petersen@ucbvax.ARPA (David A. Petersen) (07/05/85)
When is somebody going to come out with a good stereo picture tv system? It seems like all that is needed is an electrically controllable polarizer in front of the screen, say a fast liquid crystal panel. Alternate images would be visible to each eye if you wore polarized glasses. I've seen color stereo slides using a projector which projected two images of different polarizations and a special screen which did not disturb the polarization. It was pretty impressive; sometimes everyone in the lecture hall just about went cross-eyed trying to look at something that was six inches from their noses. I saw slides of cave passages and it was the closest thing to being there I've ever seen. It seems like this would be great for a big color projection tv. There is still the inconvenience of having to wear polarized glasses, but it seems like a small price to pay. Ditto for the extra bandwidth required to send twice as many images for the same picture quality. Maybe it will have to wait until after the high resolution tv standard is developed and in place. Paul Bradley
gwyn@brl-tgr.ARPA (Doug Gwyn <gwyn>) (07/05/85)
> When is somebody going to come out with a good stereo picture tv system? > > ... > > There is still the inconvenience of having to wear polarized glasses, but > it seems like a small price to pay. Ditto for the extra bandwidth required > to send twice as many images for the same picture quality. > > Maybe it will have to wait until after the high resolution tv standard > is developed and in place. I'm a big fan of 3-D (stereo or holographic) imaging. However, you're not going to see much stereoscopy in commercial video because people do not like to wear goggles. Also, about 30% of the general population seems to get nothing out of stereo pictures; I don't know whether this is due to eye dominance or what. Stereoscopy puts high demands on resolution, so you should definitely use high-definition video for it. Unfortunately, if no provisions are made for the second channel in the basic specs, it will be too much trouble for broadcasters and tape/disk producers to hassle with (remember AM-FM stereo simulcasts?).
rl@ucsfcgl.UUCP (Robert Langridge%CGL) (07/06/85)
In article <8794@ucbvax.ARPA> bradley@esvax.UUCP (Paul Bradley) writes: >When is somebody going to come out with a good stereo picture tv system? > It seems like all that is needed is an electrically controllable >polarizer in front of the screen, say a fast liquid crystal panel. >Alternate images would be visible to each eye if you wore polarized glasses. Tektronix has one. Biggest so far I believe is 12" square. Not yet commercially available. Anyone at Tek care to comment? >I've seen color stereo slides using a projector which projected two images >of different polarizations and a special screen which did not disturb the >polarization. I've used this system for years to show 3d pictures of big molecules such as proteins and DNA. It is also possible, though more complicated, to show single projector 3d movies. Bob Langridge ...ucbvax!ucsfcgl!rl (UUCP) Computer Graphics Laboratory 926 Medical Sciences rl@berkeley (ARPA) University of California San Francisco CA 94143 Phone: +1 415 666 2630
henry@utzoo.UUCP (Henry Spencer) (07/07/85)
If 3D movies don't sell consistently well (and they don't: 3D movies
are a periodic fad, with no obvious staying power), then there is no
reason to expect the heroic efforts (new standards, etc.) that would be
needed to achieve 3D TV. TV is also at a considerable disadvantage due
to its (usually) small screen and close viewing distance -- 3D TV would
give you a choice of six-inch people (between you and the screen) or a
rather narrow field of view (behind the screen, which then functions as
a porthole).
There is also a persistent technical problem, in that the 3D image one
gets from orthodox techniques (like polarization) has some tendency to
produce mild eyestrain. The difficulty is that it requires viewers to
vary the convergence of their eyes (which is a function of how near the
objects appear to be) independently of the focus of the eyes (which is
a function of how far away the light-emitting screen is). Most people
are not used to this, and I am told that it gets uncomfortable after a
while. (Either I don't have a problem with it, or I've never watched
3D material for long enough to run into trouble.)
