Discussion:
chroma offset correction?
(too old to reply)
J. P. Gilliver
2023-06-16 13:09:10 UTC
Permalink
I've just watched a clip
-
no, it wasn't to my taste) where clearly there is a significant time
offset between chroma and luma. It occurred to me that, when nowadays if
the only extant copy is a digitised version (the original analogue
having been lost, or its source unknown), it'd be quite an interesting
challenge to correct it: I don't think there's anything in the digital
domain that's analogous to the luma/chroma split, so presumably the
chroma and luma would have to be regenerated, then slewed back into
approximate alignment.

(This one is slightly unusual in that the chroma appears to _lead_ the
luma rather than the usual lag.)
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf

interracial marriage was still illegal in 17 states in 1967.
Brian Gaff
2023-06-17 09:28:23 UTC
Permalink
I know what you mean. Some VCRs tended to smear the chroma and pulsate it as
well.
Brian
--
--:
This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...
***@blueyonder.co.uk
Blind user, so no pictures please
Note this Signature is meaningless.!
Post by J. P. Gilliver
I've just watched a clip http://youtu.be/_P09xG1clcg -
no, it wasn't to my taste) where clearly there is a significant time
offset between chroma and luma. It occurred to me that, when nowadays if
the only extant copy is a digitised version (the original analogue having
been lost, or its source unknown), it'd be quite an interesting challenge
to correct it: I don't think there's anything in the digital domain that's
analogous to the luma/chroma split, so presumably the chroma and luma
would have to be regenerated, then slewed back into approximate alignment.
(This one is slightly unusual in that the chroma appears to _lead_ the
luma rather than the usual lag.)
--
interracial marriage was still illegal in 17 states in 1967.
J. P. Gilliver
2023-06-17 11:16:05 UTC
Permalink
Post by Brian Gaff
I know what you mean. Some VCRs tended to smear the chroma and pulsate it as
well.
Brian
I don't remember seeing a pulsating one! Though a multigenerational copy
could get horrendous pretty quickly. Chroma _lag_ was common - I think
most memorably on snooker; one got used to seeing a grey ball, with a
ghostly coloured one to its right, especially in a long shot (where a
fixed chroma offset was of course more noticeable).

My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).

I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf

Madness takes its toll. Please have exact change
[via Penny Mayes (***@pmail.net)]
NY
2023-06-17 13:22:32 UTC
Permalink
Post by J. P. Gilliver
Post by Brian Gaff
I know what you mean. Some VCRs tended to smear the chroma and pulsate it as
well.
Brian
I don't remember seeing a pulsating one! Though a multigenerational copy
could get horrendous pretty quickly. Chroma _lag_ was common - I think
most memorably on snooker; one got used to seeing a grey ball, with a
ghostly coloured one to its right, especially in a long shot (where a
fixed chroma offset was of course more noticeable).
My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).
I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
Digital colour photographs and videos are still stored as luminance and some
form of chrominance difference. And often the colour resolution is half the
luminance resolution in one or both directions, as for analogue video.

It ought to be fairly easy to write a program that shifts the chroma pixels
sideways a bit compared with the luminance.


If you take a still photo (eg a JPEG or PNG) and make a black and white
copy, you can subtract this from the original photo (most image-editing
software such as Paint Shop Pro or Photoshop can do this), you can see how
blurred the remaining colour info is after you've subtracted the luminance.

Here's an example: a colour photo of the famous clock on the city wall in
Chester, then a black and white version of it and then colour minus BW
(adding on 128 to each pixel to avoid "pixels with a negative value"!)
leaving just the chroma.

