) will do this. Does anyone know if the new Samsung 71 series will?

Thanks.

[necrolop]

This here is a simulation of what the Backlight will look like, including diffusion layer, which cannot be forgotten.

Also At the bottom is a Contrast deliniation image, showing a line where a brite part of the image meets up with a dark part, the lower resolution backlight distorts this.

I think everyone is getting too into the contrast ratio theoretical numbers. When this display has a half black and half white image, the contrast ratio will be nearly infinite, but as the frame becomes more complex the screen will have to brighten up dark areas to compensate and make sure there arent huge anomolies in high contrast areas of the image. What this does is kill the ratio as the image becomes more complex, and makes absolute black from a video source, a floating value.

Notice in the images how the brightest areas of the brightside are brighter, and the darkest are darker. Now pretend this image has the sun, or even better, the moon. On a samsung it would be a muted grey on the backlight, on the brightside the LEDs could be put in highest possible position, theoretically alllowing for "blooming" which is the effect of too much light hitting your retnas, sorta like looking up at the sun. Now maybe it wouldnt be that good but definatly better.

http://a314.ac-images.myspacecdn.com/images01/121/l_190d282f917edff508488f2c13093611.jpg
Credits to jksgvb for his original.

Here is an image from Star wars. Since the backligt has to be on to illuminate the X-wings, there will be light on behind areas that are also black, this means leakage will occur in those areas, but in the black areas that have nothing needing to be lit, there will be no leakage. Therefore you will have one of two things, a grey cloud around the xwings, or the TV will compensate and turn all of the pure black areas, into grey to match the leakage, therefore lowering the constrast ratio on purpose in order to avoid lighting anomolies.

http://a228.ac-images.myspacecdn.com/images01/87/l_fe2768e8b850bf65be56b8f5b51522bb.jpg
This image has a 10x6 backlight behind it, I have lowered the opacity of the overlaying scene from star wars in order to show how leakage will affect the image.

[z_toth]

Anyone got some info on who might carry the 81 series in canada?

I really don't have much choices locally anyway so I may have to drive a bit or order online if I feel lucky... Just wondering if anyone asked around at their local dealers up here.

Though it wont help your plight, a local AV Specialty Shop, Bianco's Audiotronic, in Sudbury Ontario will be carrying the 46 and 52 inch version. They knew about it, were looking forward to it and stated the Samsund Rep wasnt sure when exactly besides "Sometime in August" that they would get it. Im pumped about it and the fact the dealer said that he would give me a "great" deal on the set .... wonder how much I can shave off the MRSP ....

Ill keep you other Canadians informed as I hear more.

Zsolt

[techtvman]

i wonder if American will carry the 81 series

[taurus2007]

i wonder if American will carry the 81 series
Guarantee it (but probably not every electronic store though).

[VoodooZ]

Thanks z_toth

I'm ready to drive to Sudbury from home (Ottawa) to get one if it's at the right price...

Let's keep ourselves updated...

Stephane

[glow11]

As far as HDR images are concerned, the ones I use range from between 5 to 20 times as large as a standard image format of the same scene. It varies depending on how many exposures made up the HDR. This is usually dependent on how bright the scene is.

I believe the speed at which a HDR image is captured is limited by the time it takes to get the brightest exposure. I'm not a camera guy, but I think the larger the aperture the more light the camera can gather in a shorter period of time. If you could gather everything from completely underexposed, to completely overexposed in a 30th of a second (and store the files in that time) you could achieve a true HDR video camera.

Has someone determined what the specifications for the theoretical "perfect" display would be for displaying current video images "perfectly"? If there was a theoretical "perfect" display, it would definitely give display manufacturers something to shoot for.

Then HDR would add a whole new level, and I think we would bump into the "enough is enough" factor when sunsets filmed in HDR would actually cause instantaneous retinal fatigue. Then the displays would have to be degraded to a point where viewing them isn't a chore.

As far as interpolating HDR information from a standard signal, I'm sure it could be faked fairly well, but I'm sure there would be issues.

[glow11]

Necrolop, in your Xwing image, did you use a known value for LCD light leakage to determine how bright the halos would be? I don't think I've seen late model LCDs with no dynamic backlighting leak that much...or maybe thats just because there are no pure black areas on those particular screens. Either way, if Samsungs 81's look like that on the floor, I dont think we would have to worry about anyone buying them.

[necrolop]

Necrolop, in your Xwing image, did you use a known value for LCD light leakage to determine how bright the halos would be? I don't think I've seen late model LCDs with no dynamic backlighting leak that much...or maybe thats just because there are no pure black areas on those particular screens. Either way, if Samsungs 81's look like that on the floor, I dont think we would have to worry about anyone buying them.

No, Its an exaduration of light leakage in order to see the effect. In a lit enviroment Im sure you wont be able to tell, but in a dark unlit room you certainly will be able to. Albiet not as bad as my image shows.

I beleive Samsung will use compensation to avoid halos, but this will make black levels at worst on par with other Samsung LCDs. So im still pretty confident that these sets will look great, and usually better than others, but while displaying complex material, it may at times look no different than a regular Samsung LCD. This inconsistency may be troubling to some, shall see.

just to reiterate, The contrast ratio will always be atleast as good as the best samsung LCDs that exist, but with complex scenes, the ratio will be knocked down to near non dimmed levels.

[GrantR]

Bear in mind that the human eye is limited to perhaps 1000:1 contrast ratio in a small field of view area - but can detect a lot more than that as the visual distance between bright and dark gets further apart.

The size of one of these 64 areas/clusters/regions when you're sitting at a reasonable viewing distance may actually fit within this region of close contrast ratio eye limits, so even though the local blacks might be brighter next to the x-wings (in the example image necrolop created) our eyes may not really notice that the black is brighter because the bright stuff adjacent overwhelms the eye's local sense of black.

I'm curious to see one of these sets in action in a dark room to see how all this works in practice :)

[Flash01]

Here is an image from Star wars. Since the backligt has to be on to illuminate the X-wings, there will be light on behind areas that are also black, this means leakage will occur in those areas, but in the black areas that have nothing needing to be lit, there will be no leakage. Therefore you will have one of two things, a grey cloud around the xwings, or the TV will compensate and turn all of the pure black areas, into grey to match the leakage, therefore lowering the constrast ratio on purpose in order to avoid lighting anomolies.
image.

Thank you for your efforts into trying to explain what LD is (Although I suspect as contrary to what was posted, I think more than just a handful of us understand the ramifications of LD) but you're exaggerating things a bit to prove your point. Fact is independent media folks saw this TV running demos and movies and if it was not convincing, at least on some level, we would've heard so.

The difference between 3000:1 and 100000:1, while very substantial, has to be picked up by the eye, and Samsung probably has complex blending algorithm that makes this less noticeable. Thus by using its own static CR, the TV can blend the backlight dimming gradually. I'd be VERY surprised if they didn't do something like that. That's also why a TV with a good static CR will still be better even using LD technology. Ya, it trumps some of the advantage of LD... Guess we'll have to wait and see.

What I would be worried about is motion, which has to potential for having these *blending* algorithms creeping up. Someone said movie credits and that's one key thing I'll be personally looking at.

On a funny note, using the X-wings to illustrate is a rather good choice. The original movie had horrible, quite noticeable greyish squares around the X-wings during the final battle. They fixed it good in the later versions but I remember being quite upset the first time I saw the movie at home :)

[dicey]

Just to reiterate, the contrast ratio will always be at least as good as the best Samsung LCDs that exist, but with complex scenes, the CR will be knocked down to near non-dimmed levels.
Very well put. Plus, you also get the +100% NTSC color gamut that only LEDs can provide.


On a funny note, using the X-wings to illustrate is a rather (ironic) choice. The original movie had horrible, quite noticeable greyish squares around the X-wings during the final battle. They fixed it in the later versions but I remember being quite upset the first time I saw the movie at home.
Too true! Hopefully, the new 81 series won't replace those grey squares with grey circles! :eek: :D JK!

[bpt8056]

Very well put. Plus, you also get the +100% NTSC color gamut that only LEDs can provide.

I might have missed this, but is there an official number for the NTSC color gamut on the 81 series?

[vtms]

I might have missed this, but is there an official number for the NTSC color gamut on the 81 series?
105%

http://www.macnn.com/articles/07/01/07/samsung.lcd.with.led.light/
At the head of the announcements is the company's 81 series. The company says the new LCDs, ranging between the 40-inch LN-T4081F and the 57-inch LN-T5781F, all mark the first LED-backlit TVs with local dimming. The new lighting technique dramatically improves the color gamut to 105% of the NTSC range, promising images that perfectly match the source material.

[lipcrkr]

I might have missed this, but is there an official number for the NTSC color gamut on the 81 series?

I believe it's 142%.

