Popular Post pkane2001 Posted April 9, 2019 Author Popular Post Share Posted April 9, 2019 4 hours ago, esldude said: I too think the gain matching needs improvement, and should use the whole file. I've also noticed some tracks have error concentrated either at both ends or toward the middle depending upon the track. And that stays consistent with different gear in use. I've one that seems to usually produce a great null at the beginning and then get worse gradually toward the end across multiple pieces of gear. I need to sit down and go over this again to make sure that is consistent. It does seem related to gear having a drift in gain levels. But you only see it once the nulls are pretty good to begin with. I do believe there's a drift in gain, at least in the files I've seen so far with this problem. If it's a linear drift, I should be able to measure and correct it. Will do some more testing soon. Interestingly, I was able to improve gain calculation by using the whole track, but also by eliminating very large differences from the calculation, using mostly the places where the two waveforms differ by a tiny amount. I have a theory as to why this may be working like this, but I'm afraid to generalize this to all cases. The alternative is using a few different gain computation methods, and then determining the one that results in the best null. It'll slow things down a bit, but should help improve the quality of the null. esldude, fas42 and Arpiben 3 -Paul DeltaWave, DISTORT, Earful, PKHarmonic, new: Multitone Analyzer Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 @fas42 Audacity 2.0.5 has the enhanced speed and gain changes. You can change the speed percent by as little as .00000003 while it will let you enter more digits if you get past this number nothing actually changes. So 7 zeroes and a 3 or higher will change speed. Gain changes can be as little as .000001 db. Again it will let you input more zeroes, but nothing actually happens. The current version only allows 2 zeroes past the decimal instead of 5. You can download 2.0.5 here. https://code.google.com/archive/p/audacity/downloads I have it running on an old laptop with Win 8. Don't know how it runs with Win 10. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
fas42 Posted April 10, 2019 Share Posted April 10, 2019 Thanks! I'm using 8.1, so should be no problem ... Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 Here is something interesting. Not sure what to make of it. The first two drift graphs are from a loopback on my Zen Tour vs the original music file. Done several days apart. These two are a Zen Tour loopback but with external clocking by a Focusrite 18i20 and then a Musical Fidelity Vlink. This one is with a March DAC feeding the Zen Tour so no loopback. Looks like the content of the file itself is creating a common pattern in the corrected drift plot. Below are two plots using a different snippet of music. First a ZT loopback and then March feeding the ZT. I've also checked with external clocking seeing the same pattern, but it is redundant to post them all. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
fas42 Posted April 10, 2019 Share Posted April 10, 2019 Very interesting results, Dennis ... Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 Here is more I don't know what to make of it. Using loopback for the same music file vs the original I see the same drift pattern shape. Only notice the scale of that pattern is 5 times greater. And using loopbacks from the 18i20 once again the same pattern only at an even larger scale. Not sure what if anything this means. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
pkane2001 Posted April 10, 2019 Author Share Posted April 10, 2019 4 hours ago, esldude said: Here is more I don't know what to make of it. Using loopback for the same music file vs the original I see the same drift pattern shape. Only notice the scale of that pattern is 5 times greater. And using loopbacks from the 18i20 once again the same pattern only at an even larger scale. Not sure what if anything this means. Can you try the same test with different music track? And sorry, I missed it, but what was different between these two captures? DAC only or DAC and ADC? -Paul DeltaWave, DISTORT, Earful, PKHarmonic, new: Multitone Analyzer Link to comment
Arpiben Posted April 10, 2019 Share Posted April 10, 2019 I have been out and busy those last days but should have more time left for a few tests. @esldude Dealing with your last tests and queries keep in mind that DW only correct linear drifts in time domain . In other words it is correcting only Frequency Offsets (steady) in Frequency Domain. Clocks have frequency offset, frequency aging, frequency accuracy, etc.... A linear frequency drift ( Frequency Domain) is a parabolic curve in time domain. Forgetting temperature, warming time (30min), DC fluctuations, vibrations, etc, when you compare two clocks or sampled signals based on them you need to take into account the linear frequency drift ( clock aging) or differential one ( 2 clocks ) especially when comparing signals recorded with big time lapse ( day(s), month). Hope it helps. Rgds. Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 3 hours ago, pkane2001 said: Can you try the same test with different music track? And sorry, I missed it, but what was different between these two captures? DAC only or DAC and ADC? I'm sorry I left out part of the info I intended to include in the last post. In the prior post I had loopback results vs the original file for the Zen Tour loopback, and the same with external clocking plus one with a separate DAC feeding the Zen Tour. The corrected drift had a consistent pattern in all of those. In my most recent post the first one is loopback results for a Focusrite Forte and loopback for a Focusrite 18i20. So everything was different. The scale of the corrected drift differed, but had the same pattern in it. Also in the post with all the drift graphs the last two were with a completely different track of music. Time of that track was nearly the same at just over a minute. With the other track the pattern of corrected drift was a different pattern. Though I only showed two of them like the other that pattern remains under several conditions and even using different DACs with different total drift. Which looks like the content of the file tested has a unique pattern that shows up in the corrected drift curve even with different gear. Which seems like the content of the file is effecting DW results. Then again, maybe there is something I don't understand. I'll post the results for a third piece of music in a few minutes. