tomw@orac.esd.sgi.com (Tom Weinstein) (09/06/90)
In article <2982@amc-gw.amc.com>, kenb@amc-gw.amc.com (Ken Birdwell) writes: > PS: I saw something from MIT that does true 3D display by using a grid of > piezo-electric and acousto-optic light modulators to generate a normal > holographic interference pattern that can be recomputed on the fly (if you > have a connection machine :) but thats something different, and far too > complicated. I believe it's far from real time. Their (unnamed) supercomputer was about 50 times too slow for that. They could, however, store a two second, 20 frame loop in memory and play that continuously. Also, they were only able to manage horizontal parallax. -- Tom Weinstein Silicon Graphics, Inc., Entry Systems Division, Window Systems tomw@orac.esd.sgi.com Any opinions expressed above are mine, not sgi's.
kenb@amc-gw.amc.com (Ken Birdwell) (09/06/90)
With all this talk about video walls (walls of CRT's) got me thinking about real wall sized TV. Around 1980 a friend of mine was working on a project for video projection that used a laser as the scanning beam. The idea was that since you can get lasers in the exact frequency you want you can get great colors as bright as you want (well, until the reflecting surface melts). The goal was to use it as a table sized CAD-layout system (5' by 3' and 16k by 16k pixels) but as a first step they built a wall sized (8' by 6' and 512 by 480 pixels) TV projector. Well, the first time I saw it I was blown away by the intensity of the colors (they were showing Barbrella) and the brightness (shown in a well lighted room). Unfortunitly the next thing I and everyone else noticed was speckel (sp?). For those of you who dont know what speckel is, If you've ever seen a laser show its the fuzzyness around where the beam hits. Anyway, it totally ruined the picture and they were unable to get rid of it. They had reams and reams of physicists come in and try to solve the problem. Each one knew what is was, (something different every time), and each one had a solution that didnt work. The most popular explanation was that it was a second order quantum effect??? but no one knew what to do about it. One of the worst problems was that it seemed to get bigger the farther away to where from the reflecting surface. They tried it in a vacuum, with polarizing glasses, with diffuse surfaces, with just about everything and no luck. Well the project died after a few years and all the other companies that were trying it (about 4) ran into the same problem so all eventually had to quit. Does anyone out there know if there's work being done on the problem? The reason Im asking is that lately Ive seen some really strange transmitting surfaces, mainly CRT tube screens made out of millions of fiber-optic stands (about 1/8" long) packed side by side and melted (well, fused somehow) together. The end effect is that the image from one side of the glass appears to be on the surface of the other side. This means that the glass on you CRT appears to have zero thickness, a really cool effect. It also has a bunch of weird diffusion properties that I thought would work great for a laser TV. (if youre in WA and have a laser, give me a call. I can get a small sample of this stuff and try it out) Has anyone out there seen a working laser TV system? PS: I saw something from MIT that does true 3D display by using a grid of piezo-electric and acousto-optic light modulators to generate a normal holographic interference pattern that can be recomputed on the fly (if you have a connection machine :) but thats something different, and far too complicated. --
musgrave-forest@cs.yale.edu (F. Ken Musgrave) (09/06/90)
I did my undergrad thesis on exactly this topic back in '77; two years ago I
saw such a device in operation at a company around Moffet Field in the Silicon
Valley (can't remember the name).
While the resolution was spectacular and the color was theoretically-optimal,
there were three significant problems: first, the interference-speckle charac-
teristic of coherent monochromatic light, second, a problem with interlaced
video and zero-decay time pixels (phosphor-equivalents), third, and strangest
of all, a sort of crawling of the image reminiscent of a heavy dose of psyche-
delic drugs.
The fist problem is straightforward and well-understood. There may be a
phase-randomizing technique to defeat it, but I suspect that it will always
occur with monochromtic light sources - it indeed is caused by phase shifts
in such light.
The second problem showed up as disappearance of alternate scan lines when
the action moved vertically. I was mystified by it at the time, but have since
hypothesized that it is a result of intantaneous decay of the image on the
"screen" (whatever that is chosen to be). This might have the result of the
image decay being relegated to the optic nerves, which are moving (in image
space) with the action, thus causing interlaced scanlines to overlap at criti-
cal vertical motion speeds. On the other hand, it may have been a strange
electronic problem with the system; impossible to tell, since I'm working from
memory.
The third problem was truly bizarre. The image appeared to be swimming,
crawling, billowing, like it was slowly melting. The company rep said that it
had something to do with the vitreous humor of the eye, but the explanation
didn't fit any physics that I knew at the time. It was, however, clearly
an effect generated in the human visual system, and therefore perhaps very
difficult to solve.
The overall effect was quite impressive - a 40 foot wide image with amazing
color and resolution. But I think that these problems would make it unaccept-
able to the general consumer. At any rate, the device was for sale, at
~80,000 dollars per unit. If anyone really wants, I think I can find the name
of the company...
Ken
--
*===============================================================*
F. Kenton ("Ken") Musgrave arpanet: musgrave-forest@yale.edu
Yale U Depts of Math and CS (203) 432-4016dlou@dino.ucsd.edu (Dennis Lou) (09/06/90)
In article <2982@amc-gw.amc.com> kenb@belltown.amc.com (Ken Birdwell) writes: [stuff about wall size laser tv with "spreckels" problem] >can get a small sample of this stuff and try it out) Has anyone out there >seen a working laser TV system? Where I work (Naval Ocean Sys Cen), they are working on a 3D laser imaging system. (I'm not sure how much I can tell you. They gave me a Secret level security clearance and all...) They aim a laser at a metal device that moves around on the X-Z plane to get depth. What is the "spreckels problem"? Is it that the pixels don't come out sharp? I noticed that the 3-D images this thing put out were kind of fuzzy, but I always attributed that to an inaccurate scan... I think they used diffraction crystals as reflectors, though... -Dennis -- Dennis Lou Disclaimer: I don't use lame disks. dlou@dino.ucsd.edu "But Yossarian, what if everyone thought that way?" [backbone]!ucsd!dino!dlou "Then I'd be crazy to think any other way!"
ron@vicorp.com (Ron Peterson) (09/11/90)
In article <2982@amc-gw.amc.com> kenb@belltown.amc.com (Ken Birdwell) writes: > >Unfortunitly the next thing >I and everyone else noticed was speckel (sp?). For those of you who dont know >what speckel is, If you've ever seen a laser show its the fuzzyness around >where the beam hits. Anyway, it totally ruined the picture and they were >unable to get rid of it. They had reams and reams of physicists come in and >try to solve the problem. Each one knew what is was, (something different >every time), and each one had a solution that didnt work. The most popular >explanation was that it was a second order quantum effect??? but no one knew >what to do about it. One of the worst problems was that it seemed to >get bigger the farther away to where from the reflecting surface. They tried >it in a vacuum, with polarizing glasses, with diffuse surfaces, with just >about everything and no luck. Speckle is caused by the coherence of a laser beam. You can think of it as all the waves in the laser light moving in step with each other. When light with this property reflects off a surface interference patterns are set up that creates an effect that looks like sparkling light. It only looks sparkly if you (or the surface) is moving however. If you can stay perfectly still there will be a fixed pattern. So the way to get rid of speckle is get rid of coherence. One way is to use a poorly tuned laser. Another way is to use a light source other than a laser. There might be other ways to remove coherence also, like a thin wedge of glass to create an uneven phase distribution in the laser beam. Try asking in sci.physics.