[darkninja67]

I might have missed this, but is there an official number for the NTSC color gamut on the 81 series?
100%

[necrolop]

Well Contrast ratios are not really what Im worried about. Its black levels, which are the first component of Contrast ratios. Anyone who has seen an LCD in a dark room knows that the blacks can be horrible, even on my XBR2 I am annoyed by how grey they are. So Im not entirely worred about ratios here, i want an inky black that appears to be empty, with no illumination. Since the array of LEDs is of low resolution, the display will have to compensate, and up the brightness of black fields in order to avoid lighting anomolies as I showed in the star wars image. This will be completely noticable, this is not something that is out of the ability of the human eye to notice. The Prototype Brightside Display has black that bacically appear as if there is no screen present in black fields, the samsung will not be able to acheive this in many types of situations, although close in many.

I have seen the Locally dimmed Samsung in person, in a well lit enviroment, which I reckon just about everyone else has seen it in. It looked great, but high contrast for a nice LCD in a lit room is not hard to acheive, nor are good black levels. LCDs have very good blacks in lit enviroments because leakage is drowned out by ambient light and there is minimal reflective light. But in a dark room, and only in a dark room, is the leakage light fully visable, and only in a dark room can the advantages of Locall Dimming and IMLED be fully seen. So in a dark room, this samsung WILL show some flaws. Either in form of Grey blacks, or halos, or both. How acceptable they are will be up to the consumer.

[lipcrkr]

For those people who are a little lost on what "necrolop" is saying let me put it a little simpler. The 81 series will probably have the best overall CR of any LCD, however, on normal viewing the CR numbers will come further down, sometimes a lot further. Here is the reason:
Although the LED's will be able to turn off in certain portions of the screen, something a CCFL can't do, allowing you to greatly lower the blacks, there will be situations where a lighted object in front of a black background, for instance, will have to cheat a little in order to properly give you a contrasting image. So some of those "blackened" areas need some light to get thru thereby lowering the CR. Movies would probably lower the CR more so than a sporting event or a HD video program. My question is will this present a situation similar to what industrial Panny plasmas had a couple of years ago with "floating" blacks. Will the eye percieve a change in black level due to some pixels being lit inside a predominently black area?

[studdad]

I might have missed this, but is there an official number for the NTSC color gamut on the 81 series?

I don't care what it is, just don't let it have a green hue, grrrrrrrrrrrrr.

[necrolop]

For those people who are a little lost on what "necrolop" is saying let me put it a little simpler. The 81 series will probably have the best overall CR of any LCD, however, on normal viewing the CR numbers will come further down, sometimes a lot further. Here is the reason:
Although the LED's will be able to turn off in certain portions of the screen, something a CCFL can't do, allowing you to greatly lower the blacks, there will be situations where a lighted object in front of a black background, for instance, will have to cheat a little in order to properly give you a contrasting image. So some of those "blackened" areas need some light to get thru thereby lowering the CR. Movies would probably lower the CR more so than a sporting event or a HD video program. My question is will this present a situation similar to what industrial Panny plasmas had a couple of years ago with "floating" blacks. Will the eye percieve a change in black level due to some pixels being lit inside a predominently black area?

correct, See I think you could trick the eye, I dont think halos are inherently bad. I think at a viewing distance, the contrast would make your eye not see the halos, just as long as the halo is small enough. But in a star filled sky, the stars wouldnt just have grey halos around them, the whole sky would have to be grey, thus no advantage of local dimming because the clusters are too big.

NOTE:
When I say the word grey, I mean the Pixel is in BLACK position, and the backlight is lit. When I say Black, I mean Both Pixel and Backlight are in BLACK position. Using this logic, the darkest color a normal LCD makes is Grey. Just to avoid confusion.

[mike123abc]

Well Contrast ratios are not really what Im worried about. Its black levels, which are the first component of Contrast ratios. Anyone who has seen an LCD in a dark room knows that the blacks can be horrible, even on my XBR2 I am annoyed by how grey they are. So Im not entirely worred about ratios here, i want an inky black that appears to be empty, with no illumination. Since the array of LEDs is of low resolution, the display will have to compensate, and up the brightness of black fields in order to avoid lighting anomolies as I showed in the star wars image. This will be completely noticable, this is not something that is out of the ability of the human eye to notice. The Prototype Brightside Display has black that bacically appear as if there is no screen present in black fields, the samsung will not be able to acheive this in many types of situations, although close in many.


They might also compensate by turning down the brightness (in the bright areas), so they do not have to turn up the back light beyond the LCD panel's ability to effectively block it out. They probably know what the eye can detect and probably will use an ambiant light sensor to make adjustments. If they sense a dark room they can use lets say 50% brightness and know the eye will see it as white when compared to the black around it. In a brightly lit environment they can turn up the average backlight and know the black is washed out anyways.

Perhaps waiting for the second generation of sets might be nescessary (although I cringe at having to wait another year, maybe just replace the bedroom plasma for now). Next year probably will see more LED drivers, giving a finer grain, and more compute power to do a better job of balancing dark and light area needs. But, I suspect that in general outer space shots are going to be the hardest, and even the full brightside backlight resolution would have reduced contrast on space shots.

But, it could be better than the 65s if indeed this is a 10 bit 8th generation panel.

[go81]

Tube tv will do worse light leakage around bright areas. (150:1 static when compared to 81 series 2000:1 static) It's really no problem with 81 series. Human eye is limiting factor here.

[necrolop]

I disagree with the assertion that the human eye cannot see a difference. In ratios maybe not, but in blacks yes, in a room with no light, its easy to see a field that as even slightly lit by leaking light.

Here is a summary of what I have said:

On complex material black levels will be similar to a normal LCD
You can definately see leakage on an LCD in a dark room, no panel will stop that.
The problems of this screen will ONLY be visable in a dark room
Local dimming will make almost no difference when in a lit room


I will probably shell out the cash for this, then do it again next year. Even though I really shouldnt, and cant really afford to. haha.

[spincut]

I disagree with the assertion that the human eye cannot see a difference. In ratios maybe not, but in blacks yes, in a room with no light, its easy to see a field that as even slightly lit by leaking light.

Here is a summary of what I have said:

On complex material black levels will be similar to a normal LCD
You can definately see leakage on an LCD in a dark room, no panel will stop that.
The problems of this screen will ONLY be visable in a dark room
Local dimming will make almost no difference when in a lit room


I will probably shell out the cash for this, then do it again next year. Even though I really shouldnt, and cant really afford to. haha.

yeah you lost me a long time ago. but basically local dimming will look bad in a dark room and no better in a lit one?

my question was going to be though that despite this issue (which i have trouble understanding, where was the star wars picture you were reffering to?) is it still going to be generally superior looking to anything out now, or will this issue make it end u pbeing worse???

[go81]

Have you seen the 81 series or why are you so certain about those issues ;) 81 series won't be perfect, but it sure does beat crt in blacks&contrast.

[petrovy]

you can wait until Fall :D :D (maybe) for the Philips 47in LED BLU, so if you guys haven't seen this yet:
http://aving.net/usa/news/default.asp?mode=read&c_num=35258&C_Code=09&SP_Num=0

[necrolop]

yeah you lost me a long time ago. but basically local dimming will look bad in a dark room and no better in a lit one?

my question was going to be though that despite this issue (which i have trouble understanding, where was the star wars picture you were reffering to?) is it still going to be generally superior looking to anything out now, or will this issue make it end u pbeing worse???

I may be being a bit overly critical here. These problems will exist in the sets, but I do beleive this thing will look amazing, better than any other LCD, just how much better, will varie.

[vtms]

Necrolop, I see what you're saying and why and I think we had talked about this in the first pages of this thread. Yes, I agree that a star-filled night sky scene might pose great challenge to IMLED displays, let alone ones with local dimming where LED clusters are even bigger. If the local dimming algorithm can handle that scene, it should handle everything else. That would be a good test for early owners/potential buyers of 81 series sets to conduct. I think we can all agree with that much.

So here's how BrightSide algorithm solves the problem of illuminating a bright spot against dark background *when the LED cluster is bigger than the bright spot*.

(Here's how bit-tech explained this. (http://www.bit-tech.net/hardware/2005/10/04/brightside_hdr_edr/7) )

http://www.bit-tech.net/content_images/2005/10/brightside_hdr_edr/image_processing.jpg

Notice that the bright region that LED array produces here is bigger than the bright spot in the HDR picture, yet after "LCD correction" the image is reproduced correctly (in theory, of course). This is when go81's statement about Samsung's high panel contrast comes into play which basically means that the algorithm will be much more effective at making those LCD corrections. The higher panel contrast, the less backlight resolution is required and the lower panel contrast the higher backlight resolution is required to produce the same picture (with the same static CR, that is).

I must admit I'm starting to believe that 81 Series' "big" LED clusters might not compromise static CR as much as I had previously thought they would. Obviously the last word on this will belong to someone who gets a chance to see how star-filled night sky scene looks like on 81s with his own eyes. Can't wait for the owner's thread. :)

[necrolop]

You are right, but just remember, correcting a black sky means illuminating it ;)


Im sorry if Im being a wet blanket haha, Im still very excited about this TV so lets all keep out hopes up.

