Digital TV: the picture really is horrible!
 

In article , Peter
Hayes scribeth thus
tony sayer wrote:

One part of the digital chain that needs improvement is the quality of
most receivers. Freeview quality can be excellent, especially
downconverted HD, but Judging by what I see in the likes of Curry's the
receivers the public are buying are complete rubbish.


I should give what you wrote there some more thought;!...

Freeview quality can be excellent, see the Reporting Scotland thread,
but if some content, "flashing lights, fast movement, detail and smooth
gradients on screen at the same time etc.", to quote the OP, fails, then
there's room for improvement.

That doesn't excuse manufacturers selling receivers that are, frankly,
junk.


Now where and why are the receivers junk?, when the real problem with
freeview as implemented is bit rates that are too low most all of the
time.. these are the cause of when you have outlined above....

When you consider of the rates used in studios and for transmission
distribution!...

Peter

--
Tony Sayer

Digital TV: the picture really is horrible!
 

In article .com, Mark
Carver scribeth thus
On Aug 30, 11:58 am, (Peter Hayes) wrote:

One part of the digital chain that needs improvement is the quality of
most receivers. Freeview quality can be excellent, especially
downconverted HD, but Judging by what I see in the likes of Curry's the
receivers the public are buying are complete rubbish.

I'm not sure. Yes, 20 quid DTT boxes lock up, get hot, have crappy
GUIs etc, but I've not seen any dramatic differences in the basic
picture quality, between those and 100+ quid boxes ? With MPEG the
clever bits are done by the encoder, not the decoder, so in short
quality improvements can only really be achieved (and they have been
over the last 10/15 years) by improving the encoders. The only element
where the picture quality can suffer in a DTT box is after the D-A
converter, and of course the interface to the display device.


Encoders can only do so much with so many bits..

...Or rather lack of so many bits!...
--
Tony Sayer

Digital TV: the picture really is horrible!
 

"tony sayer" wrote in message
...
In article , Peter
Hayes scribeth thus
tony sayer wrote:

Now where and why are the receivers junk?, when the real problem with
freeview as implemented is bit rates that are too low most all of the
time.. these are the cause of when you have outlined above....

When you consider of the rates used in studios and for transmission
distribution!...

Yes, how much variation would you expect between the picture from a good TV
and/or STB and a poor one, if they both get fed with the same aerial signal?
Does the quality of the decoder hardware and software have any effect?

Is a "good" decoder less prone to freezing and glitches due to local
interference such as fridge switching on/off?

I'd have thought that most of the variation would be in the analogue
electronics between the D-A converter and the tube: amount of ringing and
bandwidth (and edge-enhancement to try to compensate for loss of HF).

What is regarded as an acceptible bit rate to provide no visible blockiness
on fast-moving or gradually changing gradients such as smoke? I presume that
it can be lower than the bit rate used internally within the studio because
that has to allow for multi-generation copies without noticeable
degradation.

Digital TV: the picture really is horrible!
 

tony sayer writes:

Encoders can only do so much with so many bits..

..Or rather lack of so many bits!...

But if they reduced the number of channels on each MUX then the bitrate
for each channel could be higher, and therefore also the quality. Do we
really need so many +1 channels? If the number of repeats (especially of
programmes which were only shown days or weeks previously) were reduced
then not so many channels would be needed to show the same
material. Come analogue switch-off, if they were to keep the current
digital MUXes and convert the current analogue channels to digital then
not only could they increase the number of channels on DTT but could
increase the bitrate of all the channels.

Digital TV: the picture really is horrible!
 

On Aug 30, 2:37 pm, "Mortimer" wrote:

Is a "good" decoder less prone to freezing and glitches due to local
interference such as fridge switching on/off?

Some are better than others I think ?

I'd have thought that most of the variation would be in the analogue
electronics between the D-A converter and the tube: amount of ringing and
bandwidth (and edge-enhancement to try to compensate for loss of HF).

Yep, see my other post.

What is regarded as an acceptible bit rate to provide no visible blockiness
on fast-moving or gradually changing gradients such as smoke?

Well, a DVD's max bit rate is 9.99 Mb/s, but that's not really a fair
comparison as they are often mastered using multipass encoding, not
'on the fly' as used in DVB. Take a look at C5 *analogue* on any
transmitter outside the London region. That's fed to the transmitters
by an 8.3 Mb/s MPEG2 stream. MPEG4 and other new schemes reduce the
bit rate required for the same quality as MPEG2.

I presume that
it can be lower than the bit rate used internally within the studio because
that has to allow for multi-generation copies without noticeable
degradation.