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 44 minutes ago, Arpiben said: I have been out and busy those last days but should have more time left for a few tests. @esldude Dealing with your last tests and queries keep in mind that DW only correct linear drifts in time domain . In other words it is correcting only Frequency Offsets (steady) in Frequency Domain. Clocks have frequency offset, frequency aging, frequency accuracy, etc.... A linear frequency drift ( Frequency Domain) is a parabolic curve in time domain. Forgetting temperature, warming time (30min), DC fluctuations, vibrations, etc, when you compare two clocks or sampled signals based on them you need to take into account the linear frequency drift ( clock aging) or differential one ( 2 clocks ) especially when comparing signals recorded with big time lapse ( day(s), month). Hope it helps. Rgds. But I am seeing that pattern in results several days apart, less than two minutes apart, with completely different gear. All the gear in these tests have been on for several hours. Some of the external clocks applied were on just a few minutes. Arpiben 1 And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 Third piece of music. First a loopback measure on the Zen Tour interface, and then a separate DAC feeding the ZT ADC. A pattern consistent across conditions with this piece of music. But a third pattern with the third piece of music. These two are not quite identical, but almost. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
pkane2001 Posted April 10, 2019 Author Share Posted April 10, 2019 5 minutes ago, esldude said: I'm sorry I left out part of the info I intended to include in the last post. In the prior post I had loopback results vs the original file for the Zen Tour loopback, and the same with external clocking plus one with a separate DAC feeding the Zen Tour. The corrected drift had a consistent pattern in all of those. In my most recent post the first one is loopback results for a Focusrite Forte and loopback for a Focusrite 18i20. So everything was different. The scale of the corrected drift differed, but had the same pattern in it. Also in the post with all the drift graphs the last two were with a completely different track of music. Time of that track was nearly the same at just over a minute. With the other track the pattern of corrected drift was a different pattern. Though I only showed two of them like the other that pattern remains under several conditions and even using different DACs with different total drift. Which looks like the content of the file tested has a unique pattern that shows up in the corrected drift curve even with different gear. Which seems like the content of the file is effecting DW results. Then again, maybe there is something I don't understand. I'll post the results for a third piece of music in a few minutes. It would be interesting if the content somehow has an effect on clock drift. Can't think of any mechanism that could cause that, inside DeltaWave or in the physical world. -Paul DeltaWave, DISTORT, Earful, PKHarmonic, new: Multitone Analyzer Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 27 minutes ago, pkane2001 said: It would be interesting if the content somehow has an effect on clock drift. Can't think of any mechanism that could cause that, inside DeltaWave or in the physical world. I cannot think of anything either. I also have noticed I can slice off either end, and get the same pattern minus the part sliced off. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
pkane2001 Posted April 10, 2019 Author Share Posted April 10, 2019 2 minutes ago, esldude said: I can think of anything either. I also have noticed I can slice off either end, and get the same pattern minus the part sliced off. I'd be curious to take a look at a couple original/loopback tracks using the same content, exhibiting the same residual drift. Would it be possible for you to share them? -Paul DeltaWave, DISTORT, Earful, PKHarmonic, new: Multitone Analyzer Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 Just now, pkane2001 said: I'd be curious to take a look at a couple original/loopback tracks using the same content, exhibiting the same residual drift. Would it be possible for you to share them? Yes, it might be later. Uploads on my pitiful internet are iffy this time of day. pkane2001 1 And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 https://www.dropbox.com/s/jrmx9khrajsxmpp/guy.7z?dl=0 https://www.dropbox.com/s/hp9wlnpbkxo35tc/Caram.7z?dl=0 https://www.dropbox.com/s/xwu8dlx9ujtnn3l/Fleck.7z?dl=0 Three different bits of music. Each one has the original, a loopbacked copy and a copy with separate DAC and ADC. Each bit of music will leave a very similar drift plot in the two copies. Each bit of music has a different pattern in that plot that seems to follow the music and not the device, clocks used etc. pkane2001 1 And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
blue2 Posted April 10, 2019 Share Posted April 10, 2019 52 minutes ago, esldude said: https://www.dropbox.com/s/jrmx9khrajsxmpp/guy.7z?dl=0 https://www.dropbox.com/s/hp9wlnpbkxo35tc/Caram.7z?dl=0 https://www.dropbox.com/s/xwu8dlx9ujtnn3l/Fleck.7z?dl=0 Three different bits of music. Each one has the original, a loopbacked copy and a copy with separate DAC and ADC. Each bit of music will leave a very similar drift plot in the two copies. Each bit of music has a different pattern in that plot that seems to follow the music and not the device, clocks used etc. What about using a different computer to play the music? 🎸🎶🏔️🐺 Link to comment
