OLED TVs: Technology Advancements Thread - Page 432 - AVS Forum | Home Theater Discussions And Reviews
Baselworld is only a few weeks away. Getting the latest news is easy, Click Here for info on how to join the Watchuseek.com newsletter list. Follow our team for updates featuring event coverage, new product unveilings, watch industry news & more!


Forum Jump: 
 845Likes
Reply
Thread Tools
post #12931 of 13035 Old 02-09-2016, 04:00 AM
Member
 
Zellio2009's Avatar
 
Join Date: Dec 2008
Posts: 18
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 11 Post(s)
Liked: 13
Quote:
Originally Posted by rogo View Post
1988 was the first LCD TV, it was 14 inches. It was awful but it was a TV. Pretty sure the first active-matrix laptop was also 1988 with the Macintosh Portable, which didn't even have a backlight and was monochrome. None of the stuff that came before.

According to this, 1989 was the first year 10-inch TFT LCD monitors existed: http://www.personal.kent.edu/~mgu/LCD/index.htm

I feel pretty good about arguing 1988 was the first real year of commercialization of LCD as we understand it today. I'm also comfortable with using the 2002 date as the analogue to what we are talking about with PDP and OLED. Referring to passive matrix laptops or worse laptops with some cash-register sized LCD display as comparable confusing things. By that logic, there was a digital camera with an OLED in 2000. And a monochrome plasma in 1983 with a full-color monitor-sized one in 1992.

If we're compared apples to apples, we should just ignore the 14-inch LCD TV milestone entirely and stick with 2002. The only reason I mentioned the 1988 TV was that it technically qualifies as "commercialization of a television." That's also analogous to the milestones I listed for PDP and OLED, even though it was a tiny LCD TV. The early laptop computers are not remotely similar in that regard.
So tvs are the baseline for mature lcd production?

Alright, that lcd was tiny, but what about the 1984 data general one?


Large backlit lcd that also did cga, machine that was comparable to desktops at the time, AND the lcd still works today!

Can you say that this isn't mature enough for mass production?
Zellio2009 is offline  
Sponsored Links
Advertisement
 
post #12932 of 13035 Old 02-09-2016, 06:29 AM
AVS Special Member
 
tgm1024's Avatar
 
Join Date: Dec 2010
Posts: 9,331
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 2022 Post(s)
Liked: 1909
Quote:
Originally Posted by Zellio2009 View Post
So tvs are the baseline for mature lcd production?

Alright, that lcd was tiny, but what about the 1984 data general one?



Large backlit lcd that also did cga, machine that was comparable to desktops at the time, AND the lcd still works today!

Can you say that this isn't mature enough for mass production?
AIUI, he was saying that we should set aside the passive matrix stuff because it's confusing. Like the display in that DG.

Thought #AIA:
There is only one kind of quantum physicist in the world.
Those that understand it, and those that don't.
tgm1024 is offline  
post #12933 of 13035 Old 02-09-2016, 02:17 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
Quote:
Originally Posted by Zellio2009 View Post
So tvs are the baseline for mature lcd production?

Alright, that lcd was tiny, but what about the 1984 data general one?

Large backlit lcd that also did cga, machine that was comparable to desktops at the time, AND the lcd still works today!

Can you say that this isn't mature enough for mass production?
Quote:
Originally Posted by tgm1024 View Post
AIUI, he was saying that we should set aside the passive matrix stuff because it's confusing. Like the display in that DG.
Also, it was monochrome.

I think the point should be clear: Nothing goes from "first device" to much of anything commercial is <15 years. It doesn't go from first device to a big TV in <30 years. We're barely at first device with anything that isn't LCD or OLED. We can safely rule out a new kind of TV anytime soon.

Also, read this if you never have: http://www.amazon.com/gp/product/B00...ie=UTF8&btkr=1

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12934 of 13035 Old 02-10-2016, 07:41 PM
Newbie
 
lbjack's Avatar
 
Join Date: Apr 2010
Posts: 10
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 3 Post(s)
Liked: 10
Quote:
Originally Posted by rogo View Post
Also, it was monochrome. .

Also, read this if you never have: http://www.amazon.com/gp/product/B00...ie=UTF8&btkr=1
2011...isn't that kind of, um, outdated?
lbjack is offline  
post #12935 of 13035 Old 02-10-2016, 10:36 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
Quote:
Originally Posted by lbjack View Post
2011...isn't that kind of, um, outdated?
The book is from 30 years earlier.

#kidstoday

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12936 of 13035 Old 02-11-2016, 07:56 AM
Member
 
ynotgoal's Avatar
 
Join Date: Jun 2012
Posts: 199
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 81 Post(s)
Liked: 136
Quote:
Originally Posted by fafrd View Post
The Digitimes article I linked to earlier ended with this statement: "BOE will not be ready to produce OLED TVs in 2016, but it will deliver sample panels to TV makers in China - and will still consider whether to commit to a Gen-8.5 production fab."

So between Samsung and BOE, I think the other thing we are agreeing on is that by 2020, we will need to see at least 2-3 OLED TV panel manufacturers in the market to be certain OLED TV has passed the point of no return...
Another supplier is talking about orders from outside Korea for large size OLED. I would assume it is BOE. If they make orders in the next quarter or so they could be in the market in 2018.

http://seekingalpha.com/article/3883...pt?part=single

Have you received orders or do you anticipate orders yet in the OLED market outside of Korea?

Yes, we did, mainly in the small-to-mid. But in 2016, we expect to see that also in the large displays.
Cooters likes this.
ynotgoal is offline  
post #12937 of 13035 Old 02-13-2016, 12:17 AM
Member
 
mypretty1's Avatar
 
Join Date: Dec 2012
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 18
Forgive the naivety of my question but a thought crossed my mind about LG's 55" screens. Are there 2 factories, one making screens for the EF9500 and another for the 2016 range? I am assuming the 2015 screens are being phased out so could the EF9500 TVs being assembled this year be using the 2016 screens?
mypretty1 is offline  
post #12938 of 13035 Old 02-13-2016, 10:14 AM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
Quote:
Originally Posted by mypretty1 View Post
Forgive the naivety of my question but a thought crossed my mind about LG's 55" screens. Are there 2 factories, one making screens for the EF9500 and another for the 2016 range? I am assuming the 2015 screens are being phased out so could the EF9500 TVs being assembled this year be using the 2016 screens?
Not right now from what we know. It seems that the older facility is still making the 1080p models and the newer facility is making everything 4K. That is likely going to change sometime in 2016 but no one has intel suggesting that has happened.

But remember that even within one facility it's possible to make multiple products. It's not like there is a single assembly line that performs every step for every TV.
mypretty1 likes this.

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12939 of 13035 Old 02-13-2016, 12:14 PM
Member
 
mypretty1's Avatar
 
Join Date: Dec 2012
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 18
Quote:
Originally Posted by rogo View Post
Not right now from what we know. It seems that the older facility is still making the 1080p models and the newer facility is making everything 4K. That is likely going to change sometime in 2016 but no one has intel suggesting that has happened.

But remember that even within one facility it's possible to make multiple products. It's not like there is a single assembly line that performs every step for every TV.
Thank you, rogo.
mypretty1 is offline  
post #12940 of 13035 Old 02-13-2016, 12:26 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by rogo View Post
Not right now from what we know. It seems that the older facility is still making the 1080p models and the newer facility is making everything 4K. That is likely going to change sometime in 2016 but no one has intel suggesting that has happened.

But remember that even within one facility it's possible to make multiple products. It's not like there is a single assembly line that performs every step for every TV.
It is still unclear to me whether M1 has been converted to full-sheet manufacturing or remains a half-sheet manufacturing facility.

