| Author |
Message |
Jim Easterbrook
Guest
|
 Posted:
Thu Oct 21, 2004 12:31 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In sci.engr.television.advanced John S. Dyson <toor@iquest.net> wrote:
| Quote: | In article <cl7shf$e2g$2@nntp0.reith.bbc.co.uk>,
Jim Easterbrook <jim.easterbrook@rd.bbc.co.uk> writes:
In sci.engr.television.advanced Doug McDonald <mcdonald@snpoam_scs.uiuc.edu> wrote:
The fact is, NTSC is completely decodeable at the
consumer level, while PAL is apparently only decodeable
only by a very few devices that only the BBC has. This
makes a tremendous difference on the quality of the
image seen on actual home TVs.
Let me get this straight. You claim, as fact, that consumer NTSC
equipment is completely free of cross colour and cross luminance, under
all circumstances, with all source material?
Please refer to your patent, and how it doesn't solve all of those
problems either...
|
I never claimed it did. However, you keep claiming that NTSC is
"completely decodeable".
| Quote: | Note also that your decoder is VERY dependent upon the double sideband
signal scheme, which does limit some of the opportunities.
|
Indeed, this might be a weakness. However, tests designed to probe this
didn't show up any problem. Note that both PAL and NTSC will suffer if
one of the chroma sidebands is attenuated.
| Quote: | Note that the decoder in my TV does about the best that I have seen:
no comet-tailing, no visible cross color (except on significant movement,
solvable by motion compensation), and no visible cross luma. Even when
viewing PAL recently in the UK (upconverted in the studio to the digital
transmission), I saw MUCH MUCH more cross color than I have recently
seen in the US. Geesh, and my decoder is just a cheap one (anything in
a consumer TV set has been aggressively cost-reduced) :-).
|
Thanks for describing what you have seen. Perhaps someone else will be
able to explain how a consumer level decoder can achieve such amazing
results.
--
Jim Easterbrook
BBC Research & Development <http://www.bbc.co.uk/rd/>
*** All opinions are mine and might not be shared by the BBC ***
|
|
| Back to top |
|
 |
Ian Mackenzie
Guest
|
| Posted:
Fri Oct 22, 2004 3:01 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
John Dyson wrote:
| Quote: | The bottom line: it cannot be 3D decoded in the home, and that
is sad, because it would have significantly improved the analog OTA image
quality. If the NTSC color encoding techniques and gamut were generally
used (instead of PAL encoding), you'd see little or no difference on
an A/B comparison, unless you would use a full 3D decoder (not likely
for PAL) and/or have full gamut NTSC phosphors.
|
It is not unlikely, many "Home" PAL TVs in Australia actually have a
true
comb 3d decoder and give the same results as NTSC.
My 29 inch Mitsubishi gives the same decoding capabilities as my
professional
adaptive comb filter 3D decoder used with standards converters and in
an AB test on the set show no difference between the inbuilt and the
external professional decoders unless I turn off the comb filter
functions in the menu. |
|
| Back to top |
|
|
John S. Dyson
Guest
|
| Posted:
Fri Oct 22, 2004 3:18 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In article <cl8a6g$itp$1@nntp0.reith.bbc.co.uk>,
Jim Easterbrook <jim.easterbrook@rd.bbc.co.uk> writes:
| Quote: | In sci.engr.television.advanced John S. Dyson <toor@iquest.net> wrote:
In article <cl7shf$e2g$2@nntp0.reith.bbc.co.uk>,
Jim Easterbrook <jim.easterbrook@rd.bbc.co.uk> writes:
In sci.engr.television.advanced Doug McDonald <mcdonald@snpoam_scs.uiuc.edu> wrote:
The fact is, NTSC is completely decodeable at the
consumer level, while PAL is apparently only decodeable
only by a very few devices that only the BBC has. This
makes a tremendous difference on the quality of the
image seen on actual home TVs.