--
Henry Spencer @ U of Toronto Zoology
{allegra,ihnp4,linus,decvax}!utzoo!henryken@turtlevax.UUCP (Ken Turkowski) (07/08/85)
In article <8794@ucbvax.ARPA> bradley@esvax.UUCP (Paul Bradley) writes: >When is somebody going to come out with a good stereo picture tv system? > > It seems like all that is needed is an electrically controllable >polarizer in front of the screen, say a fast liquid crystal panel. >Alternate images would be visible to each eye if you wore polarized glasses. This is exactly what a company in Mill Valley (I believe), California is marketing. One problem is that the refresh rate per eye is only 30 Hz, whereas normal television is 60. The flicker is noticeable, and you do have to wear glasses with electrical wires attached, but I guess you could get used to both of them. -- Ken Turkowski @ CADLINC, Menlo Park, CA UUCP: {amd,decwrl,hplabs,nsc,seismo,spar}!turtlevax!ken ARPA: turtlevax!ken@DECWRL.ARPA
king@dciem.UUCP (Stephen King) (07/08/85)
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Stereo TV implemented using polarizing glasses will require higher resolution than currently employed in order that the 3-D image has the same resolution as a conventional television. Simply switching alternate fields between eyes will produce flicker, as the rate to each eye will be half that normally available. HDTV now under development in Japan will offer the required resolution and (presumeably) frame rate. Tektronix (Beaverton, Ore) makes a liquid crystal shutter which may be suitable for such a project. I am also looking forward to such developments, but bear in mind that TV producers will not spend the money to produce 3-D shows until there is a substantial audience and this will require a standard for the broadcast industry and HDTV manufacturers. -sjk
doug@terak.UUCP (Doug Pardee) (07/08/85)
> When is somebody going to come out with a good stereo picture tv system? >... > It seems like this would be great for a big color projection tv. > There is still the inconvenience of having to wear polarized glasses, but > it seems like a small price to pay. Movie audiences aren't buying it -- I think it'd be an even harder sell for TV. I saw a polarized 3-D movie once. It was great, as long as you don't need eyeglasses, and as long as you hold your head upright. I watch most of my TV from a reclining position, and polarized glasses would drive me crazy (a very short trip, for me :-). -- Doug Pardee -- Terak Corp. -- !{ihnp4,seismo,decvax}!noao!terak!doug ^^^^^--- soon to be CalComp
dgary@ecsvax.UUCP (D Gary Grady) (07/08/85)
> When is somebody going to come out with a good stereo picture tv system? > > It seems like all that is needed is an electrically controllable > polarizer in front of the screen, say a fast liquid crystal panel. > Alternate images would be visible to each eye if you wore polarized glasses. > Paul Bradley I suspect the technique suggested won't work very well for a number of reasons. First, phosphors don't decay rapidly enough. Second, even if they did, experiment has determined that the images for both eyes need to "flicker" in synchronization to avoid unpleasant sensations in the viewer (I'm personally quite surprised the human eye is that sensitive, but stereo systems using to projectors locked together require even that their shutters be synchronized). Lenny Lipton has written an excellent overview of 3-D techniques in film that would be excellent reading for anyone interested in the subject. By the way, broadcast stereo TV has been quite successful as a novelty item. About 6 or 8 years ago a syndicator made a lot of money with a godawful 3D movie "Gorilla At Large". Viewing glasses were sold at 7-Elevens and the thing was heavily promoted. It used color anaglyph 3-D, which means the two images were separated by the "lenses" of the viewing glasses which were different colors. Many people (me included) find this type of 3-D very uneffective. Polaroid anaglyph is MUCH better, but still requires excellent technical support to be really acceptable. There are ways of producing 3-D images that don't require glasses. Most people have seen 3-D postcards that use a lenticular screen to separate two images (the way lenticular screens have been used to, say, produce a plaque that alternates between saying VISA and Master Card). There is a projection analog of this that has been used in the USSR. (See the SMPTE Journal, about 14 years ago.) Another system used with computer graphics involves a mirror that wiggles in and out. I haven't seen this but I understand it is only of use for doing computer generated images at the moment. Like holograms, these images are full 3-D and not just two-eyeball stereo. (It's possible to produce a similar kind of lenticular stereogram, called a parallax panaramogram, by the way, using a special camera that slides the back and the lens during exposure.) Then there are holograms. Sometimes stereo pairs are holgraphically encoded (as in the famous "Kiss" moving hologram), but of course true holograms are incredible. Unfortunately, they are not projectable and I'm unaware of a successful movie system using true holograms. A final note: How about 4-D pictures? A computer graphics researcher whose name eludes me used to go around showing stereo pairs that were of 4-dimensional objects (computer generated, of course). How? The third dimension comes from the angling of the eyes in and out (as with normal 3-D) and the fourth from tilting the head from side to side. Supposedly one could train the mind to actually perceive 4-D images... Stand by for Buckaroo Banzai Part 4, in 4-D, I suppose. -- D Gary Grady Duke U Comp Center, Durham, NC 27706 (919) 684-3695 USENET: {seismo,decvax,ihnp4,akgua,etc.}!mcnc!ecsvax!dgary
sean@ukma.UUCP (Sean Casey) (07/09/85)
Now what I'D like to see is a 3-d system where you didn't have to wear glasses.