Loading Image...
Loading Image...
Loading Image...
J. P. Gilliver
2023-06-17 22:34:46 UTC
Permalink
[]
Post by NY
Post by J. P. Gilliver
My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).
I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
Digital colour photographs and videos are still stored as luminance and
some form of chrominance difference. And often the colour resolution is
half the luminance resolution in one or both directions, as for
analogue video.
I didn't realise that was still the case.
Post by NY
It ought to be fairly easy to write a program that shifts the chroma
pixels sideways a bit compared with the luminance.
If you take a still photo (eg a JPEG or PNG) and make a black and white
copy, you can subtract this from the original photo (most image-editing
software such as Paint Shop Pro or Photoshop can do this), you can see
how blurred the remaining colour info is after you've subtracted the
luminance.
Here's an example: a colour photo of the famous clock on the city wall
in Chester, then a black and white version of it and then colour minus
BW (adding on 128 to each pixel to avoid "pixels with a negative
value"!) leaving just the chroma.
https://i.postimg.cc/7Y1Q6h1C/Chester-Colour.jpg
https://i.postimg.cc/1XhJR1fc/Chester-BW.jpg
https://i.postimg.cc/gjV4nth9/Chester-Colour-minus-BW.png
That last one looks a lot sharper than what I used to see on a colour TV
when viewing only the colour information (I forget how - I think just
turning the contrast right down did it). That gave you no sharp edges at
all - just very smeary.
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf

Only dirty people need wash
John Williamson
2023-06-18 07:35:48 UTC
Permalink
Post by J. P. Gilliver
[]
Post by NY
Post by J. P. Gilliver
My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).
I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
Digital colour photographs and videos are still stored as luminance
and some form of chrominance difference. And often the colour
resolution is half the luminance resolution in one or both directions,
as for analogue video.
I didn't realise that was still the case.
Only true for compressed formats and some cameras.

I use a Fuji camera and RAW format to store images, and that file is,
basically, a dump of the data off the sensor, so the chrominance
definition depends on the layout of the filters on the sensor. Not all
of them use an RGBRGB pattern.

The normal resolution in compressed files is 4:2:2 (Luminance, and two
colour difference bytes.)
Post by J. P. Gilliver
Post by NY
Here's an example: a colour photo of the famous clock on the city wall
in Chester, then a black and white version of it and then colour minus
BW (adding on 128 to each pixel to avoid "pixels with a negative
value"!) leaving just the chroma.
https://i.postimg.cc/7Y1Q6h1C/Chester-Colour.jpg
https://i.postimg.cc/1XhJR1fc/Chester-BW.jpg
https://i.postimg.cc/gjV4nth9/Chester-Colour-minus-BW.png
That last one looks a lot sharper than what I used to see on a colour TV
when viewing only the colour information (I forget how - I think just
turning the contrast right down did it). That gave you no sharp edges at
all - just very smeary.
It was even worse when you were using VHS or Betamax rather than
broadcast signals, as the colour bandwidth was very low. The VHS colour
subcarrier had a frequency of 629 kHz.
--
Tciao for Now!

John.
Brian Gaff
2023-06-18 10:24:35 UTC
Permalink
Yes the colour definition was much less than the b/we and the brain
compensated for this smear to some extent.
Somebody showed me a Umatic recorder when used in playback, not only did
it handle multi standard tv signals and colour systems, but had a kind of
preset knob that moved the colour delay about as well. I imagine there were
quite a few machines or boxes like this to make tapes that came from all
over the globe look presentable for broadcast. I do remember tithe old
colour Andy Williams shows that had issues with yellow or green faces or a
kind of judder in the frame rate when fast motion was on the screen, due one
supposes to primitive standards conversion not being able to keep up.
I of couse cannot see the current state of the art, but I understand that
when these oldies TV music video stations play, say for instance the video
of Village People YMCA, you can clearly notice the judder and an effect on
mouth movements and teeth almost look like a cheap cartoon effect, where the
white of the teeth are seen but the actual mouth movement is not!