esldude Posted April 10, 2019 Share Posted April 10, 2019 6 minutes ago, blue2 said: What about using a different computer to play the music? Actually some of these were done that way. Some were with a Win10 machine and some with a Macbook pro. The loopbacks on the Zen Tour were done both ways. Those several days apart were first Mac and then later Win. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
Jud Posted April 10, 2019 Share Posted April 10, 2019 3 hours ago, esldude said: Third piece of music. First a loopback measure on the Zen Tour interface, and then a separate DAC feeding the ZT ADC. A pattern consistent across conditions with this piece of music. But a third pattern with the third piece of music. These two are not quite identical, but almost. Or - oh dear! 👹 - the clocking done with the recording somehow comes through in the results! 😳 fas42 1 One never knows, do one? - Fats Waller The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein Computer, Audirvana -> optical Ethernet to Fitlet3 -> Fibbr Alpha Optical USB -> iFi NEO iDSD DAC -> Apollon Audio 1ET400A Mini (Purifi based) -> Vandersteen 3A Signature. Link to comment
fas42 Posted April 11, 2019 Share Posted April 11, 2019 Dennis, just tried a speeding up and slowing down of the Marley track, in 2.0.5 Audacity - and it gives me the identical results, for a null. So, the algorithm is identical to the current one used - just, that I can now go for finer adjustments in terms of the numbers - and prod DW with that, . esldude 1 Link to comment
esldude Posted April 11, 2019 Share Posted April 11, 2019 10 hours ago, Jud said: Or - oh dear! 👹 - the clocking done with the recording somehow comes through in the results! 😳 I don't think that is what is happening. But I have a way to check it out actually. I'll see what it looks like. If clocking in the original recording is getting thru to the 1st generation copy, then I should see the same result comparing 1st and 2nd generations. My guess is the software has some relation to the file effecting how well it lines up, but even that doesn't seem likely. I'd like to know what is going on however. EDIT:OK the pattern of drift and the magnitude of it does get thru each subsequent generation. All the way up to the 8th generation. I don't suppose that says clocking of the recording is getting thru. It could be an interaction of the recording clock and the file data. And I don't know what the original clocking drift is. It is rather interesting and I don't know how that could happen unless the sample values interact with the alignment algorithms in some way. I don't know enough about it beyond that to be of much help. I don't think it is clocking actually because previously I've changed clocks and three of them give the same pattern on one of my ADCs. I feel sure all three clocks aren't drifting in the same pattern. So something else is going on. Also in these 8th gen files I had a separate DAC feeding my ADC so the two clocks aren't going to always synch up the same way or drift from the same starting point 8 times in a row. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
esldude Posted April 11, 2019 Share Posted April 11, 2019 Ok using Bob Marley in this case. Starts with original vs 1st generation, and finishes with 7th gen vs 8th generation. The pattern isn't identical, but the basic pattern of drift makes it all the way through them all. March DAC feeding Zen Tour ADC. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
esldude Posted April 11, 2019 Share Posted April 11, 2019 Here is original Phil Woods vs 1st gen followed by 7th gen vs 8th gen. March DAC feeding Zen Tour ADC. And always keep in mind: Cognitive biases, like seeing optical illusions are a sign of a normally functioning brain. We all have them, it’s nothing to be ashamed about, but it is something that affects our objective evaluation of reality. Link to comment
Jud Posted April 11, 2019 Share Posted April 11, 2019 6 hours ago, esldude said: I don't think it is clocking actually because previously I've changed clocks and three of them give the same pattern on one of my ADCs. My original remark was more than a little tongue in cheek (thus the emojis). But as long as you feel like exploring this a little, could you do it with some recordings you have made from "live" sources with different ADCs, to see whether characteristic patterns per ADC emerge? (Yes I realize there should be no distinction between a recording of a live source and any other analog source.) One never knows, do one? - Fats Waller The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein Computer, Audirvana -> optical Ethernet to Fitlet3 -> Fibbr Alpha Optical USB -> iFi NEO iDSD DAC -> Apollon Audio 1ET400A Mini (Purifi based) -> Vandersteen 3A Signature. Link to comment
pkane2001 Posted April 11, 2019 Author Share Posted April 11, 2019 6 hours ago, esldude said: It is rather interesting and I don't know how that could happen unless the sample values interact with the alignment algorithms in some way. I don't know enough about it beyond that to be of much help. I'll take a look at these, ASAP, as I also find the results very curious. The drift detection algorithm is fairly simple, the only way I can see that it might get fooled is if there are some long, near-perfectly repeating patterns in the audio files that can cause an alignment error. This is due to the way the cross-correlation algorithm works -- it finds one or more ways in which the samples can be aligned resulting in the best possible match. This can be fooled, but only if a large portion of the waveform repeats perfectly, meaning there are multiple ways to align the two waveforms with no way to distinguish between these multiple alignment points. This is extremely unlikely in any real event recording. In any case, DW will flag those as having multiple possible matches. This is most likely to happen with synthetic/generated tones. -Paul DeltaWave, DISTORT, Earful, PKHarmonic, new: Multitone Analyzer Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now