A Sky, cable and digital tv forum. Digital TV Banter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » Digital TV Banter forum » Digital TV Newsgroups » uk.tech.digital-tv (Digital TV - General)
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

uk.tech.digital-tv (Digital TV - General) (uk.tech.digital-tv) Discussion of all matters technical in origin related to the reception of digital television transmissions, be they via satellite, terrestrial or cable. Advertising is forbidden, with no exceptions.

Timecode on programme-as-broadcast archive material



 
 
Thread Tools Display Modes
  #21  
Old October 15th 17, 10:06 AM posted to uk.tech.digital-tv
Roderick Stewart[_3_]
external usenet poster
 
Posts: 2,131
Default Timecode on programme-as-broadcast archive material

On Sun, 15 Oct 2017 11:11:31 +0100, "NY" wrote:

I notice on https://www.youtube.com/watch?v=aA8WFaVt0fo the timecode has a
letter after it which cycles through A, B, C, D, A, ... with every frame. I
presume this is to identify the 4-frame boundaries on which PAL VTRs must
make an edit if it is to be seamless. Is the rule that you must only make an
edit on one specific boundary (eg D - A) or can you edit on any boundary
you like, as long as you use a corresponding boundary at the other end of
the edit - eg you can edit on A - B if the other edit is also on A - B,
likewise for matched B - C or C - D edits, so the A, B, C, D sequence is
preserved.


Fascinating. Though I worked in television engineering for many years
I never saw this. The nearest I ever saw was a dash that appeared on
the timecode for every alternate field, to signify the odd/even
sequence. I think you must be right about the purpose of the four
letters.

The need to preserve PAL frame sequence was really only considered for
what was called an "invisible edit", where the picture content, or the
majority of it, was identical between the shots being joined, such as
the same shot with and without a caption or effect of some sort.
Playback timebase correctors had to maintain continuous subcarrier
phase by adjusting the delay of the whole signal (because luminance
and chrominance did not follow separate paths as in domestic
recordings), and if the PAL sequence was incorrect, the whole image
would appear to jump sideways at the join by some multiple of a
quarter of a subcarrier cycle. When editing for action, it was
generally disregarded because a change between two shots with
different content would conceal the sideways jump.

Rod.
  #22  
Old October 15th 17, 10:11 AM posted to uk.tech.digital-tv
NY
external usenet poster
 
Posts: 1,187
Default Timecode on programme-as-broadcast archive material

"The Other John" wrote in message
news
On Thu, 12 Oct 2017 12:07:21 +0100, NY wrote:

This is probably a weird, obsessive question...

I've noticed that if you see a programme from a tape archive (eg BBC's
Windmill Road) with burnt-in timecode, the TC often begins at 10:00:00
rather than 00:00:00. An example is
https://www.youtube.com/watch?v=Yf3zE0x_dkI Is there a reason why they
chose to begin the TC at a "zero" of 10:00:00?


Early time code based video tape editors were unable to cue up a tape for
an edit at 00:00:00:00 because a 10 second pre-roll would need to be at
23:59:50:00 which the edit controller would see as 23+ hours into the
tape. In other words it couldn't handle a negative pre-roll, I don't know
if later ones could but of course now it's all non-linear editing they can
handle anything.


I notice on https://www.youtube.com/watch?v=aA8WFaVt0fo the timecode has a
letter after it which cycles through A, B, C, D, A, ... with every frame. I
presume this is to identify the 4-frame boundaries on which PAL VTRs must
make an edit if it is to be seamless. Is the rule that you must only make an
edit on one specific boundary (eg D - A) or can you edit on any boundary
you like, as long as you use a corresponding boundary at the other end of
the edit - eg you can edit on A - B if the other edit is also on A - B,
likewise for matched B - C or C - D edits, so the A, B, C, D sequence is
preserved.

Are there any restrictions for matching frames in modern editing of digital
recordings, or can you join together any two frames, in the same way as you
can with film? Certainly on an amateur basis, editing MPEG or H264 off-air
recordings, VideoRedo seems to be able to make seamless edits at any
boundary.

  #23  
Old October 15th 17, 10:35 AM posted to uk.tech.digital-tv
Roderick Stewart[_3_]
external usenet poster
 
Posts: 2,131
Default Timecode on programme-as-broadcast archive material

On Sat, 14 Oct 2017 22:52:39 +0100, "NY" wrote:

I thought there wasa time when film cameras had an "add on" video camera
so
that the director could see the shots. he could then tell the camera
operator to start rolling.

No it was so he could review the take and decide if they needed another.


Yes, "video assist" I think it was called. It really only made sense
at a time when the attainable picture quality on film was much better
than what was possible on video, and the film was intended for the
cinema and not broadcast. But those times are long gone.


