BBC2 lousy DTT compression ATM
 

"just me" wrote in message
...
OK, so it's never terrific, but at this precise moment it's awful! Loads
of
fizzy noisy artefacts and blocky effects.
TX is Winter Hill, programme is TOTP2.

Anyone else having trouble?


BBC4 today when they reran to documentary on the eruption of Mt Vesuvius.
One inch square blocs filled the entire screen (28") during the smoke
scenes. This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke. There is no reason why the encoder
could not have interpolated the image to smoothen it out since there was
plenty of available bandwidth.

On 16/2/04 10:29 pm, in article ,
"Agamemnon" wrote:


"just me" wrote in message
...
OK, so it's never terrific, but at this precise moment it's awful! Loads
of
fizzy noisy artefacts and blocky effects.
TX is Winter Hill, programme is TOTP2.

Anyone else having trouble?


BBC4 today when they reran to documentary on the eruption of Mt Vesuvius.
One inch square blocs filled the entire screen (28") during the smoke
scenes. This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke. There is no reason why the encoder
could not have interpolated the image to smoothen it out since there was
plenty of available bandwidth.


Err - smoke and rippling water are VERY difficult for MPEG2 to encode - as
there is lots of near-random motion, coupled with quite a lot of luminance
variation. This makes motion-estimation and tracking very difficult - and
thus reduces the temporal redundancy of the GOPs quite a lot. IME smoke and
water seldom compress well - whilst it may seem slow moving - it contains
quite a lot of "change" ...

Steve

On 16/2/04 10:29 pm, in article ,
"Agamemnon" wrote:


"just me" wrote in message
...
OK, so it's never terrific, but at this precise moment it's awful! Loads
of
fizzy noisy artefacts and blocky effects.
TX is Winter Hill, programme is TOTP2.

Anyone else having trouble?


BBC4 today when they reran to documentary on the eruption of Mt Vesuvius.
One inch square blocs filled the entire screen (28") during the smoke
scenes. This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke. There is no reason why the encoder
could not have interpolated the image to smoothen it out since there was
plenty of available bandwidth.


Err - smoke and rippling water are VERY difficult for MPEG2 to encode - as
there is lots of near-random motion, coupled with quite a lot of luminance
variation. This makes motion-estimation and tracking very difficult - and
thus reduces the temporal redundancy of the GOPs quite a lot. IME smoke and
water seldom compress well - whilst it may seem slow moving - it contains
quite a lot of "change" ...

Steve

"Agamemnon" wrote in message
...
This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke.

Wow, what a quick diagnosis of all our digital TV problems! It all seems so
difficult, but he shows it's all really easy!

Agamemnon seems to have inside knowledge of all the kit used at the BBC. For
example, that the MPEG-2 coders that it uses are "15 years out of date".
Would you care to give the make and type numbers then? I'd like to know more
about this "defect" now. (Note that for the coders to be "15 years out of
date", they have to be at least 15 years old, and I seem to remember that
the MPEG-2 standard wasn't even finalised until 1993 or thereabouts - just
over 10 years ago!).

Others have already pointed out that smoke is difficult to code, not easy.

It's such a shame Agamemnon has not yet been hired by broadcasters to sort
out all the problems they've ever head or ever will have, both with radio
and television.

"Agamemnon" wrote in message
...
This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke.

Wow, what a quick diagnosis of all our digital TV problems! It all seems so
difficult, but he shows it's all really easy!

Agamemnon seems to have inside knowledge of all the kit used at the BBC. For
example, that the MPEG-2 coders that it uses are "15 years out of date".
Would you care to give the make and type numbers then? I'd like to know more
about this "defect" now. (Note that for the coders to be "15 years out of
date", they have to be at least 15 years old, and I seem to remember that
the MPEG-2 standard wasn't even finalised until 1993 or thereabouts - just
over 10 years ago!).

Others have already pointed out that smoke is difficult to code, not easy.

It's such a shame Agamemnon has not yet been hired by broadcasters to sort
out all the problems they've ever head or ever will have, both with radio
and television.

Can I add:

Fire,
Sky,
Water,
Earth (basically the four Greek elements!)

but I hear it does code some things very well.


Zach

"Stephen Neal" wrote in message
...
On 16/2/04 10:29 pm, in article ,
"Agamemnon" wrote:


"just me" wrote in message
...
OK, so it's never terrific, but at this precise moment it's awful!
Loads
of
fizzy noisy artefacts and blocky effects.
TX is Winter Hill, programme is TOTP2.

Anyone else having trouble?