Let me get this straight. You claim, as fact, that consumer NTSC
equipment is completely free of cross colour and cross luminance, under
all circumstances, with all source material?
Please refer to your patent, and how it doesn't solve all of those
problems either...
I never claimed it did. However, you keep claiming that NTSC is
"completely decodeable".
Which it is. The much touted transform decoder seems to be a way |
to mitigate the traditional problems with PAL decoding (equvalently,
NTSC decoding is best done using another method.) I need another day
or so of 'recovery' to be able to make sense and to hold a reasonable
conversation without being more rude than I normally am :-).
(I am incredibly exhausted, with an incredibly fragile sleeping pattern,
which has temporarily damaged my ability to think clearly -- while
my judgement about the quality of my analysis is also quite flawed. This
is embarassing... I am one of those people who is very willing to admit
when I am wrong, but on the basic facts I am not, some of my recent responses
are incredibly distorted. I have never been so logically inconsistent, while
not able to judge the inconsistency... In a way, part of my self awareness
is weak!!! The human brain/mind is quite an interesting bit of machine
and programming.)
I need another day or so of recovery from the timezone change and incredible
amount of sleep deficit.
John |
|
| Back to top |
|
|
Jim Easterbrook
Guest
|
| Posted:
Fri Oct 22, 2004 7:10 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In sci.engr.television.advanced John S. Dyson <toor@iquest.net> wrote:
| Quote: | In article <cl8a6g$itp$1@nntp0.reith.bbc.co.uk>,
Jim Easterbrook <jim.easterbrook@rd.bbc.co.uk> writes:
I never claimed it did. However, you keep claiming that NTSC is
"completely decodeable".
Which it is.
|
Then will someone please explain how. Not by giving model numbers of
decoders that are claimed to be perfect, or reporting how happy they are
with their own TV sets, but by quoting or pointing to a published paper
or patent that reveals how NTSC can be perfectly decoded.
Ideally such a publication should consider signals with difficult
characteristics. A couple of ideas off the top of my head:
1/ motion at critical speeds, such as 2/455 of a full line period per
frame period.
2/ A colour pattern of stripes such that I & Q are square wave signals
in quadrature. E.g. stripes of (+ve I, +ve Q), (+ve I, -ve Q), (-ve I,
-ve Q), (-ve I, +ve Q).
| Quote: | I need another day or so of recovery from the timezone change and incredible
amount of sleep deficit.
|
I sympathise. I hope you are feeling on top form again soon.
--
Jim Easterbrook
BBC Research & Development <http://www.bbc.co.uk/rd/>
*** All opinions are mine and might not be shared by the BBC *** |
|
| Back to top |
|
|
Doug McDonald
Guest
|
| Posted:
Fri Oct 22, 2004 9:59 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
Jim Easterbrook wrote:
| Quote: | In sci.engr.television.advanced Doug McDonald <mcdonald@snpoam_scs.uiuc.edu> wrote:
The fact is, NTSC is completely decodeable at the
consumer level, while PAL is apparently only decodeable
only by a very few devices that only the BBC has. This
makes a tremendous difference on the quality of the
image seen on actual home TVs.
Doug MCDonald
Let me get this straight. You claim, as fact, that consumer NTSC
equipment is completely free of cross colour and cross luminance, under
all circumstances, with all source material?
|
No, it is perfect only for static material. But it is
exceedingly good for all ordinary program material.
Doug McDonald |
|
| Back to top |
|
|
Doug McDonald
Guest
|
| Posted:
Fri Oct 22, 2004 10:02 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
Jim Easterbrook wrote:
| Quote: | Thanks for describing what you have seen. Perhaps someone else will be
able to explain how a consumer level decoder can achieve such amazing
results.
|
An interesting question indeed. I don't have any idea. All I actually
know is that the decoder in my TV set is better than the decoders
used at some of the TV stations I watch.