I have seen video assist used extensively in filming of made-on-film TV
programmes such as Lewis.

[...]

I've seen this sort of thing too. It always struck me as utter madness
to use video to assist a film recording that would eventually become
video because it was for broadcast. If you're pointing a video camera
directly at the action anyway, why not just use that? The quality of
television cameras has been considered good enough for television
broadcasting since the beginning of television broadcasting, because
that's what they've been used for, and lately they've become a lot
smaller too. There seems no point in using any other medium as long as
the end product is intended for broadcast, and in broadcasting what
other purpose should there be?

I suspect a lot of directors preferred to use film because it was what
they were accustomed to, and were indulged by their bosses as long as
the practicalities and cost comparisons were not too prohibitive, but
these things are changing rapidly. The cost of just the film stock and
processing necessary for even a modest production can run into
thousands, while the smartphones in your pocket can probably record
hours of high quality video on memory cards the size of a fingernail.
It would make no more sense for a TV production to use film today than
it would for the crew to travel to work on horseback.

Rod.
  #24  
Old October 15th 17, 12:32 PM posted to uk.tech.digital-tv
NY
external usenet poster
 
Posts: 1,187
Default Timecode on programme-as-broadcast archive material

"Roderick Stewart" wrote in message
...
I've seen this sort of thing too. It always struck me as utter madness
to use video to assist a film recording that would eventually become
video because it was for broadcast. If you're pointing a video camera
directly at the action anyway, why not just use that? The quality of
television cameras has been considered good enough for television
broadcasting since the beginning of television broadcasting, because
that's what they've been used for, and lately they've become a lot
smaller too. There seems no point in using any other medium as long as
the end product is intended for broadcast, and in broadcasting what
other purpose should there be?

I suspect a lot of directors preferred to use film because it was what
they were accustomed to, and were indulged by their bosses as long as
the practicalities and cost comparisons were not too prohibitive, but
these things are changing rapidly. The cost of just the film stock and
processing necessary for even a modest production can run into
thousands, while the smartphones in your pocket can probably record
hours of high quality video on memory cards the size of a fingernail.
It would make no more sense for a TV production to use film today than
it would for the crew to travel to work on horseback.


In days gone by, video (especially outdoors) gave a very different look to
film that had been converted to video with a telecine. More vivid colours,
worse handling of overexposure (requiring more even lighting or reflectors).
I remember To Serve Them All My Days, which the BBC made in about 1980, and
some of the outdoor shots had lurid cyan or canary-yellow skies. I remember
when I was studying As You Like It for Eng Lit O level, so probably 1977-78,
the BBC made a lot of adaptations of Shakespeare plays, entirely on video,
and the reviewer in one of the newspapers referred to "the unsubtle
holiday-brochure colours" or some such disparaging phrase.

Things have evolved a *lot* since then, to the situation now where it's a
lot harder to distinguish between video and super 16 film: the lurid
colours, blooming and lag of tube video cameras went away when they switched
to solid state sensors, and the drab, grainy film went away with slightly
larger film frame and finer-grain film. Gamma of video cam be tweaked to
give a "look" that is more like film. And maybe the video will be shot as
25p rather than 50i, giving the different (more jerky!) motion of film.

I agree: if you are shooting for TV and the technology is good enough (eg
HD), it probably doesn't make sense to shoot on film, with the extra cost of
the film stock and processing. Lewis was one of the dramas that (AFAIK)
continued to use film right up to the end. I didn't notice clapperboards
being used, so presumably modern technology allows the film to be timecoded
to match the timecode recorded along with the sound (no doubt as a wav-type
file rather than on mag tape).

I wonder if any modern TV drama is still shot on film?

  #25  
Old October 15th 17, 12:58 PM posted to uk.tech.digital-tv
NY
external usenet poster
 
Posts: 1,187
Default Timecode on programme-as-broadcast archive material

"Roderick Stewart" wrote in message
...
On Sun, 15 Oct 2017 11:11:31 +0100, "NY" wrote:

I notice on https://www.youtube.com/watch?v=aA8WFaVt0fo the timecode has a
letter after it which cycles through A, B, C, D, A, ... with every frame.
I
presume this is to identify the 4-frame boundaries on which PAL VTRs must
make an edit if it is to be seamless. Is the rule that you must only make
an
edit on one specific boundary (eg D - A) or can you edit on any boundary
you like, as long as you use a corresponding boundary at the other end of
the edit - eg you can edit on A - B if the other edit is also on A - B,
likewise for matched B - C or C - D edits, so the A, B, C, D sequence is
preserved.


