This is totally different. I'm only interested in microseconds true impulsive pops and not wide areas where music has to be replaced. I can't relate to the comment below with a modern CPU (they also talk about running out of memory). 80 bit floats, in my experience do not help but sounds impressive.
"If there is a particularly problematic area, then Type-E manualinterpolation (perhaps left to run overnight) can help clean up small
regions."
Last edited:
Aah, that's The Art of Audio, which is somewhat related to another fair skill, of Electronics ...Hi, how do you explain that, sometimes, even with thicks and pops, an old LP can make you hear a REAL acoustic instrument (keyboard) in front of you ? The surface noises become less sensitive, as a sort of disturbance played BEHIND the music , far behind the scene ??
Simply put, the ear/brain mechanism in our heads is pretty sharp, on the ball most of the time. So when it hears simultaneous acoustic events, streams of sound, which are completely unrelated it separates them quite easily; in our mind we perceive them as distinct "things". So, a noisy LP sends to our hearing system two messages: the recorded musical performance, and the noise of the imperfect record surface. These are completely unconnected in any meaningful way, so if the information is clearly received then the mind pushes the uninteresting acoustic event, the ticks and pops, into the background; all you subjectively tune into is the musical event.
However, if the information is not clearly received, the system is not "transparent" enough, there's too much distortion, then the two acoustic messages get mixed up, they intermodulate, and the poor brain struggles to separate the streams of sound data. Similar to being tired, at a cocktail party, and trying to listen to one person amongst noisy talkers; you end up with a headache ...
So, the Art is to get the LP replay good enough so your mind has no trouble continuing to focus on what's interesting you at that moment, and disregarding extraneous 'noise' ...
Last edited:
This is totally different. I'm only interested in microseconds true impulsive pops and not wide areas where music has to be replaced. I can't relate to the comment below with a modern CPU (they also talk about running out of memory). 80 bit floats, in my experience do not help but sounds impressive.
interpolation (perhaps left to run overnight) can help clean up small
regions."
LOL. This manual was written back in the days when this particular process ran native on a 68k Mac. Later manuals don't go into details, that's why I posted this version. I didn't expect you to start nit-picking.
Sorry just joking, I originally said something to that effect that but erased it. I used to program on an old MAC and they were very pround of the native 80bit floats on the MOTO FPU.
Not much of a surprise, imo. Same same for graphic stuff, becomes pretty for 3d order b-splines.
Like Wadia, based on pretty looking output with some undesireble frequency content. In this case the frequency domain artifacts are very limited by minimizing the fix up window which is almost always the case.
For example 2msec ~.025" wide pop OUCH!
Sheesh. I just used the Creative DAC program. I'd find the two lobe impulse, and cut out one cycle of the dominant frequency wave...if 1Khz was playing at the time, I'd cut 1 mSec out, guessing at the boundaries which gave the smoothest look to the area. I did have to get creative at times when both channel content didn't line up well.
If I hadta guess, I's say it was 9th order interpolation..
jn
Fas42, you explained it extremely well. Going through these messages, and comparing it to my own personal experience with vinyl playback, shows that my critics aren't even in the same listening space as I have.
They LISTEN to clicks and pops, count them, and then go through all kinds of DIGITAL manipulations to smooth them over, and live with the shell of what is left, after digitization.
Now, here is where I might be able to relay some expertise, as to what works and what doesn't.
First, I have analyzed clicks on a vinyl record to a much higher degree than most here. Not by COUNTING, but by spectrum and transient analysis, with 'modern'(at the time) Tektronix test equipment, that I got, on loan from JBL (HK) back 35 years ago.
Of course, a simple record player, with a cheap cartridge, static prone mat, and dirty, abused records are going to click and pop, as well as audibly distort almost all the time.
You know, like when they demo 'digital vs vinyl' on the commercials.
But that is not MY reality, when I play a record, and I don't have the very best out there.