It does seem clear that M1 remains limited to 55" 1080p OLEDs while all 4K WOLED production is coming off of M2 currently.

So the most likely scenario is that once LG decides to cease 1080p WOLED production, M1 will be converted to manufacting of 4K WOLEDs, most likely 55" (and limited to 55" if it remains a half-sheet facility).

That means M1 will probably go down for a period of 3-4 months at some point this year while it is converted to 4K WOLED production, and this down-time is probably one of the factors contributing to LGs reduced 2016 production forecast (1M OLED panels in 2016, down from 1.5M that had been forecasted for 2016 last year).

My guess is that once the C6 and B6 are up and running at full volume later this year, we'll probably see the end of 1080p WOLEDs and some additional capacity for 55C6 and 55B6 soon after. When the dust has cleared, I expect all of M1 to be devoted to 55" WOLED panel production for 55C6 and 55C6, all 65" WOLED demand for 65C6, 65B6, 65E6 and 65G6 to be coming off of M2, with remaining M2 production capacity being used for additional 55" panels (55C6, 55B6). To the extent that there is any actual number of 77G6s sold this year beyond prototype levels, these will also come off of M2.

Early indications are that the 65G6 panel employs the same 'anti-vignetting' (lightened edges) engineering fix as has been seen in December and beyond 65EF9500s, and this suggest that there are not going to be any significant yield improvements before 2017 OLEDs hit production at the end of this year. How much of a yield hit has resulted from the tighter QC and testing criteria LG has been forced to adopt is anyone's guess, but it is almost certainly the biggest contributor to the reduced 2016 production target and 10-15% additional yield loss to levels of 65-70% until the causes of Vignetting have been further addressed at the root level seem like a reasonable guess.

LG has stated that market demand forcing a change in the mix to 40% 65" OLEDs is the reason for the reduced 2016 production forecast, so we have a 2016 production target of 40% of 1M or 400,000 65" WOLEDs (all from M2).

If we take worst-case assumptions for M1, meaning a full 4 months out of production and yields of only 65%, that still means the 9000 sheets-per-month of capacity it has will result in production of 35K OLEDs per month or about 280K 55" OLEDs this year (over 8 months accounting for down time). [compared to production of over 43K 1080p OLEDs per month at 80% yield, or more than 500,000 55" 1080p OLEDs this year if LG thought they had that level of demand for the 55EG9100). So as many as 220K out of the 'lost' 500K of 2016 production could be attributed to M1 conversion.]

400,000 65" OLEDs coming off of M2 will require 11,111 unyilelded sheets a month or 17,000 sheets a month assuming worst-case reduced yields of 65%.

M2 has a capacity of 26,000 sheets a month, so that leaves a remaining 9000 sheets a month that can produce an additional 35,000 55" OLEDs per month at reduced yields of 65%, or 420K this year.

This adds up to:
400,000 65" OLEDs off of M2
420,000 55" OLEDs off of M2
280,000 55" OLEDs off of M1
1.1M OLEDs in 2016.

If LG actually want to produce any meaningful quantity of 77G6s this year at all, that will of course cut into this number. At 50% yield (which is optimistic ), production of 10K 77" WOLEDs will consume a total of 10K sheets off of M2, reducing 2016 production to 1.07M total...

Overall, I take LGs reduced 2016 production forecast as a sign that they have learned the lesson about 'under-commit and over-deliver .
rogo, Brandon B, mypretty1 and 1 others like this.

Last edited by fafrd; 02-13-2016 at 12:30 PM.
fafrd is offline  
post #12941 of 13035 Old 02-13-2016, 08:34 PM
AVS Special Member
 
KLee's Avatar
 
Join Date: Jun 2001
Posts: 1,740
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 20 Post(s)
Liked: 28
Quote:
Originally Posted by fafrd View Post
It is still unclear to me whether M1 has been converted to full-sheet manufacturing or remains a half-sheet manufacturing facility.

It does seem clear that M1 remains limited to 55" 1080p OLEDs while all 4K WOLED production is coming off of M2 currently.

So the most likely scenario is that once LG decides to cease 1080p WOLED production, M1 will be converted to manufacting of 4K WOLEDs, most likely 55" (and limited to 55" if it remains a half-sheet facility).

That means M1 will probably go down for a period of 3-4 months at some point this year while it is converted to 4K WOLED production, and this down-time is probably one of the factors contributing to LGs reduced 2016 production forecast (1M OLED panels in 2016, down from 1.5M that had been forecasted for 2016 last year).

My guess is that once the C6 and B6 are up and running at full volume later this year, we'll probably see the end of 1080p WOLEDs and some additional capacity for 55C6 and 55B6 soon after. When the dust has cleared, I expect all of M1 to be devoted to 55" WOLED panel production for 55C6 and 55C6, all 65" WOLED demand for 65C6, 65B6, 65E6 and 65G6 to be coming off of M2, with remaining M2 production capacity being used for additional 55" panels (55C6, 55B6). To the extent that there is any actual number of 77G6s sold this year beyond prototype levels, these will also come off of M2.

Early indications are that the 65G6 panel employs the same 'anti-vignetting' (lightened edges) engineering fix as has been seen in December and beyond 65EF9500s, and this suggest that there are not going to be any significant yield improvements before 2017 OLEDs hit production at the end of this year. How much of a yield hit has resulted from the tighter QC and testing criteria LG has been forced to adopt is anyone's guess, but it is almost certainly the biggest contributor to the reduced 2016 production target and 10-15% additional yield loss to levels of 65-70% until the causes of Vignetting have been further addressed at the root level seem like a reasonable guess.

LG has stated that market demand forcing a change in the mix to 40% 65" OLEDs is the reason for the reduced 2016 production forecast, so we have a 2016 production target of 40% of 1M or 400,000 65" WOLEDs (all from M2).

If we take worst-case assumptions for M1, meaning a full 4 months out of production and yields of only 65%, that still means the 9000 sheets-per-month of capacity it has will result in production of 35K OLEDs per month or about 280K 55" OLEDs this year (over 8 months accounting for down time). [compared to production of over 43K 1080p OLEDs per month at 80% yield, or more than 500,000 55" 1080p OLEDs this year if LG thought they had that level of demand for the 55EG9100). So as many as 220K out of the 'lost' 500K of 2016 production could be attributed to M1 conversion.]

400,000 65" OLEDs coming off of M2 will require 11,111 unyilelded sheets a month or 17,000 sheets a month assuming worst-case reduced yields of 65%.

M2 has a capacity of 26,000 sheets a month, so that leaves a remaining 9000 sheets a month that can produce an additional 35,000 55" OLEDs per month at reduced yields of 65%, or 420K this year.

This adds up to:
400,000 65" OLEDs off of M2
420,000 55" OLEDs off of M2
280,000 55" OLEDs off of M1
1.1M OLEDs in 2016.

If LG actually want to produce any meaningful quantity of 77G6s this year at all, that will of course cut into this number. At 50% yield (which is optimistic ), production of 10K 77" WOLEDs will consume a total of 10K sheets off of M2, reducing 2016 production to 1.07M total...

Overall, I take LGs reduced 2016 production forecast as a sign that they have learned the lesson about 'under-commit and over-deliver .

That's all fine and dandy for LG products but what about their supply of WOLED panels to their partners like Skyworth and Panasonic?