[vtms]

you can wait until Fall :D :D (maybe) for the Philips 47in LED BLU, so if you guys haven't seen this yet:
http://aving.net/usa/news/default.asp?mode=read&c_num=35258&C_Code=09&SP_Num=0
Yup, still waiting for these. Let's take bets on how many clusters this thing will have. Just kidding. :p

[go81]

This photo of sharp mega contrast (1200:1 native panel) gives a good idea of local dimmed starfield.

http://i7.tinypic.com/66nllxg.jpg

As you can see, black levels will rise. Nothing ugly though...

tip: cover the centre if your eyes can't see the difference :p

[vtms]

Here's simulated brightness difference between brightside and samsung 81 series displaying that Agatha Peak picture:
http://i13.tinypic.com/53z8ggp.jpg
White color means that there's no difference. Black would mean 100% difference. As good as brightside? Almost. :D
If that's the case then I might not wait for 2nd gen of this technology. Incredible.

[vtms]

This photo of sharp mega contrast (1200:1 native panel) gives a good idea of local dimmed starfield.

Wait, I thought Sharp's Mega Contrast LCD didn't use LEDs or local dimming.

[go81]

Wait, I thought Sharp's Mega Contrast LCD didn't use LEDs or local dimming.


It's a led lcd.

[vtms]

It's a led lcd.
Perhaps it is, even though it's been reported to have CCFL backlight.

http://www.digitaltvdesignline.com/news/171203011
Sharp introduced a new structure for the panel and also reviewed various aspects from light scattering to polarization, but Mizushima declined to go into detail about the technology that enabled the high contrast ratio. The backlight uses conventional CCFLs (cold cathode fluorescent lamps).

[go81]

Perhaps it is, even though it's been reported to have CCFL backlight.

http://www.digitaltvdesignline.com/news/171203011

Well at least it looks like one. Black level would remain uniform if the panel had 1M:1 native contrast. Another option could be "sandwich" lcd, but again, black level would remain uniform. When they first showed the mega contrast lcd, you had to wait for the next frame to show up.(dimming algorithm wasn't fast enough for real time viewing?) Perhaps they will bring it to the consumer market next year for obvious reasons...

[Dominus]

you can wait until Fall :D :D (maybe) for the Philips 47in LED BLU, so if you guys haven't seen this yet:
http://aving.net/usa/news/default.asp?mode=read&c_num=35258&C_Code=09&SP_Num=0

I like that "Transformer" bezel. :)

[CMRA]

If that's the case then I might not wait for 2nd gen of this technology. Incredible.

Yes, indeed. So...where will Sammy display these? That $3k + price tag isn't looking so bad now.

[Flash01]

You are right, but just remember, correcting a black sky means illuminating it ;)


Im sorry if Im being a wet blanket haha, Im still very excited about this TV so lets all keep out hopes up.

As I said 4-5 days ago, it is fun to see people speculating and arguing about this technology. You have done so not only in a very polite manner, but have worked to provide example of your points. I may not completely agree, but your logic behind it all seems accurate, for all I know. Just not sure that the algorithms Samsung uses to correct backlight variations will be that easily noticeable, but that is something we can continue discussing later on...

Can't wait to see the darned thing in a store!

[AndyW35]

I was following links on the Internet and found this thread, a very enjoyable read. I see the matte or gloss screen certainly divides opinion ! Currently I still use an old 32inch 100Hz Sony CRT widescreen but am finally going to go to LCD and looking forward to the reviews to see if this tv is as good as the specs suggest.

My CRT obviously has reflections but what I find more annoying about it is the silver colour of the screen surround which has annoyed me more and more over the years, it distracts from the image for me. So you can tell that to me the ergonomics and design of this Samsung are a lot better already because it is black.

I am looking forward to my first steps into the LCD tv world, I just wish they did a 37inch one of this if it is the real deal, sizes tend to cater towards bigger rooms in the US than the UK, but then again the US does buy a lot more LCD's than the UK :)

So I am looking forward to how good it is, hopefully reviews will be around soon.

Many thanks for teaching me things I did not know over the last 60 pages.

Andy

[Admiral Ackbar]

you can wait until Fall :D :D (maybe) for the Philips 47in LED BLU, so if you guys haven't seen this yet:
http://aving.net/usa/news/default.asp?mode=read&c_num=35258&C_Code=09&SP_Num=0


Yeah that thing is MIA. I was really hoping to have choices when it came to LED backlit LCD TVs.

[spincut]

I may be being a bit overly critical here. These problems will exist in the sets, but I do beleive this thing will look amazing, better than any other LCD, just how much better, will varie.

yeah but i guess what makes me wonder is, these are problems unique to LED, so presumably, they would have to be less noticeable than whatever issues one gets from using CFFL, in other words, the net result is a picture better than the current CFFL ones in every way still, right?

also, what was this picture you had as en example? i think i'd still like to see it.

[Alex the Great]

Technically speaking LD can be done with current CCFL. IMO there are just not too many sources (24-28?) for efficient dimming.
BTW the dimming option presents in SM of each Sammy LCD TV.

[go81]

Technically speaking LD can be done with current CCFL. IMO there are just not too many sources (24-28?) for efficient dimming.
BTW the dimming option is presents in SM of each Sammy TV.

CCFL is practically useless for local dimming. CCFL isn't instant on technology and lacks various brightness levels of LED lamp.

[punchwalk]

Can anyone here comment on the long term viability of LED BL? How long of a functional life can one expect the LED arrays to have?

[Alex the Great]

Can anyone here comment on the long term viability of LED BL? How long of a functional life can one expect the LED arrays to have?

AFAIK 70-80K H

[Alex the Great]

CCFL is practically useless for local dimming. CCFL isn't instant on technology and lacks various brightness levels of LED lamp.

Im not quite agree with you. Lets look at BFI technology:
http://www.xbitlabs.com/articles/other/display/lcd-parameters_4.html
Coz this is not the LD technology, but it is quite similar and based on dimming of the BL lamps (presumably CCFL). Dimming time for each lamp is 1 msec. (1/60/16)

P.S. I'd recommend to read the quote article Contemporary LCD Monitor Parameters - I've found it very useful.

[go81]

Im not quite agree with you. Lets look at BFI technology:
http://www.xbitlabs.com/articles/other/display/lcd-parameters_4.html
Coz this is not the LD technology, but it is quite similar and based on dimming of the BL lamps (presumably CCFL). Dimmind time for each lamp is 1 msec. (1/60/16)

Turning CCFL off is not good for color accuracy. LED doesn't have this problem.

[coolsixnine]

Hello all,

Not a newb here as far as general experience goes...just planning the purchase of a high end panel with-in the next 3 months.
Having read just about this entire thread. I have found the majority of it extremely technical and a good but sometimes confusing learning experience.
My question is,,,, what is the significant differences between the 71 series and the 81 series when finally released. I have searched for relevant threads on the 71 series and the majority of the time it brings me back to this one with little explanation. Is the 71 series currently available in the USA?
Links or quotes appreciated.

Thanks in advance everyone,

Brian S

[Alex the Great]

CCFL dimming is not good for color accuracy. LED doesn't have this problem.

Wrong!
Dimming does not affect CCFL or LED color accuracy.
CCFL Intrinsically has poor color accuracy with or with out dimming.
Go and read the document I've quote before.

[go81]

what is the significant differences between the 71 series and the 81 series when finally released.

71 series = normal yearly evolution
81 series = lcd revolution

81 series is a bargain, don't even consider 71 series.

[go81]

Wrong!
Dimming does not affect CCFL or LED color accuracy.
CCFL Intrinsically has poor color accuracy with or with out dimming.
Go and read the document I've quote before.

Turning off CCFL does affect color accuracy. It's not instant on tech and requires some time(30minutes) to warm up and reach full accuracy.

[darkninja67]

81 series is a bargain, don't even consider 71 series.
I for one am waiting for the street prices to determine if the 81 is worth the premium.

[talman]

71 series = normal yearly evolution
81 series = lcd revolution

81 series is a bargain, don't even consider 71 series.

++

As someone who bought the original Sony XBR LCD right before the 1080p panels started to hit the market I'd absolutely wait and see what the early reviews are on the 81 series.

71 series = CCFL backlight and 25,000:1 contrast ratio
81 series = LED Backlight and 100,000:1 contrast ratio

[jksgvb]

I'm wondering about the economics of LED backlight resolution. We assume that this Samsung has only 60 to 70 modulated backlight arrays because to modulate more would be prohibitively expensive, but is that really the case? For example, I can buy an entire video camera retail for $250 that has a video display with 114K pixel resolution. How much would it cost for the viewfinder display only--$20? And how much would it cost the manufacturer for the display control circuitry only?

I understand that Samsung already has more than 2000 LEDs for this model. That, to me, is the big expense, and they've already paid it. So why not modulate all the backlight LEDs individually?

[go81]

I understand that Samsung already has more than 2000 LEDs for this model. That, to me, is the big expense, and they've already paid it. So why not modulate all the backlight LEDs individually?

2000+ led elements. It's not processing issue, more like led driver cost issue.