Fascinating. Though I worked in television engineering for many years
I never saw this. The nearest I ever saw was a dash that appeared on
the timecode for every alternate field, to signify the odd/even
sequence. I think you must be right about the purpose of the four
letters.

The need to preserve PAL frame sequence was really only considered for
what was called an "invisible edit", where the picture content, or the
majority of it, was identical between the shots being joined, such as
the same shot with and without a caption or effect of some sort.
Playback timebase correctors had to maintain continuous subcarrier
phase by adjusting the delay of the whole signal (because luminance
and chrominance did not follow separate paths as in domestic
recordings), and if the PAL sequence was incorrect, the whole image
would appear to jump sideways at the join by some multiple of a
quarter of a subcarrier cycle. When editing for action, it was
generally disregarded because a change between two shots with
different content would conceal the sideways jump.


I hadn't realised that the 4-frame rule could be broken in some situations.
Books about NTSC and PAL always make it sound as if it it mandatory to edit
on a 2-frame boundary (NTSC) or a 4-frame boundary (PAL). I'd assumed that
it was because of the continuity of the colour carrier, and that this is why
PAL was less forgiving because you had to match the alternating phase of the
CSC. I hadn't realised that there was also a lateral shift in the picture if
you didn't match. It sounds as if the lateral shift was more noticeable than
any CSC error, so as long as you can disguise that in a shot change, any
colour errors were less noticeable.

What caused the coloured blotches that you got when joining (insert or
assembly editing) two VHS recordings? Was that a loss of continuity of
colour sub carrier?

I did a night-school course in the 90s about making videos, and we shot on a
normal domestic camcorder, on VHS rather than Hi-8. The guy who ran the
course edited it for us, and he said that if he was doing it properly he'd
dub everything to U Matic and edit on that because results were better, even
with the extra stages of VHS-to-U Matic and then U Matic-to-VHS for our
copies. However because he was doing to for free rather than charging us
normal rates for the hire of his equipment, he used "pro" VHS (S-VHS?)
equipment rather than U Matic. It was weird to see for real how an insert
edit worked - being able to keep the sound track from a wide-shot take, and
drop in a mute close up reaction shot as the sound continued.

When I transferred my copy of the final results from VHS to MPEG, I could
see a very noticeable *vertical* shift and coloured blotches at most shot
changes. By the miracle of modern digital editing, I was able to take out
the two affected frames at each edit (except if there was any dialogue at
that point) and it removed the noticeable glitches.

  #26  
Old October 15th 17, 01:50 PM posted to uk.tech.digital-tv
Roderick Stewart[_3_]
external usenet poster
 
Posts: 2,131
Default Timecode on programme-as-broadcast archive material

On Sun, 15 Oct 2017 13:58:03 +0100, "NY" wrote:

I hadn't realised that the 4-frame rule could be broken in some situations.
Books about NTSC and PAL always make it sound as if it it mandatory to edit
on a 2-frame boundary (NTSC) or a 4-frame boundary (PAL). I'd assumed that
it was because of the continuity of the colour carrier, and that this is why
PAL was less forgiving because you had to match the alternating phase of the
CSC. I hadn't realised that there was also a lateral shift in the picture if
you didn't match. It sounds as if the lateral shift was more noticeable than
any CSC error, so as long as you can disguise that in a shot change, any
colour errors were less noticeable.


The subcarrier phase and colour burst sequence had to be maintained in
order to keep the signal in spec. They also had to have the proper
relationship to sync pulses, so if this relationship was ever broken
by joining shots with different PAL sequences, it was the subcarrier
phase that had to be maintained in the interests of stability. The
sync timing and burst sequence would then be wrong on the raw signal
from the tape, but new sync and burst would be added in the playback
processing circuitry. This would result in a completely kosher signal,
although the picture content would not have the same horizontal
relationship to the new sync pulses as to the original ones. Only in
those cases where the picture content was substantially the same
before and after the cut would this be noticeable, but mostly you'd be
editing to a different shot so for this it was disregarded.

What caused the coloured blotches that you got when joining (insert or
assembly editing) two VHS recordings? Was that a loss of continuity of
colour sub carrier?


I'm not sure, but alternate head sweeps on VHS overlapped slightly
with different azimuth angles, so it wouldn't be possible to make a
100% erasure of an individual field. If the flying erase head was wide
enough to include the overlap, then the next field would be slightly
noisier on account of the track being narrower, and if the flying
erase head only wiped its own material on its own track, there would
still be a remnant of it on the next one. That's without even
considering any possible misalignment of head sweeps with the recorded
tracks. Therefore you'd expect a bit of extra noise on every edit no
matter what you did, and only a system with completely isolated tracks
could handle this properly. VHS was such a departure from broadcast
standards I've always thought it a wonder it worked at all.

