Sony Crystal LED Display - new display tech - Page 6

On 2Q revenue basis:
Windows and Windows Live 4,736
Server and Tools 4,772
Online Services Division 784
Microsoft Business Division 6,279
Entertainment and Devices 4,237

No price for guessing on profitability basis.


Neither was Obama responsible for the deficit. But the buck has to stop somewhere. "There's nothing I can do about it" is not a viable excuse for someone at the apex of the hierarchy.

What makes you so sure about this?

Very interesting that Sony actually showed a led array display now ...bold move
this unit probably costed a few hundred grands ...
what was sony's goal in prototyping this screen and showing it off at ces ??
they didn't have any ready OLED prototypes to show and they wanted to get some attention or ?

on another hand,
this type of display is exactly what i wanted to do a few years back looking at diy possibilities ..

problem is LED size and of course price

the smallest usable LED i've found are RGB modules of 1.5mm X 1mm from ROHM company...
http://www.rohm.com/ad/picoled-rgb_p...duo/index.html

that would work quite good for a 130-140" wide screen at 1080p

problem is still the cost ... 2073600 LEDS to purchase
how much each ?? even at that quantity i'm pretty sure it can't be under 10cents each
that's still 200 000$ only for the leds

so again..why is sony even looking into that ?
even at 1 cents each led , still would cost 20 000$ only for the leds to produce a 1080p
display
and how the !@#@#^%$% they found leds that are small enough to product 1080p on a 55" screen ?? ( would require approx 0.25mm of width per pixel ..that inclueds LED , its lens and the space in between each led .... ) i'm perplex that something that small actually exists ...


so dudes, i will be receiving my 5'X10' CNC router later this month
maybe we can all put some $$ together
and fund my new 130-140" LED array prototype screen ???

also about the heat ...

if they state 70W
the panel can't output more than 70W of heat
( which it won't since some of this energy is used for light and some for heat ...
don't remember the output efficiency of current led technology ..but it is probably 10 times better than the ~10% of incandescent bulbs .. )

let's assume 100W of power is used on a 50-55" panel
50% efficient ....so 50W of heat
dissipated on the 50" ?? that's like nothing at all


Sony might not be a large LED player at all,
but it might also have research focus on a single led chip design solely for this kind of display product ...
probably have found a way to spray or electro deposit led materials on a waffer
else i don't see how they can even think of producing a display with 2-6 million individual leds ... even if the leds are on large chunk chips

the only way to get near the 0,25mm spacing required is spraying or similar ...
can't see it possible with another way


tought, if you look at the rohm website i've linked before this post,
the led chips are larger than the actual led spots we see
maybe if large numbers of such are produces, they can make them much closer

Sony would use chips that each contain ~10,000 LEDs or more. The chips would be made like chips... I'm not sure how much you know about semiconductors, but simple ones can be made very cheaply.

Hey wass up Rogo ? haven't posted here since a few years ... ur still daily working around here?

i do not know much about manufacturing of semi and leds...

but logic to me is that even at pretty large chips,
the price per pixel would be much higher than it is with printed stuff for sure
as SS inorganic leds require a bit more complex materials and setup ..


though as i said, it is unlikely that sony has made up this prototype just for fun,
so they might have invested in the RD and manuf. of special led chips

i've done some more research with many led manufs, and i can't even find ones that produce the required pixel size for the sony display


also, at that count, the size will play a major role in the design and manufacturing , so it will probably be impossible to manufacture smaller screens with 1080p rez

anyone done any searches on sony's patents yet ?
they usually register before showing off the products so that competitor doesn't steal off the patent while they are working on it

cheers

"it will probably be impossible to manufacture smaller screens with 1080p rez"

Absolutely true. At least initially.

As for what they can and cannot make, the actual size of the LEDs is just not especially small when seen as a semiconductor feature. We did some math in the thread and it was something like 150 microns. Whatever it was, it's huge on a chip-level scale. The point is that no one makes an LED that small because it's an otherwise useless LED, but if you are making thousands for a TV like this, it's a very doable size.

Sony already makes some microdisplay-type products based on this technology. That's a far cry from making these TVs, so I wouldn't assume anything is going to happen, but it's certainly plausible they could decide to go down this path.