Brian
--
--:
This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...
***@blueyonder.co.uk
Blind user, so no pictures please
Note this Signature is meaningless.!
Post by NY
Post by J. P. Gilliver
Post by Brian Gaff
I know what you mean. Some VCRs tended to smear the chroma and pulsate it as
well.
Brian
I don't remember seeing a pulsating one! Though a multigenerational copy
could get horrendous pretty quickly. Chroma _lag_ was common - I think
most memorably on snooker; one got used to seeing a grey ball, with a
ghostly coloured one to its right, especially in a long shot (where a
fixed chroma offset was of course more noticeable).
My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).
I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
Digital colour photographs and videos are still stored as luminance and
some form of chrominance difference. And often the colour resolution is
half the luminance resolution in one or both directions, as for analogue
video.
It ought to be fairly easy to write a program that shifts the chroma
pixels sideways a bit compared with the luminance.
If you take a still photo (eg a JPEG or PNG) and make a black and white
copy, you can subtract this from the original photo (most image-editing
software such as Paint Shop Pro or Photoshop can do this), you can see how
blurred the remaining colour info is after you've subtracted the luminance.
Here's an example: a colour photo of the famous clock on the city wall in
Chester, then a black and white version of it and then colour minus BW
(adding on 128 to each pixel to avoid "pixels with a negative value"!)
leaving just the chroma.
https://i.postimg.cc/7Y1Q6h1C/Chester-Colour.jpg
https://i.postimg.cc/1XhJR1fc/Chester-BW.jpg
https://i.postimg.cc/gjV4nth9/Chester-Colour-minus-BW.png
J. P. Gilliver
2023-06-18 12:41:34 UTC
Permalink
Post by Brian Gaff
Yes the colour definition was much less than the b/we and the brain
compensated for this smear to some extent.
[]
The brain's higher luminance than chrominance resolution (or it might be
the eye's - rods and cones) has been known from well before television;
if you look closely at old black-outline prints (I mean from printing
blocks, not photographic), that have had colour added, often
watercolour, you'll see that the colours can be quite sloppily applied,
and the brain sorts them out. (My mum dabbled at colouring such old
prints, and was quite surprised how careful you _didn't_ have to be.)
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf

"quidquid latine dictum sit, altum viditur". ("Anything is more impressive if
you say it in Latin")
Brian Gaff
2023-06-18 10:14:25 UTC
Permalink
I seem to recall the pulsating colour was mainly on the old Philips vcr
systems, of any of the speeds. There were on some colours a static
vertical striping in saturation effect. Seen moor on highly saturated
colours. It looked odd as it was static so when the camera panned it stayed
in the same place.
Some video 8 recordings had the leading colour effect, but as the format
never really caught on a body bothered. Video 2000 of course was known for
its video noise effects.
Vhs standard was crap definition and betamax was good if you used basf or
Fuji tapes on it.
I actually miss video recorders, despite their many moving parts, they
worked surprisingly well.
Brian
--
--:
This newsgroup posting comes to you directly from...
The Sofa of Brian Gaff...
***@blueyonder.co.uk
Blind user, so no pictures please
Note this Signature is meaningless.!
Post by J. P. Gilliver
Post by Brian Gaff
I know what you mean. Some VCRs tended to smear the chroma and pulsate it as
well.
Brian
I don't remember seeing a pulsating one! Though a multigenerational copy
could get horrendous pretty quickly. Chroma _lag_ was common - I think
most memorably on snooker; one got used to seeing a grey ball, with a
ghostly coloured one to its right, especially in a long shot (where a
fixed chroma offset was of course more noticeable).
My post was really just thinking - if all we have is a digitised copy,
it might be an interesting task how to correct such material, whether
it's chroma lag or lead (presumably both of similar difficulty).
I think the days of retrieving e. g. lost Dr. Who episodes from private
copies were mostly still in the analogue days, or at least where the
machinery in which the two signals were present was still in use, even
if the actual re-syncing was done digitally. (I think the most
fascinating aspect of that work was when someone realised that
black-and-white film copies might still contain the colour subcarrier.)
--
Madness takes its toll. Please have exact change
J. P. Gilliver
2023-06-18 12:52:20 UTC
Permalink
Post by Brian Gaff
I seem to recall the pulsating colour was mainly on the old Philips vcr
systems, of any of the speeds. There were on some colours a static
vertical striping in saturation effect. Seen moor on highly saturated
colours. It looked odd as it was static so when the camera panned it stayed
in the same place.
Some video 8 recordings had the leading colour effect, but as the format
never really caught on a body bothered. Video 2000 of course was known for
its video noise effects.
Can't say I was aware of it: I do remember V2000, with its dynamic
following, was much _better_ at all sorts of things - including colour -
than the other two (especially VHS). I remember when one of mine was
newish, going from fast rewind play to fast forward play (or it might
have been the opposite): it slowed down, passing through still, then
speeded up in the other direction, without any tearing or noise bars -
IIRR, it didn't even lose colour. (IIRR VHS always had noise bars at
anything other than ×1 forward speed, and _possibly_ still frame on
later machines.)
Post by Brian Gaff
Vhs standard was crap definition and betamax was good if you used basf or
Fuji tapes on it.
I remember reading somewhere that Betamax and V2000 were (for PAL,
anyway) 3 MHz bandwidth, and VHS 2½ (that must have been luminance only,
of course). I had an old Philips reel-to-reel that I think was 2½ with
proper tape (worked with computer tape if you backed off the video gain
- I think that reduced the bandwidth; but that tape wore the heads).
[I've still got a couple of them, but they aren't in working order.]
(They weren't colour.)
Post by Brian Gaff
I actually miss video recorders, despite their many moving parts, they
worked surprisingly well.
Brian
I often thought the manufacturers missed a chance by not releasing one
or two models with transparent housings so you could see them working;
OK, most people probably _wouldn't_ want to, but I'd have thought there
would be enough of us that would to justify one or two models. After
all, people love steam trains, and I'm sure that's part of the
attraction.
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)***@T+H+Sh0!:`)DNAf

