I just want to get one thing straight.
If the source of a sound is mp3 the visual representation (I use AnalFreq 1.8) lack any sound levels above about 14kHz. Right? I mean ANY, straight line?
Or is it possible that it still shows SOME sound levels?
If the frequency is cut at or below 16khz and there is no real line after it it's mp3.
So yes, it's possible that there is MINOR sound after a sound dropout due to lossy format. But that doesn't make it lossless.
It can also be cut after 14khz, 17khz is quite normal also. As long as it has cuts from 19khz and downwards, it's probably mp3 sourched.
I guess there can still be some sort of power after the cut, but can anyone verify this?
Basically I'm asking about one thing- if there's an irregular line around 14 / 16 and after- still something happening, but pretty low- what does that mean? I checked some recording I had- veeeeeery old and veeeeeery bad quality- and it dropped almost in the middle, not at 14! NOW- does this simply represent the fact that it's a copy of a copy of yet another copy, OR- the age... Or... what? It is kind of cut in the middle, but the line is definitelly still irregular, there's still sth happening...
That's why I'm wondering- is it possible that the analysis also represents how low (or high) generation recording this is...? Does the line have to be pretty much straight and at that point it's dead certain it's mp3?
Thanks Barry, I'll try this one!
But after reading quite few things... Went through google and all that... and all I'm getting is that apart from dead certain ones (perfect cut, dead certain mp3) it is at times difficult to say whether it really is lossy format or not... I've read somewhere that it depends even on the way the show was miked... especially in 70s... Well, that's exactly the kind of recordings I need to check :)
Fenderek wrote: NOW- does this simply represent the fact that it's a copy of a copy of yet another copy, OR- the age... Or... what? It is kind of cut in the middle, but the line is definitelly still irregular, there's still sth happening...
That's why I'm wondering- is it possible that the analysis also represents how low (or high) generation recording this is...??
No, the frequency response doesn't tell you anything about the quality of the recording, the generation of it or the way it was recorded. I remember someone once claimed here that the frequency response of a recording might cut off, because the mics that were used to originally tape the show only a had a frequency response up to 16 khz. This is definately not true. Even though these high frequencies were not originally recorded, they are shown in the FR, but of course only if the show is lossless and hasn't been compressed after it's been digitized. A terrible sounding audience tape can have a 'better' frequency response than a top notch SBD tape.
The FR does tell you through what formats the recording went. For example if it's sourced from a FM broadcast, there will be a visible stripe at around 16 khz in the FR. And of course it tells you if a show has been digitally compressed somewhere in the lineage. Therefore it's possible to spot mp3 or MD sources.
Fenderek wrote: .. and all I'm getting is that apart from dead certain ones (perfect cut, dead certain mp3) it is at times difficult to say whether it really is lossy format or not...
You can always take a screenshot of the frequency analysis and then upload it, so that others can have a look at it. Usually it is possible to spot if it's lossy or lossless.
We've been through this so many times before, people! Unless you know exactly what's happened to a recording (including which mics were used, because every mic will have a high frequency roll-off at some point, and onto what medium it was recorded), you can't reliably tell if it's ever been through a lossy process by running a frequency analysis on it.
I know people will argue with this, but surely if your ears (and speakers/monitors) aren't good enough to HEAR the difference between lossless and lossy formats (and you have to run material through f**king software to check the lineage!), then there's no need get so upset about mp3-sourced material! If you can't hear the difference, what's the problem? May as well use mp3s!
Please guys, use common sense: your ears are more accurate than ANY software will ever be. Do yourself a favour and sit down and LISTEN. Convert some wav files to mp3 and LISTEN to them back and forth - you WILL hear the difference. And if you can't, you never need worry about whether something is lossy again...
Ah, Mr. sound engineer himself again ;-)
Well, no need to argue again about spreading mp3 sourced shows, just one thing: Trading means to circulate the best available version of a show and this is always the lossless one and never the degenerated mp3 version. Even if your ears aren't able to tell the difference between lossy and lossless maybe someone other's ears(or his stereo) can.
tilomagnet wrote: Ah, Mr. sound engineer himself again ;-)
Well, no need to argue again about spreading mp3 sourced shows, just one thing: Trading means to circulate the best available version of a show and this is always the lossless one and never the degenerated mp3 version. Even if your ears aren't able to tell the difference between lossy and lossless maybe someone other's ears(or his stereo) can.
That's exactly the point and the reason I'm asking this. Even if I was fine with it (BTW- I'm not)- someone else may not be... And I really don't want to send someone mp3 sourced shows when they don't want those... Which did happen- I was completely unaware I was spreading two mp3 sourced boots. I'm trying to avoid that...
I bet 90% of the people on this board who claim there was no audible difference between mp3 and lossless are listening to the shows on some crappy computer speakers. They are likely right- you can't tell the difference then. However when listening to both versions on a good stereo or some high quality headphones the difference becomes obvious. Not even talking about mp3s that have been encoded multiple times or anything like that....