where you will find a sample of CD audio, along with MP2 encodings at 96,
128, 160, 192 and 256kbps. In order to avoid any elements of bias, I won't
pass any comment right now. I'd be interested to know what people think is
"acceptable", however.
Oh, and please don't bitch about my taste in music. (If you want a different
genre, let me know and I'll chuck samples of something else up)
I'll try again -
Last time Tried to play they samples they didn't work - could have been a
browser problem at my end. I'll give it another whirl, though however the
only way (in practice) to evaluate DAB at 128kbps or any other rate, is by
listening to it on a DAB radio. Satan has already explained previously that
the BBC R&D have gone to great endeavours to 'optimise' their encoders for
lower bit rates, so we are only *really* testing your encoders algorithms -
which will differ from that of BBC R&D's implementations
Therefore, apart from being an interesting experiment in bit rates will
offer no purpose to the current state of BBC DAB transmissions in relation
to a .wav
Jonnie
> ...I point you to http://samples.ibrokethe.net
>
> where you will find a sample of CD audio, along with MP2 encodings at 96,
> 128, 160, 192 and 256kbps. In order to avoid any elements of bias, I won't
> pass any comment right now. I'd be interested to know what people think is
> "acceptable", however.
I'd convert the MP2 encodings back to WAV, and give them a codified file
name and then ask for feedback on them.
--
Good idea, I'll do that with another track at some point.
Yes, we're testing tooLAME. For the purposes of a simple bitrate
demonstration, though, this should do: a basic encoder with none of the
frills of BBC R&D's implementations. Yes, you can do cunning psychoacoustic
stuff, which will make it sound better - but not necessarily more like the
original sound.
> Therefore, apart from being an interesting experiment in bit rates will
> offer no purpose to the current state of BBC DAB transmissions in relation
> to a .wav
I never said it bore any relation to the actual quality of BBC DAB
transmissions. Nor do I claim, because of the dynamic range compressors and
so on that are used, that it bears any relation to the audio that you will
hear on any DAB multiplex. This is examining the difference between *what
is received* and *what goes into the MP2 compressor*: a direct comparison of
MP2 encodings (as are used on DAB and Freeview) at various bitrates. Seeing
as bitrate wars seem to be a regular occurrence on this newsgroup, I'm
surprised that people haven't found it more useful.
No matter which MP2 encoder you use, because of the amount of information
that's being stored, you will experience similar degradation with bitrate.
You really are an incompetent twat.
--
Steve - http://www.digitalradiotech.co.uk/ - Digital Radio News & Info
DAB sounds worse than Freeview, digital satellite, cable, broadband
internet and FM
I'll burn - and evaluate them!
I'm still - honestly - at a loss as to why Satan classes 128 kbps as
shite! - It ain't perfect but I'd hate to have it improved in the short
term, as a sacrifice, to the Asians and listeners to BBC6 music to satisfy a
few audiophiles who all - by accounts - have Freeview and good FM tuners :-)
Jonnie -
I must get to bed - But Looooove my new monitor ;-)
> I'm still - honestly - at a loss as to why Satan classes 128 kbps as
> shite!
Jonnie, I do not wish to be called Satan, so if you persist do you mind me
referring to you as faggot? I'm not homophobic, but it seems only fair that
if you're going to call me a derogatory name then I'll have to call you one
back.
> Jonnie wrote:
>
>
>>I'm still - honestly - at a loss as to why Satan classes 128 kbps as
>>shite!
>
> Jonnie, I do not wish to be called Satan, so if you persist do you mind me
> referring to you as faggot? I'm not homophobic, but it seems only fair that
> if you're going to call me a derogatory name then I'll have to call you one
> back.
And in doing so, you'll annoy a lot of people. I'd advise against it.
--
> "Nick J." <n.s.j...@durham.ac.uk> wrote in message
> news:bqjbov$64a$1...@sirius.dur.ac.uk...
>
>>I'd convert the MP2 encodings back to WAV, and give them a codified file
>>name and then ask for feedback on them.
>
> Good idea, I'll do that with another track at some point.
I did http://compsoc.dur.ac.uk/~nsj/Audio/StereoTest/ a while ago -- all
using psy model 1 in tooLAME.
--
Further to Nick's comment, I have now put up a set of six randomly named WAV
files: one is the original WAV; the others are 96, 128, 160, 192 and 256kbps
encodings of that WAV. They're all about 6MB long (and all the same length;
no clue there).
I'd be interested to know which people think is which, such that we will
know exactly how much bitrates matter.
(At the time of writing, the files are still uploading; give it 20 minutes
and they should all be there.)
You *did* decode them in a random order? Otherwise the date-stamps will
be sequential from high to low or low to high, and this will (should) be
preserved in the FTP transaction.