[Alex the Great]

Turning off CCFL does affect color accuracy. It's not instant on tech and requires some time(30minutes) to warm up and reach full accuracy.

You are wrong again. I am talking about dimming and you are arguing about could start. This is not the same. Also please note that every CCFLs in each computer monitor or TV set are “dimming” or blinking if you will, with about 200 Hz frequency anyway. But you cannot see it because this beyond human ability.

BTW I am not promoting CCFL LD, my point is that it is technically feasible.

[go81]

You are wrong again. I am talking about dimming and you are arguing about could start. This is not the same. Also please note that every CCFLs in each computer monitor or TV set are “dimming” or blinking if you will, with about 200 Hz frequency anyway. But you cannot see it because this beyond human ability.

BTW I am not promoting CCFL LD, my point is that it is technically feasible.

Doing BFI by turning off CCFL will screw up color accuracy. LEDs have also much wider dimming range/steps. CCFL is slow and has limited dimming range, it's no good for local dimming lcd.

[irkuck]

Local dimming future is bright:

http://money.cnn.com/news/newsfeeds/articles/primenewswire/122928.htm

[mark_1080p]

71 series = CCFL backlight and 25,000:1 contrast ratio
81 series = LED Backlight and 100,000:1 contrast ratioThe 71 series is a major step forward if it really does 25K or by the usual scaling factor 5000 static. That is nothing to sneeze at.

However, I see no reason why the 100,000 number could be STATIC, as long as that frame is not too bright, i.e., has lots of dark patches. After all, one has complete control over sectors of LEDs. Again, my hope for LED is that it will significantly mask panel imperfections with local dimming.

Another factor not much discussed here is form factor. It should be possible to make these displays significantly thinner and lighter, a major positive. I would assume for this generation Samsung is not going to make a special box for the 81, but it will be nice having still thinner displays.

Now if they made a matte version, but not as matte as the 61, put it in a 61 series box, and installed a CableCARD slot, I'd be very pleased :D .

[go81]

71 and 81 series(w/o local dimming) have 2000:1 static contrast. Pioneer 8G (768p) has 6000:1 static contrast. However Pioneer static contrast decreases to 3000:1 with bright material like hockey. 81 series can have 50000:1+ static contrast on certain scenes when local dimming is turned on.

[Alex the Great]

Doing BFI by turning off CCFL will screw up color accuracy. LEDs have also much wider dimming range/steps. CCFL is slow and has limited dimming range, it's no good for local dimming lcd.

Each and every CCFL goes OFF and ON 200 times per second all the time. So what? How come your color accuracy still OK?

[lorenmc]

Do you think that local dimming is something you are going to want to turn on/off depending on the material you are viewing?

[talman]

...and installed a CableCARD slot, I'd be very pleased :D .

You sound like a prime candidate for a Tivo Series 3! :D

I love my series 3!

[Flash01]

Each and every CCFL goes OFF and ON 200 times per second all the time. So what? How come your color accuracy still OK?

Well, CCFL aren't the fastest technology but I think it may be doable.

Here's an sample CCFL Driver.
http://www.microsemi.com/datasheets/ps_lxm1614e-14-11_rev0_2.pdf
I'm by no means an expert in CCFL inverters (Although I know other inverters quite, quite well), they say you should use a PWM signal to dim the CCFL, frequency should be 400Hz to 4K. Sounds mighty fast to me but it prolly take a longer time for the CCFL to react, but that is out of my expertise.

[mark_1080p]

71 and 81 series(w/o local dimming) have 2000:1 static contrast. How do you get to that number? Given the usual dynamic/static ratio of 5/1, its not even close. Sharp d92's are speced at 3000:1 and they are not THAT much different from the current Samsungs, given professional reviews, so it is reasonable to spec the current 65 series at 3000:1. That would give a 5/1 ratio.

[jksgvb]

2000+ led elements. It's not processing issue, more like led driver cost issue.

Ahhh right. I had forgotten about the drivers. They would probably have to be buck converters, or at least charge pumps, to be efficient enough for the job.

[JoeSony]

How will Samsung price the 81 series in relation to Sony's XBR 4 & 5 models?

[soloist3]

One additional problem, or inherent quality of these new LED backlit LCD's, is that they will now be susceptible to uneven screen wear. Not quite the same thing as burn-in but since there are now multiple illumination sources if you were to leave a video game on pause for awhile or have something with very high contrast on for awhile you would theoretically be wearing out certain clusters of LED's. This is very negligible but it does mean that LCD's will not be as indestructible as they once were.

Also, someone keep me up to speed, what other problems still exist with LCD technology, the vertical and horizontal angular issues (gamma changes on vertical angles, color purity on horizontal angles). LED clusters will inherently have some "haloing", they naturally put out polarized light, and there is still a motion blur issue that has yet to be resolved; 120hz refresh rate is not going to change this, it is all about the twisting action of the liquid crystal material, LCD's sort of "morph" into each new frame, that is what causes the blur; the only way of fixing that is to decrease the twisting/untwisting time of the liquid crystal material, though this results in a much poorer quality picture.

I still cannot wait for these new LCD's to come out but it seems that LCD still has quite a way to go (though they are getting there, or so it seems). Two big questions I still have is, how much is the 40" 81 series Samsung going to cost (street price) and WHEN WILL THIS TECHNOLOGY BE AVAILABLE FOR SMALLER SCREEN SIZES, LIKE 32" AND 37". These new LCD displays would be the perfect things to replace bedroom TV's, which in my home are still dominated by CRT's, just because the contrast is so incredible (and they are pretty close to indestructible even when playing video games; because the phosphors are not being driven that hard). Anyway, if anyone can help me out with some of this info that would be great.

[coolsixnine]

71 and 81 series(w/o local dimming) have 2000:1 static contrast. Pioneer 8G (768p) has 6000:1 static contrast. However Pioneer static contrast decreases to 3000:1 with bright material like hockey. 81 series can have 50000:1+ static contrast on certain scenes when local dimming is turned on.
Ok...reading this thread and learning,learning,learning. Can you please explain what "local dimming" means?

TIA,

[Alex the Great]

How do you get to that number? Given the usual dynamic/static ratio of 5/1, its not even close. Sharp d92's are speced at 3000:1 and they are not THAT much different from the current Samsungs, given professional reviews, so it is reasonable to spec the current 65 series at 3000:1. That would give a 5/1 ratio.

IMO 5:1 ratio is only common for conventional LCD w/o LD. LD does change the situation and dynamic CR can be infinite (with 100% dimming). I think Sammy is claiming kind of average dynamic CR - can be as big as you like :)
IMHO

P.S. Most likely the degree of dimming will be user adjustable and/or limited by say 80% or even 50%. But this is speculation, we have to wait for manual availability

[Admiral Ackbar]

@coolsixnine

Basically it means the TV can shut portions of the backlight off to create total blackness. To go into more detail, current LCDs have a backlight that is always on. To make portions of a screen black they simply turn the LCD to full power (full power on a LCD crystal means black, but some light still gets through, hence, non-infinite contrast ratios). If you can turn off portions of the backlight, you can make whole areas completely black.

[Flash01]

One additional problem, or inherent quality of these new LED backlit LCD's, is that they will now be susceptible to uneven screen wear. Not quite the same thing as burn-in but since there are now multiple illumination sources if you were to leave a video game on pause for awhile or have something with very high contrast on for awhile you would theoretically be wearing out certain clusters of LED's. This is very negligible but it does mean that LCD's will not be as indestructible as they once were.


Aside from incorrect driver design causing LED Burn (which would be surprising given the maturity of LED driver technology), I wouldn't be concerned about degradation. There are FEW lighting technology that are more robust than traditional LEDs. LEDs are inorganic semiconductors (though there exists Organic LEDs, called OLEDs) encapsulated in plastic and thus not as prone to localized heatup, do not have high transient dissipation (it is perfect for signal transmission, as Optocouplers and IR sources, like in remotes), are not susceptible to shock.

I will however point out that your are right assuming they *could* wear out slightly, although if I recall the data I saw a while back, it takes quite a bit of time for that to become noticeable. Maybe my data was out of data or based on specific LED materials. I think the lifespan of the TV will be complete when the LED wear out, bout you bring an interesting point... Hmm maybe I can check this out...

[LaserEdge]

Talked to my local Magnolia Sales Manager and here is the info he passed along to me. Their current ETA in store for the 81 is September 10. No discount off the MSRP with them. Still fits my budget. I will preorder it with them in the next day or so. Don't want to risk not going with a B&M on this one as it is a new technology.

FYI planning to use this as a computer monitor and hooking it up to an XBox 360. I don't watch TV at all.

[LaserEdge]

One additional problem, or inherent quality of these new LED backlit LCD's, is that they will now be susceptible to uneven screen wear. Not quite the same thing as burn-in but since there are now multiple illumination sources if you were to leave a video game on pause for awhile or have something with very high contrast on for awhile you would theoretically be wearing out certain clusters of LED's. This is very negligible but it does mean that LCD's will not be as indestructible as they once were.