However, I do hear a 'tick' once in awhile, when playing a record, especially records that I have scratched, or the people who previously owned them scratched the records, but that is the usual extent of the diversion. You see, almost all of my records are USED, either from a used record store, or from somebody's thrown away collection put out on the curb. No how could I possibly enjoy vinyl playback under such conditions? That will be for another segment of this topic.
They LISTEN to clicks and pops, count them, and then go through all kinds of DIGITAL manipulations to smooth them over, and live with the shell of what is left, after digitization.
Now, here is where I might be able to relay some expertise, as to what works and what doesn't.
First, I have analyzed clicks on a vinyl record to a much higher degree than most here. Not by COUNTING, but by spectrum and transient analysis, with 'modern'(at the time) Tektronix test equipment, that I got, on loan from JBL (HK) back 35 years ago.
Of course, a simple record player, with a cheap cartridge, static prone mat, and dirty, abused records are going to click and pop, as well as audibly distort almost all the time.
You know, like when they demo 'digital vs vinyl' on the commercials.
But that is not MY reality, when I play a record, and I don't have the very best out there.
However, I do hear a 'tick' once in awhile, when playing a record, especially records that I have scratched, or the people who previously owned them scratched the records, but that is the usual extent of the diversion. You see, almost all of my records are USED, either from a used record store, or from somebody's thrown away collection put out on the curb. No how could I possibly enjoy vinyl playback under such conditions? That will be for another segment of this topic.
Thin and shallow scratches are in the 40-80us. Thick and deep ~ x2-3 of that. Debris and dirt ~ x5 but have the highest amplitude, thus cause high dV/dt.
Below are some light to moderate scratches (the far right image shows groove wall damage)
If I hadta guess, I's say it was 9th order interpolation..
11th +
George
Attachments
If I hadta guess, I's say it was 9th order interpolation..
jn
Hmmm, how do you figure that? IIRC that means a ninth order polynomial fit. As you see in George's first pics there is essentially no frequency content in the region of the pops so a third order b spline is nearly perfect.
ummm, Scott?
9th order...by eye?? I had considered saying 42nd order, but that was too obvious..
jn
Much as I hate to admit it, I do the same thing as jn, by eye. I'm too stupid to write 3rd order spline interpolation software. Goldwave has some terrific features to allow removal of sharp artifacts, whether ticks and pops on LPs or (ummm) overenthusiastic applause too near the microphone in digital recordings.
The reality is that the duty cycle of most ticks is really small, and there's a lot of sonic latitude when removing them and smoothing that specific part of the waveform. I suppose it's worse with typical MCs and their high Q resonances, but I avoid those sorts of things.
The reality is that the duty cycle of most ticks is really small, and there's a lot of sonic latitude when removing them and smoothing that specific part of the waveform. I suppose it's worse with typical MCs and their high Q resonances, but I avoid those sorts of things.
Hi, how do you explain that, sometimes, even with digital distortion , a CD can make you hear a REAL acoustic instrument (keyboard) in front of you ? The surface noises become less sensitive, as a sort of disturbance played BEHIND the music , far behind the scene ??
May the devil take my soul
May the devil take my soul
Wait...you apply third order interpolation to the walls of the vinyl?? No coffee that day!!
Sheesh, I thought I was crazy just using the mouse...
You are not the first, nor will you be the last, to call me unstable..
Get in line..
jn
Hi, how do you explain that, sometimes, even with thicks and pops, an old LP can make you hear a REAL acoustic instrument (keyboard) in front of you ? The surface noises become less sensitive, as a sort of disturbance played BEHIND the music , far behind the scene ??
I'd say that clicks and pops are more a foreground noise, owing to the lack of reverberation, and separable by the mind-brain thus.
A good recording to demonstrate the keying-in to a sense of recorded source distance: Stockhausen, Kontakte. It's anything but realistic, an early tape piece, but has some passages that emphasize the effect of variable reverb.
You should try hull design software.