Doug McDonald |
|
| Back to top |
|
|
Doug McDonald
Guest
|
| Posted:
Fri Oct 22, 2004 10:07 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
Jim Easterbrook wrote:
| Quote: |
Then will someone please explain how. Not by giving model numbers of
decoders that are claimed to be perfect, or reporting how happy they are
with their own TV sets, but by quoting or pointing to a published paper
or patent that reveals how NTSC can be perfectly decoded.
Ideally such a publication should consider signals with difficult
characteristics. A couple of ideas off the top of my head:
1/ motion at critical speeds, such as 2/455 of a full line period per
frame period.
2/ A colour pattern of stripes such that I & Q are square wave signals
in quadrature. E.g. stripes of (+ve I, +ve Q), (+ve I, -ve Q), (-ve I,
-ve Q), (-ve I, +ve Q).
|
I really don't know, as I said. But on the motion test patterns
on the DVE test disk, the one in my set is essentially perfect in
removing color from static or moving luma bars. It is most emphatically
not perfect in removal of interlace artifacts. I don't remember whether
there are moving diagonal luma bars that could cause color banding
with imperfect decoding, however.
Doug McDonald |
|
| Back to top |
|
|
John S. Dyson
Guest
|
| Posted:
Fri Oct 22, 2004 11:03 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In article <clbi0f$uhb$2@news.ks.uiuc.edu>,
Doug McDonald <mcdonald@SnPoAM_scs.uiuc.edu> writes:
| Quote: | Jim Easterbrook wrote:
Thanks for describing what you have seen. Perhaps someone else will be
able to explain how a consumer level decoder can achieve such amazing
results.
An interesting question indeed. I don't have any idea. All I actually
know is that the decoder in my TV set is better than the decoders
used at some of the TV stations I watch.
Actually, that is my experience -- the worst 3D decoders seem to be |
used by the TV stations. There are differences in the requirements,
however. For example, a consumer level decoder can withstand a small
amount of movement smear, while a broadcast level decoder would not
allow that. (Concatenation effects, where multiple passes could
start looking very bad.)
John |
|
| Back to top |
|
|
John S. Dyson
Guest
|
| Posted:
Sat Oct 23, 2004 1:23 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In article <VA.000007fd.0008820a@escapetime.nospam.plus.com>,
Roderick Stewart <rjfs@escapetime.nospam.plus.com> writes:
| Quote: | In article <clc3jr$a2t$1@news.iquest.net>, John S. Dyson wrote:
An interesting question indeed. I don't have any idea. All I actually
know is that the decoder in my TV set is better than the decoders
used at some of the TV stations I watch.
Actually, that is my experience -- the worst 3D decoders seem to be
used by the TV stations. There are differences in the requirements,
however. For example, a consumer level decoder can withstand a small
amount of movement smear, while a broadcast level decoder would not
allow that. (Concatenation effects, where multiple passes could
start looking very bad.)
Broadcast decoders, if they are part of monitoring equipment, are designed
to *show* any faults in the signal, whereas domestic ones in TV sets are
designed to *conceal* them.
But the equipment that I am speaking of is the decoding used before |
encoding the digital transmission... In essense, this isn't monitoring
equipment, but the online equipment used for the real broadcast.
This 'online' equipment is what I had complained about on the upconverted
PAL broadcasts in the UK (the color flashing resulting from luma to
chroma interference.) The NTSC 3D decoders that I have seen from composite
to component conversion before MPEG encoding -- one example provides
nearly perfect (probably close to perfect in an absolute sense) seperation
of the chroma and luma, but also leaves significant interference from
chroma into the luma in moving (changing) video. I suspect that it (the
pattern effect of 3.58MHz into the luma) was
deemed to be almost invisible for most people, but I see the artifacting
during quick movement. A simple dynamic notch (dsp algorithm) that is
active only during changing content would have mitigated the pattern
effect entirely, without a visible effect against detail -- because the
detail would have been mostly obscured by movement anyway. In the
specific case that I cited, the decoder appeared
to be near perfect otherwise. Perhaps they just ran out of 'dsp'
CPU budget? :-). Of course, the more irritating color flashing effect
is effectively totally removed (even for movement.) It is actually
reasonable that they ran out of DSP CPU resources, or even that the
decoder was improperly set-up by TV station 'engineers' who didn't
understand the subtile effects of modifying various parameters.
