From: brucec@phoebus.labs.tek.com (Bruce Cohen)
Subject: Re: Integrated laser arrays for Eyephones
Date: 1 Nov 91 17:44:15 GMT
Organization: Computer Research Lab, Tektronix Inc.



In article <1991Oct31.225456.3495@milton.u.washington.edu> 70353.3056@
CompuServe.COM (Christopher Fry) writes:

> The beauty of the fixed array is no moving parts and the whole thing made by 
> VLSI technology. How to you make a laser scan with those constraints?

Same way you scan an LCD array or an electroluminescent panel: you have
some (well, a lot, in terms of number of transistors) switching
electronics which controls when current gets sent to each pixel; the
switching is timed by the synchronization pulses in the video signal.

> I'm thinking just a little ahead where putting a few ram cells under each 
> laser would be, fabrication-wise, easy, and oh so conveinient. That means 
> the OUTPUT of the RAM is in parallel. ie right out to the laser. We wouldn't
> need any lines to carry the output of a ram cell though the ram matrix. 

The problem I see here is one of fabrication technology: it's still not
easy to incorporate visible light emitters and VLSI logic on the same
chip because they have different requirements for semiconductor doping,
device geometries, wafer processing, etc.  Developing new process
technology is incredibly expensive (tens to hundreds of millions of
dollars for each new process generation that the commercial memory and
microprocessor vendors develop, for instance).  In any case, you still
need to get the data to the RAM cells, and you still need an output from
the RAM for readback to the central processor unless you're building a
central compositing frame buffer which is copied into from other frame
buffers where the images are built.  The compositing buffer is a nice
solution technically, but is very expensive in terms of amounts of
memory and cost of the high-speed interconnect bus.

> The shift-register that you described is indeed a neat trick, but remember, 
> for decent resolution eyephones we've got a LOT more pixels than any display 
> yet made. And besides, you still have to have the frame buffer somewhere so 
> you're not saving hardware in total. 

I don't have the time right now to go deeply into the design of
shift-register and video-ram based display controllers, but take my word
for it, there has been a *lot* of clever engineering put into this area.
I'll try to post something about this in a week or two, but I can't
promise; I'm still tied to my day job, which isn't VR :<(.

In the meantime, you might want to ask your local video-ram chip
salesman (Texas Instruments would be a good place to start) for some
application notes.  Or is someone else in this group willing to take the time
to explain this stuff now?

I don't agree with your statement about the required number of pixels.
First, the requirement for high-fidelity images is not more than about
2000-2500 lines in the horizontal; this was achieved in very high-cost
CRT displays years ago, and building framebuffers of this size is just a
question of whether you can afford it (if you have to ask, you can't).

Also eyephones have a major advantage over fixed screen displays: they
move with the eye, so they can be designed with more resolution at the
center than the edges to match the variance in resolution in the eye.
Of course, you can't be as parsimonious with resolution as you could if
you quarantee that the fovea of the eye was always aimed at the center
of the display, but there are still some gains to be made there.

> Including the RAM on the chip will increase the reliability problems,
> but probably not more so than a moving part scanner.

Oh, certainly less.  But there's still a density problem.  We are at
least 6, and possibly 12 years away from having commercially available
IC fabrication processes which are capable of the densities required for
an entire frame buffer plus output emitter array.

> Hope you're right. I believe in parallel development. Good LCD displays have 
> been "just around the corner" for so long that I've lost faith. Laser arrays 
> are the new kid on the block.

Good LCD arrays are getting here a lot faster now that there's a
large-scale profit to be made from them.  Some of the new active-matrix
displays are as good we need both in speed and resolution, but they're a
tad expensive yet.  When the TV manufacturers or the laptop computer
makers see a significant market demand, they'll use the better displays.

------------------------------------------------------------------------
Speaker-to-managers, aka
Bruce Cohen, Computer Research Lab        email: brucec@crl.labs.tek.com
Tektronix Laboratories, Tektronix, Inc.                phone: (503)627-5241
M/S 50-662, P.O. Box 500, Beaverton, OR  97077