Ideas for Version Seven

[Craig Maier]

Yeah, Tim - - - that's the ticket! R U sure that you do not work in our marketing department?

[Audyossey]

What is scrubbing

what's 'scrubbing?'

Scrubbing is a type of timeline playback that gives you precise control over the speed and direction of playback. The effect is like manually rocking analog audio tape across the playback head. You can hone in on an anomaly which doesn't leave an easily identifiable fingerprint in the waveform.

It can be implemented in multiple ways; for example a button that you move left or right in relation to a reference mark. The file plays forward or backwards starting at a low playback speed and the further you move from the reference, the faster it plays. A more eloquent approach involves playback following the cursor as you manually drag it along the waveform. Other alternatives include trackwheels, the "J-K-L" keys, or special keyboards.

HB

{Whoops. Somehow I came into the middle of the thread and didn't see the more concise answer already given. Pardon the redundancy.}

HB

[MBwa]

Monophonic record declick feature for DC7

There is one feature that I have regularly wished for when working on mono records.

I record them in stereo and often find it helpful to use the File Conversion to get L+R and L-R. The L+R has all the music and some of the noise while L-R has just the noise. The L-R gives a beautiful indication of where all the clicks and pops are, but there seems to be no way to make use of that information other than to manually look at L-R and then interpolate the L+R channel :-(

So what I think I would like is an addition to EZ-CLean and the Declickers that would consist of a Mono button so that when I turn it on, it would take my L and R and use L-R to detect clicks and pops and then correct or interpolate those locations to give the corrected (L+R)/2 back in L and R.

Perhaps something similar could be done with the hiss in the EZ-Clean or Continuous Noise Filter to identify the noise spectrum from L-R and use that to correct L+R data and put the signal back into L and R.

Or have I been overlooking a current feature of DC6?

[Dan McDonald]

I think L-R leaves just the noise that is only on the left channel; if the noise is on both channels, L-R takes it out (along with the music), so that if you clean the file, it's not as clean as cleaning both channels. At least, that's what I've found.

[Craig Maier]

Hi MBWA -

Yeah, that does sound good in theory, but it just does not work in practice. Our very first version of the software developed about 10 years ago allowed that, but it does a terrible job of detection getting only a few of the clicks. Each channel must be used as its own reference to effect a good click detector. I think that the reason for this is that the signal used in the detector must be in the proper phase as well as containing the basic amplitude event in order to work.

BTW - one of the file conversions will give you L+R on one channel and L-R on the other if you just want to see the difference signal.

- The L+R signal represents lateral movements of the phonograph stylus
- The L-R signal represents vertical movements of the phonograph stulus

[Dan McDonald]

One of the things that I've found that works nicely with mono is to save residuals from the CNF and use that as one channel, then go back to the original channel as the left channel, the residual as the right channel, and then do the subtraction. It works for me because it gives me a lot more freedom to figure out how much of the noise is optimal to remove. I just adjust the l-r on the channels, and that controls how much noise is taken out. I find I'm more conservative this way because I hear the introduction of artifacts a little more clearly, and make adjustments that way.

Anyway, I find that works for me with mono material sometimes when other things just don't seem to help.

[ramiro]

DC7-Expanding Punch and Crunch, Bugfixes

Hi DJBohn,

...I guess you want a dual thresholding expander and compressor. I am also still trying to figure out if the existing Dynamics Processor could be made to do that since it has independent Expansion and Compression systems. What I am also wondering about is how to avoid the pumping effect as it slides between the two thresholds. Maybe it needs to have a "soft knee" at both switchover points. Maybe Rick has some ideas.

If you want to see an elegant implementation of this function, take a look at x.

It uses a graph of user-adjustable points, akin to the one in DC6's continuous-noise filter, so that you can expand or compress any point on a dynamic curve in one operation.

When the slope is 1:1, there is no change. For example, you can taper off the top, and put a bump near the bottom, to simultaneously compress the louder parts, and boost the soft parts, without disturbing the middle-level volumes. Continuously adjustable, and no problems with discontinuities, (I think that's what you meant by "pumping effect"?).

It does not, however, allow you to select different frequency bands the way DC6 does-it works on the entire spectrum at once.

Other possible suggestions for DC7:
Fix the help function that should pop up the relevant help entry for each filter.
Fix the binary zooming out bug
Fix the program-crashes-when-saving bug

[Craig Maier]

ramiro,

Your inputs are noted. Thanks.

[ramiro]

Recording Timer

How about a timer option, so you can start a tape playing, and walk away while it records so you don't have to worry about eating your HD?

VOX more or less works for that, but a timer would be more secure. Set it for, say, 40 min. and walk away.

For other uses, being able to set it for a particular time would give VCR-like capabilities.

[Craig Maier]

You already have a Timer Recorder which is located under the Edit Menu. See, we are way ahead of you.

[TimG]

Buddy have ya got a time(r)?

Though, a simplified version within the Record File window (ie. direct entry of the duration) could be beneficial in recording tracks of known time-length.

[Craig Maier]

OK, we'll look at it. I guess it would be sort of like the "sleep" mode on an alarm clock, right?

[Dan McDonald]

Where you could put an lp on and set record for 22 minutes or something like that... very handy.

Dan

[Dan McDonald]

Auto-Spectrum CNF

First - a compliment about the auto-spectrum CNF. I hadn't used it seriously before, but found it did a really nice job on some messy cylinders.

I noticed that when you 'freeze' it, the controls stay lit (not grayed out). I assume they don't have any effect, though, right? If not, can you make them gray out in the new version?

And, a slight suggestion for #7 - would it make sense to make the attenuation control still work, even though the other parameters are frozen? I think that would be helpful in 'fine-tuning' the filter after you've let it figure out the noise.

[Craig Maier]

As I recall, I think that freezing it just freezes the noise fingerprint stopping the adaptation process. It may be that we left the rest of the controls operative, thus not greyed out, although I am not 100% on that one. Been a while.