Originally Posted by davegow
This illustrates one reason why front-projection has technical advantages in large sizes, since the yield issue is confined to tiny chips inside the projector, not spread out over the screen.
Front projection has a lot of advantages. It's disadvantages remain completely fundamental
to the technology and experiments in "daylight viewable" screens haven't done much (anything?) to cahnge the equation.
Fact is, front projection into the home is not expanding as a fraction of the market. And things like 90" Sharp LCDs are going to eventually lead to its presence shrinking. There was a push a few years with the "daylight viewable" screens to change things, but what happened was the screens cost thousands instead of $200-300 like a regular screen and therefore projectors still cost a lot to implement.
(There are other reasons people don't want projectors. They need to be installed in ceilings or have cables draped around. You'd be surprised to learn that many people cannot realistically have a projector in their family room even if they wanted one as no install is viable.)
Anyway, you are certainly spot on about yield. In the long run, however, yield rises high enough for whatever size of flat panel being made to be reasonably viable. The Sharp 80" on an older 40" line shows that kind of progression.
Very large commercial displays, for example at stadiums, often consist of multiple panels bolted together with individual datafeeds. That gets around the yield issue, since a faulty panel can be replaced without touching the remaining ones. I wonder if such a model might be tried in some future high-end home TVs. If so, bring your checkbook.
The problem with this is the inter-panel spacing has to be less than one-pixel. No current panel design makes this realistic. Anything greater, and the screen will have a large grid where the panels are joined. I have envisioned a system by which OLED panels could work this way, but the alignment would still be challenging and the manufacturing of the "tiles" would be difficult as well. At the end of the day, it's possible that even an 8G fab can turn out 2 100" OLED panels at a time.
Should the OLED revolution take off, eventually you'll see larger sizes, especially if manufacturing techniques get radically altered. That said, the logistics of moving rigid displays are what they are, which is why ideas like "tiled" displays and roll up OLEDs are popular here. "How can I go even bigger
someday?" Legitimate questions, but also why the projector market still exists.