Nowadays, there really is NO excuse for color flashing except in the
most extreme cases (probably never), or when nonstandard video is being
used (which is very bad anyway.)
John |
|
| Back to top |
|
|
Roderick Stewart
Guest
|
| Posted:
Sat Oct 23, 2004 3:12 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
In article <clc3jr$a2t$1@news.iquest.net>, John S. Dyson wrote:
| Quote: | An interesting question indeed. I don't have any idea. All I actually
know is that the decoder in my TV set is better than the decoders
used at some of the TV stations I watch.
Actually, that is my experience -- the worst 3D decoders seem to be
used by the TV stations. There are differences in the requirements,
however. For example, a consumer level decoder can withstand a small
amount of movement smear, while a broadcast level decoder would not
allow that. (Concatenation effects, where multiple passes could
start looking very bad.)
|
Broadcast decoders, if they are part of monitoring equipment, are designed
to *show* any faults in the signal, whereas domestic ones in TV sets are
designed to *conceal* them.
Rod. |
|
| Back to top |
|
|
Doug McDonald
Guest
|
| Posted:
Sat Oct 23, 2004 6:42 pm Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
Roderick Stewart wrote:
| Quote: | Broadcast decoders, if they are part of monitoring equipment, are designed
to *show* any faults in the signal, whereas domestic ones in TV sets are
designed to *conceal* them.
|
We are not talking MONITOR equipment here ... we are talking
actual broadcast chain NTSC decoders. The output of these goes
into an HDTV upconvertor and encoder, so the result is viewed
at home on an HDTV set ... a far more demanding application.
Doug McDonald |
|
| Back to top |
|
|
Paul Ratcliffe
Guest
|
| Posted:
Fri Oct 29, 2004 2:41 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
On Mon, 18 Oct 2004 00:14:28 +0000 (UTC), John S. Dyson <toor@iquest.net>
wrote:
| Quote: | All the crap about chromaticity is waffle designed to confuse the ignorant
reader.
If a 'very good green' isn't in your gamut, you cannot display it. The
'Green' in the NTSC gamut is certainly better. It seems like you are
talking beyond your competency. Just because there is a too-complex
idea for you, this doesn't mean that the idea is wrong.
|
You fucking arrogant little shit. Typical attitude of a lot of Yanks.
And you wonder why the world wants to blow you up?
| Quote: | Bandwidth affects how much detail you get in the colour information. It
does not affect what the colour is.
Well, actually it can affect the color in the transitions.
|
Well yes, obviously. But once it has settled it is irrelevant.
| Quote: | Along with
'bandwidth', actually the phase shift that occurs with most filter
designs can also cause hue problems. Eventually, the hue will settle
to as correct as it can be (assuming that the bandwidth is adequate at
|
Oh God. Get real, as you would say.
| Quote: | all, and you have an adequate gamut provided by your phosphors or filters,
the matrix is correct, etc.)
|
Yawn. |
|
| Back to top |
|
|
Paul Ratcliffe
Guest
|
| Posted:
Fri Oct 29, 2004 2:46 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
On Wed, 20 Oct 2004 11:29:31 -0500, Doug McDonald
<mcdonald@SnPoAM_scs.uiuc.edu> wrote:
| Quote: | Apparently the Europeans have been brainwashed. It's
apparently a congenital ability, bred into them
during the 20th century.
|
God, you're just as fucking arrogant and ignorant as Dyson.
| Quote: | They have had it repeated that PAL is superior
so many time ... for one and only onme obsolete reason ...
that they actually believe it.
|
Because it's true. Why would we not have implemented NTSC otherwise?
| Quote: | The fact is, NTSC is completely decodeable at the
consumer level, while PAL is apparently only decodeable
|
You never actually seem to define what this 'decodable' means.