--
- Sean Casey UUCP: sean@ukma.UUCP or
- Department of Mathematics {cbosgd,anlams,hasmed}!ukma!sean
- University of Kentucky ARPA: ukma!sean@ANL-MCS.ARPA zrm@prism.UUCP (07/09/85)
You also have the problems of channel bandwidth and or video-recorder bandwidth. High definition TV will have to be digital TV or it will never be broadcast in the same broadcast band as TV is now because it will take twice as much of the TV band to carry a single channel. Add to that a second whole image and you are requiring everything in the system to process four times the information as present consumer TVs. Expensive!
dgary@ecsvax.UUCP (D Gary Grady) (07/09/85)
> There is also a persistent technical problem, in that the 3D image one > gets from orthodox techniques (like polarization) has some tendency to > produce mild eyestrain. The difficulty is that it requires viewers to > vary the convergence of their eyes (which is a function of how near the > objects appear to be) independently of the focus of the eyes (which is > a function of how far away the light-emitting screen is). Most people > are not used to this, and I am told that it gets uncomfortable after a > while. (Either I don't have a problem with it, or I've never watched > 3D material for long enough to run into trouble.) > -- > Henry Spencer @ U of Toronto Zoology The focus/convergence problem is often cited as a cause of eyestrain and sounds quite plausible, but Lipton and others claim that experiment has shown it to be false. A well-shot and well-projected 3-D film has no such problems (see the article on the unreleased Owensby feature "Rotweiler" in The American Cinematographer about 1982). Indeed, I have never experienced eyestrain clearly attributable to that from watching 3-D films or from using a Viewmaster. A much more serious problem is camera convergence. This is a surprisingly difficult and counterintuitive problem (how to set the distance between the two taking lenses and whether and how much to toe them inward). Most eyestrain in 3-D projection probably comes from dirty glasses, out-of-sync flicker (much rarer with modern one-projector systems), and poor convergence. As Henry notes in another paragraph of his posting, there is a "frame" problem in 3-D. If you have a closeup of a person, you have a problem with the edge of the screen (at a perceived distance of several meters) cutting off something much closer, which gives your brain all sorts of trouble. One way around that is to use a huge screen, possibly curved, so that there is no perceived frame. That runs into some technical problems as well (using lenses that wide is liable to put each lens in the other's field). I haven't had a chance to see it yet, but I'm told ultra-widescreen 3-D has been used with considerable success at Disney's Epcot Center. I suspect we'll see touring ultra-widescreen 3-D before too long. Already we've had Imax tours and 3-D seems to be the next step. Wouldn't you pay to see a top-quality, eye-filling 3-D tour of the Amazon complete with snakes jumping into your face? Well, suppose I promise to leave out the snakes? How about a trip aboard the Space Shuttle? How about ... [here the author was found gazing off into space and thinking about a Certain Actress] ... -- D Gary Grady Duke U Comp Center, Durham, NC 27706 (919) 684-3695 USENET: {seismo,decvax,ihnp4,akgua,etc.}!mcnc!ecsvax!dgary
hutch@shark.UUCP (Stephen Hutchison) (07/10/85)