Read in a white paper some where engineers are accounting for this. Over time the drivers to the LEDs increased voltage to compensate. It does so through a sensor that detects the decrease in brightness. No clue if the 81 includes this self-correcting circuit.

[LaserEdge]

I'm wondering about the economics of LED backlight resolution. We assume that this Samsung has only 60 to 70 modulated backlight arrays because to modulate more would be prohibitively expensive, but is that really the case? For example, I can buy an entire video camera retail for $250 that has a video display with 114K pixel resolution. How much would it cost for the viewfinder display only--$20? And how much would it cost the manufacturer for the display control circuitry only?

This is just idle speculation based upon the white papers I have read. The perspective comes from being a digital hardware engineer. It does not make much sense that a significant cost increase would occur in the digital hardware as the design moves towards IMLED. Suspect the real cost occurs in the driving circuits become much more complicated.

[glow11]

Just saw a blurb about the 81 on another site stating

"It has 10-bit processing, which means 8-bit color, which means 16 million colors."

I thought I saw something mentioned about this in this thread, but I couldnt find it.
Is his statement correct?

[LaserEdge]

Do you think that local dimming is something you are going to want to turn on/off depending on the material you are viewing?

Seriously doubt local diming is something that one will be able to turn off in the 81 seriers. Local diming is not the same thing as dynamic contrast setting on an LCD panel. They work entirely differently. The on screen menu for the 81 series is going to be different in some areas as compared to the 71. The details will be apparent when we have access to the manuals.

[vtms]

This is just idle speculation based upon the white papers I have read. The perspective comes from being a digital hardware engineer. It does not make much sense that a significant cost increase would occur in the digital hardware as the design moves towards IMLED. Suspect the real cost occurs in the driving circuits become much more complicated.
I recall reading somewhere that this is indeed the reason why 81s don't have IMLEDs.

[mrjgkelly]

Just saw a blurb about the 81 on another site stating

"It has 10-bit processing, which means 8-bit color, which means 16 million colors."

I thought I saw something mentioned about this in this thread, but I couldnt find it.
Is his statement correct?
Yes.

[LaserEdge]

Just saw a blurb about the 81 on another site stating

"It has 10-bit processing, which means 8-bit color, which means 16 million colors."

I thought I saw something mentioned about this in this thread, but I couldnt find it.
Is his statement correct?

For local diming tech 8-bit color does not necessarily equal 16 million colors. It doesn't have to be engineered that way.

Why does 10-bit processing equal 8-bit color? That is not clear either.

[mark_1080p]

IMO 5:1 ratio is only common for conventional LCD w/o LD. LD does change the situation and dynamic CR can be infinite Alex, I agree, I was just objecting to his implication that the 71's will only come in at 2000 static and since they claim 25,000 dynamic and given continuity in Samsung's specs that would imply the 65 series is only 2000(15000/25000)=1200 spec. So I am not buying his 71 series number.

[vtms]

Just saw a blurb about the 81 on another site stating

"It has 10-bit processing, which means 8-bit color, which means 16 million colors."

I thought I saw something mentioned about this in this thread, but I couldnt find it.
Is his statement correct?
This is confusing. 16 million colors is for 24-bit color (8 bits per channel). Now, does 10-bit processing mean that there's a total of 18 bits per channel?

[LaserEdge]

For local diming tech 8-bit color does not necessarily equal 16 million colors. It doesn't have to be engineered that way.

I want to make this clear with a simple example. Let us take the color red. With 8-bit resolution in the glass for red we already have 255 possible shades of red. Not 256 because a value of 0 is obviously black. The LED back lighting for the 81 series has 6-bit resolution of light intensity. So the maximum possible shades of red with the technology is 63 * 255 = 16065. Now this doesn't translate to 4.1 trillion colors on any one given image. But it does offer the potential of displaying much more than 16 million colors. The actual possible color range with 6-bit LED processing is 1 billion colors.

Wether or not the engineers at Samsung are using the tech in this way to achieve 10-bit color resolution even if the glass is 8-bits remains to be seen. Only saying what is possible with this technology.

From engineering perspective I have to say this technology is going to mature much faster than other display technology's like OLED and plasma. Suspect by 2010 LED BLU will likely be the dominate tech pushing the industry forward.

[yanks1927]

I want to make this clear with a simple example. Let us take the color red. With 8-bit resolution in the glass for red we already have 255 possible shades of red. Not 256 because a value of 0 is obviously black. The LED back lighting for the 81 series has 6-bit resolution of light intensity. So the maximum possible shades of red with the technology is 63 * 255 = 16065. Now this doesn't translate to 4.1 trillion colors on any one given image. But it does offer the potential of displaying up to 4.1 trillion colors.

Wether or not the engineers at Samsung are using the tech in this way to achieve 10-bit color resolution even if the glass is 8-bits remains to be seen. Only saying what is possible with this technology.

From engineering perspective I have to say this technology is going to mature much faster than other display technology's like OLED and plasma. Suspect by 2010 LED BLU will likely be the dominate tech pushing the industry forward.

since when does brightness = color? light intensity does not translate to color. color is the wavelength of light reflected, not the amount.

[LaserEdge]

since when does brightness = color? light intensity does not translate to color. color is the wavelength of light reflected, not the amount.

It does when the light source is behind an adjustable 8-bit filter. LCD pixels are light filters.

[glow11]

since when does brightness = color? light intensity does not translate to color. color is the wavelength of light reflected, not the amount.

In terms of computer graphics (something I do understand) 24 bit color is 8 bits per channel (R/G/B) creating 16+ million colors. This 16m color gamut is composed of hue, saturation, and value, so in CG, a dark red of a hue is a different color than the a light color of the same hue.

[studdad]

I want to make this clear with a simple example. Let us take the color red. With 8-bit resolution in the glass for red we already have 255 possible shades of red. Not 256 because a value of 0 is obviously black. The LED back lighting for the 81 series has 6-bit resolution of light intensity. So the maximum possible shades of red with the technology is 63 * 255 = 16065. Now this doesn't translate to 4.1 trillion colors on any one given image. But it does offer the potential of displaying much more than 16 million colors. The actual possible color range with 6-bit LED processing is 1 billion colors.

Wether or not the engineers at Samsung are using the tech in this way to achieve 10-bit color resolution even if the glass is 8-bits remains to be seen. Only saying what is possible with this technology.

From engineering perspective I have to say this technology is going to mature much faster than other display technology's like OLED and plasma. Suspect by 2010 LED BLU will likely be the dominate tech pushing the industry forward.

The new Toshiba specs say it has 14bit/12bit,,,,,,what does this mean in relation to the 10bit/8bit of the Sammy 71? Is it more color possibilities, or deaper colors, or what?

[yanks1927]

It does when the light source is behind an adjustable 8-bit filter. LCD pixels are light filters.

okay, but each pixel still has a red, green, and blue (subpixels)....twisting the subpixel just allows light to pass through. i guess im still not seeing how light intensity changes the actual color and not just the brightness

[LaserEdge]

since when does brightness = color? light intensity does not translate to color. color is the wavelength of light reflected, not the amount.

Let me break it down a little more for clarity. We can generate 255 shades of red with the LED on-value of 1. We can generate another 255 shades of red with the LED on-value of 2. Etc.

Sure there is overlap in the process and probably color accuracy issues with the nonlinearity of the design. Just saying this is possible.

[yanks1927]

Let me break it down a little more for clarity. We can generate 255 shades of red with the LED on-value of 1. We can generate another 255 shades of red with the LED on-value of 2. Etc.

Sure there is overlap in the process and probably color accuracy issues with the nonlinearity of the design. Just saying this is possible.

this still is not making sense, sorry. turning the LED backlight up or down is increasing intensity...it has no effect on the color of light the LCD pixel is allowing through. the color you see is the balance of the RGB. if you increase the intensity of the backlight, all you are doing is making each subpixel brighter, but the color remains unchanged. does anyone else want to weigh in?

[LaserEdge]

The new Toshiba specs say it has 14bit/12bit,,,,,,what does this mean in relation to the 10bit/8bit of the Sammy 71? Is it more color possibilities, or deaper colors, or what?

Both are CFFL based so the Toshiba is limited to a total of (2**12)**3 = 68 billion colors and the Sammy 71 is 16 million colors. Now color accuracy is very important. You could have 12-bit processing, but poor accuracy. The 71 Sammy could still look better with 8-bit color resolution if it had better color accuracy. In other words wait for the reviews and find out which one has better color accuracy first.

[spincut]

++

As someone who bought the original Sony XBR LCD right before the 1080p panels started to hit the market I'd absolutely wait and see what the early reviews are on the 81 series.

71 series = CCFL backlight and 25,000:1 contrast ratio
81 series = LED Backlight and 100,000:1 contrast ratio

And just to make sure though, the 81 series in the end will still look better than the 71 or anything else, even with new and unique issues, the end result will be a superior display i assume? I would say thats really all that matters, unless a really bad new issue makes it a tossup.