Will these guys be left out in the cold?
KLee is offline  
post #12942 of 13035 Old 02-13-2016, 11:32 PM
Member
 
mypretty1's Avatar
 
Join Date: Dec 2012
Posts: 68
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 18
Thank you, fafrd. As I am someone with zero knowledge of the manufacturing processes, your reply was most interesting and clearly explained.
fafrd likes this.
mypretty1 is offline  
post #12943 of 13035 Old 02-13-2016, 11:40 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 23,403
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1093 Post(s)
Liked: 1597
Quote:
Originally Posted by fafrd View Post
LG has stated that market demand forcing a change in the mix to 40% 65" OLEDs is the reason for the reduced 2016 production forecast, so we have a 2016 production target of 40% of 1M or 400,000 65" WOLEDs (all from M2).
About how many of those would you expect in Q42016?

And how many 65" OLEDs do you think were produced in Q42015 for their 200k overall claim?

Thanks,
Darin

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is offline  
post #12944 of 13035 Old 02-14-2016, 10:06 AM
AVS Special Member
 
gadgtfreek's Avatar
 
Join Date: Aug 2012
Location: Bama
Posts: 6,076
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 632 Post(s)
Liked: 1478
Ex plasma guy here who recently got all 65" 9500 and it's quite amazing.

I was just curious if someone could briefly explain how the panel is made up since I am behind the curve. I understood what made up a plasma panel between the glass, phosphors and filter. The 9500 is amazing to me because it is so thin and rejects ambient light well. Just kinda curious of a brief tutorial of what makes up the "panel" and how it rejects light so well. I figured you guys would be up to speed on this.

Thanks.

LG 65EF9500 / UBD-K8500
AV7702mkII / Model 5000
Klipsch RF-7II's / RC-64II
Dual Rythmik FV15HP's
gadgtfreek is offline  
post #12945 of 13035 Old 02-14-2016, 10:21 AM
AVS Special Member
 
gadgtfreek's Avatar
 
Join Date: Aug 2012
Location: Bama
Posts: 6,076
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 632 Post(s)
Liked: 1478
Think I found some tech stuff, kinda what I thought maybe they were, at least the white oled LG's






LG 65EF9500 / UBD-K8500
AV7702mkII / Model 5000
Klipsch RF-7II's / RC-64II
Dual Rythmik FV15HP's
gadgtfreek is offline  
post #12946 of 13035 Old 02-14-2016, 01:52 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by darinp2 View Post
About how many of those would you expect in Q42016?

And how many 65" OLEDs do you think were produced in Q42015 for their 200k overall claim?

Thanks,
Darin
To be clear on your question, I believe you were asking about about how many of the 400,000 65" OLEDs LG has planned for 2016 will be produced in Q4'16, right?

This is complete speculation, but just to flesh-out the logic:

First, if LG is now running M2 at capacity (and at reduced yield due to the improved QC), 100,000 per quarter would be the default production rate.

Second, since the product mix can easily be shifted, the short answer is essentially however many LG wants.

In terms of your question about the 200,000 OLED TVs sold in Q4'15, I don't see any reason not to take LG's claims at face value (especially since they were communicated to the financial markets, so inaccurate claims could get them into trouble .

40% of 200,000 = 80,000, so that is at least a swag of what sold in Q4'15.

Q4'15 was the quarter where they had to make adjustments to QC after ramping to full-volume in Q3, so that is going to introduce transients, and there is a lag between production of OLED panels by LGD and production of OLED TVs by LGE, so that's another reason that Q4'15 does not necessarily represent steady-state.

But the most important point out of all of this is the fact that, except for possible down-time associated with conversion of M1 from 1080p to 4K, LG appears to now be running flat-out at a production rate of 250,000 OLEDs per quarter and there is no reason to expect that they are going to have a higher production rate than that until the new LCD->OLED conversion they have announced for mid-2017 is complete.

The only other possible rationale for increased production would be recovery of the lost yield caused by improved QC, but that is highly unlikely to be possible before the 2017 OLEDs are in production.

As a placeholder, once they get 4K OLED yields back to 80% (unlikely before 2017), and if we assume continue production of 40% 65" and 60% 55", they will be able to produce 48,000 65" OLEDs per month (using 20,000 sheets) and 72,000 55" OLEDs per month (using 15,000 sheets per month), for a total production rate of 130,000 OLEDs per month 1.5M per year (which is spot-on the original 2016 production target).

Between the new conversion coming online in 2017 and hoped-for yield improvements associated with the 2017 OLEDs, it is a certainty that OLED TV prices will be declining significantly in 2017. Until then, if LG is truly able to sell-through available production at current discounted pricing (~$4500 for 65"), it is difficult to see a reason to expect much decline from current OLED TV pricing levels (basic economics of supply and demand).
fafrd is offline  
post #12947 of 13035 Old 02-14-2016, 10:37 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 23,403
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1093 Post(s)
Liked: 1597
Quote:
Originally Posted by fafrd View Post
[In terms of your question about the 200,000 OLED TVs sold in Q4'15, I don't see any reason not to take LG's claims at face value (especially since they were communicated to the financial markets, so inaccurate claims could get them into trouble .

40% of 200,000 = 80,000, so that is at least a swag of what sold in Q4'15.
With the prices on the 1080p units and timing of releases I doubt that 40% of unit sales in Q4 are 65" 4K units. Maybe going forward, but I doubt it was that high unless LG said it was that high and not that they learned it should be that high. I don't recall exactly how they put it, especially how much room they left for reading between the lines.

It seems to me that you've made a certain amount of assumptions to get to the same after QC production rate in Q4 of 2016 for 4K units as 2015 and I think the odds of that are low. Do you believe there were no temporary slowdowns in Q1 for the extra QC issues that won't improve in 8 months or slowdowns in Q1 to switch production to 2016 models?

We clearly see some of the same data different ways because I highly doubt LG is selling 250k units in Q12016 or that their current plan is to sell the same number of 4K units of either size in Q4 of 2016 as Q4 of 2015.

Just because they said 1 million total doesn't mean it is even per quarter, especially when more slowdowns in production happen during some quarters than others. I don't consider assumptions that every quarter will have the same number of post QC units to be good assumptions with products at this kind of point in their production.

I can just imagine how well it would go if a manager in production told somebody at the top of the company that production after QC won't improve in the next 8 months. The odds that those are LG's current internal plans are close to zero IMO.

--Darin
darinp2 is offline  
post #12948 of 13035 Old 02-14-2016, 10:41 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
I'd also add that it's very possible (probable?) that finished goods inventory was very unusually high headed into late 2015 as pricing was mostly ridiculous throughout the year and TV sales always go up -- a lot -- into Q4. So even if we assume sales were 200K, we're not learning a lot about production in Q4 of last year.

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12949 of 13035 Old 02-14-2016, 11:05 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 23,403
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1093 Post(s)
Liked: 1597
Quote:
Originally Posted by rogo View Post
I'd also add that it's very possible (probable?) that finished goods inventory was very unusually high headed into late 2015 as pricing was mostly ridiculous throughout the year and TV sales always go up -- a lot -- into Q4. So even if we assume sales were 200K, we're not learning a lot about production in Q4 of last year.
Good point. I hadn't paid much attention to what the 200k represented, but you are right. It is even possible that production was only 150k in Q4, which would definitely mean production would rise for this year to hit 1 million.

I do think it is likely we will see stronger competition from LCD this year (like hopefully Vizio P-series) and so the same pricing as last year could be tough to hit the same volume.

I think the main issues might be higher production and the mix moving to a higher percentage of 4K units though.

--Darin
darinp2 is offline  
post #12950 of 13035 Old 02-15-2016, 12:28 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by darinp2 View Post
With the prices on the 1080p units and timing of releases I doubt that 40% of unit sales in Q4 are 65" 4K units. Maybe going forward, but I doubt it was that high unless LG said it was that high and not that they learned it should be that high. I don't recall exactly how they put it, especially how much room they left for reading between the lines.