My TV decodes PAL into R, G and B just fine. There will always be
cross-luminance and cross colour (assuming that's what you are
talking about) whatever the system. The actual hue of the colours
will NEVER NEVER EVER be wrong on PAL. It will be on NTSC. This is
the point. Do you get it?
| Quote: | only by a very few devices that only the BBC has. This
makes a tremendous difference on the quality of the
image seen on actual home TVs.
|
Crap. |
|
| Back to top |
|
|
Paul Ratcliffe
Guest
|
| Posted:
Fri Oct 29, 2004 2:51 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
On Mon, 18 Oct 2004 12:02:16 +0000 (UTC), John S. Dyson <toor@iquest.net>
wrote:
| Quote: | Of course -- you help by agreeing with my point -- the fact is that the PAL
system isn't helping at the point of failure...
|
Whatever that is. You never actually seem to stick to one point, but go
rambling off. What ARE you talking about?
| Quote: | PAL is tending to solve a problem that no longer effectively occurs
as often as the failure of studio set-up.
|
What do you mean by "the failure of the studio set-up"? It is meaningless
garbage. Why can't you actually describe what you are talking about?
| Quote: | Sure, it is possible for there to be phase errors in transmission
|
Bloody hell. Are you sure?
| Quote: | that appear through the NTSC system,
|
And cause hue errors in NTSC and not in PAL. Hmmm.
| Quote: | but that problem is NO LONGER as common as some PAL
advocates might think.
|
So how is a system that doesn't suffer at all from this phenomenon worse
than one that does?
| Quote: | complex. The bottom line: it cannot be 3D decoded in the home,
|
What does 3D decoded actually mean? You've never defined it? And what
difference does it make to the colour anyway? |
|
| Back to top |
|
|
Pete Fraser
Guest
|
| Posted:
Fri Oct 29, 2004 4:12 am Post subject:
Re: Origins of PAL: 1956 radio engeenering airticle from UK |
|
|
"Paul Ratcliffe" <abuse@orac12.clara34.co56.uk78> wrote in message
news:slrnco2tla.c5i.abuse@news.pr.network...
| Quote: | On Wed, 20 Oct 2004 11:29:31 -0500, Doug McDonald
mcdonald@SnPoAM_scs.uiuc.edu> wrote:
They have had it repeated that PAL is superior
so many time ... for one and only onme obsolete reason ...
that they actually believe it.
Because it's true. Why would we not have implemented NTSC otherwise?
|
It *was* true. Back when it was difficult to design a signal chain with low
diff phase.
Now it is easy, the one-time advantages of PAL do not apply, and NTSC
is the superior system.
| Quote: |
The fact is, NTSC is completely decodeable at the
consumer level, while PAL is apparently only decodeable
You never actually seem to define what this 'decodable' means.
My TV decodes PAL into R, G and B just fine. There will always be
cross-luminance and cross colour (assuming that's what you are
talking about) whatever the system. The actual hue of the colours
will NEVER NEVER EVER be wrong on PAL. It will be on NTSC. This is
the point. Do you get it?
|
The appearance of cross luminance on PAL is different. In my opinion
PAL has the advantage here. However, the 15 Hz cross color of
NTSC is much less objectionable than the mixed 6.25 / 18.75 Hz
crosscolor of PAL.
The overwhelming problem with PAL (well System I or G
or whatever) is the 50 Hz flicker. It's way worse than 60 Hz System
M flicker. Of course, this will go away with non-CRT displays and
frame-store processing, but at the moment NTSC is the better system.
| Quote: |
only by a very few devices that only the BBC has. This
makes a tremendous difference on the quality of the
image seen on actual home TVs.
Crap. |
|
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in