Linux Tools for Serious Photographers

Len Ovens len at
Tue Aug 7 21:13:52 UTC 2012

On Tue, August 7, 2012 8:14 am, Ralf Mardorf wrote:

> I remember similar prices for the Sony Betacam, 150000DM.
> Betamax already was pro in the beginning 80s, but only Umatic high-band
was pro too. I started my "carrier" in the 80s, at the video studio of
the University Essen in the age of 17/18 (I thought I was 16/17, but I
was mistaken). At that time the first Sony Betacam was introduced.

Ya the betacam was there, the government channel (CBC) had them. We ran
hitachi cameras and a sony bvu200 if I remember correctly. The whole price
was about 40k I think, a lot less than the betacam.
> I attache photos from that time. I'm the one at the Bosch camera, I
wonder about my short hair, usually I had and I have long hair.

Our RCA studio cameras were a bit bigger, but the peds and controls were
similar. Control room layout was much the same too.

>> stream in the building) Video streams don't have to be synced and timed
any more, frame store can fix that.
>> Audio was a whole separate chain.
> No sync for video does work? At home I don't have sync for my RME audio
card and an ADAT device, when syncing by ADAT only. I still don't have a
wordclock thingy for my RME card. The ADAT device has got an additional
wordclock connector.
> I try to avoid to have un-synced devices when ever possible.

Video is not like audio. It is not always true, but most often only one
video source is used at a time except for the relatively short time of
switching between sources. At the video switch the video has to be in sync
at least so as the vertical and horizontal sync lines up, but that is the
same from a frame from a minute ago or a year ago. In our studio, all the
local cameras, telecine and tape machines where hard synced and delayed to
match up using a studio wide master clock and delay lines (often long
loops of video cable), But we also used satellite input of uncertain
timing. For that we had a frame store device that recorded a full frame
(two subframes) and fed the output out synced to our master clock so we
could switch to it without a picture roll. This would even work for a
fade, but would probably not be enough storage for long chromakey stuff
unless frames could get dropped at some point. I was in engineering not
operations BTW. (No I am not a papered engineer :)

Anyway, for that kind of switching between two sources, frame store
technology could work with unsynced sources. You would have one fader and
could add as many outboard hard switchs as wanted so long as you didn't
try to fade too soon after switching the inputs... One frame wait may be
needed. (1/30th sec or so) For a computer this would mean at least two
video inputs. I think there are 4 input cards out there though. USB video
in would probably be a problem as you would likely already be using those
for extra monitors (three monitors total with one of them having controls
and small preview windows and one full size preview and one full size
output monitor.) Another USB would get used up for input from a control
surface IF (think MIDI controller). So your video ins would need to be
PCIe or something. Starting to get quite specialized, but when you
consider the cost of some of the RME stuff, not outrageous for a serious
amateur. (actually it would take a group of amateurs so you would have
camera men sound people etc.) I don't know how the cost would vary from
purpose built HW either.

Still for most people (pro or not) computer video means capture on one or
more cameras and editing and mixing after the video file has been
transfered to the computer. Much the way most electronic news gathering is
done. (and has been done for ages... microwave links are not cheap, even
just from a licensing point of view only the rich networks do it and only
for special stories)

Len Ovens

Len Ovens

More information about the Ubuntu-Studio-devel mailing list