Philosophy is questions that may never be answered. Religion is answers that
may never be questioned.
NY
2023-06-18 19:52:19 UTC
Permalink
Post by J. P. Gilliver
I often thought the manufacturers missed a chance by not releasing one
or two models with transparent housings so you could see them working;
OK, most people probably _wouldn't_ want to, but I'd have thought there
would be enough of us that would to justify one or two models. After
all, people love steam trains, and I'm sure that's part of the attraction.
Yes it would have been good to see the complicated tape-threading
process that takes place when a cassette is inserted and the unthreading
when it is ejected.

Maybe even illuminate the video head with an LED running at exactly 50
Hz, and see the head speeding up and slowing down as it tries to get the
heads in phase with the tape tracks so the heads hit the centre of the
track.

The very first VHS machine that I used, a Ferguson like this
Loading Image... didn't lace
the tape when the cassette was loaded. Instead it left the tape unlaced
in case you wanted to REW or FF, and then laced when you pressed play -
so there was a long pause before playing began. Then if you wanted to
wind a bit further and resume playing, there was an unlace pause and
then (after you'd finished winding) a lace-up pause. I suppose it caused
a bit less tape wear, but the fact that every VHS machine I've ever see
since then threaded as soon as the tape was loaded suggests that it
can't have been a major problem.

VHS was surprisingly rugged. Picture quality varied a lot with tape
speed: SP was the best, LP (SP/2) seemed subjectively to be the worst,
and EP (SP/3) seemed to be somewhere in between. My Panasonic VCRs could
do a stable, colour freeze-frame on SP and EP, but it was monochrome on
LP and often I couldn't get the picture stable despite going from one
extreme of tracking adjustment to the other.

I have "fond" memories of a tape jamming (probably tape=up spool jammed)
and the tape looped all over the inside of the tape-transport area
before I managed to press STOP. It was fun extricating the tape and
winding the undamaged section back into the cassette. As the damaged
section was very close to the end of the tape, I just cut it out and
spliced the clear leader onto the end so it still operated the
end-of-tape sensor. Sadly the tape had fouled the video heads so the
picture was very noisy and the hifi sound didn't work, even with a good
cassette. So I bit the bullet and dabbed the all heads on the drum with
a cotton-wool bud soaked in IPA (no, not the beer!) to remove any
clogged oxide. Interestingly, I got the pictures to play without any
noise without too much effort, but it took a long time before I got hifi
sound to work again.