It seems to me that you've made a certain amount of assumptions to get to the same after QC production rate in Q4 of 2016 for 4K units as 2015 and I think the odds of that are low. Do you believe there were no temporary slowdowns in Q1 for the extra QC issues that won't improve in 8 months or slowdowns in Q1 to switch production to 2016 models?

We clearly see some of the same data different ways because I highly doubt LG is selling 250k units in Q12016 or that their current plan is to sell the same number of 4K units of either size in Q4 of 2016 as Q4 of 2015.

Just because they said 1 million total doesn't mean it is even per quarter, especially when more slowdowns in production happen during some quarters than others. I don't consider assumptions that every quarter will have the same number of post QC units to be good assumptions with products at this kind of point in their production.

I can just imagine how well it would go if a manager in production told somebody at the top of the company that production after QC won't improve in the next 8 months. The odds that those are LG's current internal plans are close to zero IMO.

--Darin
Just in case it was not already posted, here is the link to the information we are discussing: http://www.oled-info.com/lg-display-...s-2015-half-q4

I stated in my earlier post where I was entering into the domain of pure speculation, so hopefully that was clear. In terms of your questions in bold, let's focus on the facts (assuming LGD is being truthful ):

-LG sold 200,000 OLED panels in Q4'15. While these panels certainly could have been stashed into inventory at the parts or finished goods level, as suggested by Rogo, the fact that LGE had a discounted 'sale' that ended at year-end and resulted in an increase in prices back to MSRP levels suggests that they were not.

-LG has reduced the 2016 OLED panel sales target from 1.5M to 1.0M with a mix of 40% 65" panels (400,000).

-LG has an installed capacity of 35,000 Gen8 sheets per month and this will not increase until Q2'17 when an additional 25,000 Gen8 sheets per month is scheduled to come online due to the new conversion.

-if we assume the full 35,000 sheet production capacity being used for production of 400,000 65" and 600,000 55" OLEDs, the resulting yield is 56% (raw unyielded production is 720,000 65" and 1.08M 55" OLEDs)

-LG has stated that they are only going to be producing 4K OLEDs once the 55EG9100 has reached end of life, which suggests that M1 will be converted from production of 1080p OLEDs to 4K OLEDs and any downtime associated with such conversion would reduce total raw unyielded production output and increase effective yields required to achieve stated production targets of 1M OLEDs in 2016.

-Aside from better quality-control/final test criteria used to screen out defective product, the only changes that have been made appear to be the brightened edges or 'anti-vignette' which is almost certainly a change at the drive electronics and/or algorithmic level which would not impact OLED panel production.

-Early indications from the first 2016 65G6 tested is that it appears to have the same 'anti-vignette' edge-brightening, so this is a strong indication that all aspects of yield/production rate for 2016 OLEDs will be similar to recent-builds of EF9500 OLEDs. The 2016 OLED panels appear to use improved phosphors for wider color gamut and increased brightness, but it is unlikely that there have been any additional changes made that will result in increased yields (while maintaining the same QC).

-LG has reduced the 2016 production target from 1.5M to 1M and is suggesting that increased demand for 65" OLEDs is the cause, but this does not add up. If the 35,000 substrates were used to produce a mix of 20%/80% 65"/55" at stated yield levels of 80%, the resulting total production would be 336,000 65" and 1.344M 55" for a total of 1.68M OLEDs. Keeping yields constant at 80% and doubling the mix of 65" to 40%/60% results in 576,000 65" and 864,000 55" for a total 1.44M OLEDs (ergo, yields are either significantly worse than 80% or there is significant down-time for conversion or a combination of the two, as I have speculated).

So first, it seems exceedingly likely that LG is now running steady-state and that there are not going to be any further changes (other than temporary down-time for conversions) that would be rewsons to expect increased production output before Q2'17 (when 2017 OLEDs should be in production and the additional capacity from the conversion should be online).

And second, as we see continued evidence of LG occasionally running discounted 'sales' followed by a return to full MSRP pricing, that is strong evidence of them finding and following the pricing needed to maintain market balance.

So back to speculation, I will repeat my contention that if we do see reduced pricing over the course of 2016 for the EF9500 volume-runners (and eventually 65B6 volume runners), it will be because LG has satisfied pent-up premium TV demand from plasma owners and needs to offer lower prices to continue to find the demand for their current production level.

This is a strategy that LG has already demonstrated a proven capability for. I purchased 4 55EC9300s last year, all through Costco (and all returned due to the dimpling defect). I paid $2300 for the first one, $2100 for the second one about 3 months later, $1900 for the third one after another 3 months, and $1700 for the last one (and you can now find the 55EG9100 being on-and-off available for $1400 through Adorama: http://www.oled-info.com/adorama-off...0-oled-tv-1399).

So LG knows how to match demand to supply, supply is now going to be relatively steady-state for 1-year+ (in my view ), and we will see as this year unfolds whether demand is maintained at current discounted pricing levels or if LG is going to need to go lower to maintain appetite.

And the other certainty is that by mid 2017, prices will certainly be significantly lower than they are today (at the basic entry-level / volume runner level for OLED). As we progress through this year, it is likely we will see limited initiatives by LG to probe the pricing levels they will need to offer to drive that higher level of sales (anyone remember the MicroCenter 'mistake' on the 9800 or the Fry's 'close out sale' ).

But there will not be another driver for sustained lower pricing like we had with M2 coming fully online in late Q3'15 until mid 2017...
fafrd is offline  
post #12951 of 13035 Old 02-15-2016, 01:44 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 23,403
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1093 Post(s)
Liked: 1597
Quote:
Originally Posted by fafrd View Post
-LG has stated that they are only going to be producing 4K OLEDs once the 55EG9100 has reached end of life
I think your own story supports that a lot of the 200k units sold in Q415 were 1080p units which had lower prices than 4k units.

How do you propose that LG sell significantly more 4k units in Q416 with the same or higher prices than 4k units of the same size in Q415?

Do any of these not match your current positions about LG's history and plans:

- LG sold 200k units in Q415 and a significant portion were 1080p units.
- LG plans to sell 250k units total in Q416 with the vast majority being 4k units.
- Street prices for the lowest level 55" 4k unit from LG will be at least as high in Q416 as Q415?
- LG plans to sell significantly more 55" 4k units in Q416 than 55" 4k units in Q415.
- Street prices for the lowest level 55" 4k unit from LG will be at least as high in Q416 as Q415?
- LG plans to sell significantly more 65" 4k units in Q416 than 65" 4k units in Q415.

--Darin

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is offline  
post #12952 of 13035 Old 02-15-2016, 01:50 PM
Advanced Member
 
slacker711's Avatar
 
Join Date: Apr 2006
Posts: 977
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 347 Post(s)
Liked: 329
Some thoughts on capacity.

1) There is no reason to think that switching M1 from 1080p to 4K is going to result in months of downtime. The only real change is increasing the density of transistors in the backplane. If that change involved months of downtime, mobile fabs would never run anywhere near capacity as they must supply a mix of resolutions.

2) OTOH, any fab must have some amount of downtime and this likely is one factor in why our theoretical capacity calculations never match reality. LGD must also still be running R&D runs occasionally on some of this capacity which would also impact total output.

3) While LGD may be testing and correcting for vignetting, this is unlikely to dramatically impact throughput. The limiting factor in a display fab is the number of substrates. Testing adds costs and a particularly cumbersome test would require more test equipment/personnel but that is a cheaper solution than utilizing a fab at less than its full capacity.

4) The more I read, the more I think that the biggest factor is that we think of bringing a new line or fab on line as flipping a switch on capacity. I believe we are underestimating the time it takes to bring a new fab up to full capacity and mature yields. Here is an excerpt of some commentary from a display analyst (Andy Abrahms) on the theoretical ramp of a flexible OLED fab to support Apple.