I'm wondering about the economics of LED backlight resolution. We assume that this Samsung has only 60 to 70 modulated backlight arrays because to modulate more would be prohibitively expensive, but is that really the case? For example, I can buy an entire video camera retail for $250 that has a video display with 114K pixel resolution. How much would it cost for the viewfinder display only--$20? And how much would it cost the manufacturer for the display control circuitry only?

I understand that Samsung already has more than 2000 LEDs for this model. That, to me, is the big expense, and they've already paid it. So why not modulate all the backlight LEDs individually?

i've been seeing this talked about alot, about how there is "only 60 something dimmable clusters" or somesuch. Well does that mean there are only that many blocks, the whole thing of which has to be dimmed? That sounds really bad, because most of the time you arent looking at a very seperate black and non black picture, often times there is color or detail within the black.

basically what i'm saying is people are complaining that it should be alot more, otherwise the dimmable and non dimmable areas wont even be close to correct, are these people far overexagerating the problem? will it be ok enough to work and be superior looking overall to the CCFL screens, or will this supposed limited about of dimmable sectors only make it look bad? and please spare me too much jargon, my head is spinning at this point.

[LaserEdge]

this still is not making sense, sorry. turning the LED backlight up or down is increasing intensity...it has no effect on the color of light the LCD pixel is allowing through. the color you see is the balance of the RGB. if you increase the intensity of the backlight, all you are doing is making each subpixel brighter, but the color remains unchanged. does anyone else want to weigh in?

Thought about it a little more are and you are correct in that it is limited to 16 million colors if the glass is 8-bits. The tech has localized digital resolution on the luminescence of those 16 million colors versus CFFL tech.

Is the human eye more sensitive to changes in luminesence than it is hue and saturation?

[yanks1927]

Thought about it a little more are and you are correct in that it is limited to 16 million colors if the glass is 8-bits. The tech has localized digital resolution on the luminescence of those 16 million colors versus CFFL tech.

Is the human eye more sensitive to changes in luminesence than it is hue and saturation?

ha, i was thinking about it a little more and was thinking u might have a point....what is the difference if u have RGB of 5 5 5 versus 10 10 10 or 200 200 200? is that considered a different color of gray? would adding different intensities give u different values?

[Alex the Great]

I want to make this clear with a simple example. Let us take the color red. With 8-bit resolution in the glass for red we already have 255 possible shades of red. Not 256 because a value of 0 is obviously black. The LED back lighting for the 81 series has 6-bit resolution of light intensity. So the maximum possible shades of red with the technology is 63 * 255 = 16065. Now this doesn't translate to 4.1 trillion colors on any one given image. But it does offer the potential of displaying much more than 16 million colors. The actual possible color range with 6-bit LED processing is 1 billion colors.

Wether or not the engineers at Samsung are using the tech in this way to achieve 10-bit color resolution even if the glass is 8-bits remains to be seen. Only saying what is possible with this technology.

From engineering perspective I have to say this technology is going to mature much faster than other display technology's like OLED and plasma. Suspect by 2010 LED BLU will likely be the dominate tech pushing the industry forward.

10 bit processing can lead to 8 bit image if panel is limited by 8 bit only. This is very brave assumption that LD LED technology has ability to individually control R, G and B LEDs in each domain. This is triple the complexity of electronics and push computing envelop way beyond the limit. In addition it will screwup all color settings... big time!

As I mentioned early in this thread Sammy has 10 bit processing in xx6x HDTVs. New Sony HDTVs V, W and XBR 4/5 have 10 processing and 10 bit panel. Sammy and Sony have shared production facility and we may see 10 bit panels from Sammy very soon.

[vtms]

BrightSide had 16 bits per color channel (8 bits for color, 8 bits for luminance). 81 Series has 18 bits per color channel (8 bits for color, 10 bits for luminance)? Can anyone confirm this?

[LaserEdge]

ha, i was thinking about it a little more and was thinking u might have a point....what is the difference if u have RGB of 5 5 5 versus 10 10 10 or 200 200 200? is that considered a different color of gray? would adding different intensities give u different values?

The best way to describe it is you can control hue, saturation and luminescense with 24-bit resolution on the LCD pane for each pixel. Then you have another 6-bits of luminescense resolution with the localized LED. So there are two ways to control luminescense on an LED BLU panel with localized diming. I am sure there is overlapping in the two controls. Couldn't even possibly say by how much. Just take the gradient from black to white. LED BLU is capable of more shades of grey than CFFL even if they have the same LCD glass.

It still remains to be seen if the engineers at Sammy are utilizing this possiblity. A good side by side comparision of the 71 vs 81 would reveal the answer.

[Alex the Great]

In terms of computer graphics (something I do understand) 24 bit color is 8 bits per channel (R/G/B) creating 16+ million colors. This 16m color gamut is composed of hue, saturation, and value, so in CG, a dark red of a hue is a different color than the a light color of the same hue.

WOW!!! 16M CG is nonsense !! You have no idea what you are talking about. Color Gamut ( color space) has no connection with color bitness (color depth) what so ever.

[Alex the Great]

The new Toshiba specs say it has 14bit/12bit,,,,,,what does this mean in relation to the 10bit/8bit of the Sammy 71? Is it more color possibilities, or deaper colors, or what?

FYI New Tosi has 12 bit processing and 8 bit panel. This is just an oversampling.

[LaserEdge]

10 bit processing can corresponds to 8 bit image if panel is 8 bit only. This is very brave assumption that LD LED technology has ability to individually control R, G and B LEDs in each domain. This is triple the complexity of electronics and push computing envelop way beyond the limit. As I mentioned early in this thread Sammy has 10 bit processing in xx6x HDTVs. New Sony HDTVs V, W and XBR 4/5 have 10 processing and 10 bit panel. Sammy and Sony have shared production facility and we may see 10 bit panels from Sammy very soon.

Sure it is pushing it. I am just saying what is theoretically possible LED BLU localized diming. Wether we see it in the Sammy 81 or the next year. I don't have a clue.

Toshiba is making big advancements in LCD TV processing because they are making SOCs (system-on-chip) to handle these complex issues. The have been using some intellecutal property to speed up their SOC designs. Lets just say I had a hand in helping design that IP. There is no reason why a well designed SOC can't do what we are talking about here.

[Alex the Great]

BrightSide had 16 bits per color channel (8 bits for color, 8 bits for luminance). 81 Series has 18 bits per color channel (8 bits for color, 10 bits for luminance)? Can anyone confirm this?

NO WAY

[LaserEdge]

WOW!!! 16M CG is nonsense !! You have no idea what you are talking about. Color Gamut ( color space) has no connection with color bitness (color depth) what so ever.

Yep. Color gamut is just a meausre of accuracy. Just because the panel is digitally set to display a certain color doesn't mean that is what actually comes out of the panel. The reason for the increased color gamut on the 81s is from using LEDs instead of CFFL for the light source.

[studdad]

FYI New Tosi has 12 bit processing and 8 bit panel. This is just an oversampling.

Ok, but then what does the following statement mean from their spec sheet, and how does that compare to what a 10bit VP/8 bit panel can do?

Sorry, forgot to paste. The info. is below.

[studdad]

FYI New Tosi has 12 bit processing and 8 bit panel. This is just an oversampling.

Ok, but then what does the following statement mean from their spec sheet, and how does that compare to what a 10bit VP/8 bit panel can do?

PixelPure 3G™14-Bit Internal Digital Video Processing

Toshiba’s 3rd Generation digital video processor runs at a fast 333Mhz, with high 14 bit internal processing for accurate 12 bit output. PixelPure™ 3G is capable of producing an amazing 4,096 levels of gradation (16 times* greater than an 8 bit processor) for a smooth, natural picture without image banding. PixelPure 3G continuously monitors the incoming video signal, adjusting multiple picture quality parameters including real speed progressive scanning, dynamic gamma, and now dynamic back-light control and improved dynamic MPEG noise reduction. Superior processing creates a superior picture. With PixelPure 3G, all 2007 REGZA models set themselves apart with incredibly deep, vibrant, sharp, clean and natural images, for home theater beyond your imagination.
Benefit: This advanced third-generation digital video processing system is capable of producing 4,096 levels of gradation for a smooth, natural-looking picture without image banding.

Oh, and if it is a 8bit panel, will the 12bit output be of any benefit?

[LaserEdge]

BrightSide had 16 bits per color channel (8 bits for color, 8 bits for luminance). 81 Series has 18 bits per color channel (8 bits for color, 10 bits for luminance)? Can anyone confirm this?

Can't confirm, but you should be careful to conclude that the LED BLU simply adds more colors. There are to many variables involved. Don't even recall BrightSide having explored that possiblity. Even if they did their video processing chip likely limited them to much less.

The other thing is its really 24-bit color resolution with 8-bits of localized luminescence. That does not equate to 16 bits per color channel.

