RE: 3D SPEX (and LCS in general)

[P3D <URBANIC@xxxxxxxxxxxxxx>]

I haven't had the chance to keep up in this forum as much as I would like,
but if Greg Marchall's otherwise well-informed reply (sounds like he's been
around awhile) reflects the current conventional wisdom in this group, I
have a pleasant surprise.  Greg stated the rather overlooked fact that:

> (1) Refresh rate is usually limited by the monitor ... The max
> refresh rate (vertical sweep rate) itself is usually not the limiting
> factor unless you are using relatively low resolution.  The hori-
> zontal rate max may limit the refresh rate for a given resolution.

which is so very true, but then (as most LCS vendors have) he went on to miss
the major implication of this seemingly trivial technical note when he said:

> I share your dream that someday we'll be able to show great 3D
> on common display devices, but you should be aware of some
> of the problems:  (1) High refresh rate and high resolution translate
> pretty directly to high cost with CRTs - I don't think this can be got
> around.

NO, NO, NO!  You can look at my companies "3D Theory" section on our Web
pages (www.neotek.com), but basically, as Greg noted, you can get just about
any monitor to go to 120Hz if you keep the total bandwidth (i.e. resolution)
the same.  This is the basis of the system that we have been selling to schools
(Universities, Colleges and now High Schools) for years.  You CAN NOT get a
high-quality display if it flickers, and it will flicker in a normal room on a
normal monitor if it is not at least 120Hz.  We have to guarantee that every
student in a class can view the images for hours at a time (graduate medical
students usually use it for 2-3 hours at a stretch) without eye fatigue (a
nice way of saying headaches and nasuea).  And, we have to be able to deliver
that on their own computers and monitors, many of which were purchased years
ago.

The way we do that is to include hardware which "synch-doubles" or adds an
extra vertical refresh signal to the video cards signal.  This allows the
monitor to refresh at twice the old rate without changing the bandwidth, or
total resolution of the signal as delivered from the video card.  This
technique has been around for years and is used in all the serious high-end
stuff out there.  The only two long-standing LCS companies,
StereoGraphics and 3DTV Corporation, both use this technique in their top-
of-the-line equipment although they pretty much sell cheap "flicker" glasses
like all the other companies to the more frugal hobbyist market that
predominates here (although our one distributor, Jon Gross' Catalyst Corp at
http://www.skypoint.com/~catalyst/ has had some success selling to the more
discriminating amongst the NSA; I would be surprised if there weren't some
of his customers reading out there now).

Why doesn't everyone use this technique?  Simple, the signal processor to
add in the extra synch costs money.  People using novelty 3D devices can't
justify spending such money.  Hence the parade of companies that have come
and gone over the past 5 years selling low-frequency glasses to the masses.
We could only justify using synch-doubling because our initial market was
in education, where they will pay a little more for something that is usable.
Our original prices locked out the anaglyph glasses and video game crowd
(although that is changing now).

Greg also has some other astute insight:

> (2) Stereo images on CRTs exhibit ghosting due to the
> long exponential decay of the phosphors.  This can be solved
> very easily by changing the phosphor composition.

OR, and this is a big OR, reprocessing the image to mute the problem
chromal boundries.  Actually, it is a little more complex than that as the
real problem isn't the absolute phospher persistance, but the imbalance in
persistance between phosphers.  By doing some image processing to mute
this appropriatly (say a high parallax, vertical edge with a large contrast
change on the worst chroma), this problem can be eliminated in all but
pathological cases.  Again, you may want to check out our "3D Theory" page
for this issue as well as some others that are equally important.

> CRTs will eventually be replaced by various flat panel displays
> which (guess what?) are even more problematic for stereo!

If these were LCD's you would be right, but the future at this point seems
to be all Digital Micro Mirror (the Texas Instruments technology that has
quickly begun to dominate the projection market; it is used by most of the
major Japenese vendors) which is even more stereo-friendly than CRT's.  It
is a great time to be doing high-quality stereo work!

John Urbanic


------------------------------