jbn@glacier.STANFORD.EDU (John B. Nagle) (07/02/87)
This Sony 8-frame-at-a-time machine suggests some ideas. True single
frame stop-start recording requires tight tolerances on the video deck.
Why not use streaming techniques? Back up a few frames, get up to speed,
switch from play to record as the last frame goes by, cue the computer
to start sending frames, record as many frames as the computer can send
at full speed before it runs out of data, and then come to a normal stop,
waiting for the computer to generate some more frames.
This isn't exactly your home workshop kind of project, but it's an
idea for people who build VCR add-on electronics boxes. Lyon Lamb could
probably do it, but it would cost too much.
Sony also makes a machine for survellance applications which records
at rates down to 1 frame/second. This might be a useful output device for
an animation system, if one can build the frames and buffer them on disk,
then read them back at a rate fast enough to keep up with the slow video
recorder.
John Nagleking@dciem.UUCP (Stephen King) (07/07/87)
I am also disappointed that the new video-8 decks do not give us single-frame recording. I have, however, found some information on time-lapse VTRs, namely in the May 15/87 issue of _Sound and Video Contractor_. Here, Neil Heller of Gyyr Products, Anaheim CA, describes his company's approach to modifying the conventional VHS system to permit this time-lapse recording. I'm afraid I don't completely understand how it works, but it may prove to be a useful tool for the computer animator. The system was originally designed for security and surveillance systems. Does anyone have more info on these _wee beasties_? regards, sjk (Stephen J King) ...utzoo!dciem!king
shep@datacube.UUCP (07/09/87)
I've been using the Sony CCD-V110 (8mm camcorder) for a few months now. The 8 frame assembly recording works really well; even the audio is cut lickety-split. I have the service manual and have been dicking with the serial control bus to the point where I think there is a strategy for *SINGLE* frame assembly recording: The 8-frame recording seems well embedded into the machine control CPU and I have no idea why 8 frames was chosen over 1. (Is there some technical reason?). There seems to be enough smarts to make the deck index to a frame point. This is used in the "record review" mode. By backing the deck over seven of the eight recorded frames and dropping in; clean single-frame assembly edits should be possible. More the moment it works...! Shep Siegel Datacube Inc. DSP Products Group 4 Dearborn Rd. Peabody, Ma. 01960 UUCP: shep@datacube.COM VOICE: (617) 535-6644; FAX: (617) 535-5643; TWX: (710) 347-0125
doogan@acf4.UUCP (Vincent Doogan) (07/10/87)
See the new issue of Popular Science. It has an article about Fisher-Price's VTR that uses a fixed head and cassette tape to record video images. Howard Fink acf1nyu.edu
rwhite@nu3b2.UUCP (Robert C. White Jr.) (07/11/87)
Since the veido recording head spins on its axis and it's axis is
on a bias [slanted with respect to] to the direction of tape travel
a veido tape is "always in motion" even if the tape is not traveling.
Each image is a diagonal stripe on the tape so seccuessive frames are
dropped in anad apear as:
frame #: 1 2 3 4 5 6 7 8
---------------------
\ \ \ \ \ \ \ \ \ \
\ \ \ \ \ \ \ \ \ \
...
\ \ \ \ \ \ \ \ \ \
---------------------
on a segment of tape. The problem is that the jog [forward|backward] until
sync is dificult to maintain/produce. Even good consumer level VCR's
have some problem keeping an acceptable "still frame" [stop the tape and
keep the head engaged and spining] Multiple "heads" on a single "head
drum" are used to catch the slack and skew.
Extreemly good veido tape editors have a "multi head" drum and
two extra little arms which travel the tape in increments of millimeters.
The controll circuts which preform this act are extensive and expensive.
Beacuse a computer would be more upset than a human eye about the loss of
bits most vedio/data medimum write seccuessive frames of identical data.
A system could easily be designed to "still frame" data but TWO sets of
"travel arms" would be required 1) alignment on amilimeter basis, and a
data-frame-by-frame set to account for/remove startup skew and tracking.
Such a device is easily possible, but quite expensive without quantity
purchases [thousands]. It can't really be added on to an existing VCR.
Further Details Available. [I know a lot about veido from my high school
minor]
Robert.
Disclaimer: My mind is so fragmented by random excursions into a
wilderness of abstractions and incipient ideas that the
practical purposes of the moment are often submerged in
my consciousness and I don't know what I'm doing.
[my employers certainly have no idea]