Expansion versus Normalization

[adaywayne]

Hi Paul. You can always use the corrected and uncorrected playback buttons to compare your processed and unprocessed tracks. I suspect you may have over processe a bit going for zero clicks. You know, one set of play back buttons plays only the current screen and the other set starts from the beginning of the track or tracks if "all tracks" are displayed below.
Stan

I would concur absolutely with this. It is not usually a good idea to keep reprocessing until the detected clicks drop to zero. This is because of the problem of 'false positives' The process of discriminating between clicks and music is never perfect and therefore, the more times you process, the more is the likelihood that you will pick up a lot of false positives. This may be what is causing the dull sound you are reporting.

I thionk the best approach is to process once and then to audition as citguy suggests. Only re-process if there are still audible clicks to be removed. And even then, it is better to remove them manually or to mark a block around them and use Block-SuperScan. That way it is only the region immediately around then rogue click that will be reprocessed.

Having said all that, another user has reported that he successfully does what you do but at detection threshold 2. If you do decide to do multiple superscans, then threshold 2 is to be preferred as the likelihood if false prositive is much reduced.

.

~~~~~~~~
Derek....let me clarify that a little. I scan first at a setting of 2. Then I audition and then superscan only if there are still audible clicks. After the first superscan I do the same again. I NEVER aim for 0 clicks, more in the region of fewer than 20 to 30, or so. My point was that I have never found it necessary to use a click detection setting higher than 2 to produce the desired result. The only filter I use is the rumble filter since I solved the hum problem at its source by proper grounding etc.

[Glenn]

Hi,

Arnie wrote:

Afraid I have to disagree. The way A/D converters work it is mathematically impossible to go to a higher signal level than 0dB (dBV or dBu).

Yes, I knew I fumbled that explanation when I hit the submit button. What I meant to say was that with peak limiting, the worst effects of digital clipping are avoided, allowing one to raise the perceived rms volume of the recording. The limited sound never actually reaches 0db, though the tips of the waveform are visibly clipped. Why this doesn't sound worse than it actually does is beyond me, but it appears to be done routinely on commercial pop CDs. The sound of these CDs is characterized by an exceedingly loud playback volume with very little dynamic subtlety. Incidently, without peak limiting, the sound is not listenable at all (loud clicks and pops), never mind whether it sounds better or worse than analogue clipping.

Glen, can you explain a bit more about the 50 to 60dB to spare? I thought all CDs were standardized at a peak of -! dB though I believe some pop music CDs now have a peak of 0dB. But 50 to 60???????

In this case, I'm simply refering to the wider dynamic range of digital recordings (approx. 0db to -90db, or 90 db overall) as opposed to vinyl, which has a range of approx. 0db to -30 to -40db (or 30-40 db) to the noise floor. With digital's much wider dynamic range, it would seem possible to record an LP to hard-disk at -50db and apply 50 db of gain normalization afterwards without compromising the signal quality. In practice however, the sound quality is nowhere near as good, and it is still (as it always was) better to record as close to 0db as possible.
Glenn

[adaywayne]

Hi,

Arnie wrote:

Afraid I have to disagree. The way A/D converters work it is mathematically impossible to go to a higher signal level than 0dB (dBV or dBu).