When you think about the compromises that were made to fit the luminance
and chroma onto the tape (chroma bandwidth was only about 0.5 MHz) it
was amazing that results with VHS were not worse.

I always wondered how the later VHS decks were able to maintain a tape
library, so that when you inserted a given tape, the deck would display
all the programmes that were recorded on that tape. I realise that the
list was maintained, per tape, in non-volatile memory on the deck, and
was not stored on the tape itself. But how did the deck manage to
distinguish between one tape and another as soon as the tape was
inserted, so as to present the correct tape-contents list? Did all VHS
tapes contain a unique RFID tag, or was there some other way that a
unique tape ID was recorded all the way along the control track?
Adrian Caspersz
2023-07-22 07:44:59 UTC
Permalink
On 6/18/23 20:52, NY wrote:

<snip>
Post by NY
I always wondered how the later VHS decks were able to maintain a tape
library, so that when you inserted a given tape, the deck would display
all the programmes that were recorded on that tape. I realise that the
list was maintained, per tape, in non-volatile memory on the deck, and
was not stored on the tape itself. But how did the deck manage to
distinguish between one tape and another as soon as the tape was
inserted, so as to present the correct tape-contents list? Did all VHS
tapes contain a unique RFID tag, or was there some other way that a
unique tape ID was recorded all the way along the control track?
Dunno. First I heard of it :)

Panasonic NV-FJ760 Tape Library


https://www.manualslib.com/manual/117527/Panasonic-Nv-Fj710-Series.html?page=26

Control track apparently.
https://en.wikipedia.org/wiki/Control_track
--
Adrian C
NY
2023-07-22 20:08:03 UTC
Permalink
Post by Adrian Caspersz
<snip>
Post by NY
I always wondered how the later VHS decks were able to maintain a tape
library, so that when you inserted a given tape, the deck would display
all the programmes that were recorded on that tape. I realise that the
list was maintained, per tape, in non-volatile memory on the deck, and
was not stored on the tape itself. But how did the deck manage to
distinguish between one tape and another as soon as the tape was
inserted, so as to present the correct tape-contents list? Did all VHS
tapes contain a unique RFID tag, or was there some other way that a
unique tape ID was recorded all the way along the control track?
Dunno. First I heard of it :)
Panasonic NV-FJ760 Tape Library
http://youtu.be/TNgaUd2CIfY
https://www.manualslib.com/manual/117527/Panasonic-Nv-Fj710-Series.html?page=26
Control track apparently.
https://en.wikipedia.org/wiki/Control_track
Yes that makes sense. If you put in a brand new tape it presumably allocates
a tape ID based on time/date (so it is guaranteed to be unique) and then
uses that ID for the whole tape.

As long as you insert a half used (partially virgin) the tape at a point
where there is a recording, it will pick up the ID and use for subsequent
recordings.

The only problem would come if you inserted the tape at an unrecorded point
because the VCR would not pick up the existing ID and so subsequent
recordings would get a new ID rather than continuing the earlier one. I
never tried that. The difficulty now is finding a VHS tape anywhere that has
a virgin section to test that "split personality" tape logical flaw. I
imagine every tape I possess has been recorded end-to-end so doesn't have a
long enough virgin section to see what would happen.

I never knew the control track contained so much information: I imagined it
to be a dumb "electronic sprocket hole", with maybe a blip to denote the
start of each new recording, but not to have space to accommodate a few
bytes for an ID.