ROGO : you have an example of a sony product that uses this type of display ?

please link/explain the 150micron minimum size ??

for a ~55" i get 250um of space for each pixel ... 150 micron minimum size + the spacing in between ... not much left


i'd say if they figure out a way to have the actual die to take most of the front space, and all else connected under it

not too sure about how the phospor/die thing is applied at manufacturing leds,
but could it be like printed or sprayed ?

i remember some paper talking about some company X tinkering
with some spraying technique ( was more like gravity deposition )
of bioluminescent materials on a substrate layered with a grid
( let's assume some kind of plate with holes on it..or a tight wire grid )
the grid was then removed after some backing or i don't remember what
and what was left was an almost perfect pattern of bioluminescent pixels
neway

another very small one i found

http://www.lumex.com/images/Lumex_Qu...ctBrochure.pdf

1mmX1mm for an RGB
that makes approx 330um for each pixel

but most importantly ..
if you look at the package
it is not very space wise

the led dies only occupy 1/3rd of the space
and all the remaining is for connective puposes
and they are form the sides

so if we look at those and the ROHM's ...
it must not be very hard for sony to come up with an array block
of multiple RGB leds that are even smaller than the required size
if they use connections at the back

( from the link .. i'd estimate quickly that the dies use approx 300umX 400um
let's use 500um X500um for the complete RGB unit if properly designed on an large array
with rear connection ... )

now all this makes me want to try to make my own even MORE!!!

how much you guys think i could get 2million of those for ?

would probably make a very nice 100" screen

playing with the net to see what exists in terms of rgb arrays
here a quick pic form ebay just to show how small
the dies can be ... could even be closer together

this is probably the type of array sony is using on larger chips
( this is all paralell ..but just to show what it could be .. )

ROGO : the back of the chip would require to be black for a display right ??

Read back through the thread, but I calculated the dimensions based on a separate R G and B per pixel so like 6000 pixels into 1200mm I get about .21mm per LED, incuding the white space. Again, I doubt you're going to find the part available. If they made this, they'd use something that was at least 20 x 20 RGB triads I believe.

I dont think the problem is size or heat. The problem is that the blue and green LED's are made using GaN. As far as I can tell, the LED manufacturers are still trying to create GaN on silicon based LED's. There has been a ton of work done in this field so maybe Sony has solved it but if so, a television would be an odd way to announce it. This would be revolutionary in bringing down costs in the solid state lighting market.

Maybe they have made enough progress that they can create low-power LED's but not high-power LED's? That is the best scenario I can come up with. If the LED's are still made using a sapphire substrate, I dont see how the costs are going to make sense.

@rogo (or possibly anyone is free to answer)

1) What do you think is Sony's purpose for this TV??? (this Crystal Display thing)

2) How long do you think till we see this for the next-gen BRAVIA? (estimate)

3) What about existing OLED plans of Sony? Will they ditch OLED and pursue this instead? Or would they continue to work on both?

ROGO : yes this size is what i calculated also

personally, i want to build a 120"dia + screen
so i'd be working with roughly <=1.6mm of pixel pitch ( for a complete rgb )
and the products already exist for that
my problem would be cost right now
as the best price i've found for a rgb single pixel was around 0.30$usd with a purchase of 5000+
but i'm looking at 2million +
ouch


slacker711: PLEASE elaborate..why is GAN material a problem ? cost?


Blackraven: 1 : sony since about 10 years, rarely show a fully working prototype without having a defenite plan for a final product
if they have indeed found a way to manufacture large arrays of leds that could be used with the correct pitch on this kind of display, you can expect the products to be available within 2-3 years at max
could even be sooner if they are dumping part of the oled program in favor of this technology

i don't believe sony will ditch compeltly oled
but as they did with many many products in the past,
sony tend to invest in technology patents just as much as in the products
( blur ray, trinitrion, beta and many many more .. )
they might not be a keyholder in the oled patent domain
so maybe they are trying to go with a seperate technology so they can register it for themselves and charge others for the use of it

maybe not .. but they've been doing that for a few decades so

we would need to search patents to check it out
i personally have no experience in looking for patents..maybe a more skilled member could do it out of curiosity !
'cheers all

Slaker and ROGO :
do you think a display requires high power leds or low power ?

i'd think that low power ( efficiency vs size? ) is more than enough

That's good insight. Honestly, I have no clue about what the chemistry is, but it wouldn't surprise me to learn they'd achieved success at low power and not high power.


(1) To seem relevant
(2) Probably never, but we can dream
(3) Sony talks a good game, but is neither building an OLED fab for anything larger than their tiny broadcast monitors nor doing anything that actually indicates they have any plans to commercialize OLED TVs. When a company uses wiggle words like Sony does with regard to OLED, it's hard/impossible to take them seriously as a player.


That's a pricey TV. How's your latest rap CD selling?

We can wish.


Low power should be fine. Each LED needs to output very little light. But they need to pack a lot of them together on a substrate.