In article <5760@utzoo.UUCP> henry@utzoo.UUCP (Henry Spencer) writes: >There is also a persistent technical problem, in that the 3D image one >gets from orthodox techniques (like polarization) has some tendency to >produce mild eyestrain. The difficulty is that it requires viewers to >vary the convergence of their eyes (which is a function of how near the >objects appear to be) independently of the focus of the eyes (which is >a function of how far away the light-emitting screen is). Most people >are not used to this, and I am told that it gets uncomfortable after a >while. (Either I don't have a problem with it, or I've never watched >3D material for long enough to run into trouble.) >-- > Henry Spencer @ U of Toronto Zoology > {allegra,ihnp4,linus,decvax}!utzoo!henry Separating convergence from focus as a long-term thing is *VERY BAD*. I have mild systemic damage to my optic chiasm (which organizes much of the binocular information from the eyes) because I have accidentally done this (a result of adult ambliopia). What you get is the inability to correctly track objects moving toward you quickly, and difficulty determining exactly how far away things really are. The real result of this is a drastic reduction in my response to depth-cued stimuli; it takes longer to figure out what's happening, ergo I have up to a 3-second (measured) lag in response over some other folks. Convergence and focus are interdependent systems, and it would be very very bad to train them to operate independently. Unfortunately this is what would happen if people watched a lot of 3-d that works that way. Hutch
nzm10@amdahl.UUCP (Neal Macklin) (07/10/85)
> In article <8794@ucbvax.ARPA> bradley@esvax.UUCP (Paul Bradley) writes: > > > > It seems like all that is needed is an electrically controllable > >polarizer in front of the screen... > > This is exactly what a company in Mill Valley (I believe), California is > marketing. One problem is that the refresh rate per eye is only 30 Hz, > whereas normal television is 60. The flicker is noticeable, and you do > have to wear glasses with electrical wires attached... > > Ken Turkowski @ CADLINC, Menlo Park, CA I think the vertical resolution is also halved, because they transmit both left and right frames within one vertical sweep interval. (To use standard TV bandwidth). You then modify your TV's vertical sweep to go twice as fast and their box connects the glasses to automatically switch left and right as the appropriate half is being scanned. I saw the system at NCGA two years ago and they were showing "House of Wax" (?). It looked pretty good to me. The flicker wasn't annoying. You could see the spaces between the scan lines, but they claimed you could play with the focus to make the scan spot taller. They also showed a still image of a 3D line drawing which took some staring at (at least by me) until I saw it in the proper perspective. I don't remember the name of the company, but I remember the president's name: Lenny Lipton. He wrote a book on stereoscopy that I've been meaning to send for. I remember him because he would pull the glasses out of your hand if you tried to touch the lens. He also didn't like it if you held the glasses up to your ear so you could hear them buzzing. I've heard he's been better behaved at other trade shows. I think the glasses need quite a high voltage to work, so maybe he was worried I'd get a shock. -- Neal Macklin (408) 737-5214 ...{hplabs,ihnp4}!amdahl!nzm10 [There are no opinions expressed in this article].
jbn@wdl1.UUCP (07/11/85)
About two years ago American Cinematographer had an issue on 3D,
in which the subject of making 3D movies that don't cause eyestrain was
discussed in much detail. This is a non-trivial problem, because we
want to put images on the screen that aren't in 1:1 scale with reality,
and this implies that certain arbitrary choices must be made about the
viewing plane. But there's a custom calculator available to help make
them.
By the way, the most common cause of eyestrain in 3D movies is
misadjusted projector optics. Look at the image without glasses; if there
is any horizontal misalignment visible between the two images,
the 3D splitter isn't mounted straight. There is a 3D alignment leader
available through SMPTE which if projected through the 3D splitter but
viewed without polarized glasses makes it instantly obvious if things are
misaligned. Mention this to your theater manager as you are demanding
your money back.