On VCRs from (I'm guessing) the mid/late 80s onwards, the control track also
operated the real-time counter, with pulses still being readable even when
the tape was rewound or fast-forwarded at high speed. That made *such* a
difference over the arbitrary 4-digit mechanical "mileometer" counters which
were non-linear (a difference of 10 units meant different things at
different ends of the tape!) and was not at all reproducible. When I was in
the sixth form at school in 1980-1, my prefect duty was in the AV room. We
had a couple of these beasts
Loading Image... and the counters
were terrible. We got into the habit of setting the counter to zero at the
beginning of the tape and then noting the position on the tape index info on
the cassette box for each subsequent recording. That meant that a teacher
who wanted to use a given programme in a lesson would wind the tape to the
beginning (if it hadn't been left in that state), zero the counter and then
wind to position 1275 (for example) and know he was at the start of
Tomorrow's World which was the third programme on the tape and started at
the counter number. Fine, except that the counter could be several minutes
out either way if you repeated the operation several times :-(

But it got you in the right locality: you hoped that if you didn't hit the
exact changeover, you could at least identify whether you were near the end
of the previous programme or a few minutes after the beginning of the
programme you wanted. We tended to deliberately *not* record all episodes of
a series consecutively on the same tape, but to deliberately make dissimilar
programmes follow each other: this had the disadvantage that various
episodes were dotted over a variety of tapes, but it made it easier to
distinguish between "end of wrong programme" and "start of desired
programme" ;-)

Of course a wise teacher came up and prepared the tape and got it to exactly
the right place, before the lesson, rather than doing it on the fly with a
tape that he'd never seen before.

The VCRs that we had had one thing that the one in the photo doesn't appear
to have: a *wired* remote control which allowed the tape to be played at
various speeds: something like freeze frame, 1/15 .. normal speed, 2x speed.
The 2x speed still played with sound (IIRC) though it sounded like Pinky and
Perky. Just about intelligible, even if it made the boys crack up with
laughter.

Nowadays with software players like VLC you can play at any speed you like,
with sound that preserves the correct pitch, so speech is intelligible over
a fairly wide range and even music sounds OK from about 0.75x to 1.25x. I
presume it takes short bursts of consecutive sound samples which it plays at
the correct speed (to maintain pitch) but ditches a variable number of
samples in between depending on playback speed. Technology which we'd have
killed for in the early 1980s. ;-)

One thing about those early VCRs: they unlaced the tape every time you went
from playback to rew/ff, so there was always a few seconds unlacing delay at
the start of the fast movement and a few seconds re-lacing delay at the end.
Subsequent VCRs (maybe all those which used the control track for the
counter) kept the tape permanently laced both for play and for rew/ff so
there was only a delay when you first inserted the tape or ejected it.

The top-loader mechanism on those early VCRs was a PITA. You had to be
*very* careful to press down very gently and evenly on both sides of the lid
as you inserted it, otherwise the lid mechanism would jam. I have very fond
(blush) memories of demonstrating this (using that sort of wording) to a
stunningly attractive female teacher who was probably only a few years older
than me, and her responding "Oooooooo, Matronnnnnnnnnn!" in her best Kenneth
Williams voice, which cracked us both up so I couldn't continue for several
minutes. Happy times...

Graham.
2023-07-21 16:32:34 UTC
Permalink
On Fri, 16 Jun 2023 14:09:10 +0100, "J. P. Gilliver"
Post by J. P. Gilliver
I've just watched a clip http://youtu.be/_P09xG1clcg -
no, it wasn't to my taste) where clearly there is a significant time
offset between chroma and luma. It occurred to me that, when nowadays if
the only extant copy is a digitised version (the original analogue
having been lost, or its source unknown), it'd be quite an interesting
challenge to correct it: I don't think there's anything in the digital
domain that's analogous to the luma/chroma split, so presumably the
chroma and luma would have to be regenerated, then slewed back into
approximate alignment.
(This one is slightly unusual in that the chroma appears to _lead_ the
luma rather than the usual lag.)
We SECAM 'ere,
We SECAM there,
We SECAM Frenchies everywhere.
--
Graham.
%Profound_observation%
Loading...