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: Pre-echo on digital (mp3) files? (Read 8441 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Pre-echo on digital (mp3) files?

Reply #1
It might be an analog tape recording where a print-through occured. Or it is ripped from an vinyl where the grooves were too narrow...

Pre-echo on digital (mp3) files?

Reply #2
I despise having to give away personal information and my email address to yet another organization just to download one thing...so I haven't heard the actual file you're referring to.

However, it sounds like your definition of "pre-echo" might be a little bit off here.

Pre-echo as used in the context of lossy compression refers to very sharp transients being "smeared" upon playback because the attack is shorter than the block length used to encode it, and therefore the leading edge of the transient starts very slightly (milliseconds) before it should.  It does not mean that you literally hear an echo of a sound.

From your description, it sounds more like you're referring to analog tape print-through, where you hear a sound very faintly shortly before you hear the actual sound itself at full volume (exactly how long "shortly" is depends upon the tape speed and the size of the reel).

Do they say when these recordings were made?  I didn't see anything pertaining to recording dates on the website.

Edit:  Sunhillow was faster, and more succinct.
"Not sure what the question is, but the answer is probably no."

Pre-echo on digital (mp3) files?

Reply #3
They say it is recorded in 1990, so it might be digital.
Anyway, I own a lot of CDs remastered from analog recordings and due to my age  I have some experience with vinyl. Clear print-through from analog tape is really really rare.
Stephen, maybe you can upload a short snippet < 30s? If it is clearly audible, LAME V2 would be good enough I suppose

Pre-echo on digital (mp3) files?

Reply #4
...or use mp3directcut to trim out that section without re-encoding (decoding the entire thing to WAV and then trimming to a FLAC would also work, but could be a bit tedious, depending on your computer).
"Not sure what the question is, but the answer is probably no."

Pre-echo on digital (mp3) files?

Reply #5
oh sorry mixminus1, I did not (want to) see that the file already is mp3

Pre-echo on digital (mp3) files?

Reply #6
Here's the Sample

Pre-echo on digital (mp3) files?

Reply #7
I am no pro in evaluating the recording technics but this is indeed no pre-echo in the sense of an encoding error mp3 produces.
The trumpet starts to play silent ~2sec before the real entry. This sounds of cause like an Echo Pre the signal. Lots of noise in the recording, it must be taped.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Pre-echo on digital (mp3) files?

Reply #8
Can you verify that you simply cut a portion out using something like mp3direct cut and that you did not transcode it?

It's a shame that there is no lossless source for comparison, otherwise I'd feel more free in commenting.

I suspect that I do in fact hear pre-echo and that it has been exacerbated by lossy-lossy transcoding, even if the sample was losslessly extracted from the original file.

Pre-echo on digital (mp3) files?

Reply #9
greynol,

Apologies for posting this in the wrong part of the forum and yes, I used mp3direct cut to extract the sample and did not  transcode it.

Pre-echo on digital (mp3) files?

Reply #10
I can quite clearly hear two adjacent 'pre-echos' on the horns, each with a 1.8 second time period. 1.8 seconds just so happens to be the time for one revolution of a 33 RPM record. Hmm. Interesting.

Pre-echo on digital (mp3) files?

Reply #11
Oh, yes you are right, I do hear the faint horns coming in early, but I also believe that I hear smearing.

The spectral view (not frequency response!) of the track looks pretty hideous for 320 kbit mp3.  It looks like it may have lived its life once on minidisc (just a guess).

Pre-echo on digital (mp3) files?

Reply #12
IMO there is too little background noise for a vinyl rip... or it is excessively filtered.

If the original was tape, at a speed of 38cm/s the diameter of the spool when the print-through occured must heve benn approx. 22cm. In this case the tape has been stored wound forward - end of tape outside.
But anyway, 2 audible pre-echos are quite bad.


Pre-echo on digital (mp3) files?

Reply #14
How do you know that?


The total length is ca. 35 minutes, so if the position in question is a ca. 30 min there is not much tape following to produce a spool diameter of 20 cm
... but I do not know, I only guess

Pre-echo on digital (mp3) files?

Reply #15
If the original was tape, at a speed of 38cm/s the diameter of the spool when the print-through occured must heve benn approx. 22cm. In this case the tape has been stored wound forward - end of tape outside.

Actually, that's backwards - print-through from "tails out" storage will be post-echo, not pre, hence the common practice of storing analog tape tails out (which, evidently, was not done in this case).

Here is an article that explains it nicely.
"Not sure what the question is, but the answer is probably no."


Pre-echo on digital (mp3) files?

Reply #17
That odd distortion?

It sounds like either improperly calibrated noise reduction or a record bias issue.
"Not sure what the question is, but the answer is probably no."

Pre-echo on digital (mp3) files?

Reply #18
Does anyone care to comment on the transient sounds of the trumpet itself?  Am I imagining what I think I hear?


I think i hear what you mean. The poster asked about the 'coming horns' that is something different. I think you mean the added noise that sounds like a distortion or smearing, especialy at 10.5 or 11.2 sec. I hear it as a noise out of the audience, like someone has dry cough.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!


Pre-echo on digital (mp3) files?

Reply #20
These trumpets sound damn clear for a mp3 encoding. Without a lossless counterpart i doubt we will find a bad pre-echo here. Can you try to give a position where you think it is most obvious?

Guru! Where are you when needed?
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Pre-echo on digital (mp3) files?

Reply #21
As I said earlier...
It's a shame that there is no lossless source for comparison, otherwise I'd feel more free in commenting.

These trumpets sound damn clear for a mp3 encoding.
Careful with the sweeping statements; TOS #8 is looming.

Pre-echo on digital (mp3) files?

Reply #22
As I said earlier...
It's a shame that there is no lossless source for comparison, otherwise I'd feel more free in commenting.

These trumpets sound damn clear for a mp3 encoding.
Careful with the sweeping statements; TOS #8 is looming.


We often found artifacts especially on trumpets and other wind instruments with lame mp3 when you remeber.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

Pre-echo on digital (mp3) files?

Reply #23
I remember that artifacts were found on only one particular killer sample (which sounded very amateurish, apologies to the performer) and not in the context of music.  I also remember people making sweeping generalizations about how one should configure the encoder for an entire library based on this one sample and not being very happy about it. I also remember the issues with that sample getting fixed.

However, to support your point, has there been any improvement with harpsichord?

Pre-echo on digital (mp3) files?

Reply #24
I didn´t follow any mp3 development anymore since ~3.97 when i totally went lossless. Trumpet/wind instrument samples were still and issue with vbr then. Guruboolez once offered even several in his testing library. The sandpaper noise i found with voices were on these also. I also remember Master dibrom using a trumpet sample in the very beginning of lame tuning.
Most likely these are no problem anymore for modern encoders but it must be of the toughest sounds to encode.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!