Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Identify what you can't create. (Read 7082 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Identify what you can't create.

An important problem seems to stay unsolved.

I create my mp3 with best quality I can:
- EAC + lame 3.98 SECURE as recommended on this fantastic site
- VBR V2 etc.

But nothing can guarantee the same quality in downloaded files.

There are some programs ( I use audio identifier... I don't know if there's something better or not ) that can tell me bitrate and coder... ( for example Fgh 192 CBR .. or Lame 3,97 V0....).

But if a 3 mb file coded @ 128 with the worst coder will be re-encoded with lame 3,98 VBR V0 it will be presented like an 6 mb high quality file!
I think you have to fight every day with this problem.

So.. can someone uptate me about the better way to recognize real quality?

Only personal listening?
An analysys file by file with some spectum analyzer?
Some rare linux program?
A voo-doo pratice? 

Is there something new?
Is there someone WHO KNOWS A SECRET WAY? 

Identify what you can't create.

Reply #1
But if a 3 mb file coded @ 128 with the worst coder will be re-encoded with lame 3,98 VBR V0 it will be presented like an 6 mb high quality file!
I think you have to fight every day with this problem.

Ehh, does LEGAL music sellers on the interwebs do this? Really! I can't help but thinking that this could be a problem with warez music.. which I don't condone.

Identify what you can't create.

Reply #2
Ehh, does LEGAL music sellers on the interwebs do this? Really! I can't help but thinking that this could be a problem with warez music.. which I don't condone.

On many countries, recordings of live shows are legal. There are also artist that upload music for free.

But i see your point.

Identify what you can't create.

Reply #3
Yeah I know that there's perfectly legal content circulating the net (not much though IMHO compared to illegal), it was the "I think you have to fight every day with this problem." which changed the situation a bit; I've not seen this discussed in HA (or anywhere else) before.. ok, there's always a first time but I think that we would have seen that already considering how many here are anal retentive about the quality of music/sound (those who listen to brown/blue/whatever noises on their 200k Hi-Fi).

Identify what you can't create.

Reply #4
An analysys file by file with some spectum analyzer?

yeah, it's actually easy when you have trained eye.

Focus on the high frequency part, especially above 16kHz. If the file you got was encoded with Lame V2 or any Lame >192kbs, it should have kept nearly all frequencies up to 19kHz in the case of V2. If you see lower lowpass or drop-outs, it is a re-encode.

Identify what you can't create.

Reply #5

An analysys file by file with some spectum analyzer?

yeah, it's actually easy when you have trained eye.

Focus on the high frequency part, especially above 16kHz. If the file you got was encoded with Lame V2 or any Lame >192kbs, it should have kept nearly all frequencies up to 19kHz in the case of V2. If you see lower lowpass or drop-outs, it is a re-encode.

That is probably the easiest without special tools.  You may have to look at quite a bit of the file though as encoders will vary bandwidth frame by frame.

Identify what you can't create.

Reply #6


An analysys file by file with some spectum analyzer?

yeah, it's actually easy when you have trained eye.

Focus on the high frequency part, especially above 16kHz. If the file you got was encoded with Lame V2 or any Lame >192kbs, it should have kept nearly all frequencies up to 19kHz in the case of V2. If you see lower lowpass or drop-outs, it is a re-encode.

That is probably the easiest without special tools.  You may have to look at quite a bit of the file though as encoders will vary bandwidth frame by frame.



Somewhere I already read that a good-quality file seems to be a file with 19-20 kHz instead of 15-16.
I think it is sometheing intuitive... we hear  20-20.000 Hz frequences, so a more complete file has to be better.
Now.. pratically...there are some ways to view it.

We can watch with our tired eyes the spectrum of a file, represented for example in bars..and try to see if  extreme-right columns seems to move or not... Right?

Otherwise we can try to find a software that works for us... telling... I don't know.. "the present file is 3% between 20 and 50 Hz.... 6% 50-100 ...........  3% between 17-20 KHz.

Now I ask you... which softaware do you use? How does it work?


@ Akkurat
If we are here, our target is quality. Independently of legal questions.
I like quality, so 90% om my mp3 files I ripped personally. By an original CD.
Unfortunately, there's for example some rare or strange versions I only find in p2p programs.
I KNOW, an original CD is the best source.. of course!

There thousands of media players, rippers, encoders, codecs, editors, burners.... and no-one can tell something useful about it?

Identify what you can't create.

Reply #7
Now I ask you... which softaware do you use? How does it work?
I use my ears to listen with. If I can't ABX a lossy encoding from the original irrespective of where the upper cut-off frequency is then it's totally irrelevant what anything else tells me visually. Do you listen with your eyes or your ears?

Cheers, Slipstreem. 

Identify what you can't create.

Reply #8
If we are here, our target is quality. Independently of legal questions.
I really don't think that the Terms Of Service or the Moderators would agree with you on that particular point.
lossyWAV -q X -a 4 -s h -A --feedback 2 --limit 15848 --scale 0.5 | FLAC -5 -e -p -b 512 -P=4096 -S- (having set foobar to output 24-bit PCM; scaling by 0.5 gives the ANS headroom to work)

Identify what you can't create.

Reply #9
Some reasons why p2p can never replace ripping an original cd yourself.

1. You don't know if the mp3s have been transcoded. With a spectrum analyzer you can see if the lowpass matches the preset. This won't work on tanscodes like -V0 to -V2 or -V2 to -V2.

2. You don't know if the rip was secure. EAC logs can be faked.


Identify what you can't create.

Reply #11
A typical example: notice the FHG encoding has very little content (just a few "spikes") above 16kHz even though I set the lowpass at 18kHz. This is typical for most low-bitrate encodings because encoding into sfb21 (>16kHz) needs a lot of bits which should better be saved for the lower frequencies.

[Anyone to help me to resize that image/make it into thumbnail?]




(edit- added frequency measure to the graphs)

Identify what you can't create.

Reply #12
If we are here, our target is quality. Independently of legal questions.
I really don't think that the Terms Of Service or the Moderators would agree with you on that particular point.



         

I agree

I'm sorry for my english language...

What I wanted to say... it was that I look for the difference between high and low quality in these files.
I don't want to speak about legal questions because I'm not interested of illegal uses.

Identify what you can't create.

Reply #13
It's strange that moderators didn't jump yet.  Even the topic is difficult to decipher.

I would like to remind everyone that graphics don't help here,  (people wanting to encode all the frequencies even at 128kbps, or the opposite: the original being already with little content on high frequencies. Add to that the stereo vs joint stereo, the long blocks vs short blocks...)

There is no way to know what an mp3 was originated from, more so, if you don't have the original to compare to.

Identify what you can't create.

Reply #14
It's strange that moderators didn't jump yet.  Even the topic is difficult to decipher.

I would like to remind everyone that graphics don't help here,  (people wanting to encode all the frequencies even at 128kbps, or the opposite: the original being already with little content on high frequencies. Add to that the stereo vs joint stereo, the long blocks vs short blocks...)

There is no way to know what an mp3 was originated from, more so, if you don't have the original to compare to.

I thought the OP was asking about identifying re-encoded files. I provided my answer how do I do it. I don't see anything wrong with it (notice I said "for most low-bitrate encodings"). I've seen a lot of 128k encodins randomly from the net and most of them look like the 128k encoding on my spectrum, many of them are even hard-limited to 16kHz. Lame V2 and >192k CBR files usually leave A LOT more content above 16kHz.