It isnt the GaN itself. The problem is that you need to grow GaN film on top of a substrate. Silicon hasnt worked due to various mismatches between the materials. The LED industry has instead had to use sapphire and silicon carbide. Both are much less mature than silicon and are far more expensive.

Sony needs 6 million LED's for a single television. I dont have enough of a background on this stuff to speculate on what they might be doing, but they cant be using the existing processes. There is just no way that they are economical.

Actually, there has been some recent success in the power semiconductor industry with growing GaN on silicon to produce high performance rectifiers and transistors. It could be that some of this has flowed down to the LED ndustry.

Slacker : understood..as rogo pointed out ..thanks for the chem insight on led
i dind't know it was an issue with the substrate price


ROGO: i hate rap music ..don't think i'd sell any !!
but i have my money sources ( cie WORK HARD )

i wouldn't mind paying 10 000$ for the required leds to make up my display
but that still brings down at around half a cent for rgbs and 0.0016$ each for individual colors ..

would be SUPER fun to assemble all 2million individual leds by hand with my kids
aahah

We discussed this earlier. Do quote which LED manufacturer managed to commercialize silicon LED.


As for Sony mass producing CLED, my simple observation is just that they are not even a major LED maker, nor do I hear any collaboration or JV soon.

specuvestor: pardon my ignorance
but please explain : " CLED " ??

they might also only buy the leds from another company ?

whatever it is ...they used something to create that prototype
and the resolution looks as announced from the pictures we've seen
so the size of the pixels must be in the range we calculated

as of now , i've browsed through almost all large known led manufacturer websites without finding any array or package small enough to work on that
particular display

so either it is a product unavailable on manuf websites
( which is doubt would be logical if it is a current product ..they want to sell it no ? )
or
sony has done some led investments in its factory nobody seems aware of
or
sony has ordered a custom product form a manufacturer
that is not available to other's than sony

i still think that if sony developped a led array specific for display applications,
they are going to put a patent on it as they always do

also m if you look at large pictures from CES , when the screen shows black parts, we can't see any details on the leds and spacing
so they must have worked alot on the spacing, color of the holding substrate
and color of the led substrate so that when it is off, we do not see reflected
light on the lenses or on the mirror under the dies, colors from the dies etc ...


maybe they installed 6 million tiny 2 legged leds in holes, connected them all by hand, and sprayed everything with mat black paint from the front
that would certainly explain the dark black

CLED as in Crystal LED... which is this thread

CLED is just a LED billboard shruck to 55". The challenge is of course as you stated how to get the chips small enough, arranged properly (which is an issue because I just remembered RGB is manufactured separately so they cannot be on the same wafer) and how to get them powered individually, within the constraints of heat dissipation. However as discussed previously in this thread, I doubt they will be packaged like normal LEDs, so no 2 legs

Hence the key technology is the LED layer... which Sony is not a key player. Whereas in LED LCD, there are still key technologies in the backlight unit (BLU) and VP processing to synchronise scanning etc. So yes I think it will remain prototype until someone figure out how to manufacture the panel cheaply, rather than in the lab

The only way to do this would be to manufacture R G and B on the same wafer, repeatedly. Honestly, the part I'd try to invent would be a 20 x 20 LED "chip". It would seemingly be about 1/2" square and you'd need approximately 500 of them to make the display. If you built a backplane with an alignment grid, you could just mount the chips directly to the backplane and then slap a glass on the front.

Honestly, if you could make a 100 x 100 LED chip, that would be better, but I don't know which is harder.. yielding good parts, or cutting and aligning more parts. So depending on that, it would dictate the approach.

I also am no expert on LED-on-silicon technology, but given that TI figured out how to make DMD devices using fairly standard silicon-wafer technology, and Sony, JVC, et al. eventually developed LCOS technology, I find it difficult to believe this is an insoluble problem. Of course, Switch Lighting still can't ship an LED light bulb, so perhaps LED is trickier than I think it is.

^Wow, on a side note, just read a blurb on Switch. Did you see their bulbs @ CES? I presume their bulbs are designed in such a manner that they can be used in fully enclosed fixtures without a significant reduction in their lifespans?

Switchlighting ... never heard of that company before
their bulbs look neats ..but they don't even state CRI ???
and what is the price of those ?

only 25K hours ?


i personally bought 60 CR6 and LR6 by CREE
and i can tell you guys that i don't regret any cents i invested in those ...

been testing with the LR6 for a few years ...installed already a few CR6 in my new home the light quality has been the issue with all "warm white" leds and lighting products for years

the CR6 excell brilliantly ( using many leds of many different colors combined to cover the spectrum as it should be )
and i'd go as far as to say that they even surpasses the halogen lamps i would've used in their places

the good thing.. in my bathroom for example, i use 2 10W units
instead of 4-6 50-60W halogens + the diffuse lighting
( disc shape ) instead of spot lihghting ... wow

i know i might sound like a seller here,
but i'm only trying to spread my contentment with this product ..
i invested quite a sum in those of money and time ( research and testing prior to purchase )

cheers all

CLED = CRYSTAL LED /// OMG !! i am sooooo slow minded //// ouch ...too much skyrim
must stop playing skyrim and go to sleep now !!! ...