Quote:
Thus far, our scenario time line is 12-18 months for basic construction, 2 -3 months for equipment delivery and tune-up, and a minimum of 6-9 months to bring yields up to ~75%+. This puts us between 20 months and 30 months to actual commercial level mass production. Making the assumption that a full 19,000 sheet Gen 5.5 line was installed, we can now assume that the fab, after 20 months, can produce the full 19,000 sheets (100% factory efficiency) and 75% product yield,
5) LGE's sales to end customers are likely to be far more lumpy than LGD's OLED output. The television market is seasonal and the fact that inventories and distribution still need to be filled means that the a significant amount of production early in the year will be used to fill out LGE's sales channels. There is very little chance that LGE is selling 200,000 units this quarter.

6) It will be far easier for LGE to maintain pricing on their OLED's if Samsung never ships a FALD set this year. If I was looking for a set this year I would be hoping that Samsung ships the KS9800 for a reasonable price.

FWIW, I saw the G6 today. It was a brightly lit environment but damn did some of the scenes look beautiful. While I was there, I was surprised by the fact that more people stopped to look at the 65" OLED than the ~120" Samsung that was nearby.
slacker711 is online now  
post #12953 of 13035 Old 02-15-2016, 01:58 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
I don't disagree with your analysis, but you make a lot of assumption. They each contain an error margin and therefore it's pretty easy to conclude something very different... even if that conclusion doesn't fundamentally alter the total production numbers for the year. I'm going to highlight just a few:

Quote:
Originally Posted by fafrd View Post
J
-LG sold 200,000 OLED panels in Q4'15. While these panels certainly could have been stashed into inventory at the parts or finished goods level, as suggested by Rogo, the fact that LGE had a discounted 'sale' that ended at year-end and resulted in an increase in prices back to MSRP levels suggests that they were not.
First, I'm not saying anywhere near 200K of the 200K were finished goods. But your conclusion that "because sale" none were is a huge assumption that I suspect is in error. Your belief that "because sale started and ended, supply and demand balanced" is also mostly belief. Sales often coincide with year-end targets to hit a certain volume, often end soon after, often prove nothing about supply (or really even demand given seasonality).
Quote:
-if we assume the full 35,000 sheet production capacity being used for production of 400,000 65" and 600,000 55" OLEDs, the resulting yield is 56% (raw unyielded production is 720,000 65" and 1.08M 55" OLEDs)
You've now set a baseline based on the sales goal that you're going to use for a lot of important math: That yields are the same/similar between M1 and M2, that the 1M is really the goal and that production is going to magically meet the goal (I believe production will need to wildly overproduce the goal because I believe the chain from factory --> store --> sale is far longer than I think you do).
Quote:
-Aside from better quality-control/final test criteria used to screen out defective product, the only changes that have been made appear to be the brightened edges or 'anti-vignette' which is almost certainly a change at the drive electronics and/or algorithmic level which would not impact OLED panel production.

-Early indications from the first 2016 65G6 tested is that it appears to have the same 'anti-vignette' edge-brightening, so this is a strong indication that all aspects of yield/production rate for 2016 OLEDs will be similar to recent-builds of EF9500 OLEDs. The 2016 OLED panels appear to use improved phosphors for wider color gamut and increased brightness, but it is unlikely that there have been any additional changes made that will result in increased yields (while maintaining the same QC).

-LG has reduced the 2016 production target from 1.5M to 1M and is suggesting that increased demand for 65" OLEDs is the cause, but this does not add up. If the 35,000 substrates were used to produce a mix of 20%/80% 65"/55" at stated yield levels of 80%, the resulting total production would be 336,000 65" and 1.344M 55" for a total of 1.68M OLEDs. Keeping yields constant at 80% and doubling the mix of 65" to 40%/60% results in 576,000 65" and 864,000 55" for a total 1.44M OLEDs (ergo, yields are either significantly worse than 80% or there is significant down-time for conversion or a combination of the two, as I have speculated).
Wait, fafrd, what? They changed the fundamental nature of the panels! They are much brighter somehow, maybe better OLED deposition, maybe new transistors, maybe a bunch of other stuff. This wasn't a small change to fix vignetting, it sounds like they changed almost everything.

I would suggest that a great deal of the reduced target is down time and that by year end yields will, in fact, be every bit as high. What you seem to be interpreting as lower yield across the board, I strongly suspect is much lower yield at first with wholesale changes followed by a return to the same yields as before.
Quote:
So first, it seems exceedingly likely that LG is now running steady-state and that there are not going to be any further changes (other than temporary down-time for conversions) that would be rewsons to expect increased production output before Q2'17 (when 2017 OLEDs should be in production and the additional capacity from the conversion should be online).
And therefore I very much doubt LG is running anywhere near steady state. The dribbling into the market of only the expensive models, the summer (delayed from original announcement) arrival of the cheaper ones, the possibility Europe doesn't get the new models till fall, the imminent conversion of M1... I see all of this as almost entirely different than you: Much lower production at first from M2, tons of finished goods still sloshing around, M1 1080p panels pushing any volumes at all in 1H2016 (especially in Europe), and a sincere hope by LG that steady state is achieved as soon as possible. This will allow them to have confidence in the 2017 conversions to very, very similar product to the 2016 stuff (marginal changes only, no more mucking with backplane, etc.). I imagine the output curve looks a lot more like an "S" than a line right now.
Quote:
And second, as we see continued evidence of LG occasionally running discounted 'sales' followed by a return to full MSRP pricing, that is strong evidence of them finding and following the pricing needed to maintain market balance.
Sure, this is possible. But it could easily be incorrect. It could easily be true that LG is testing, gathering data, delivering results within -20% of some spreadsheet and absolutely not worried about "market balance" except to ensure FGI doesn't explode on them. But for example, I'm sure they know well how many 55-inch 1080p sets sell by now. As that market is shrinking, they'll have to erode prices, but not to balance magically some set of Econ 101 supply and demand curves. Once the FGI is sold to retailers, it mostly stops being LG's problem. They just have to offer stuff at wholesale at attractive prices: That's changing less than what you see at retail.
Quote:
So back to speculation, I will repeat my contention that if we do see reduced pricing over the course of 2016 for the EF9500 volume-runners (and eventually 65B6 volume runners), it will be because LG has satisfied pent-up premium TV demand from plasma owners and needs to offer lower prices to continue to find the demand for their current production level.
See, this is true and again, to me, not. They aren't looking to satisfy demand for a production level, they're looking to sell more than 1M OLED TVs. When you see prices going up and down, I doubt you're seeing LG's hand at work. When you see prices going only down -- and by lots -- you're almost certainly seeing only LG's hand at work.
Quote:
This is a strategy that LG has already demonstrated a proven capability for. I purchased 4 55EC9300s last year, all through Costco (and all returned due to the dimpling defect). I paid $2300 for the first one, $2100 for the second one about 3 months later, $1900 for the third one after another 3 months, and $1700 for the last one (and you can now find the 55EG9100 being on-and-off available for $1400 through Adorama: http://www.oled-info.com/adorama-off...0-oled-tv-1399).
Exactly, that's LG. It wants to sell more of those every month and yet faces more and more competition. Prices have to fall and volumes need to rise. Production needs to be constantly rising against the falling prices unless demand for that product is disappearing (likely along with the whole category).
Quote:
So LG knows how to match demand to supply, supply is now going to be relatively steady-state for 1-year+ (in my view ), and we will see as this year unfolds whether demand is maintained at current discounted pricing levels or if LG is going to need to go lower to maintain appetite.
So your possibly 100% correct and yet, again, I suspect supply is a very, very lumpy "S" curve. I don't think it even remotely resembles what you think it does.
Quote:
And the other certainty is that by mid 2017, prices will certainly be significantly lower than they are today (at the basic entry-level / volume runner level for OLED).

