Toshiba to show no-glasses 3D TV at CES

[gisabun]

"Toshiba will show its no-glasses 3D television for the first time outside of Japan in January at the International Consumer Electronics Show in Las Vegas.

Unlike other 3D televisions that require viewers wear a pair of glasses, Toshiba's TVs achieve the illusion of depth thanks to a thin sheet of small lenses in front of the display. The lenses split light from the screen to one of nine points in front of the TV. If the viewer sits in one of these sweet spots they see the 3D illusion."

More at source: http://www.itbusiness.ca/it/client/en/c ... p?id=60613" onclick="window.open(this.href);return false;

[cybereality]

Sounds like its lenticular-based. Nothing to see here, move along.

[Likay]

Nine point views seems similar to philips wowvx technology. Quite old one too.

[mediavr]

These are for sale now in Japan -- the 12" model at least, the 20" on the 25th Dec.
http://www.engadget.com/2010/12/21/tosh ... row-no-gl/" onclick="window.open(this.href);return false;
There is actually a lot of new technology in these screens (new for a released display that is).
It is the first 3d TV that does realtime interpolation from 2 view into 9 views (using the Cell processor in the TV)
It has a new vertical aligned subpixel display to remove the necessity for diagonal lenticles for moire reduction.
They are based are superhires base displays.

Anyway US reviewers will be able to judge at CES.

PeterM

[Likay]

Philips already tried and trashed it and there's still no proper or good looking support for 2d+depth images.
Interpolating 2d+depth into 9 views was necessary in the philips as well so i don't think toshiba was first with this so except for a few minor stuffs it's all old tech and the final result (actually quite holographic but with mostly depth) even if gimmicky not close to true stereoscopy.
My initial experiences with this tech are really bad and the results are so far from good that i doubt it ever will give true stereoscopy a match. It probably doesn't matter if they manage to polish the tech much further.
Pros with 2d+depth is that autostereoscopic displays are relatively easy to make and the interpolation into the necessary number of views doesn't require so much resources. Zecotek for instance uses 100 views for their displays...
Cons with 2d+depth is that it will never come close to a true stereoscopic experience especially in close up situations.

I'm still very curious when crysis2 is released with the 2d+depth imaging. The pros with that should also be that autostereoscopic multiview displays like philips, zecotek, toshiba etc will be quite easydriven and once crytek are done i think they would have gone far in the way of polishing the tech.
Let's see what happens but 2d+depth does not make me hopeful.

[mediavr]

my understanding it is not 2D+ depth .. it is interpolation from standard 3d (side by side) tv broadcast (interpolation like
speed adjustment in video editing -- just image based morphing)

[mediavr]

the japanese guy explains it here from 0.57
http://www.youtube.com/watch?v=xzZk03O0 ... ature=fvwp" onclick="window.open(this.href);return false;

[cybereality]

Well I am not saying that multi-view cannot work. I mean, even a simple parallax barrier can look great in certain circumstances. And I've seen lenticular movie posters that had a nice depth to them (with maybe 4 or 5 views). The principles are sound. Its just that its not really anything new. People have tried it before and, for various reasons, it wasn't market viable. Low resolution, limited viewing angles and fixed viewing distance are among the major issues. Now it looks like Toshiba does have a decent solution, and the video didn't look that bad. But I just don't think its ready yet. For example, where is there 9-view stereoscopic content? Nowhere. So the TV has to interpolate at least 7 views. Now with 2 views it is possible to generate a depth map. And from there you can use the original images to texture an angle-adjusted depth layer. So this will solve the major issue with 2D+depth, in that it wouldn't have to recreate [as much] missing content. However its still not an exact science and could produce anomalies. So I am willing to give them the benefit of the doubt, but I still don't think its ready for prime-time use in a living-room setting.

[Likay]

the japanese guy explains it here from 0.57
http://www.youtube.com/watch?v=xzZk03O0 ... ature=fvwp" onclick="window.open(this.href);return false;

Ok! This makes it a little more interesting so i reserve some of my statements in previous posts. I wonder how good the algorithm works and is curious of the result.

[mediavr]

http://money.cnn.com/video/technology/2 ... .cnnmoney/" onclick="window.open(this.href);return false;

autostereo laptop a winner, narrow sweet spot for 3d main issue with larger 56" screen according to this video review of Toshiba autostereo tech
at CES

PeterM

[cybereality]

Actually, now that I see a video of this display I take back what I said. It does indeed look like a step forward. The key point is that it is a 4K display, so they have a little more room to do the multiple views (9 views on a 1080P display is not as hot). From the video, it appeared pretty high resolution, certainly 1080P quality. I would like to see that myself.