Yes, I knew I fumbled that explanation when I hit the submit button. What I meant to say was that with peak limiting, the worst effects of digital clipping are avoided, allowing one to raise the perceived rms volume of the recording. The limited sound never actually reaches 0db, though the tips of the waveform are visibly clipped. Why this doesn't sound worse than it actually does is beyond me, but it appears to be done routinely on commercial pop CDs. The sound of these CDs is characterized by an exceedingly loud playback volume with very little dynamic subtlety. Incidently, without peak limiting, the sound is not listenable at all (loud clicks and pops), never mind whether it sounds better or worse than analogue clipping.

OK, Glen, I understand what you meant now. You are talking about peak limiting before the input signal reaches the computer soundcard A/D converter. Yes, that does prevent clipping but it also reduces the dynamic range, of course.

Glen, can you explain a bit more about the 50 to 60dB to spare? I thought all CDs were standardized at a peak of -! dB though I believe some pop music CDs now have a peak of 0dB. But 50 to 60???????

In this case, I'm simply refering to the wider dynamic range of digital recordings

Oh- I read that as though you were saying that CDs had 50 to 60dB headroom over and above -6dB!

[Quote (approx. 0db to -90db, or 90 db overall) as opposed to vinyl, which has a range of approx. 0db to -30 to -40db (or 30-40 db) to the noise floor. With digital's much wider dynamic range, it would seem possible to record an LP to hard-disk at -50db and apply 50 db of gain normalization afterwards without compromising the signal quality. In practice however, the sound quality is nowhere near as good, and it is still (as it always was) better to record as close to 0db as possible.
Glenn[/quote]

A lot has been written about the dynamic range of CDs versus LPs. The argument is by no means settled,but the following article does, in my opinion, present a very good summary.


http://users.bigpond.net.au/christie/comparo/part4.html

Cheers,
Arnie

[Glenn]

Hi Arnie;
Last week you wrote about limiting-

Yes, that does prevent clipping but it also reduces the dynamic range, of course.

I agree, and as such it is a form of compression, but one than can provide a significant increase in the S/N ratio because of it's 'top-down' nature. My implementation of it has little effect on the audio, as I use it to control clipping only if it occurs. Dynamics expansion also increases S/N but the benefit of using it is offset by the lower record levels necessary. It also reduces the audibility of quieter instuments by further reducing their levels. My approach is to capture the original dynamics as closely as possible, using a time-honored recording technique. My CDs sound better this way.
Thanks for the link; it does give one much to think about.
Glenn

[adaywayne]

Hi Arnie;
Last week you wrote about limiting-

Yes, that does prevent clipping but it also reduces the dynamic range, of course.

I agree, and as such it is a form of compression, but one than can provide a significant increase in the S/N ratio because of it's 'top-down' nature. My implementation of it has little effect on the audio, as I use it to control clipping only if it occurs. Dynamics expansion also increases S/N but the benefit of using it is offset by the lower record levels necessary. It also reduces the audibility of quieter instuments by further reducing their levels. My approach is to capture the original dynamics as closely as possible, using a time-honored recording technique. My CDs sound better this way.
Thanks for the link; it does give one much to think about.
Glenn