But there will not be another driver for sustained lower pricing like we had with M2 coming fully online in late Q3'15 until mid 2017...
If we're banking on the 2017 conversions for better pricing, I suspect it'll be a similar time frame that additional supply could allow for more aggressive pricing to capture more share. I suspect it could also be 2018. But if we're seeing supply mostly increasing for the next 4 quarters (more likely I think) and then increasing again starting in Q2/Q3 2017, I think the pricing will decline somewhat more steadily. That said, in-year pricing action tends to be small on real products. That LG didn't do anything (apparently) to lower the base pricing suggests very high pricing vs. my expectations in 2016. And it suggests, therefore, minimal production and minimal share taking. This further leads me to believe that supply will increase throughout the year. If it were already high, prices would already be lower.

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12954 of 13035 Old 02-15-2016, 02:04 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by darinp2 View Post
I think your own story supports that a lot of the 200k units sold in Q415 were 1080p units which had lower prices than 4k units.

How do you propose that LG sell significantly more 4k units in Q416 with the same or higher prices than 4k units of the same size in Q415?

Do any of these not match your current positions about LG's history and plans:

- LG sold 200k units in Q415 and a significant portion were 1080p units.
- LG plans to sell 250k units total in Q416 with the vast majority being 4k units.
- Street prices for the lowest level 55" 4k unit from LG will be at least as high in Q416 as Q415?
- LG plans to sell significantly more 55" 4k units in Q416 than 55" 4k units in Q415.
- Street prices for the lowest level 55" 4k unit from LG will be at least as high in Q416 as Q415?
- LG plans to sell significantly more 65" 4k units in Q416 than 65" 4k units in Q415.

--Darin
You raise a good point - if the EG9100 ceases production before the end of this year (which is an unknown - we'll know when it dissapears ), then conversion of M1 will result in increased 55" 4K OLED production which may drive lower pricing for those models.

And it is also true that we do not know what % of the 65" OLEDs sold in Q4 were 65" - the safe bet would be at least 40-50% of the 4K sales.

The one counter argument is that we do not have any idea what level of sales the 55" 1080p OLEDs had in Q4. At least in my local Best Buy, all of their sales were of 4K OLEDs. And the fact that the prices on 55" 1080p OLEDs have dropped by 50% over the past 12 months does not suggest stable demand or market balance.

So we'll just need to sit tight and see. The one indisputable fact, however, is that LG's reduction in 2016 production target from 1.5M to 1M is not good news for continued price decreases - whatever ends up happening this year as far as pricing, it will be less aggressive than it would have been had LG had an additional 0.5M / 50% of annual production to sell...
fafrd is offline  
post #12955 of 13035 Old 02-15-2016, 02:12 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
"So we'll just need to sit tight and see. The one indisputable fact, however, is that LG's reduction in 2016 production target from 1.5M to 1M is not good news for continued price decreases - whatever ends up happening this year as far as pricing, it will be less aggressive than it would have been had LG had an additional 0.5M / 50% of annual production to sell..."

I think that's the most important point. But looking at the number as monolithic still points out the challenge of guessing, say, whether the 65-inch ends the year at $5000, $4500, or $4000.

I imagine LG has to (a) sell close to 400K units in Q4 this year, not 200K, to get anywhere near 1M units and (b) will be able to sell very few 55-inch 1080p units even in Europe, where I suspect many of them sold last year though more of them likely sold in the U.S. than you and I would spitball a guess at.

I also imagine you'll tell me how this is not realistic based on production capabilities. I'll not only respectfully disagree but point out that it would be twice as many as they sold last year with a much greater mix of 4K than last year. Can we still bet on the $4000 65-inch B6?

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12956 of 13035 Old 02-15-2016, 02:27 PM
AVS Addicted Member
 
darinp2's Avatar
 
Join Date: Sep 2003
Location: Seattle, WA
Posts: 23,403
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 1093 Post(s)
Liked: 1597
Quote:
Originally Posted by rogo View Post
Can we still bet on the $4000 65-inch B6?
My position seems pretty simple. If we don't see that for street and/or a $2500 street for a 4k 55-inch B6 before the end of the year then they won't sell 1 million total units in 2016. And even at those exact street prices if those are only rare sales then they aren't hitting 1 million this year either. IMO of course.

So, I think that if they hit their current production targets street prices at those levels are lower will be necessary to keep inventory from stacking up. Unless they divert units to other manufacturers, those units take the lower end, and that is how they get to 1 million.

--Darin

This is the AV Science Forum. Please don't be gullible and please do remember the saying, "Fool me once, shame on you. Fool me twice, shame on me."
darinp2 is offline  
post #12957 of 13035 Old 02-15-2016, 03:01 PM
AVS Addicted Member
 
rogo's Avatar
 
Join Date: Dec 1999
Location: Stop making curved screens
Posts: 31,713
Mentioned: 13 Post(s)
Tagged: 0 Thread(s)
Quoted: 1357 Post(s)
Liked: 1533
Quote:
Originally Posted by darinp2 View Post
My position seems pretty simple. If we don't see that for street and/or a $2500 street for a 4k 55-inch B6 before the end of the year then they won't sell 1 million total units in 2016. And even at those exact street prices if those are only rare sales then they aren't hitting 1 million this year either. IMO of course.
I completely share this belief.
Quote:
So, I think that if they hit their current production targets street prices at those levels are lower will be necessary to keep inventory from stacking up. Unless they divert units to other manufacturers, those units take the lower end, and that is how they get to 1 million.
Possibly, but I doubt those other OEMs can be counted on for anywhere near 100K this year.

And you may ask yourself well how did I get here?
And you may ask yourself am I right? Am I wrong?
And you may say to yourself yourself: What have I done?!
- - - - - - - - - - - - - - - -
Once you let the trolls run the bridge, you have a bridge run by trolls. I'll be back.
rogo is online now  
post #12958 of 13035 Old 02-15-2016, 03:14 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by rogo View Post
I don't disagree with your analysis, but you make a lot of assumption. They each contain an error margin and therefore it's pretty easy to conclude something very different... even if that conclusion doesn't fundamentally alter the total production numbers for the year. I'm going to highlight just a few:
Can't argue with that statement. I'd summarize the analysis I have laid out as the 'best case' as far as the level of maturity of OLED. If LG is still suffering from growing pains in M2, my entire analysis will be incorrect as a result. Having 'lived through' both the launch of the 55EC9300 and more recently the launch of the 65EF9500, and having some background in volume manufacturing of consumer products, my gut tells me LG is now in a more stable situation than you are suggesting (also supported by the announced investments both in P10 and the nearer-term 2017 conversion).

Quote:
First, I'm not saying anywhere near 200K of the 200K were finished goods. But your conclusion that "because sale" none were is a huge assumption that I suspect is in error. Your belief that "because sale started and ended, supply and demand balanced" is also mostly belief. Sales often coincide with year-end targets to hit a certain volume, often end soon after, often prove nothing about supply (or really even demand given seasonality).
It is correct that the 200K was panels sold and do says nothing about TVs sold. Sales could have been higher due to inventory build-up earlier in the year, but since M2 had just come up to full production and the EF9500 had just launched both in Q3, I think the more reasonable read would be that 200,000 is probably an upper-limit on Q4'15 sales.

