Topic: Is a MIX of lossy SBD stream with lossless AUD still lossy?

[Emile]

Just wondering what people think about this one.

One of my recordings was just banned on DIME. It's a MIX of 2 sources:
1) a LOSSY webcast stream rip
2) a LOSSLESS AUD recording

The result is a really solid sounding recording, which I would love to share with people.. however..

Do people consider the result of this MIX still lossy? I would argue that it's lossless, but then again.. I'm not an expert on this.

Would love to hear your opinions. (if it is still lossy, can someone explain why?)

Best & Love

[ScoobieKW]

You can't add back the bits that are thrown away. Lossy + Lossless = Lossy

Dime set the Lossless rule to stop MP3 copies spreading and being considered the definitive source, in their zeal to stay pure they'll ban an excellent recording.

Seed it to another tracker, the proof is in the pudding, if it's good it's good.

[page]

You can't add back the bits that are thrown away. Lossy + Lossless = Lossy

ehhh, I don't think it's that cut and dry. Was the majority of the signal from the lossy source? If you had 8 tracks and 2 were lossy, is it still considered lossy after you do the mixdown? I think one defining characteristic is whether or not you re-compress it to MP3 after all is said and done. By technical accounts, the new mix is lossless as you didn't re-compress it post-render. It might have been sourced from lossy information, but that's a different distinction.


Yeah, DIME has a blanket ban which helps them run the site, makes sense and I don't dispute that, but for the "angels on the head of a pin" discussion, I think there is more nuance.

[flipp]

Were you given a reason for the banning? It may not be because it is a mix. See the following sections of Dime's FAQ for other possible reasons.

http://wiki.dimeadozen.org/index.php/DimeFAQ:Frequently_Asked_Questions#I_have_a_digital_satellite_radio_broadcast_recording.2C_a_webcast_recording.2C_or_audio_material_taken_from_a_digital_TV_broadcast._Is_it_allowed_to_seed_this_recording_on_DIME.3F

http://wiki.dimeadozen.org/index.php/Caclas#lossystreams

< There was a topic here on ts that I see has been closed. If you want to know a specific reason for the banning, PM the dime lady as suggested in the final post of the thread. http://taperssection.com/index.php?topic=152617.0  < or to keep open a can of worms you might pose your question in the following: http://taperssection.com/index.php?topic=154702.0 >>

[Emile]

Here's the reason:

"Source 1: Webcast Stream Video Capture (SBD) > WAV > Traders Little Helper > FLAC"

Sourced from a lossy webcast and bloated to .flac, which is not allowed here:

Please read http://wiki.dimeadozen.org/index.php/Caclas#lossystreams

[yltfan]

Weird--they let me post a similar mix one time. I accidentally did the SBD lossy. For my money, the mix sounded fine, and I imagine that no one would have been able to tell if I hadn't said anything.

[Emile]

You can't add back the bits that are thrown away. Lossy + Lossless = Lossy

Dime set the Lossless rule to stop MP3 copies spreading and being considered the definitive source, in their zeal to stay pure they'll ban an excellent recording.

Seed it to another tracker, the proof is in the pudding, if it's good it's good.

According to this logic, if I have a 1000 hour lossless recording and I add 1 second of lossy audio.. the entire thing is lossy

Or similarly, if I have a 1000 channel (lol) recording, which has 1 lossy channel.. its lossy

I guess this is what Page is getting at.. what (technical) evidence is there to call it one or the other?

[yousef]

I think this will become a more prevalent issue due to the growing popularity of digital mixers that can output a stereo mix directly to a memory stick in MP3...

I think there has to be an element of common sense here: if the "lossy" source is the least lossy version in the wild (eg if it was sourced from a webcast, digital broadcast or a desk as above) then surely that must be a mitigating circumstance?

I mean, everything is "lossy" in the final analysis, isn't it? Even a 'perfect sound forever' 44.1Hz/16bit wav if that wasn't the original format...

Actually, doesn't Dime allow lossy audio from, say DVB-T/DVB-S, as long as it isn't 'bloated' to a bigger file?

[runonce]

I think there has to be an element of common sense here: if the "lossy" source is the least lossy version in the wild (eg if it was sourced from a webcast, digital broadcast or a desk as above) then surely that must be a mitigating circumstance?

Agree - if something was mastered to mp3 and thats all that exists - convert it to wav...flac it and fingerprint - - that is the defacto master now.

There was time when all mp3s were probably derived from wavs - no longer.

If Im running low on memory at a long show - hello 320K!

[Gutbucket]

Better that the SBD portion is lossy sourced rather than the AUD, in the opinion of this pin-head angel.

Agree - if something was mastered to mp3 and thats all that exists - convert it to wav...flac it and fingerprint - - that is the defacto master now.

I don't get that.  What value could their be in converting it rather than simply leaving it as mp3?  Seems bad practice in several ways to me.

[page]

Better that the SBD portion is lossy sourced rather than the AUD, in the opinion of this pin-head angel.

my first reaction was to agree; the reverb and stereo cues that are valued in the audience tape are more likely to be damaged by mp3 (pre-echo, stereo construction, etc). But if you have an excellent sbd tape (e.g. a custom mix) and you're using the aud as a reverb/transient warming instrument I'm not so sure I'd pick the same result. My primary source I'd want as a lossless copy if at all possible. If they are close to an even split (by RMS, not just peak or existing levels), then I'd probably pick the aud.

[Gutbucket]

I can see that.  I also think most compression algorythms are better tuned for up-front sources which are generally more SBD than AUD-like (similar to the reverb ambience & stereo cues point).. and since one of the typical failings of them is with transient rich material, applause is perhaps the primary source which will tend to suffer most from audible compression artifacts.  Yeah, many don't care about the stuff between the songs, but I do.  For the same reason, I don't let the applause clip on my quieter recordings where it is the highest level source by far. 

Hmmm, I guess I might choose the lossless SBD for a concert of solo harpsichord. 

[page]

Hmmm, I guess I might choose the lossless SBD for a concert of solo harpsichord. 

agreed. any type of super complicated instrument with harmonics I'd look at it.