h a l f b a k e r yIt's the thought that counts.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
3D LCD display
How to build your own 3D display cheaply. (Baked, I know!) | |
Before I start, I apologise for posting
what I already know to be a fully-baked
idea - but it's not a well-known idea
and the instructions here will let anyone
build a neat piece of kit that's fun to use.
By the way I built this back in 2001
and sat on it thinking I would patent it. An
exhaustive
patent search at the time showed
no prior art. But I was too lazy and never
did the paperwork, and now I've seen at least
two of these home-brewed at exhibitions in the
last year or two. That's my only justification
for entering this as a half-baked idea, because
like most of my hobby projects it never went
past the prototype :-)
One of the earliest 3D displays was Edwin Land's
system (US Patent No: 02084350 from 1935) with
two TV tubes and two sheets of polarizing film,
plus a semi-silvered mirror. The displays
are placed 90 degrees to each other with the
mirror at 45 degrees between them so they overlap
when viewed through polarized glasses.
My design is to use two LCD flat panels, and
*no* polarizing filter, because the panels
are already polarized. And fortunately almost
all flat panels are polarized at about 45 degrees
(actually nearer 39 but it's close enough for
this to work) so when placed at 90 degrees
to each other with the semi-silvered mirror
between them, by a wondrous piece of luck
they are almost perfectly cross-polarized.
Back in 2001 I mailed the authors of several
3D graphics drivers and asked them to support
this format (one of the displays needs to be
mirror inverted - either horizontally or
vertically, it doesn't matter - just rearrange
the panels and the mirror to suit) and much to my
surprise several of them did so, so you ought to
find it relatively easy to make a home-brew
system like this work.
The glasses you use are the standard passive
ones as used at old-style (pre shutter-glass)
IMAX cinemas. They're easily obtainable from
Stereophotography vendors.
Any two-headed VGA card will work, like the
very old Matrox I used in 2001, or cheap modern
cards like the GEForce 5500 which can be had
for $50.
I'ld love to see a new generation of arcade games
use this technology. Many stand-up arcade boxes
already use a 45 degree mirror to conserve space
with large glass monitors. They'd be easy to
tweak into using a semi-silvered mirror and two
flat panel LCD displays.
Graham
Original posting by me in 2001
http://www.stereo3d.../643.html?994777197 with link to original 1935 Land patent. [gtoal, Jan 12 2006]
Planar display
https://web.archive...nding/stereomirror/ Commercial version now on sale (not by me!) [gtoal, Sep 08 2006, last modified Dec 25 2019]
HD version
https://web.archive...d.de/3dvideo-e.html And another one [gtoal, Sep 08 2006, last modified Dec 25 2019]
Folding Stereoscopic computer display
http://patft.uspto....282&RS=PN/5,598,282 Damn, beaten to the punch again :-( Patent filed 7 years before [gtoal, Nov 14 2006]
Please log in.
If you're not logged in,
you can see what this page
looks like, but you will
not be able to add anything.
Annotation:
|
|
PS When I first thought of this I also thought of 3D contact lenses as a joke, not realising it was possible. According to the entry for "3D Contact Lenses" (qv) it is actually possible to design lenses which align themselves correctly in a vertical direction, so this should be very realisable. With big bright LCD TVs around nowadays which didn't exist when I built my prototype, a cheap home 3D IMAX ought to be very feasible. Unfortunately I rented one of the IMAX 3D DVDs last month to see how it could be displayed on such a system and it turned out that the DVDs are actually *2-D*!!! Grrr. |
|
|
PPS (is it considered uncool here to follow up your own postings?) before anyone points out the ridiculous high cost of lab-quality semi-silvered mirrors - you don't need anything that good. Lab-quality mirrors can be very precise and give 50% reflection and 50% transmission; however glass sold for office windows etc is pretty damn close at a ratio of about 45 to 55. You can tweak the brightness on one of the panels to make them even. Google for a product called "MirroPane". Any local glazier ought to be able to order it for you. |
|
|
i don't understand why is this called 3D. as i see it, all you get is a planar image projection, apparently in thin air. ingenious, but still, planar. in my opinion, real 3D imagery has to involve lasers. |
|
|
Long ago people used to view 2 photographs simultaneously in a handheld viewer. All you need is 2 displays as that is probably cheaper than alternating views to each eye through a filter and 1 display. Also makes refresh requirements slower. |
|
|
I toyed around with this idea myself, except with flatscreen CRTs and polarizing film, but like you, couldn't find a driver that would send separate images to each screen. |
|
|
One setup that I did get to work passably was FlightSim 2002 with WidevieW to display multiple views on multiple monitors. I just set up two views in the same direction, but with slightly varied eyepoints. It worked well as long as the onscreen motion wasn't too fast, but it pointed out a few graphical shortcuts that the coders took, such as drawing the airports as separate planar images on top of the 3-D render. While the terrain was visibly 3-D, the airports were always displayed identically on each screen, resulting in what looked like cutouts on a window, decidedly NOT 3-D. |
|
|
If, as you say, drivers have been made to do this, I may just have to try it again next time I upgrade my computer. (no dual-head currently, can't justify buying a new card for my AGP-only motherboard just yet). |
|
|
no [+] from me, since it is baked, and the concept has been baked for a long time (using a device with built-in polarizers instead of adding your own filter is not really a novel idea, although the grant-everything mentality of today's patent system may disagree) |
|
|
But no [-] either, since I've been wanting a workable 3-D system for a while. |
|
| |