The Foonly F1
Omnibus had simply bought the system from Triple-I (Info Worldwide Inc.) in
Culver Metropolis, CA. Omnibus had huge plans to make films with computer systems, and the F1 had
been utilized in making TRON and quite a few different excessive decision movie tasks. Omnibus was
a Canadian firm, primarily based in Toronto, and had lately performed gone public. They’d huge
plans to open workplaces in Hollywood and New York Metropolis and be the primary huge digital animation
firm within the movie enterprise. Omnibus president, John Pennie noticed the F1 as a machine
with a confirmed observe document. Once I first had a have a look at it, I noticed it as a catastrophe! It was an interesting R&D mission, however was by no means actually designed to have the help, spares, and experience wanted to stick with it and operating in a critical manufacturing surroundings.
The F1 was initially constructed by Triple-I in hopes of getting a big contract
with the Authorities for an Optical Character Recognition system.
Its design was performed by some graduate college students (Stanford?) named Poole, Pettit, Holloway, and Dyer (?).
As I perceive it, their design grew to become the DEC KL-10, however I might have that flawed.
The CPU was constructed on 5 wire-wrap pages, that have been machine wrapped.
The design was performed utilizing one of many first CAD programs I had seen, (SUDS?) which I imagine ran
on a PDP-15 clone that Triple-I utilized in its PFR-80, which was initially meant to be a movie recorder.
The PFR had a vector-tube graphics show, and the person might graphically join numerous
logic components from libraries and have wire-wrap lists generated for machine wrapping.
The PFR, by the way, was the explanation the F1 grew to become a film-maker.
Triple-I had discovered the right way to get a smaller, higher managed spot out of a CRT than
anyone else on the time, and used that know-how to print movie for newspaper pages.
Writer’s Notice: I’ve a number of of the high-rez CRTs used within the III movie printers, if anyone is occupied with them, please contact me, (dave (at) zfx.com).
The F1 had a bizzarre bus interface to the PFR that allowed it to dump excessive decision
imagery to the PFR and have it print to a normal 35mm digital camera. The photographs have been
ship in Pink, Blue, and Inexperienced information and uncovered by Wratten filters onto Kodak 5247 shade destructive movie.
One other PFR be aware, somebody at Triple-I had programmed a easy graphics sport to run on
the PDP-15 and the vector tube. You used front-panel dials and buttons to regulate
acceleration and rotation vectors for spaceships that might shoot particles at every
different. I am unsure of the precise origin of this program, however I imagine it was
the primary interactive vector graphics sport, which finally gave rise to a gaggle of
vector-displayed videogames within the mid-eighties.
The F1 had sufficient in widespread with “customary” DEC-10 peripherals that it might use
a disk channel controller( RH-10?). This was connected to three RP-04 (~50MB) disk drives,
one for the system, and two striped collectively to carry a complete of two high-res pictures.
One could be computed and saved as separate RGB recordsdata, whereas the earlier one was
being printed on the PFR. The F1 had an enormous Ampex core reminiscence unit that I imagine
held 256K 36-bit phrases. That turned out to be barely sufficient to do high-resolution
laptop graphics, so Triple-I had constructed the MMM, Film Machine Reminiscence, which
was constructed out of 4K static RAM chips. The reminiscence was organized in order that it might
be displayed as an RGB picture at 1024 x 1024 on a raster high-resolution monitor.
In that mode, the MMM held 6 bits per shade, 18-bits per pixel. It might additionally
be used as a bulk 512K phrase reminiscence for the TRANEW graphics rendering program, which
used it to retailer hash tables. It was fascinating to “watch” this system execute
on the MMM monitor.
I even have one of many MMM reminiscence boards…. someplace in my junk pile.
The F1 required a DEC KA-10 as a “console processor”.
Because it was a improvement machine, the console was used to load and run the
microcode. It was additionally an vital debugging device when (not if!) the system
did one thing screwy. There was a modem connected to the KA-10, and in an
emergency, Dave Poole might normally be persuaded to dial into it and run diagnostics.