[mark_1080p]

this still is not making sense, sorry. turning the LED backlight up or down is increasing intensity...it has no effect on the color of light the LCD pixel is allowing through. the color you see is the balance of the RGB. if you increase the intensity of the backlight, all you are doing is making each subpixel brighter, but the color remains unchanged. does anyone else want to weigh in?I think you are right. I think the whole idea of 24 bit color is that there are not 24 bits of different colors, but 24 bits of different shades and colors. If this is the case, they should change the nomenclature to "color shades" or something better.

[mark_1080p]

Also, I'd think that Samsung would be using RGB LED's in "grey mode". I can't imagine them individually controlling r,g,b LED's to change the color defined by the liquid crystal states.

[glow11]

WOW!!! 16M CG is nonsense !! You have no idea what you are talking about. Color Gamut ( color space) has no connection with color bitness (color depth) what so ever.

Excuse me, perhaps I should have said palette, or range, or spectrum so you could understand. Since I've been making my living in computer graphics for the last 20 years, thanks for telling me I dont know what I'm talking about....let my clear this up for you.

R0, G0, B0, is black. It is a color.

R255, G0, B0 is pure, bright, saturated red.

R255, G255, B255 is pure white.

R127, G0, B0 is medium red.

The fact that the 2 reds are stored as 2 separate colors by the computer means that these are 2 different colors.

This is what I was explaining to the poster who asked if the value variance (not hue or saturation) was indeed a different color in terms of digital recreation.

Oh, and please refer to the second definition of gamut on the attached link, then you'll be able to understand what I'm saying by utilizing a deeoer understanding of our language.

http://mw1.merriam-webster.com/dictionary/gamut

[mark_1080p]

The fact that the 2 reds are stored as 2 separate colors by the computer means that these are 2 different colors.Yea, 2 different colors as the term applies to displays, but not in the traditional definition of color, e.g. isn't red "red" :D ?

[glow11]

Yes, but as I stated, that's how it works when you're working with 24 bit color on a computer. Just tried to clear that up for the original poster before I was attacked...maybe I'll just not post.

[LaserEdge]

I preordered the LN-T4081. I will do some in depth testing/calibration (displayMate, various games, etc) once I get it home.

[vtms]

The other thing is its really 24-bit color resolution with 8-bits of localized luminescence. That does not equate to 16 bits per color channel.
Okay. I based my question on this bit of info on bit-tech site:
The picture is analysed 60 times per second and the brightness value for every LED in the array is dynamically adjusted from zero light (0 cd/m²) right up to 4,000 cd/m², and everything in between. BrightSide call this technology Individually Modulated LED, or IMLED. This is the foundation of BrightSide's core Intellectual Property. A full 8-bits of the display are dedicated to this luminance channel which, when combined with the 8-bits of information going to the LCD panel, gives a display capable of receiving and displaying 16-bits per colour channel.
BTW, what does "10-bit processing" mean exactly? I was under impression it meant 1024 shades of gradation.

[conan48]

This thread is going crazy. First glossy vs. non glossy. Then LED clusters and how they effect picture and now colour bit depth. Man I need a university degree just to understand half the stuff on this thread :D Good stuff. Can't wait to get the first reports on this set from actual users.

[LaserEdge]

Also, I'd think that Samsung would be using RGB LED's in "grey mode". I can't imagine them individually controlling r,g,b LED's to change the color defined by the liquid crystal states.

Yes. They will always be at the same intensity for one drver.

[mark_1080p]

Yes, but as I stated, that's how it works when you're working with 24 bit color on a computer. Just tried to clear that up for the original poster before I was attacked...maybe I'll just not post.Agreed, and keep those posts coming. We have our little battles here and there to spice up the forum :D.

[go81]

How do you get to that number? Given the usual dynamic/static ratio of 5/1, its not even close. Sharp d92's are speced at 3000:1 and they are not THAT much different from the current Samsungs, given professional reviews, so it is reasonable to spec the current 65 series at 3000:1. That would give a 5/1 ratio.

By actually measuring displays with test patterns. D92 has 2000:1 static. 65 series about 1200:1. LCD static remains always the same, but plasma static decreases when there's more white on screen. This is because light output is greatly reduced. So think plasma static contrast more as dynamic range.

[studdad]

This thread is going crazy. First glossy vs. non glossy. Then LED clusters and how they effect picture and now colour bit depth. Man I need a university degree just to understand half the stuff on this thread :D Good stuff. Can't wait to get the first reports on this set from actual users.

I know, i can't read for too long before I have to go to other threads just to "unwind", lol

[darkninja67]

I know, i can't read for too long before I have to go to other threads just to "unwind", lol
The Sharp threads always help me relax. Seeing people still trying to get a clean set puts me at ease. :)

[studdad]

The Sharp threads always help me relax. Seeing people still trying to get a clean set puts me at ease. :)

Roflamo, you are very bad man,,,lol, but I know what you mean,,,,it makes you feel good just knowing you missed that train wreck.

[darkninja67]

Roflamo, you are very bad man,,,lol, but I know what you mean,,,,it makes you feel good just knowing you missed that train wreck.
There have been quite a few train wrecks in the last year here.
I just want someone to make a tv that I can plug in, get it adjusted, and watch it. Not really asking all that much.

I have a feeling you will be grabbing a Samsung dude.

[LaserEdge]

Okay. I based my question on this bit of info on bit-tech site:

BTW, what does "10-bit processing" mean exactly? I was under impression it meant 1024 shades of gradation.

That bit-tech quote is actually misleading. Even if the video processing was done at 16-bits per channel the panel would not be able to display the entire 16- bit range for each color within a localized region. Each localized region would only be able to display less than 1/255ths of the total 16-bit range. Further more there should be overlapping of the possible colors. How much I don't know. It is a complex relationship between the LCD and the LED.

10-bit processing is just saying that the electronics in the panel does video processing using 10-bits in each color channel.

[studdad]

There have been quite a few train wrecks in the last year here.
I just want someone to make a tv that I can plug in, get it adjusted, and watch it. Not really asking all that much.

I have a feeling you will be grabbing a Samsung dude.

Yup, I have that feeling too. And I agree, all the new technology has not caught up with itself yet,,,,,and maybe never will with the rate of change, so it is really getting difficult to find the tv you requested,,,,at least without going back to tubes :rolleyes:

[studdad]

I still haven't gotten an answer to post 1866. Anyone want to take a shot at it??

[jbrazjr]

I too am looking at this set. When are you being told on your preorder that the set will be available ? Thanks for the info !


I preordered the LN-T4081. I will do some in depth
testing/calibration (displayMate, various games, etc) once I get it home.

[vtms]

I still haven't gotten an answer to post 1866. Anyone want to take a shot at it??
My similar question would be why have 10-bit processing if the panel is limited by 8 bits? And how do we know if it's 8 bits? Also, how do we know the cluster luminance range is 0-255?

[LaserEdge]

I too am looking at this set. When are you being told on your preorder that the set will be available ? Thanks for the info !

I was told it would be in store September 10. The Sales manager at Magnolia said it could be +/- 2 weeks around that date.

[glow11]

Found out today the only Magnolia Audio/Video in my area is going out of business on August 10th....I dont know of any other high end video dealers in my area. If anyone knows a dealer that is high end enough to sell 81's in the Portland area, let me know! TIA!

[go81]

Ok, but then what does the following statement mean from their spec sheet, and how does that compare to what a 10bit VP/8 bit panel can do?

Oh, and if it is a 8bit panel, will the 12bit output be of any benefit?

If processing is done in 8bit domain with 8 bit input, there will be loss in data. If processing is done in higher bit domain, there will be no loss in data.

[mark_1080p]

By actually measuring displays with test patterns. D92 has 2000:1 static. 65 series about 1200:1.OK, so you are referring to actual measured CR. I am surprised that the 65 tested so much lower. Who did the testing?

[studdad]

If processing is done in 8bit domain with 8 bit input, there will be loss in data. If processing is done in higher bit domain, there will be no loss in data.

Ok, let me get this straight.

The Samsung 71/65 has 10 bit processing and outputs to an 8 bit panel, correct?

The Quote from the manufacturer says the Toshiba has 14 bit processing and 12 bit output.

However, Alex said the Toshiba has an 8bit panel. So,,,,,,,,,,,here is what i am getting from all this:

Samsung: 10bit processing, 8bit output, 8bit panel (correct or not?)

Toshiba: 14bit processing, 12bit output, 8bit panel (correct or not?)

Now, if the above is correct, then what is the advantage of the Toshiba's extra processing and output power??? Will the 8bit panel be able to display those advantages? If the above is not correct, then what is correct, and how will that effect the display, if at all?

[go81]

OK, so you are referring to actual measured CR. I am surprised that the 65 tested so much lower. Who did the testing?

One isf calibrator :p 65 series looks so black because of super clear panel. 65 series has also screwed gamma, so it will make blacks even blacker and whites whiter. 81 series doesn't need any tricks apart from super clear to look black. It will look black in bright environment and black in a darkened environment.

[Orta]

If anyone knows a dealer that is high end enough to sell 81's in the Portland area, let me know! TIA!

Are these really not going to be available at places like Best Buy or Circuit City?

