Mac Mini, iTunes & high resolution audio

[kirkmc]

Two things. First, are you sure that, in Audio Hijack, you've got the sample rates set correctly? Second, I've just opened a high-res file in Audacity and used the Analyze > Plot Spectrum feature that displays the Frequency Analysis window. Is this what you're using to look at the files? Many of my files - 24/96 files - show hardly anything above 7 KHz. (These are original files, not files that I've played through iTunes and captured.) Curiously, one file shows nothing from about 17 KHZ and then a hump around 28-24 KHz. That looks almost artificial.

 

If this is the process you're using, then I'll do some testing with playback and capture and see what I can find out.

[kirkmc]

I had a discussion with someone at a record label, who looked at some of the files I tested - they were his label's files - using Adobe Audition. He sees totally different information. For example, the file that peaked in Audacity at 7 KHz shows a full range up to 48 KHz in Audition. The same for other files.

 

Audacity is clearly not the right tool to test this. I would therefore suggest that your testing is wrong, or at least flawed; not your fault, but your tool isn't reliable. One would have to use Audition or something similar to analyze the files.

[goldsdad]

Two things. First, are you sure that, in Audio Hijack, you've got the sample rates set correctly? Second, I've just opened a high-res file in Audacity and used the Analyze > Plot Spectrum feature that displays the Frequency Analysis window. Is this what you're using to look at the files? Many of my files - 24/96 files - show hardly anything above 7 KHz. (These are original files, not files that I've played through iTunes and captured.) Curiously, one file shows nothing from about 17 KHZ and then a hump around 28-24 KHz. That looks almost artificial.

 

If this is the process you're using, then I'll do some testing with playback and capture and see what I can find out.

 

Well, I'd have gladly gone to the trouble of answering your questions and demonstrating how to use the tools correctly, but seeing your subsequent post where you condemn my testing and tools before I've answered, I'll not waste the time and effort.

[goldsdad]

I had a discussion with someone at a record label, who looked at some of the files I tested - they were his label's files - using Adobe Audition. He sees totally different information. For example, the file that peaked in Audacity at 7 KHz shows a full range up to 48 KHz in Audition. The same for other files.

 

Audacity is clearly not the right tool to test this. I would therefore suggest that your testing is wrong, or at least flawed; not your fault, but your tool isn't reliable. One would have to use Audition or something similar to analyze the files.

 

I suggest your use of Audacity is flawed, rather than Audacity.

[goldsdad]

Just to show the readers that Audacity does not lose content above 7 kHz (unless you misuse the software somehow), here are screenshots of a 192 kHz fs file (from Barry Diament's Soundkeeper Recordings) in Audacity and Adobe Audition. Click images to enlarge.

 

 

[kirkmc]

Because I point out that the software isn't working as you say, you get tetchy? Seriously?

[kirkmc]

I tried with several files in Audacity, and a number of them don't display the full frequency range. It's not my use of Audacity, it's simply selecting a menu item. The software is clearly flawed.

 

It does correctly display some files, but for many it doesn't.

[goldsdad]

Because I point out that the software isn't working as you say, you get tetchy? Seriously?

 

Lol! The software does work as I say. I've shown an example of it working as well as Adobe Audition. Now, since you know the software doesn't work correctly, let's see your example of that.

[kirkmc]

Here:

 

 

I no longer have the screenshot from Audition, but it shows what it should, frequencies up to 48 KHz.

 

And I have a few others from Audacity that show frequencies stopping will below where they should.

[goldsdad]

I tried with several files in Audacity, and a number of them don't display the full frequency range. It's not my use of Audacity, it's simply selecting a menu item. The software is clearly flawed.

 

It does correctly display some files, but for many it doesn't.

 

Really? You'll have no trouble providing examples then...

[kirkmc]

Don't you see one just above?

[goldsdad]

Here:

 

[ATTACH=CONFIG]20550[/ATTACH]

 

I no longer have the screenshot from Audition, but it shows what it should, frequencies up to 48 KHz.

 

And I have a few others from Audacity that show frequencies stopping will below where they should.

 

 

That does look peculiar, but I've honestly never seen anything like it out of literally hundreds of lossless music files examined in numerous versions of Audacity over the last five years. That's not to say that I don't take your evidence of a problem seriously. It is strange that you've encountered several examples and I've encountered none. Have you seen this in multiple versions of Audacity? Has it happened with OS X other than Yosemite?