After all, the console machine had its personal disk controllers, disks, tape drives,
even a line printer, and it ended up holding person accounts and appearing as basic
non-job person disk storage. The KA-10 itself booted from punched paper tape, so
a cold-start for your complete system was fairly a journey up from floor zero!
When the F1 was up, it in all probability ran at one thing like 6 MIPS. That appears actually
sluggish by as we speak’s customary, however the code had been extremely optimized. The unique
Fortran program, TRANEW had been written by Gary Demos with assist from Jim Blinn,
the JPL man that did the pioneering computer-generated fly-by animations for NASA.
They’d optimized interior loops in meeting code, and had eliminated each final bit
of “extraneous” code from the working system, TOPS-10. It was actually extra like
Swiss Cheese than an OS.
A second PFR unit acted as an “encoding station”, and was connected to an enormous Calcomp
digitizing pill. Exterior vector shows had been added, as had one other puck,
in order that the operator might digitize from engineering-type drawings to construct a
3D database for the objects used within the animations. There have been some pretty
subtle packages for extruding solids, sweeping shapes by 3D paths,
and different wonders, all executing on a PDP-15!
So once I arrived and noticed this method in operation, my preliminary awe for the quantity
of pioneering work that had gone into making the system do all of the issues it had
performed shortly became a way of dread given the duties I had earlier than me, which
have been to: (1) re-assemble sufficient of the unique team-members that had used the
system to have the ability to do characteristic movie manufacturing, (2) find out how the system functioned
in order that it might be maintained and stored engaged on a grueling manufacturing schedule,
(3) try to seek out areas the place its efficiency or reliability might be improved,
(4) MOVE the system to new amenities that have been being constructed at Paramount Studios,
and (5) keep sane!
Jim Rapley, who was one of many upkeep guys who knew the Foonly, and had
labored with most of the folks because it was developed at Triple-I got here onboard first,
adopted quickly by Artwork Durinski, a designer who had labored on TRON, Adam the Juggler, and
many different Foonly tasks. Wealthy Schroeppel, a guide for Triple-I had performed
the TOPS-10 hacks and sometimes consulted, as did Jim Dungan, who knew TRANEW
and a whole lot of the code that ran within the PFR.
After all, Gary Demos and John Whitney, having gotten Triple-I began within the film
enterprise, noticed all of the shortcomings of the Foonly. They’d pushed for Triple-I
to construct the DFP, the primary (that I do know of) high-resolution digital movie printer
for movement photos. This was the subsequent technology PFR, utilizing an 8″ CRT which had
fast-decaying phosphors in order that it might be used for scanning in movie (utilizing photomultiplier
tubes constructed right into a particular digital camera) in addition to printing. The imagery was wonderful.
After they left Triple-I, they put collectively a deal the place they obtained Management Information Corp.
to purchase a Cray supercomputer, hire the DFP from Triple-I, attache Ramtek shows
and IMI and Evans & Sutherland vector tubes for encoding and movement. That was
Digital Productions, which Omnibus ultimately purchased from CDC, however I am getting
forward of myself.
The purpose is that Triple-I simply needed to unload the Foonly and all its peripherals
and software program as one huge package deal, and Omnibus and John Pennie have been the right sucker.
So Jim Rapley knew a man named Serge Polevitzki, who was on the Naval Ocean (?NOSC) down
in San Diego. NOSC apparently had a bunch of stuff that wanted to run on a TOPS-10
system and since DEC not made them, Dave Poole and Foonly, Inc. obliged them
by developing with the F4, a single rack-unit 36-bit laptop that ran TOPS-10.
I flew as much as Berkeley to see Poole and talk about upgrades to the F1.
Poole was a personality. He had lengthy hair and an extended beard and glasses. I met him
at Foonly, Inc. which was a small warehouse that had been the meeting line for
some thirty F4’s. He wore ragged clothes and bluejeans, and had a pet parrot
that rode on his shoulder more often than not. When the parrot wasn’t on his
shoulder, it was perched in one of many “unfinished” F4’s. There was a precision
formed piece of cardboard duct-taped into this explicit machine in such a
means that the chicken’s droppings have been deflected away from the internals of the
machine. Poole’s suggestion was that we’ve him construct us an SMD disk
controller in order that we might use greater, quicker disks, just like the Fujitsu Eagles
that have been widespread SMD drives on the time. Sadly, his schedule to do that
was going to take months, and our manufacturing schedule was already overfull.