Quote:
You've now set a baseline based on the sales goal that you're going to use for a lot of important math: That yields are the same/similar between M1 and M2, that the 1M is really the goal and that production is going to magically meet the goal (I believe production will need to wildly overproduce the goal because I believe the chain from factory --> store --> sale is far longer than I think you do).
Yes, my analysis is a best-case for LG's level of maturity - if 2016 production does not meet the goal of 1M OLEDs shipped, that would be very bad news for LGs position and the outlook for OLED. My analysis assumes that LG is now 'over the hump' and in the process of consolidation / expansion (again, supported by P10 and the 2017 conversion). If I am wrong on this and LG has not recently achieved steady-state in M2, all bets are off (along with my analysis ).

Quote:
Wait, fafrd, what? They changed the fundamental nature of the panels! They are much brighter somehow, maybe better OLED deposition, maybe new transistors, maybe a bunch of other stuff. This wasn't a small change to fix vignetting, it sounds like they changed almost everything.
There are new phosphors being used that have been credited with both increased color gamut and increased brightness (as well as increased lifetime) but no indication whatsoever of any other changes. The fact that the G6 is literally launching on the heels of the EF9500 is all but proof that little else has changed in the OLED panel manufacturing process itself. With a relatively simple change to phosphor formula, LG can be reasonably confident of a relatively smooth launch of the 2017 models (for once ) - more significant changes will likely result in a similar bungled launch such as we have seen with virtually every WOLED launch to date. Again, my analysis is the optimistic case for maturity / stability.

As far as Vignetting, there has been no engineering fix at the OLED panel level. Excessive vignette appears to impact roughly 50% of panels and LG (E or D, can't tell which) appears to have made an engineering change to either drive electronics and/or algorithm to enhance the brightness of the last 4-6" of either end of the screen.

This change unfortunately screws up the uniformity of 'perfect' panels (because they now have visibly lightened edges - anti-vignette) but no doubt results in a higher % of panels exhibiting mild vignette to be shippable. You can see some screen shots where the screen is suddenly brighter farthest from the edge and then gets darker close to the edge as the underlying vignette kicks in. This was not a fix, it was a hack.

And the most dissapointing thing is that the one informed review/feedback we have seen regarding the 65G6P indicates that it has this same hack.

So first, I don't think the engineering changes to address Vignetting are anywhere close to being as significant as you are hoping. And second, I have a strong suspicion that when it comes to underlying OLED panel technology, I believe the 2017 panels will prove to be far more similar to the 2016 panels than you are assuming (change in phosphors only is the most likely assumption).

Quote:
I would suggest that a great deal of the reduced target is down time and that by year end yields will, in fact, be every bit as high. What you seem to be interpreting as lower yield across the board, I strongly suspect is much lower yield at first with wholesale changes followed by a return to the same yields as before.
This is a perfect statement to bring the difference in our assumptions into focus: you are assuming that LG remains in an engineering / problem-solving mode for another year, while I am assuming that with the salvaged launch of the EF9500, they are now in consolidation/ramp-up/incremental-improvement mode.

If you are right and Ibam wrong then LG is highly unlikely to sell even 1M OLED panels in2017...

Quote:
And therefore I very much doubt LG is running anywhere near steady state. The dribbling into the market of only the expensive models, the summer (delayed from original announcement) arrival of the cheaper ones, the possibility Europe doesn't get the new models till fall, the imminent conversion of M1... I see all of this as almost entirely different than you: Much lower production at first from M2, tons of finished goods still sloshing around, M1 1080p panels pushing any volumes at all in 1H2016 (especially in Europe), and a sincere hope by LG that steady state is achieved as soon as possible. This will allow them to have confidence in the 2017 conversions to very, very similar product to the 2016 stuff (marginal changes only, no more mucking with backplane, etc.). I imagine the output curve looks a lot more like an "S" than a line right now.
Again, a helpful paragraph to focus on the difference in our views - I am assuming that the maturity of LG's OLED initiative is essentially one year ahead of what you have stated above. I think they achieved maturity/stability with 1080p OLEDs in M1 at the beginning of 2015 and they achieved maturity/stability with 4K OLEDs at the beginning of this year (subject to unfortunate yield loss due to tightened QC).

I see the launch of the Flagship G6 and Premium E6 that LG has developed through much of last year as a reflection of that maturity and believe they will use the launch of those low-volume enthusiast products to get the kinks out of their new WOLED panel production before converting over the volume-runners (EG9600, EF9500) to their next-generation replacements (C6, B6).

I am expecting to see much, much more successful launches of the C6 and B6 later this year than any OLED launch we have seen before.

And in the meantime, I am expecting to see LG continue to drive demand for the current 4K volume runners through discounted pricing as they did in Q4 and throughout last year with the 1080p models.

Again, if you are right and I am wrong, it is going to be another very painful year for LG OLED (though I agree even lower pricing may result).

Quote:
Sure, this is possible. But it could easily be incorrect. It could easily be true that LG is testing, gathering data, delivering results within -20% of some spreadsheet and absolutely not worried about "market balance" except to ensure FGI doesn't explode on them. But for example, I'm sure they know well how many 55-inch 1080p sets sell by now. As that market is shrinking, they'll have to erode prices, but not to balance magically some set of Econ 101 supply and demand curves. Once the FGI is sold to retailers, it mostly stops being LG's problem. They just have to offer stuff at wholesale at attractive prices: That's changing less than what you see at retail.
It is true that my assumptions are overly simplistic and are not going to apply to the monthly and seasonal variations. That being said, I believe they are correct on an annual basis and probably hold more true throughout the year than you are assuming. Depreciation is not seasonal and production lines are most efficient in steady-state.

Quote:
See, this is true and again, to me, not. They aren't looking to satisfy demand for a production level, they're looking to sell more than 1M OLED TVs. When you see prices going up and down, I doubt you're seeing LG's hand at work. When you see prices going only down -- and by lots -- you're almost certainly seeing only LG's hand at work.
I believe LG's hand is much more directly involved in virtually all OLED TV pricing than you are suggesting. Know that to be a fact with both Costco and Best Buy and am even suspicious that LG had a hand in the pricing 'mistake' by MicroCenter on the 9800 and Fry's 'unbelievable' closeput sale of the 9800 later in the year.

This is straying from the realm of fact into the realm of speculation, but I know what I witnessed with 55EC9300 pricing last year and I believe we both agree that LG has a keen awareness of the challenge they are facing as far as achieving pricing levels that will guarantee WOLEDs continued existence and expansion...

Quote:
Exactly, that's LG. It wants to sell more of those every month and yet faces more and more competition. Prices have to fall and volumes need to rise. Production needs to be constantly rising against the falling prices unless demand for that product is disappearing (likely along with the whole category).
Of course, LG will need to respond to overall market pricing trends along with everyone else. My analysis is assuming a stable overall market pricing environment and if that assumption is incorrect and TV prices decline significantly over the course of the year, LG will need to reduce OLED pricing accordingly. It's all relative (as LG acknowledges whenever they reference the 'premium' for OLED over LED/LCD pricing).


Quote:
So your possibly 100% correct and yet, again, I suspect supply is a very, very lumpy "S" curve. I don't think it even remotely resembles what you think it does.
We're both reading the tea leaves in our own way - a year from now there should be enough evidence to look back and see who's read was closer .

Quote:
If we're banking on the 2017 conversions for better pricing, I suspect it'll be a similar time frame that additional supply could allow for more aggressive pricing to capture more share. I suspect it could also be 2018. But if we're seeing supply mostly increasing for the next 4 quarters (more likely I think) and then increasing again starting in Q2/Q3 2017, I think the pricing will decline somewhat more steadily. That said, in-year pricing action tends to be small on real products. That LG didn't do anything (apparently) to lower the base pricing suggests very high pricing vs. my expectations in 2016. And it suggests, therefore, minimal production and minimal share taking. This further leads me to believe that supply will increase throughout the year. If it were already high, prices would already be lower.
Pretty much agree with everything you have stated here. My entire analysis is predicated on the underlying assumption that LG actually sold close to 200,000 OLED TVs in Q4'15. If they only sold a small fraction of that volume, significanly lower prices as the year unfolds is a certainty. I would say that the pricing that has been released for the G6 and E6 as well as the leaked pricing for the B6 and C6 s some evidence supporting my assumption .