John Naglenzm10@amdahl.UUCP (Neal Macklin) (07/12/85)
> I haven't had a chance to see it yet, but I'm told ultra-widescreen 3-D > has been used with considerable success at Disney's Epcot Center. > D Gary Grady > USENET: {seismo,decvax,ihnp4,akgua,etc.}!mcnc!ecsvax!dgary I'm not sure how they did it, but in the Kodak pavillion at EPCOT is a truly remarkable film. You wear polarized glasses to see it, though I don't think it is standard 3D. When you move your head sideways, the image moves with you, and it seems to be one foot in front of your face. You can see the audience all groping the air in front of their seats. I'm not sure, but in the "regular" 3D movies I've seen, I never noticed that the image "moved with you". -- Neal Macklin (408) 737-5214 ...{hplabs,ihnp4}!amdahl!nzm10 [There are no opinions expressed in this article].
dmmartindale@watcgl.UUCP (Dave Martindale) (07/12/85)
You may be interested to know that Expo '86 in Vancouver next summer will include an IMAX 3-D film. Since the screen is so large, the "frame" effect (where objects are cut off by the frame edges) should be minimized. And the large frame size and excellent stability of the IMAX system should produce images with lots of detail and minimize eyestrain due to jitter - improving realism. I believe that there will also be another (non-IMAX) 3D film.
doug@terak.UUCP (Doug Pardee) (07/12/85)
> The difficulty is that it requires viewers to > vary the convergence of their eyes (which is a function of how near the > objects appear to be) independently of the focus of the eyes (which is > a function of how far away the light-emitting screen is). Most people > are not used to this, and I am told that it gets uncomfortable after a > while. I wonder if eyestrain only results from focussing _closer_ than the convergence point. When I worked on a flight simulator, we had what is called "infinity optics", a filter for the CRTs which blocked 97% of the light, allowing only collimated light to pass through. The CRTs (and thus the point the eyes had to converge at) were located about 10 feet from the pilot. -- Doug Pardee -- Terak Corp. -- !{ihnp4,seismo,decvax}!noao!terak!doug ^^^^^--- soon to be CalComp
kenm@parallel.uucp (Ken Musgrave) (07/12/85)
In article <1944@ukma.UUCP> sean@ukma.UUCP (Sean Casey) writes: > > >Now what I'D like to see is a 3-d system where you didn't have to wear glasses. > > Perhaps you've not heard of the varifocal or vibrating mirror class of 3d display devices. The image of a CRT is relected from a piece of aluminized Mylar stretched across a woofer (speaker) which deforms it (very close to spherically) at 30hz. The focal length of the mirror changes, and the CRT's display is synchronized with it. Thus one gets a true volume-filling 3d display without the need for any special glasses. It has been implemented for both random and raster scan display modes. Groups at University of Utah and North Carolina have devoloped these devices, and there was one commer- cially marketed, primarily for seismic data and medical imaging. Advantages include variable point of view (unlike stereoscopically recorded images); disadvantages include low display brightness and transparency of images (i.e., no hidden surface removal). It is ideal for display of CAT scans and medical NMR images, but would never work for movies. Ken Musgrave CS net: ucbvax\!ucscc\!ucscv:bozo UUCP: ucscv\!bozo Computer and Information Science Board UC Santa Cruz, Santa Cruz, CA 95060
connolly@steinmetz.UUCP (C. Ian Connolly) (07/22/85)
> >Now what I'D like to see is a 3-d system where you didn't have to wear glasses. > Perhaps you've not heard of the varifocal or vibrating mirror class of 3d > display devices... In fact, there's a GE patent for one such vibrating-mirror viewing system, developed by Dr. Ed Joynson around 1972 (?) or so... -- C. Ian Connolly, WA2IFI - USENET: ...edison!steinmetz!connolly , , ARPANET: connolly@ge-crd An rud a bhionn, bionn.