He did what he might on the microcode facet, however the improve by no means occurred.
Most likely the worst second of my life with the Foonly got here after we had managed
to get it moved to Paramount Studios and have been cranking by the movie for
an extended sequence for Flight of the Navigator, a Disney movie. In these days,
the concept you could possibly have a digital camera make a seemingly infinite transfer, untethered,
fascinated movie administrators, so all of them needed lengthy digital camera strikes over an infinte
airplane, in all probability the toughest factor for rendering algorithms to handle, what with all that
depth sorting. Because the Foonly solely had sufficient disk storage to carry the body
being computed and the body being printed, the numbers labored out like this:
30 seconds of movie at 24 frames per second works out to 720 pictures every computed
and printed at 6000 x 4000 pixels. At twenty minutes rendering time every body,
that meant it took ten days to do “the run”. As we have been 5 days into our
ultimate “run” for Navigator, unknown to us, a Paramount electrician was exterior
hot-wiring a brand new circuit in an influence panel, when he dropped a wrench and shorted
two of the three part energy busses collectively. Provides within the KA and the F1
died, as did disk drives, interfaces, you title it! Once we lastly obtained the
programs to energy up, the KA system drive had crashed, and we needed to get new heads
in after which rebuild that drive. Bear in mind with 1600 bpi tape drives, restoring
an enormous system might take all day! Then the KA would not load the Foonly microcode.
I keep in mind a few three day interval once I would drive residence and attempt to sleep
for a number of hours solely to drive again and attempt to get operating once more. The actually
terrible factor was that I stored seeing huge billboard indicators on the way in which in promoting
Flight of the Navigator, saying “coming subsequent week!”. “We hope!” I’d mutter
to myself.
As Omnibus grew, we additionally have been creating our personal 3D software program that ran on Unix
and Silicon Graphics programs. In the end, the hope was to maneuver in that course
and let the Foonly die a peaceable loss of life. But it surely lived on as a result of it labored
(principally) and cranked out higher high-res movie than anything besides the DFP.
When Omnibus was contacted by CDC about shopping for Digital Productions, we ended up
with the Cray/DFP mixture, which was spectacular as hell, however each bit as
kludged collectively and one-of-a-kind because the Foonly.
Final I heard, (1999?) Dave Poole had sailed into some bay up close to Anchorage, AK, by no means to
be heard from once more. He was reportedly misplaced at sea. He had a 100′ Swede sailboat that was legendary. I can simply
about envision him crusing it, parrot on his shoulder, off into oblivion.
Here’s a closeup view of the “Guts” of the F1.
There have been 4 “Pages”, every machine wire-wrapped ECL on Augat customary
PC kinds.
Because it was all ECL, the indicators all ran twisted pair,
and there was a concept that the machine failed steadily due to
“minimize by” from the numerous wires going round sharp corners.
To the suitable you’ll be able to see all of the sign and energy cables.
Notice the distinguished air-conditioning outlet instantly beneath the pages!
The one remaining F1 mission engineer who would idiot with the F1 was Dave Poole, who lived up
close to Berkeley, and would often dial in to assist us with issues.
He had instituted an improve program that was underway when Omnibus purchased
Digital Productions.
Evidently, that buy introduced all additional work on the F1 to a halt.
I heard that Jim Rapley and Intergon had hauled the F1 away and made films
with it for some time, however I’ve not been in a position to attain Jim to seek out out
if that is so. Jim labored with me at Omnibus, and had labored with
the F1 at Triple-I, in addition to the Cray at Digital Productions.
All pictures Copyright © Dave Sieg
Final Up to date 9/23/2011
Back to Dave’s History Page
Please e mail me for more information: dave@zfx.com