 

A comparison with Audition would have been useful. Would you mind providing me with the file so I can try to determine what's happening?

[goldsdad]

Don't you see one just above?

 

Sorry, I was typing as you posted so hadn't yet seen it.

[goldsdad]

Kirk, since Audacity is not acceptable to you, here is an example illustrated by Adobe Audition screenshots of the dual resampling that could occur when using iTunes 10.2.2 and changing the output rate in Audio MIDI Setup while iTunes is open. Playback was captured by Audio Hijack Pro set to automatically capture at the system output rate.

The Audacity screenshots are to show that it can present the same information as Audition except that it cannot display levels below -145 dBFS in its frequency analysis plot, hence the minor discrepancies in stages 5 stage 6.

1. Close iTunes if it is open.

Set output rate in AMS to 192 kHz.

Open iTunes.

Play 192 kHz fs file and capture: captured audio has 192 kHz fs with content limited to Nyquist frequency of 96 kHz, of course.

 

 

2. Change output rate in AMS to 96 kHz.

Play 192 kHz fs file and capture: captured audio has 96 kHz fs with content limited to Nyquist frequency of 48 kHz.

 

 

3. Change output rate in AMS to 48 kHz.

Play 192 kHz fs file and capture: captured audio has 48 kHz fs with content limited to Nyquist frequency of 24 kHz.

 

 

 

 

4. Close iTunes if it is open.

Set output rate in AMS to 48 kHz.

Open iTunes.

Play 192 kHz fs file and capture: captured audio has 48 kHz fs with content limited to Nyquist frequency of 24 kHz, of course.

 

5. Change output rate in AMS to 96 kHz.

Play 192 kHz fs file and capture: captured audio has 96 kHz fs but content beyond 24 kHz is mirroring due to an upsampling from 48 kHz fs to 96 kHz fs.

 

 

 

6. Change output rate in AMS to 192 kHz.

Play 192 kHz fs file and capture: captured audio has 192 kHz fs but the content beyond 24 kHz is mirroring due to an upsampling from 48 kHz fs to 192 kHz fs.

 

 

 

 

 

The results of stages 5 and 6, notably the mirroring seen at 24 kHz in 5, and the mirroring at 24 kHz and 48 kHz in 6, imply that iTunes has resampled to the rate that was the system output rate when Tunes was opened, regardless of the current system output rate, and then the OS has resampled to the rate that currently is the system output rate.

[goldsdad]

Kirk, an explanation you may not like. Audacity has not misbehaving for you. Your problem has been a case of user ignorance.

Here is some solo piano analyzed by Adobe Audition:

 

The same analyzed by Audacity with default minimum level, which will look familiar to you:

 

And again analyzed by Audacity, but with minimum level of -145 dB:

 

To get the second Audacity result, i.e. a match for the AA result, make the change highlighted below:

 

 

As I said before, your usage of Audacity, and not Audacity itself, was flawed. Maybe next time you're about to condemn someone's tools, you might make sure of your facts first.

 

 

[kirkmc]

Sorry I didn't get back to you yesterday; I was busy.

 

First, I wasn't "condemning your tools;" I simply found that I couldn't reproduce what you were saying. Don't take things so personally.

 

Second, I was using Audacity's default settings, and I checked their documentation about the various "windows" you could display when viewing the spectrum. Nothing there said to change anything in the settings. But that, plus what you see in Audition, confirms that there is a difference in the sample rate according to the changes you make.

 

This said, I still hesitate to say that it's iTunes that does the resampling; I don't think iTunes has the code to do that. iTunes simply tells CoreAudio what to do (at least on a Mac). I suspect that, for some odd reason, CoreAudio is getting confused when you do the steps you took above. I don't have the technical know-how to find which app is doing that. This said, I'll ask someone who might know a bit more.

 

BTW, those screenshots in steps 5 and 6 above; is it normal that, instead of a more gradual rolloff, it's so sudden in the examples you show? I don't think that's normal; I would expect that for most audio the frequencies decrease much less suddenly. I saw one file where there's nothing, then a big hump around 30KHz.

[Jud]

This said, I still hesitate to say that it's iTunes that does the resampling; I don't think iTunes has the code to do that. iTunes simply tells CoreAudio what to do (at least on a Mac). I suspect that, for some odd reason, CoreAudio is getting confused when you do the steps you took above. I don't have the technical know-how to find which app is doing that. This said, I'll ask someone who might know a bit more.