--
<snipped>
Okay then - I'll stop but by dictionary definition I'd probably have the
least to be concerned about -- but NEVER call me Malcolm
Jonnie
They seem to have been timestamped in the order they were uploaded.
When I want your advice I'll ask for it.
Truce it is then.
> Nick J. wrote:
>
>>DAB sounds worse than FM in Cars wrote:
>>
>>>Jonnie wrote:
>>>
>>>>I'm still - honestly - at a loss as to why Satan classes 128 kbps as
>>>>shite!
>>>
>>>Jonnie, I do not wish to be called Satan, so if you persist do you
>>>mind me referring to you as faggot? I'm not homophobic, but it seems
>>>only fair that if you're going to call me a derogatory name then
>>>I'll have to call you one back.
>>
>>And in doing so, you'll annoy a lot of people. I'd advise against it.
>
> When I want your advice I'll ask for it.
I want, never get.
--
Pot, kettle, black
There's a couple of problems with your post:
(a) Angus Deayton was sacked fucking ages ago by Have I Got News for You,
and you're still using that lame "joke"?
(b) You're inferring that I'm incompetent in some way, but as you'll know
I'm hardly incompetent.
He's got a good point tho...
Mikeapollo
Then try harder ;-)
MIkeapollo
> Last time Tried to play they samples they didn't work - could have been a
> browser problem at my end. I'll give it another whirl, though however the
> only way (in practice) to evaluate DAB at 128kbps or any other rate, is by
> listening to it on a DAB radio. Satan has already explained previously that
> the BBC R&D have gone to great endeavours to 'optimise' their encoders for
> lower bit rates, so we are only *really* testing your encoders algorithms -
> which will differ from that of BBC R&D's implementations
I think the BBC are just using a modified version of psychoacoustic
model 2, joint stereo, 128kbps. That's judging by my ears and lots of
comparisons. Using tooLame with -p2 -b128 will do fine. The default
-p1 sounds much worse at 128kbps.
It's actually quite difficult to make 128kbps sound as bad as some of
the commercial DAB stations do. You need to use the inferior
psychoacoustic model, transcode the material a couple of times, and
chuck some very heavy DRC in at some point.
If you just take an average CD an encode it to 128kbps mp2 (-p2), it
will sound fine on casual listening. The reason we need higher
bitrates is mainly so it stands up to critical listening. However (and
more importantly to most people), _some_ sounds just won't encode
properly using 128kbps mp2; as you can't avoid broadcasting those
sounds, you need higher bitrates so they don't stand out as being
exceptionally bad. What's more sometimes it's difficult to avoid
transcoding and DRC, so it's good to remove the final low bitrate
bottleneck - otherwise what sounds "clean" leaving the studio can be
unexpectedly bad when it leaves the encoders!
Cheers,
David.
Have you tried a lowpass setting of 15 kHz? (or others)
This will leave more bits for the remaining frequencies, so they can sound
better.
For MP3 this definitely helps (when coding at low bitrates with Lame).
> It's actually quite difficult to make 128kbps sound as bad as some of
> the commercial DAB stations do. You need to use the inferior
> psychoacoustic model, transcode the material a couple of times, and
> chuck some very heavy DRC in at some point.
Transmission standards would be no luxury. Including a minimal bitrate.
> If you just take an average CD an encode it to 128kbps mp2 (-p2), it
> will sound fine on casual listening. The reason we need higher
> bitrates is mainly so it stands up to critical listening. However (and
> more importantly to most people), _some_ sounds just won't encode
> properly using 128kbps mp2; as you can't avoid broadcasting those
> sounds, you need higher bitrates so they don't stand out as being
> exceptionally bad. What's more sometimes it's difficult to avoid
> transcoding and DRC, so it's good to remove the final low bitrate
> bottleneck - otherwise what sounds "clean" leaving the studio can be
> unexpectedly bad when it leaves the encoders!
You would need to apply a rule that every tchnician working at a
radiostation has to listen to the DAB broadcast on his headphones ;-)
gr, hwh
Which is fine, providing that your hearing isn't sensitive to frequencies
over 15kHz. However, such frequencies are actually valuable to the quality
of the audio.
> > It's actually quite difficult to make 128kbps sound as bad as some of
> > the commercial DAB stations do. You need to use the inferior
> > psychoacoustic model, transcode the material a couple of times, and
> > chuck some very heavy DRC in at some point.
>
> Transmission standards would be no luxury. Including a minimal bitrate.
And a limit on how much DRC they can use. Have you ever heard Kerrang! on
Freeview? Put it through a crappy TV and you probably won't notice much,
because bass reproduction on your average TV isn't great. Put it through a
decent hi-fi, however, and you need to turn the bass as far down as it'll go
to get any kind of reasonable output...