~~~~~~~~~~
Glenn, could you explain a little more what you mean by "control clipping ONLY if it occurs? Surely, if it occurs (or has occurred) it is too late to correct it. Or do I misunderstand you?
Arnie

[Glenn]

Hi Arnie,
You asked-

could you explain a little more what you mean by "control clipping ONLY if it occurs?

Sure. A limited signal doesn't clip, at least not in the digital sense. It is limited before the onset of clipping. This might seem redundant if one defines digital clipping the same way as analogue clipping, but it isn't - at least not on my machine. What I get with a clipped signal is a sudden, high frequency polarity reversal that looks and sounds like a loud pop. When I apply limiting to just below 0db, this is prevented, allowing me a little leeway in case I missed the loudest peak while monitoring the vu meters. In fact, I can record at extremely high levels this way, producing the same obnoxiously loud playback levels (and distortions) that many pop recordings employ. In practice however, I use it strictly to prevent distortion in an otherwise acceptable recording.
Glenn

[adaywayne]

Hi Arnie,
You asked-

could you explain a little more what you mean by "control clipping ONLY if it occurs?

Sure. A limited signal doesn't clip, at least not in the digital sense. It is limited before the onset of clipping. This might seem redundant if one defines digital clipping the same way as analogue clipping, but it isn't - at least not on my machine. What I get with a clipped signal is a sudden, high frequency polarity reversal that looks and sounds like a loud pop. When I apply limiting to just below 0db, this is prevented, allowing me a little leeway in case I missed the loudest peak while monitoring the vu meters. In fact, I can record at extremely high levels this way, producing the same obnoxiously loud playback levels (and distortions) that many pop recordings employ. In practice however, I use it strictly to prevent distortion in an otherwise acceptable recording.
Glenn

That's what I thought, Glenn. I'm with you now.

[knighte]

Hi Glen,
I read about your recording techniques and your attempts to extract the sound you remember from old recordings. The freeware program, "Audacity" has a compressor located under "effect" in the menu. Recently I discovered that lowering the default threshold from -12 db to -6 db and checking the "normalize to 0 db" box seems to restore life to the music. I left the other two settings alone. Sounds good to these old ears, but I was wondering if others have an opinion. As far as I'm concerned, your recording as close to 0 db initially, does help the sound quality. In another thread started by dave.daniells I mentioned some results I had experienced with anti-skating. One thing I forgot was that wave corrector did not trigger on some noise if the anti-skating was out of whack. Has anyone else played with this?

[Glenn]

Hi Larry, you wrote:

In another thread started by dave.daniells I mentioned some results I had experienced with anti-skating.

and

If the tonearm is not in balance the stylus is exerting more force on one side of the groove thus making the tics or pops louder for whatever channel.

This is quite true, but what level of antiskate results in a balanced tonearm?
In the past I've used a blank track on my CBS test record provided for the purpose of setting antiskate and was quite happy with the result. Over the years I've gotten lazy and simply observed the action at the lead-in and lead-out grooves and averaged the result to set the antiskate.
Your comments prompted me to take an unusual (for me) approach to antiskate and use the vertical and lateral test tracks on the record to see what would happen. These are 100hz tone bursts at very high levels to test the tracking limits of a system. The vertical tracking test is in-phase mono and the lateral test is out-of-phase mono.
The results were quite difinitive. With antiskate adjusted as above, the vertical test produced a buzzing noise on the left channel (too much antiskate), with somewhat less noise on the lateral test. After considerable adjustment I've eliminated the tracking noise on both tests but was not expecting the setting required to do so: It turns out the correct antiskate force is virtually nil.
This may seem incorrect but consider this: While the stylus is happy to race to the spindle ahead of the groove, the pick-up and tonearm are nonetheless just as happy to follow along. As the stylus tracks the vertical undulations in the test track, any degree of antiskate causes lateral forces to be applied, pulling the stylus out of the goove, resulting in distortion. This may not be severe enough to cause noise, but the effect of the lateral force will cause phase distortion since the tracking within the pickup's magnetic field will not be perfectly vertical, as it should be.
I've since done several recordings and all of them exhibit little or no audible tracking distortion. As a bonus, the bass is quite solid, tighter and more dynamic. The dynamics as a whole are improved, resulting in better stereo separation. What I wasn't expecting is the emergence of a strong center channel, revealing vocal nuances like never before. Considering most vocals are mono in nature, I'd say this is proof of better phase coherence, and a tonarm in perfect balance.
The only drawback is that one needs to be verrrrrry careful when lowering the stylus. That, plus now I have to re-record everything I've done up to now. I'm not complaining mind you.
Glenn

[knighte]

Hi Glenn,
Your response made me laugh if only because I had gotten lazy myself. As you said, getting the stylus in the groove is touchy. This is the reason the anti-skating for my tonearm was set way to high. The higher force was making the start of playing an album easier. The force needed to insure proper tracking is going to vary from tonearm to tonearm. Mine happens to be as I would expect, right in the middle of the dial. Comparing the waveforms for three different settings on the same passage was all the convincing I needed. Also, it seems that Wave Corrector is able to do a better job at finding the impulse noise. This again was noticed while looking at the waveforms at the correction points. Sorry for setting you up with extra work.