Why a delay line? A delay line keeps the faster, luminance image inI'd forgotten about that reason for the delay line (which would be
step with the slower, chrominance image. The narrowband chrominance
signal requires more processing than the wideband luminance signal.
This processing takes a brief, but significant time. The delay line
assures that the two images start scanning across your screen at the
same time. Without the delay line, you might notice that the
monochrome picture starts before the color overlay does.
present for any system with a lower-bandwidth chroma signal - NTSC, PAL,
and presumably SECAM and any other such.
Seems odd that they didn't do that at the transmitter, on the basis that >complexity is better done there than making every set have it (like they
did with the gamma curve correction); maybe it wasn't realised it was >necessary until too much was in place, or maybe there was/is some
technical reason (fundamental physics?) why it couldn't be pre-done.
On 16/11/2021 21:29, J. P. Gilliver (John) wrote:
I'd forgotten about that reason for the delay line (which would beIn the analogue systems, the processing is done in real time, combining
present for any system with a lower-bandwidth chroma signal - NTSC, PAL,
and presumably SECAM and any other such.
the low bandwidth chrominance carriers with the high bandwidth luminance >carrier. Effectively, the green is a filtered luminance signal, and the
red and blue are generated by checking the phase differences between the
sub carrier and the chrominance information modulated on the luminance >signal.
PAL uses a delay line to match the phases of the subcarrier in each
line. Early sets used a glass block with am acoustic transmitter at one
end and a receiver at the other.
Seems odd that they didn't do that at the transmitter, on the basis that
complexity is better done there than making every set have it (like they
did with the gamma curve correction); maybe it wasn't realised it was
necessary until too much was in place, or maybe there was/is some
technical reason (fundamental physics?) why it couldn't be pre-done.
Digital broadcasting and video playback needs delays to allow for the >luminance, chrominance and audio information extraction from the digital >stream and processing, and as the time take for these is different on
each system, each playback device needs tailored delays. My favourite
media player on the phone and the PC, VLC, allows each of these delays
to be individually set by the user.
The delay line referred to was the luminance delay, about 600ns if I
remember correctly. It consisted of a length of what looked like
coaxial cable about 8" long, but with the inner conductor being wound
round a core of what I assume was a flexible ferrite material of some
sort. A metal foil cover provided a distributed capacitance to ground.
The low-pass filtering of the chrominance components inevitably
delayed them slightly and the luminance delay was to match this.
Delaying signals was a much bigger deal in the days when it had to be
done with analogue components, and then usually equalised afterwards
to make up for losses in the delay line.
Rod.
On Wed, 17 Nov 2021 10:20:17 +0000, John Williamson <johnwilliamson@btinternet.com> wrote:
On 16/11/2021 21:29, J. P. Gilliver (John) wrote:
I'd forgotten about that reason for the delay line (which would beIn the analogue systems, the processing is done in real time, combining
present for any system with a lower-bandwidth chroma signal - NTSC, PAL, >>> and presumably SECAM and any other such.
the low bandwidth chrominance carriers with the high bandwidth luminance >>carrier. Effectively, the green is a filtered luminance signal, and the
red and blue are generated by checking the phase differences between the >>sub carrier and the chrominance information modulated on the luminance >>signal.
PAL uses a delay line to match the phases of the subcarrier in each
line. Early sets used a glass block with am acoustic transmitter at one
end and a receiver at the other.
Seems odd that they didn't do that at the transmitter, on the basis that >>> complexity is better done there than making every set have it (like they >>> did with the gamma curve correction); maybe it wasn't realised it was
necessary until too much was in place, or maybe there was/is some
technical reason (fundamental physics?) why it couldn't be pre-done.
Digital broadcasting and video playback needs delays to allow for the >>luminance, chrominance and audio information extraction from the digital >>stream and processing, and as the time take for these is different on
each system, each playback device needs tailored delays. My favourite
media player on the phone and the PC, VLC, allows each of these delays
to be individually set by the user.
The delay line referred to was the luminance delay, about 600ns if I
remember correctly. It consisted of a length of what looked like
coaxial cable about 8" long, but with the inner conductor being wound
round a core of what I assume was a flexible ferrite material of some
sort. A metal foil cover provided a distributed capacitance to ground.
The low-pass filtering of the chrominance components inevitably
delayed them slightly and the luminance delay was to match this.
Delaying signals was a much bigger deal in the days when it had to be
done with analogue components, and then usually equalised afterwards
to make up for losses in the delay line.
Rod.
On Wed, 17 Nov 2021 at 19:21:04, Roderick Stewart <rjfs@escapetime.myzen.co.uk> wrote (my responses usually follow points raised):
[]
The delay line referred to was the luminance delay, about 600ns if I >>remember correctly. It consisted of a length of what looked like
coaxial cable about 8" long, but with the inner conductor being wound
round a core of what I assume was a flexible ferrite material of some
sort. A metal foil cover provided a distributed capacitance to ground.
The low-pass filtering of the chrominance components inevitably
delayed them slightly and the luminance delay was to match this.
Delaying signals was a much bigger deal in the days when it had to be
done with analogue components, and then usually equalised afterwards
to make up for losses in the delay line.
Rod.
So two reasons for delay lines in PAL sets - the fact that the lower-bandwidth chroma signal had a different processing delay than the higher-bandwidth luminance signal, and these had to be brought back into
line - presumably the case for any system (NTSC, PAL, SECAM) with a low-bandwidth chroma signal; and the one specific to PAL, that combined
the chroma signal from successive lines, to avoid "Hannover bars", where
the alternating colour errors from line pairs would otherwise combine
across odd and even fields. (PAL-S [simple was it?] versus PAL-D, IIRR,
for sets that didn't and did have that.)
[My spell-checker suggests "Hangover"!]
--
J. P. Gilliver. UMRA: 1960/<1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf
"If you have ten thousand regulations you destroy all respect for the
law." - Winston Churchill.
| Sysop: | Keyop |
|---|---|
| Location: | Huddersfield, West Yorkshire, UK |
| Users: | 295 |
| Nodes: | 16 (2 / 14) |
| Uptime: | 09:00:55 |
| Calls: | 6,642 |
| Calls today: | 2 |
| Files: | 12,190 |
| Messages: | 5,326,324 |