 

 

With respect, you have already asked someone who knows a bit more. Owen (goldsdad) has been on these forums a long time. I have found what he says to be quite reliable and to be based on careful, thorough research.

 

This isn't to discourage you from asking anyone else you want to, of course.

[kirkmc]

With respect, you have already asked someone who knows a bit more. Owen (goldsdad) has been on these forums a long time. I have found what he says to be quite reliable and to be based on careful, thorough research.

 

This isn't to discourage you from asking anyone else you want to, of course.

 

That my be the case, but it makes no sense that iTunes is able to change sample rates (or anything else, for that matter). On Macs, all that is done by CoreAudio.

[Daudio]

I wasn't "condemning your tools;" I simply found that I couldn't reproduce what you were saying. Don't take things so personally...

 

Second, I was using Audacity's default settings, ... Nothing there said to change anything in the settings...

 

Oh, come on Kirk, don't try and weasel out. You jumped on goldsdad pretty damn hard. Now you have been proved wrong, and more graciously then your accusations IMHO. So, man up, and 'Mea culpa', if not apologize !

 

 

This said, I still hesitate to say that it's iTunes that does the resampling; .. I suspect that, for some odd reason, CoreAudio is getting confused when you do the steps you took above. I don't have the technical know-how to find which app is doing that. This said, I'll ask someone who might know a bit more...

 

That my be the case, but it makes no sense that iTunes is able to change sample rates (or anything else, for that matter). On Macs, all that is done by CoreAudio.

 

 

And what makes you think that ?

[kirkmc]

Oh, come on Kirk, don't try and weasel out. You jumped on goldsdad pretty damn hard. Now you have been proved wrong, and more graciously then your accusations IMHO. So, man up, and 'Mea culpa', if not apologize !

 

Jumped on? In what way? I said, very politely, that I wasn't able to reproduce what he was seeing. Geez, you guys are awfully sensitive...

 

And what makes you think that ?

 

Efficiency. It makes no sense to do the same thing twice in an OS, which is what this would mean. it's no good for battery life of portables, and it's no good for CPU usage for all computers.

 

Plus, resampling or doing any processing to audio twice would increase the risk of audible artifacts.

[Daudio]

Jumped on? In what way? I said, very politely, that I wasn't able to reproduce what he was seeing. Geez, you guys are awfully sensitive...

 

OK, I went back and loked at your posts, and what I see is such bold insistence that you were right, and he was wrong, that I took away a very negitive attitude. No, you didn't attack goldsdad, but 'polite' ?, I never got that impression. I still think it would be good for your own character so openly admit you were wrong (IMNSHO).

 

 

Efficiency. It makes no sense to do the same thing twice in an OS, which is what this would mean. it's no good for battery life of portables, and it's no good for CPU usage for all computers.

 

Plus, resampling or doing any processing to audio twice would increase the risk of audible artifacts.

 

OK, so you just surmise that without any real evidence. Not that that's wrong, but just so we are clear

[kirkmc]

Well, yes, I was wrong, because I was using Audacity with the default settings. I plead ignorance.

 

As for my surmising, it's no more a supposition than to say that it's iTunes that's doing the resampling. I'm trying to find out, from a developer I know, if there's any way to confirm exactly where and when the resampling is done.

[Musicophile]

To calm things a bit here, let me suggest to simply buy Musicscope. It does all the necessary analytics and spits out a great looking report.

 

Sure you can get most of the same with other tools but this is all in one place. I use it all the time.

 

https://www.xivero.com/musicscope/

[kumakuma]

To calm things a bit here, let me suggest to simply buy Musicscope. It does all the necessary analytics and spits out a great looking report.

 

Sure you can get most of the same with other tools but this is all in one place. I use it all the time.

 

https://www.xivero.com/musicscope/

 

+1

 

In my opinion, anyone that buys HR music on a regular basis should own this application.

[goldsdad]

That my be the case, but it makes no sense that iTunes is able to change sample rates (or anything else, for that matter). On Macs, all that is done by CoreAudio.

 

I don't understand why you say it makes no sense that iTunes 10.2.2 (and other versions which I no longer have) was able to resample. Audirvana, Decibel, Fidelia, Adobe Audition, Audacity, etc, etc, etc, are all able to resample.