[go81]

Samsung: 10bit processing, 8bit output, 8bit panel (correct or not?)

Toshiba: 14bit processing, 12bit output, 8bit panel (correct or not?)


Samsung: 10bit processing, 8bit output, 8bit panel (correct or not?)

Toshiba: 14bit processing, 8bit output, 8bit panel (correct or not?)

Extra bits reduce the possibility of rounding errors. The panel is limiting factor here.

[studdad]

Are these really not going to be available at places like Best Buy or Circuit City?

The 71 will (LCD), the 81 will not (LED), at least not for the first six months or so.

[studdad]

Samsung: 10bit processing, 8bit output, 8bit panel (correct or not?)

Toshiba: 14bit processing, 8bit output, 8bit panel (correct or not?)

Extra bits reduce the possibility of rounding errors. The panel is limiting factor here.

Ok, but then why does Toshiba say the output is 12bit?

[LaserEdge]

Found out today the only Magnolia Audio/Video in my area is going out of business on August 10th....I dont know of any other high end video dealers in my area. If anyone knows a dealer that is high end enough to sell 81's in the Portland area, let me know! TIA!

What about the Magnolia Home Theater inside the Best Buy Gresham store? They should be able to order it for you.

[go81]

Ok, but then why does Toshiba say the output is 12bit?

Maybe the processing upsamples 8bit color to 12 bit color. That 12 bit color has to be reduced to 8 bit domain before actually displaying anything, so upsampling is just a marketing gimmick.

[studdad]

Are these really not going to be available at places like Best Buy or Circuit City?

Oh I forgot, if your BB has a Magnolia store inside it, you will be able to get it there.

[studdad]

Maybe the processing upsamples 8bit color to 12 bit color. That 12 bit color has to be reduced to 8 bit domain before actually displaying anything, so upsampling is just a marketing gimmick.

Ok, I got it now,,,thanks.

[glow11]

What about the Magnolia Home Theater inside the Best Buy Gresham store? They should be able to order it for you.

Ah yes! Forgot they were there...probably because they haven't been to obvious...maybe once the main Magnolia closes they'll ramp it up. You're right though, they should be able to order it. Thanks!

[Orta]

I see Studdad, thanks for the info. I believe mine does. The local Tweeter here just recently closed as well, I assume they would have carried them too. :(

[vtms]

Extra bits reduce the possibility of rounding errors. The panel is limiting factor here.
With that mystery solved (thanks), how many bits are dedicated to cluster luminance?

[LaserEdge]

Are these really not going to be available at places like Best Buy or Circuit City?

If a Best Buy store has a Magnolia Home Theater section inside the store they should be able to order it for you. They may not have the 81s on display thou.

Other stores like tweeter should have them too.

Beyond that AV sites like *******.com should be able to get them.

If you are looking for a 71 then yes both Best Buy and Circuit City will have them.

[LaserEdge]

Maybe the processing upsamples 8bit color to 12 bit color. That 12 bit color has to be reduced to 8 bit domain before actually displaying anything, so upsampling is just a marketing gimmick.

There is some small benefit for an extra bit or two of upsampling between the video process and the glass. From frame to frame it will help smooth the color transitions. No point in 4-bit over sampling IMHO.

[Alex the Great]

Yep. Color gamut is just a meausre of accuracy. Just because the panel is digitally set to display a certain color doesn't mean that is what actually comes out of the panel. The reason for the increased color gamut on the 81s is from using LEDs instead of CFFL for the light source.

Accuracy has no or very little sense for me. Please provide the definition of this parameter. How this accuracy is measured, what unit is used? What typical accuracy of average TV or monitor?

[Zerox_no1]

Where can I post link to pdf-files with 71 & 81 specs?

ABT has them on their site!

[westa6969]

Where can I post link to pdf-files with 71 & 81 specs?

ABT has them on their site!
WoW! Thanks that's excellent! Can't link since the censor pkg blocks it.

Right Click and choose OPEN IN NEW WINDOW! Had to compress with Adobe Professional for the second one but it works.

Alright CONAN are these fake too? ;)

[Alex the Great]

Excuse me, perhaps I should have said palette, or range, or spectrum so you could understand. Since I've been making my living in computer graphics for the last 20 years, thanks for telling me I dont know what I'm talking about....let my clear this up for you.

R0, G0, B0, is black. It is a color.

R255, G0, B0 is pure, bright, saturated red.

R255, G255, B255 is pure white.

R127, G0, B0 is medium red.

The fact that the 2 reds are stored as 2 separate colors by the computer means that these are 2 different colors.

This is what I was explaining to the poster who asked if the value variance (not hue or saturation) was indeed a different color in terms of digital recreation.

Oh, and please refer to the second definition of gamut on the attached link, then you'll be able to understand what I'm saying by utilizing a deeoer understanding of our language.

http://mw1.merriam-webster.com/dictionary/gamut

With all my respect RGB color space is just an individual case of CG. XVYCC, is another color space. Besides RGB can has different resolution 6, 8, or more bit. The fact that you spent all you life in 8 bit RGB makes you think that this is the only CG. And this is wrong.

[go81]

With that mystery solved (thanks), how many bits are dedicated to cluster luminance?

6 bits + 10 bits for panel.

[pawn45]

WoW! Thanks that's excellent! Can't link since the censor pkg blocks it.

Right Click and choose OPEN IN NEW WINDOW! Had to compress with Adobe Professional for the second one but it works.

Alright CONAN are these fake too? ;)

THANKS!

[Alex the Great]

Ok, but then what does the following statement mean from their spec sheet, and how does that compare to what a 10bit VP/8 bit panel can do?

PixelPure 3G™14-Bit Internal Digital Video Processing

Toshiba’s 3rd Generation digital video processor runs at a fast 333Mhz, with high 14 bit internal processing for accurate 12 bit output. PixelPure™ 3G is capable of producing an amazing 4,096 levels of gradation (16 times* greater than an 8 bit processor) for a smooth, natural picture without image banding. PixelPure 3G continuously monitors the incoming video signal, adjusting multiple picture quality parameters including real speed progressive scanning, dynamic gamma, and now dynamic back-light control and improved dynamic MPEG noise reduction. Superior processing creates a superior picture. With PixelPure 3G, all 2007 REGZA models set themselves apart with incredibly deep, vibrant, sharp, clean and natural images, for home theater beyond your imagination.
Benefit: This advanced third-generation digital video processing system is capable of producing 4,096 levels of gradation for a smooth, natural-looking picture without image banding.

Oh, and if it is a 8bit panel, will the 12bit output be of any benefit?

The bottom line is - 4K gradation stands for 12 bit processing. No word about panel resolution. From what I've heard Toshiba does not have 10 bit panels yet. Actually I called to Toshi tech support line and they could not confirmed more 8 bit panel. IMO it is just a marketing BS. And nobody has 12 bit panels AFAIK. Besides there are no sources of 12 bit signal anyway.
My 0.02

[gwynethh]

Well I guess that is answered. 43.7" wide?!? Hope they had a good technical reason to make it some much wider than a Toshiba 42LV144 (40.2" wide) and just not misguided aesthetics . I will have to wait another generation or get the 42" Toshiba. Thanx for the info. I will keep following this thread though.

Where can I post link to pdf-files with 71 & 81 specs?

ABT has them on their site!

[Alex the Great]

My similar question would be why have 10-bit processing if the panel is limited by 8 bits? And how do we know if it's 8 bits? Also, how do we know the cluster luminance range is 0-255?

If panel spec has 16M colors, it stands for 8 bit or 256 per channel gradation.

[vtms]

Notice "Auto Motion Plus LED" for 81 Series and "Auto Motion Plus 120Hz" for 71 Series.

[glow11]

With all my respect RGB color space is just an individual case of CG. XVYCC, is another color space. Besides RGB can has different resolution 6, 8, or more bit. The fact that you spent all you life in 8 bit RGB make you think that this is the only CG. And this is wrong.

Actually I was answering one question posted asking wheather differing luminance values constitute a different "color". I was pointing out that as far as the way computer graphics (my only area of knowlege) was concerned, the same hue of different luminance values are considered a different color.

The origin to his question was weather the varying luminance values created by the LED backlighting scheme would be considered additional colors, as far as the total number of colors the set can display. I didn't know the answer to that, but I used anecdotal evidence to state that differing luminance values of the same hue are considered different colors while working in computer graphics.

[vtms]

If panel spec has 16M colors, it stands for 8 bit or 256 per channel gradation.Yes, I'm well aware of that. My question wasn't about that, but it has been partially answered by go81.

[Alex the Great]

Notice "Auto Motion Plus LED" for 81 Series and "Auto Motion Plus 120Hz" for 71 Series.

Good news. I also noticed that LED BL add 5 extra pounds to 40" set

[kgj67]

At 43.7 inches wide it will not fit in my cabinet, 5.5" wider than the 61 series! Now looking at the Mitsubishi LT-40134 only 37" wide or the Mitsubishi LT-46144 at 42.3" wide and 120 HZ (they use the Samsung panel)