Rod.
  #27  
Old October 15th 17, 02:01 PM posted to uk.tech.digital-tv
Roderick Stewart[_3_]
external usenet poster
 
Posts: 2,131
Default Timecode on programme-as-broadcast archive material

On Sun, 15 Oct 2017 13:32:07 +0100, "NY" wrote:

Gamma of video cam be tweaked to
give a "look" that is more like film. And maybe the video will be shot as
25p rather than 50i, giving the different (more jerky!) motion of film.


It can - but why? I've never understood this philosophy. Surely the
aim of photography should be to give a "look" that looks like the
subject matter, not to imitate the artefacts of a different type of
photographic technology?

Different technologies work in different ways, but in attempting to
improve them, the only proper reference should be the original subject
matter, not an older technology, otherwise photography is chasing its
own tail instead of showing us a better representation of reality.

Rod.
  #28  
Old October 15th 17, 03:16 PM posted to uk.tech.digital-tv
NY
external usenet poster
 
Posts: 1,187
Default Timecode on programme-as-broadcast archive material

"Roderick Stewart" wrote in message
...
On Sun, 15 Oct 2017 13:32:07 +0100, "NY" wrote:

Gamma of video cam be tweaked to
give a "look" that is more like film. And maybe the video will be shot as
25p rather than 50i, giving the different (more jerky!) motion of film.


It can - but why? I've never understood this philosophy. Surely the
aim of photography should be to give a "look" that looks like the
subject matter, not to imitate the artefacts of a different type of
photographic technology?

Different technologies work in different ways, but in attempting to
improve them, the only proper reference should be the original subject
matter, not an older technology, otherwise photography is chasing its
own tail instead of showing us a better representation of reality.


Agreed. I remember when Casualty introduced a film look to the studio-based
shots of the casualty department. Everything then started to have that "this
is a studio set" look that a lot of filmed US drama had, whereas previously
it looked like an authentic shot-on-location casualty department. In other
words, they achieved (IMHO) the exact opposite of the situation they were
trying to create. The effect was so vilified that they removed it after a
few weeks, which suggested that it was applied very late in the production
process, not during recording or editing.

  #29  
Old October 15th 17, 03:30 PM posted to uk.tech.digital-tv
NY
external usenet poster
 
Posts: 1,187
Default Timecode on programme-as-broadcast archive material

"Roderick Stewart" wrote in message
...
What caused the coloured blotches that you got when joining (insert or
assembly editing) two VHS recordings? Was that a loss of continuity of
colour sub carrier?


I'm not sure, but alternate head sweeps on VHS overlapped slightly
with different azimuth angles, so it wouldn't be possible to make a
100% erasure of an individual field. If the flying erase head was wide
enough to include the overlap, then the next field would be slightly
noisier on account of the track being narrower, and if the flying
erase head only wiped its own material on its own track, there would
still be a remnant of it on the next one. That's without even
considering any possible misalignment of head sweeps with the recorded
tracks. Therefore you'd expect a bit of extra noise on every edit no
matter what you did, and only a system with completely isolated tracks
could handle this properly. VHS was such a departure from broadcast
standards I've always thought it a wonder it worked at all.


Yes when you look at what they did with the signal to get it to "fit" on the
tape, it's a wonder they got a usable signal back that TVs and another VHS
recorder (when dubbing) regarded as being good enough to treat as PAL. I
always wondered why when they brought in 1/2 speed LP and 1/3 speed EP, LP
actually gave worse picture quality (especially during fast forward/backward
shuttle and loss of colour on still) than EP.

Of the recordings that I made and have since tried to digitise to MPEG, it's
the LP ones which I have the greatest problem with as regards throbbing
colour and horizontal timing jitter. SP is pretty good and EP is more
blurred (lower horizontal resolution) but neither suffers as much from
throbbing and jitter.


Has anyone tried turning a VHS tape upside down? I had a tape that had been
severely mangled by an old VHS deck that had a habit of loading the tape and
then trying to eject the tape :-( Since that bit of the tape was ruined, I
chopped it out and re-spliced it to the transparent leader, but I decided
first of all to try an experiment: I wound the remaining tape onto the spool
upside down, still with the oxide on the same side, but so the head read the
end of each frame before the beginning.

OK so the colour was non-existent and the frame sync was a bit dodgy, but
you could just about make out... an upside-down picture :-)

 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT. The time now is 10:29 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.SEO by vBSEO 2.4.0
Copyright 2004-2017 Digital TV Banter.
The comments are property of their posters.