But we'll see - if prices come down a great deal by Holiday Shopping Season 2016, your read was right. If they don't, mine probably was.

I was once an advocate for believing that we might see 65" OLED pricing approaching $3000 by late this year, but I no longer believe that is likely because I don't believe LG has sufficient production capacity to meet demand at that pricing level.

Time will tell .
fafrd is offline  
post #12959 of 13035 Old 02-15-2016, 04:16 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by slacker711 View Post
Some thoughts on capacity.

1) There is no reason to think that switching M1 from 1080p to 4K is going to result in months of downtime. The only real change is increasing the density of transistors in the backplane. If that change involved months of downtime, mobile fabs would never run anywhere near capacity as they must supply a mix of resolutions.
I agree that 3-4 months to getting yields for finer-geometry transistors in a backplane seems conservative - coukd be faster but is certainly not going to be up and running at good yields overnight...

Quote:
2) OTOH, any fab must have some amount of downtime and this likely is one factor in why our theoretical capacity calculations never match reality. LGD must also still be running R&D runs occasionally on some of this capacity which would also impact total output.
Individual lines may have occasional downtime for maintainance but this is unlikely to be significant and certainly does not involve the entire production facility (unless there is a problem such as the gas leak we saw earlier).

Development/R&D runs are also likely to have a minor impact in a facility with a capacity of 26,000 panels per month (almost 1000 per day)...

Quote:
3) While LGD may be testing and correcting for vignetting, this is unlikely to dramatically impact throughput. The limiting factor in a display fab is the number of substrates. Testing adds costs and a particularly cumbersome test would require more test equipment/personnel but that is a cheaper solution than utilizing a fab at less than its full capacity.
Highly unlikely that there is any 'testing and correcting for vignetting' going on at the OLED panel manufacturing level. These are now running products and after the 'anti-vignetting' fix/hack (which I suspect is at the drive electronics and/or algorithmic level rather than anything to do with panel manufacturing) it is highly unlikely LG is going to be making any further changes to improve yield as they focus all of those efforts on next-generation 2017 designs...

Quote:
4) The more I read, the more I think that the biggest factor is that we think of bringing a new line or fab on line as flipping a switch on capacity. I believe we are underestimating the time it takes to bring a new fab up to full capacity and mature yields. Here is an excerpt of some commentary from a display analyst (Andy Abrahms) on the theoretical ramp of a flexible OLED fab to support Apple.
Great excerpt - would be great to see how it aligns to the ramp schedule of M2. If memory serves, M2 was originally to come online in Q3'14 and ultimately reached full production about a year later (based on my assessment) - I suspect that agrees pretty closely with your example. This could be a major factor impacting increased capacity in 2017 - if Q2'17 is first production, achieving actual 25,000 sheet volume may lag by 6-12 months...

Quote:
5) LGE's sales to end customers are likely to be far more lumpy than LGD's OLED output. The television market is seasonal and the fact that inventories and distribution still need to be filled means that the a significant amount of production early in the year will be used to fill out LGE's sales channels. There is very little chance that LGE is selling 200,000 units this quarter.
Fully agree, though I don't think those seasonal effects impact the big picture. Also, the decrease back to discounted pricing after spending January back at MSRP levels is evidence LGE wanted to stimulate additional demand...

Quote:
6) It will be far easier for LGE to maintain pricing on their OLED's if Samsung never ships a FALD set this year. If I was looking for a set this year I would be hoping that Samsung ships the KS9800 for a reasonable price.
Agree, more impressive HDR FALD LED/LCD offerings at more aggressive prices will force LGE's hand to drive lower pricing more quickly. It is all relative. As long as the most appropriate HDR FALD LED/LCD reference point is the Vizio R65 priced at $6000, the 65EF9500 at discounted pricing of $4500 seems relatively sustainable .

Quote:
FWIW, I saw the G6 today. It was a brightly lit environment but damn did some of the scenes look beautiful. While I was there, I was surprised by the fact that more people stopped to look at the 65" OLED than the ~120" Samsung that was nearby.
believe me, LG has a winner (including the EF9500) for viewing in all but the darkest of environments (which is why I believe they are going to be far more conservative with further engineering changes going forward).

Vignetting so visible that it was evident on the showroom floor was a showstopper and LG has succeeded to overcome the bleeding. Visible yellow banding evident on content was a quality problem that resulted in a large number of returns and could have damaged LG OLEDs reputation right out of the gate, which they have also addressed.

Yield has taken a hit which LG will eventually need to recover from, but with the EF9500 (and soon the G6/E6/C6/B6), LG finally has a mass-market product line for the premium/videophile 55" & 65" segment.

Only shadow detail fanatics like me who watch with the lights off are going to notice the remaining imitations surrounding near-black nonuniformity/streaking/DSE.

These LG WOLEDs are finally ready for prime-time
fafrd is offline  
post #12960 of 13035 Old 02-15-2016, 04:29 PM
AVS Special Member
 
fafrd's Avatar
 
Join Date: Jun 2002
Posts: 7,850
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 14 Post(s)
Liked: 1904
Quote:
Originally Posted by rogo View Post
"So we'll just need to sit tight and see. The one indisputable fact, however, is that LG's reduction in 2016 production target from 1.5M to 1M is not good news for continued price decreases - whatever ends up happening this year as far as pricing, it will be less aggressive than it would have been had LG had an additional 0.5M / 50% of annual production to sell..."

I think that's the most important point. But looking at the number as monolithic still points out the challenge of guessing, say, whether the 65-inch ends the year at $5000, $4500, or $4000.

I imagine LG has to (a) sell close to 400K units in Q4 this year, not 200K, to get anywhere near 1M units and (b) will be able to sell very few 55-inch 1080p units even in Europe, where I suspect many of them sold last year though more of them likely sold in the U.S. than you and I would spitball a guess at.

I also imagine you'll tell me how this is not realistic based on production capabilities. I'll not only respectfully disagree but point out that it would be twice as many as they sold last year with a much greater mix of 4K than last year. Can we still bet on the $4000 65-inch B6?
We're pretty closely aligned with everything you have written here.

400K in Q4 means at least 200K in each of the other three quarters before then, so it doesn't really change the big picture that much.

This would mean pricing is likely to remain close to current levels through Q3, with which I agree.

And in Q4, for LG to determine what pricing level allows them to double sales volume over Q4 2015 makes infinite sense, even if they don't have the production capacity to meet that level of demand in Q1'17.

So I agree that it is very likely that we will see more aggressive pricing in Q4'16 and as for the side bet as to whether we break below $4000 for the B65 or not, that is a near-certainty but not really all that significant. You can already purchase a 65EF9500 for only $250 above that level today (and $500 more widely).

At one point I believe we were agreeing that teaching $3000, not $4000, was the price that LG would need to achieve to really show they are serious about gaining share.

At a production target of 1.5M in 2016, I believed approaching $3000 late this seemed possible. With the reduced production target to 1M, that seems put f reach (this year).
fafrd is offline  
Sponsored Links
Advertisement
 
Reply OLED Technology and Flat Panels General

Tags
Led Hdtv , Lcd Hdtv , Plasma Hdtv , Oled Tv , Lg , Samsung

taboola here
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off