In article , John Phillips
writes
On 2005-07-23, Jim Lesurf wrote:
I'm currently exploiting the 'proms' to do some comparions between:

1) prom broadcasts on BBCTV4 DTTV

2) proms on BBC R3 DTTV

3) proms on BBC R3 FM

I did a quick comparison between the re-broadcast of Mahler's 5th symphony
as digitized from FM, and the earlier live performance captured digitally
from R3/DAB via the tuner's S/PDIF interface.

The dynamic range comparison looks visually quite bad for FM. If I match
levels at about -25 dBFS (referred to the normalised DAB waveform) then
the FM peaks show 10 dB or greater compression and the compression starts
noticeably somewhere around -20 dBFS. This comparison was done rather
quickly so beware that the figures may be imprecise. Also I would like
to do the comparison with live FM versus re-broadcast DAB (or better on
live FM versus live DAB but I can't do that yet).

I am sure such moderate compression on FM is euphonic in that it raises
the level of the decay of notes and the level of ambience, leading to
a greater impression of presence. However the amount of compression on
FM surprised me.


They "Optimod" it during most of the day so as car and casual listeners
don't have to keep adjusting the volume control, but they do or as far
as I'm aware take the Optimod out of circuit in the evenings and
especially for the proms.

I don't know if they have a digital processor in line as yet.

It'll come one day......
--
Tony Sayer

On 2005-07-28, tony sayer wrote:
In article , John Phillips
writes
... Also I would like
to do the comparison with live FM versus re-broadcast DAB (or better on
live FM versus live DAB but I can't do that yet).

They "Optimod" it during most of the day so as car and casual listeners
don't have to keep adjusting the volume control, but they do or as far
as I'm aware take the Optimod out of circuit in the evenings and
especially for the proms.

I had a suspicion something like this might be the case for the daytime
FM re-broadcasts. So a better comparison probably requires the FM to be
captured live in the evening.

--
John Phillips

In article ,
John Phillips wrote:
On 2005-07-23, Jim Lesurf wrote:
I'm currently exploiting the 'proms' to do some comparions between:

1) prom broadcasts on BBCTV4 DTTV

2) proms on BBC R3 DTTV

3) proms on BBC R3 FM

I did a quick comparison between the re-broadcast of Mahler's 5th symphony
as digitized from FM, and the earlier live performance captured digitally
from R3/DAB via the tuner's S/PDIF interface.

The dynamic range comparison looks visually quite bad for FM. If I match
levels at about -25 dBFS (referred to the normalised DAB waveform) then
the FM peaks show 10 dB or greater compression and the compression starts
noticeably somewhere around -20 dBFS. This comparison was done rather
quickly so beware that the figures may be imprecise.

Thanks for the above info. It seems broadly in line with a comparison of a
'midday concert' of piano music a few years ago. There the compression on
FM was less than above, but that seems understandable as the solo piano
probably had a smaller effective dynamic range than a Mahler symphony. :-)

[snip]

I am sure such moderate compression on FM is euphonic in that it raises
the level of the decay of notes and the level of ambience, leading to
a greater impression of presence. However the amount of compression on
FM surprised me.

My impression with the piano comparison was similar. That the effect was
making the 'sustain' more noticable, and hence giving a 'warmth' to the
results which was pleasant. But on repeated listening I ended up feeling
the DAB version actually gave a clearer sound as the transients were more
distinct. However the piano example wasn't very useful for testing things
like the data-reduction effects when listening to ppp massed strings, etc.
So it probably failed to test some of the limitations of DAB R3.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

"I am sure such moderate compression on FM is euphonic in that it
raises
the level of the decay of notes and the level of ambience, leading to
a greater impression of presence.. "

But for the reverberation decay (if that's what you refer to) to be
affected by the compression that would mean the compression was acting
with a timescale similar the reverberation time of the hall - typically
a few seconds. Surely the compression has a timescale of many tens of
seconds does it not?

Robert




--

In article .com,
Robert
wrote:
"I am sure such moderate compression on FM is euphonic in that it raises
the level of the decay of notes and the level of ambience, leading to a
greater impression of presence.. "

But for the reverberation decay (if that's what you refer to) to be
affected by the compression that would mean the compression was acting
with a timescale similar the reverberation time of the hall - typically
a few seconds. Surely the compression has a timescale of many tens of
seconds does it not?

Not necessarily. When I did some comparisons of DAB with FM a few years
ago, there were signs of rapid gain adjustments to lower the levels of
transients w.r.t. the sustain. Hence the level compression may be over
timescales from much less than a second, upwards.

Also, my impression was that it is the transient/sustain of the
instrumental sound that is being adjusted, not just the direct/reverberant
ratio. So it may be the decay characteristics of the instrument that
matter. However I'm hoping to re-investigate this soon - hence starting
this thread. :-)

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scot...o/electron.htm
Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html
Armstrong Audio http://www.st-and.demon.co.uk/Audio/armstrong.html
Barbirolli Soc. http://www.st-and.demon.co.uk/JBSoc/JBSoc.html

On 2005-08-01, Robert wrote:
"I am sure such moderate compression on FM is euphonic in that it
raises
the level of the decay of notes and the level of ambience, leading to
a greater impression of presence.. "

But for the reverberation decay (if that's what you refer to) to be
affected by the compression that would mean the compression was acting
with a timescale similar the reverberation time of the hall - typically
a few seconds. Surely the compression has a timescale of many tens of
seconds does it not?

I don't think you need to bring in compressor time constants into the
reasoning [1].

If the top level signals are compressed at all, by any means, that will
(I assume) lead to a perceived reduction in average level. If so,
restoring the perceived level to that of the uncompressed signal will
mean the lower level signals will be higher (and perceived as such).


[1] If so, at least the attack time constant needs to be very short to
prevent over-deviation.

--
John Phillips