PDA

View Full Version : source question killers 2005-02-25.


range_hood
2005-05-25, 05:55 AM
I recorded this show by myself, so thereīs no lossy stage in the line on my side.
How could this happen.

cable FM > hdd (wav; 44.1 kHz, 16 bit) recorded from fm4 (http://fm4.orf.at/spielzimmer)

I know they broadcast also over satellite and have a live stream. Do they store their shows on their computers in a lossy format? Or is the cable broadcast sourced from satellite? Any others ideas? Would you seed such a show, is this allowed here?

Ted
2005-05-25, 06:21 AM
The only help I could be (and I know it's not much) is to say that if it was streamed over the internet, I'm pretty certain it was compressed in one way or another. I'm not sure about cable FM though.

range_hood
2005-05-25, 07:15 AM
Thanks Ted. I tried around and compared the cable broadcast to their terrestrial. They came out with similar spectrals.
Iīll try this again with a cast which Iīm sure that it is live. (Bonus Track 00:00 tonight).

vegas_jay
2005-05-25, 08:50 AM
FM broadcasts are frequency limited. You are most likely seeing the effect of that.

range_hood
2005-05-25, 10:12 AM
What makes me feel strange about this spectrals are the blocked out areas below the frequency cut off. Also there is no straight cut off.

ssamadhi97
2005-05-25, 11:07 AM
Well, broadcasts from lossy sources (or broadcasts that go through a lossy stage - usually mp2 - before they arrive at your home) are not really uncommon.

Sad as it is, I'd say it's relatively normal for a broadcast. :(

range_hood
2005-05-25, 06:49 PM
This is a clip of a live cut of fm4 (terrestrial 103,8 MHz here in Vienna) from Bonustrack, you can call in there, itīs like talk radio. What do you think?

brainysod
2005-05-26, 01:52 AM
i'd like to ask, how would you do a spectral analysis? i know how to get it that way, but what signs do you look for to see if it was from a lossy source? Thanks

AAR.oner
2005-05-26, 03:17 AM
i'd like to ask, how would you do a spectral analysis? i know how to get it that way, but what signs do you look for to see if it was from a lossy source? Thanks

http://www.thetradersden.org/forums/showthread.php?t=4288

that should get ya started :D

AAR.oner
2005-05-26, 03:25 AM
What makes me feel strange about this spectrals are the blocked out areas below the frequency cut off. Also there is no straight cut off.

i've run across a few FM broadcasts which had the same sharp line around 19kHz...unfortunately, its just the nature of broadcast technology at this point :(

AAR.oner
2005-05-26, 03:28 AM
for spotting MiniDisc & mp3'd recordings:
http://www.audiohub.org/get/fa/fa.htm
http://www.audiohub.org/get/fa/sa.htm

Rider
2005-05-26, 12:24 PM
Yep that sharp line at 19k gives it away as an FM.

brainysod
2005-05-26, 01:04 PM
http://www.thetradersden.org/forums/showthread.php?t=4288

that should get ya started :D

Thanks!

range_hood
2005-05-26, 02:58 PM
this is a little off topic.
Do you think the same as me (itīs an other source).

Five
2005-05-26, 03:46 PM
looks the same to me, lossy and all.

http://img200.echo.cx/img200/5305/lookslossy17cr.th.gif (http://img200.echo.cx/my.php?image=lookslossy17cr.gif)

ssamadhi97
2005-05-26, 05:12 PM
that kind of "hole" can very well be part of a normal speech signal

Five
2005-05-26, 05:51 PM
It is a little fuzzy...

I thought you said this was probably mp2?

maybe closeups from another couple spots might help

4candles
2005-05-28, 07:55 AM
It's well known that FM broadcasts have a spike at 19KHz (the "pilot tone"), and are also subject to a frequency roll-off at around 15KHz.

Stereo FM signals are modulated with one part of the signel (L+R) transmitted in the 0-15KHz range, and the other part (L-R) transmitted in the 23KHz-53KHz range - as shown by this diagram:

http://hyperphysics.phy-astr.gsu.edu/hbase/audio/radio.html#c2

But back to the subject - how do you think the radio signals travel from the original broadcaster to the cable company's distribution center? Even if the cable company is rebroadcasting from a local analogue FM transmitter, how does the signal travel to that FM transmitter?

The answers to these questions are generally unknown, but I have read that France uses MP2 compression via satellite, and the UK (specifically, the BBC) uses NICAM digital signals. Someone has mentioned that the BBC have now switched to MP2 distribution for their FM transmitters, but I haven't found any secondary sources to back that up - the fact that they use (used?) NICAM in the past is well documented.

So whilst we don't know for sure how the Cable company got the radio signal, an MP2 feed is a common method, and the spectral analysis seems to confirm that. The radio station in question (FM4) broadcasts via digital satellite in Europe (at 192kbps), which would seem a likely source for the cable company to tap into.

And we still don't know how the satellite transmission is fed....

Dave.

EDIT: If anyone wants to compare the sound of this torrent with the digital satellite version, the original 192kbps MP2 data is torrented here:

http://www.dimeadozen.org/torrents-details.php?id=42814&hit=1

As I write this, there are four seeders, and no leechers, so a good time to hop on - and it's only 77MB (being the original MP2 data).

range_hood
2005-05-28, 04:51 PM
Here it is.
That`s the right channel of both at about the same track position.

range_hood
2005-05-28, 06:22 PM
This is from brighteyes2005-04-03:

fm > wav 44.1/16bps (note: no cableFM)

range_hood
2005-05-30, 03:45 PM
So, let me say... I think the fm broadcast is superior to the satellite one.
Is this allowed to be torrented here (f. e. bright eyes)? I think thereīs something lossy in it.

Next question, does anyone bother about this to be seeded or not :) :lol

Five
2005-05-30, 03:59 PM
yeah, the FM looks a little fuller in the SA.

these days it seems that more and more non-internet (or at least not exclusively net) stations are broadcasting lossy.

unfortunately lossy isn't allowed here unless its master MD or a DVD.

range_hood
2005-06-01, 05:24 AM
I could record it on MD too. They repeat the cast tonight. It would be MD(M). :)

Five
2005-06-01, 03:11 PM
:eek:

I'll have to find a way to word things more carefully. MD is only okay for something recorded at the show like sbd, mics or whatever. And no loopholes recording the radio simulcast from the audience or something silly like that! :D

truth is, when we started this site there was quite a discussion on the topic of allowing mp2/webcasts when they are unique sources. My interpretation of our final decision is that it will help surface the non-lossy sources faster. Since sometimes lossless sources never surface it is often worth it to share mp2/webcasts, just not on this site. We ended up keeping the MD exception so in the end our stance on lossy sources is pretty much the same as it was at STG (altho many other things are different).

range_hood
2005-06-01, 03:47 PM
:D ...

:hmm:
hm... so, ... what I want to share (f.e. brighteyes) is neither webcast nor mp2 stream. Itīs plain old terrestrial FM (as you get it here in vienna at 103.8MHz). As this is a soundboard recording you will get no better sbd unless itīs postFM or another soundboard patch.
Sure, I would be glad to get true lossless.

range_hood
2005-08-27, 11:43 AM
correction:
hm... so, ... what I want to share (f.e. brighteyes) is neither webcast nor mp2 stream. Itīs plain old terrestrial FM (as you get it here in vienna at 103.8MHz). As this is a soundboard recording you will get no better sbd unless itīs preFM or another soundboard patch.
Sure, I would be glad to get true lossless.

Five
2005-08-27, 02:15 PM
yeah, seed the non-mp2 version anytime. the sa looks just like the one we were discussing in the bbc thread so go for it.