Studio bit rates, (1.4 Gb/s for HD, 270 Mb/s for SD) are uncompressed.
Multigeneration is not really a problem on very mildly compressed tape
formats such as DigiBeta (155 Mb/s). Problems begin with some of the
50 Mb/s tape and 'storage' formats, where compression starts to
become significant.

Digital TV: the picture really is horrible!
 

On 30 Aug, 13:37, (Peter Hayes) wrote:
tony sayer wrote:
One part of the digital chain that needs improvement is the quality of
most receivers. Freeview quality can be excellent, especially
downconverted HD, but Judging by what I see in the likes of Curry's the
receivers the public are buying are complete rubbish.

I should give what you wrote there some more thought;!...

Freeview quality can be excellent, see the Reporting Scotland thread,

I haven't seen Reporting Scotland, either the old "bad" pictures, or
the new "good" ones, but I don't think talking heads in a news studio
are really testing of MPEG-2!

but if some content, "flashing lights, fast movement, detail and smooth
gradients on screen at the same time etc.", to quote the OP, fails, then
there's room for improvement.

That doesn't excuse manufacturers selling receivers that are, frankly,
junk.

"Compliant" MPEG decoders do not degrade the video signal (beyond
rounding errors).

If you watch on a 4:3 set in letterbox mode, there's some horrible
scaling going on, which can be good, bad, or indifferent (usually
bad). If you watch 4:3 in centre cut, or on a 16:9 TV, there isn't.

As long as the output is reasonable quality RGB (and 6MHz bandwidth,
45dB+ SNR, no screwy phase issues etc is hardy rocket science), it's
not a big issue.

What's important is the MPEG encoding, and the display.

Cheers,
David.

Digital TV: the picture really is horrible!
 

"Peter Hayes" wrote in message
news:1i3npvh.1syr85z1gf6rfhN%[email protected] com...
wrote:


Mind you, it should help push HD, if it stays at sensibly high
bitrates.

And comes down substantially in price.




it has - compare the price of a hd capable tv now to the price 2 years ago.



--
Gareth.

That fly... is your magic wand.
http://www.last.fm/user/dsbmusic/

Digital TV: the picture really is horrible!
 

"Mortimer" wrote in message
...


What is regarded as an acceptible bit rate to provide no visible
blockiness on fast-moving or gradually changing gradients such as smoke? I
presume that it can be lower than the bit rate used internally within the
studio because that has to allow for multi-generation copies without
noticeable degradation.


if freeview had a bitrate for each channel of up to 10mbit per second - as
dvd does, which is also mpeg2, it could look every bit as good as that.


--
Gareth.

That fly... is your magic wand.
http://www.last.fm/user/dsbmusic/

Digital TV: the picture really is horrible!
 

"the dog from that film you saw" wrote
in message ...

"Mortimer" wrote in message
...


What is regarded as an acceptible bit rate to provide no visible
blockiness on fast-moving or gradually changing gradients such as smoke?
I presume that it can be lower than the bit rate used internally within
the studio because that has to allow for multi-generation copies without
noticeable degradation.

if freeview had a bitrate for each channel of up to 10mbit per second - as
dvd does, which is also mpeg2, it could look every bit as good as that.

Mind you, I've seen some god-awful DVDs :-( I bought a boxed set of The
Sweeney and they've tried to squeeze four hours (four episodes) onto some
DVDs - with consequent blocky pictures on some of the more action-packed
episodes. The actual action scenes look as if they have been specially
processed to give them increased bit rate (according to the figures reported
by Power DVD) but some of the other scenes with a lot of movement are
atrocious - bit rates down to less than 2 Mbps and blocks the size of
footballs.

Digital TV: the picture really is horrible!
 

Graham Murray wrote:

tony sayer writes:

Encoders can only do so much with so many bits..

..Or rather lack of so many bits!...

But if they reduced the number of channels on each MUX then the bitrate
for each channel could be higher, and therefore also the quality. Do we
really need so many +1 channels?

We don't, the channel bean counters do. Each +1 is an additional revenue
stream at minimal cost.

If the number of repeats (especially of programmes which were only shown
days or weeks previously) were reduced then not so many channels would be
needed to show the same material.

The +1 channels should help fund better programmes, in theory... :-)

Come analogue switch-off, if they were to keep the current
digital MUXes and convert the current analogue channels to digital then
not only could they increase the number of channels on DTT but could
increase the bitrate of all the channels.

And add a few HD channels.

But will the cash strapped broadcast TV industry be able to compete for
this additional spectrum?

--

Immunity is better than innoculation.

Peter