Set it to 2nd order b-splines, and life is fine.
Pick 3d order, just a minor inconsistency, and what was a smooth shape a second ago, turns all martian.
(forget to make a backup after each fairing step, and better to replace the sugar in the coffee by valium
)
I am not going to quibble whether the click is in the foreground or in the background. Let's try to MINIMIZE the distraction and the annoyance of something that we have to expect, even with the best of set-ups.
Now, first, let me talk about a bit of experience over the last 45 years.
In 1968, I worked for the Professional Audio tape recorder division at Ampex. Because I was an employee, I could get my very own 15ips 1/2tr (with a switchable 1/4 tr head) for pre-recorded tapes, Ampex pro recorder. My associates, even at Ampex, thought I was 'over the top'. Why not just buy a consumer player with slimmer tracks and slower speeds?
Well, I thought, just like many here, that I could SAVE my vinyl records, by recording them at 15ips, 1/2tr with little or no obvious degradation of the sound quality. It did not work out. The record, virtually always, sounded better, and the tape copy was not as good. However, IF I used the tape recorder to record LIVE music, then the tape recorder master copy was better. Why?
When I was visiting the standard tape department, an old hand at recording, vinyl and tape, asked me if I had ever heard a DIRECT DISC vinyl recording? No, I had not, but I was assured by my more experienced colleague that it was BETTER than magnetic recording. It took several years before I could prove this for myself, when Sheffield came out with their direct disc recordings. I still have some, today. Of course, I have played a vast number of vinyl recordings that initially came from analog master tapes. They are not as good as true, direct disc, but they can be pretty good sounding, in any case. Well, what does this all mean to this discussion, today?
Here, I see that people, who are usually my critics, are trying what I tried to do 45 years ago! They are going to take a vinyl recording, with some 'average' playback equipment, and digitize it, to 'save' the vinyl from further abuse. THEN, they find the ticks and pops so annoying, that they actually use programs to get rid of them.
I think the best solution is to get as close to the original program as you can, and MINIMIZE the problems by looking at the problems, directly, rather than 'fixing' them by putting the program material through another layer of recording. I will try to talk about some factors that I have found to work, in a future input.
Now, first, let me talk about a bit of experience over the last 45 years.
In 1968, I worked for the Professional Audio tape recorder division at Ampex. Because I was an employee, I could get my very own 15ips 1/2tr (with a switchable 1/4 tr head) for pre-recorded tapes, Ampex pro recorder. My associates, even at Ampex, thought I was 'over the top'. Why not just buy a consumer player with slimmer tracks and slower speeds?
Well, I thought, just like many here, that I could SAVE my vinyl records, by recording them at 15ips, 1/2tr with little or no obvious degradation of the sound quality. It did not work out. The record, virtually always, sounded better, and the tape copy was not as good. However, IF I used the tape recorder to record LIVE music, then the tape recorder master copy was better. Why?
When I was visiting the standard tape department, an old hand at recording, vinyl and tape, asked me if I had ever heard a DIRECT DISC vinyl recording? No, I had not, but I was assured by my more experienced colleague that it was BETTER than magnetic recording. It took several years before I could prove this for myself, when Sheffield came out with their direct disc recordings. I still have some, today. Of course, I have played a vast number of vinyl recordings that initially came from analog master tapes. They are not as good as true, direct disc, but they can be pretty good sounding, in any case. Well, what does this all mean to this discussion, today?
Here, I see that people, who are usually my critics, are trying what I tried to do 45 years ago! They are going to take a vinyl recording, with some 'average' playback equipment, and digitize it, to 'save' the vinyl from further abuse. THEN, they find the ticks and pops so annoying, that they actually use programs to get rid of them.
I think the best solution is to get as close to the original program as you can, and MINIMIZE the problems by looking at the problems, directly, rather than 'fixing' them by putting the program material through another layer of recording. I will try to talk about some factors that I have found to work, in a future input.
- Status
- Not open for further replies.