BBC4 today when they reran to documentary on the eruption of Mt
Vesuvius.
One inch square blocs filled the entire screen (28") during the smoke
scenes. This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke. There is no reason why the
encoder
could not have interpolated the image to smoothen it out since there was
plenty of available bandwidth.


Err - smoke and rippling water are VERY difficult for MPEG2 to encode - as
there is lots of near-random motion, coupled with quite a lot of luminance
variation. This makes motion-estimation and tracking very difficult - and
thus reduces the temporal redundancy of the GOPs quite a lot. IME smoke
and
water seldom compress well - whilst it may seem slow moving - it contains
quite a lot of "change" ...

Steve

Can I add:

Fire,
Sky,
Water,
Earth (basically the four Greek elements!)

but I hear it does code some things very well.


Zach

"Stephen Neal" wrote in message
...
On 16/2/04 10:29 pm, in article ,
"Agamemnon" wrote:


"just me" wrote in message
...
OK, so it's never terrific, but at this precise moment it's awful!
Loads
of
fizzy noisy artefacts and blocky effects.
TX is Winter Hill, programme is TOTP2.

Anyone else having trouble?


BBC4 today when they reran to documentary on the eruption of Mt
Vesuvius.
One inch square blocs filled the entire screen (28") during the smoke
scenes. This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke. There is no reason why the
encoder
could not have interpolated the image to smoothen it out since there was
plenty of available bandwidth.


Err - smoke and rippling water are VERY difficult for MPEG2 to encode - as
there is lots of near-random motion, coupled with quite a lot of luminance
variation. This makes motion-estimation and tracking very difficult - and
thus reduces the temporal redundancy of the GOPs quite a lot. IME smoke
and
water seldom compress well - whilst it may seem slow moving - it contains
quite a lot of "change" ...

Steve

"David van Kemenade" wrote in message ...
"Agamemnon" wrote in message
...
This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke.

Wow, what a quick diagnosis of all our digital TV problems! It all seems so
difficult, but he shows it's all really easy!

Agamemnon seems to have inside knowledge of all the kit used at the BBC. For
example, that the MPEG-2 coders that it uses are "15 years out of date".
Would you care to give the make and type numbers then? I'd like to know more
about this "defect" now. (Note that for the coders to be "15 years out of
date", they have to be at least 15 years old, and I seem to remember that
the MPEG-2 standard wasn't even finalised until 1993 or thereabouts - just
over 10 years ago!).

Others have already pointed out that smoke is difficult to code, not easy.

It's such a shame Agamemnon has not yet been hired by broadcasters to sort
out all the problems they've ever head or ever will have, both with radio
and television.


Just ignore him, he clearly hasn't got a clue.


--
Steve - http://www.digitalradiotech.co.uk/

DAB sounds worse than Freeview, digital satellite, cable, broadband internet and FM

"David van Kemenade" wrote in message ...
"Agamemnon" wrote in message
...
This has to be a defect in the MPEG-2 encoders which the BBC is
using that are 15 years out of date, because there was nothing else to
encode except lots of slow moving smoke.

Wow, what a quick diagnosis of all our digital TV problems! It all seems so
difficult, but he shows it's all really easy!

Agamemnon seems to have inside knowledge of all the kit used at the BBC. For
example, that the MPEG-2 coders that it uses are "15 years out of date".
Would you care to give the make and type numbers then? I'd like to know more
about this "defect" now. (Note that for the coders to be "15 years out of
date", they have to be at least 15 years old, and I seem to remember that
the MPEG-2 standard wasn't even finalised until 1993 or thereabouts - just
over 10 years ago!).

Others have already pointed out that smoke is difficult to code, not easy.

It's such a shame Agamemnon has not yet been hired by broadcasters to sort
out all the problems they've ever head or ever will have, both with radio
and television.


Just ignore him, he clearly hasn't got a clue.


--
Steve - http://www.digitalradiotech.co.uk/

DAB sounds worse than Freeview, digital satellite, cable, broadband internet and FM

On Tue, 17 Feb 2004 12:02:31 -0000, just me wrote:
I've come to expect lousy picture from cable, but I've found the picture on
Freeview - while far from ideal - to be superior to Telewest, NTHell and
KYTV and feared a decrease in quality in advance of more undesirable
channels.

KYTV ... now that's a name I haven't heard in a long time! ;)

--
-- Michael "Soruk" McConnell
Eridani MailStripper -- www.eridani.co.uk/MailStripper -- Uncovers Everything!
/* Halley */ --- Halley's Comment