CRI isn't all that useful of a spec. At best, it just tells you what bulbs to avoid, rather than what bulbs are good.

It's disappointing how little technical information manufacturers put out about their bulbs, it's very difficult to find the spectral distribution for most bulbs on sale today.

LED lighting just isn't there yet in my opinion. Not too long ago, we replaced 200W of halogens with 16W of Philips LEDs. The colour of the LEDs is pleasing, at 3000K it's slightly whiter than the 2700K bulbs it replaced. But there's a pinkish hue that seems to be there with all the better LED bulbs on the market just now, colour rendering isn't great, and the thing which I notice the most is that LEDs are much more directional than incandescents, fluorescents, or halogen bulbseven when specified as having the same, or even a wider angle beam. They're also not nearly as bright as they claim to be. Perhaps when viewed directly they are, but they put out less light into the room than the bulb they are supposed to be equivalent to. (and this is comparing lumen output not their "equal to 60W nonsense") These are not just issues with the Philips bulbs, I've been looking around at a variety of LED bulbs for a while now.

Here's a comparison from Cree's website:


Now I don't know what kind of fluorescents those are, but I have to assume that they're high CRI bulbs, because they talked a lot about lighting quality in the video.

First of all, look at how even the lighting is with the fluorescents compared to the LED bulbs. There is noticeable hotspotting, and some areas are in complete shadow with the new bulbs. And look at the noticeably pink hue from the LED lights. In other photos with a closer view of colour charts, some colours were noticeably brighter than before. (but only some)

The biggest problem for me is, once you buy the LED bulbs you're stuck with them for a long time. They have a high cost, and have a very long lifetime. This would be great if the lighting quality was there, but in five years time or less, I bet we have much better bulbs for less money.

Room lighting is far more demanding than LEDs for television displays. (where it's actually beneficial to have three narrow spikes rather than a uniform spectrum)

We're sort of off-topic but let me say this, we keep seeking better lightbulbs. Yes, the tried and true incandescents put out nice color and diffuse light, but they are ridiculously inefficient and really have lousy lifespans. The quality of a modern incandescent bulb is horribly low, which is one reason they are so cheap.

We've tried CFs, and they work fine in some places in the home, but as many have noted, they don't love being turned on and off, they tend to output less light then they are supposed to, and most of them are not especially dimmable. So the LED dream lives here. And I think the reasons why relate to the reasons it lives for TVs: Good power efficiency, good color, good control over the brightness, etc. etc.

I'd also like -- in my lightbulbs and TVs -- to know they are going to be highly reliable. That's the kind of thing that only time proves.

I'm not sure what comparisons you can draw about hotspotting, since it looks to me like the LED lamps in pic2 are recessed and the flourescents are not. Hardly an apples to apples comparison.

Recently I had reason to replace bulbs in an open fixture. CFs to me look awful in this situation since you can see the bare bulb. So the alternatives were halogen and incandescent. The halogen boasted 29W as equivalent to a 40W incandescent, but a comparison of the lumens indicated that scaled to 40W, the halogen produced 465 lumens where the incandescent at 40W produced 450 lumens. The price of the halogen (per bulb) was roughly 12 times the cost of the incandescent, at twice the lifetime. So for the same lumens, halogen would cost you roughly six times the price for the same life expectancy. No "truth in advertising" on that package, IMO.

BTW my house is almost exclusively CF. Most of the CF bulbs were installed 4+ yrs ago, when I moved in. I have not replaced any yet. Several years ago, when you turned on CF bulbs, they were noticeably dim and took a few minutes to get up to speed. That problem seems much alleviated these days. I KNOW they still take a bit of time to get warmed up and outputting all they can, but the difference in lumen output between cold and warm these days seems much closer. I admit I haven't done any actual measurements though. I'd rather use LED, but damn, they are still pricey!

Chronoptimist : thanks for your input

yes these does have a light pink tint also,
the CR6 i use are less affected by this
it is from the red leds used to give the warm spectral points

they combine white yellow( amber?) and red leds to get the correct spectral
and adjust them at factory on the CR6
the LR6 are auto adjusting ( u actually see the color change while startup, adjust )


brb clients..sorry