[Elader]

OK
 
Macbook Pro
mydac, USB
preamp, amp, ML speakers
 
 
Switch on back of mydac set to 2
 
at 192 set on the mac midi, I am hearing the occasional crackle. I believe it's because neither itunes nor MOG lets me do this.
 
at 96, everytihng seems fine.
 
In general, what the heck should I set the mac to? Does it depend on the music? The player?
When MOG says they stream at 320, does this have anything to do with the MIDI setting?

 

[obobskivich]

Could you provide units with those numbers? Pretty please.

And no, 320k streaming has nothing to do with what I'm assuming to be sampling frequency (fs). Set fs to something realistic, like 48khz. This may or may not be related to the "crackle" as well.

 

[sterling1]

Set player to play the highest resolution files in your computer library at their native rate ie 24/96.

 

[Elader]

Ah yes units.... something a molecular biologist ought to remember...
 
I suppose what is confusing me is
 
a) what is the 'normal' bit depth and sampling frequency of internet streamed music? Does it matter if I listen through web browser or the spotify or mog app? For that matter, am I correct in assuming all my old crappy MP3s are sampled at 44K?
 
b)The DAC (musicstreamerII and MyDAC) 'knows' what settings to use based on the bitstream from the USB? I ask this because the MSII does have indicator lights. It stays glued on 96K when I set the computer output to 96K regardless of the source material.
 
c) Is the dac or the computer 'upsampling' - I suppose it would be nice to know. I am one of those guys who don't believe in creating something out of nothing - and would rather play music at the sampling rate it was recorded in.
 
d) Is it true that for the player to adjust its output based on the sampling rate and bit depth of the music file, you have to use another player that piggybacks on itunes? What happens if you dont? e.g. if I play a crappy MP3 through Itunes with the MIDI set at 96K/24bit? What does the mad do?
 
Sorry about all this, but if you want to learn about gene regulation, I am your guy.

 

[Elader]

and this is the source of a lot of my confusion.... does anyone really change settings in MIDI when they switch between files?
 
from micromega
 
To setup the computer go to Finder / Utilities / Audio MIDI setup.

On the above example after selecting the appropriate peripheral « MICROMEGA USB 2.0 Audio Out » and having clicked on the « Ouput » tab, you must set the sampling frequency of the file you want to play. In our example it is 96 kHz.

If you want to play a 192 kHz sampling frequency file you will need to do this setup again.

By the way, in Audioclass 1, the problem is exactly the same.

In any case, we strongly recommend that you use « Audioclass 1 » mode for all files wich sampling frequency between 44.1 kHz and 96 kHz.

 

[mich41]

Most stuff is released at 44kHz/l6bit because this format is used on CDs and because it's sufficient.

Your OS converts all streams from all applications to one common format in order to mix them. Conversion of 16bit audio to 24bit is completely harmless. Sample rate conversion can be done in few ways - the dumbest one is simple linear interpolation and it makes very little difference even if you know what to listen for, others are better and I can't hear them at all to be honest.

There ought to be some way to bypass this mixer, but I don't know how to do it in OSX and iTunes may not support it at all. This has an extra advantage/disadvantage of disabling all other sounds during music playback. If you don't care or need to be able to play other sounds, switch to 96/24 or 44/24 if you don't have any 48k/96k material.

 

[chewy4]

The normal bit depth and sampling frequency of any kind of music other than special audiophile music(like stuff from hdtracks) is 16 bit 44.1kHz. This sampling rate and bit depth allows for a dynamic range of 96dB and reproduction of frequencies up to 22kHz.
 
192kHz sampling rate can cause problems, even if that is the file's native sample rate. Devices aren't designed to produce frequencies that high up, and it puts more stress on your computer for no benefit. I suggest avoiding it.

 

[Elader]

Thanks you guys, this makes a lot of sense.
 
So unless Ive got a source of higher quality material (e.g. HDtracks), I can assume streaming audio or old crappy MP3 files can be done at the 44K rate.

 

[obobskivich]

Thanks you guys, this makes a lot of sense.

So unless Ive got a source of higher quality material (e.g. HDtracks), I can assume streaming audio or old crappy MP3 files can be done at the 44K rate.

It would be safe to assume that the bulk majority of music is 44.1khz and 16-bit, yes. DVDs tend to be 48khz, and the default sampling settings for most Windows computers (and I assume OS X as well) will be either 16/48 or 24/48, to accommodate that. I'd honestly suggest leaving it at default, unless you have a lot of high resolution files (but I'd also suggest reading up on what that "high resolution" actually means for you). SRC will be performed to bridge whatever "mismatch" occurs, and that's correct behavior (the whole "native playback" jag isn't worth the time imho). I wouldn't be so quick to dismiss mp3 or CD quality either - sure, it isn't "new and improved" and doesn't align with modern marketing speak, but the better question is "can you notice a difference?"


Internet streaming - there's no single answer. Generally yes you're talking 16/44.1, but it's usually heavily compressed to conserve bandwidth; the original might be 1411k, but the stream might only be 96k. That's somewhere where you *should* be concerned, as over-compression will produce audible artefacts. There isn't much you can do with streaming, but it's something to consider when buying downloads or ripping music.

 

[nicko7i]

Quote:

 
a) what is the 'normal' bit depth and sampling frequency of internet streamed music? Does it matter if I listen through web browser or the spotify or mog app? For that matter, am I correct in assuming all my old crappy MP3s are sampled at 44K?
 
b)The DAC (musicstreamerII and MyDAC) 'knows' what settings to use based on the bitstream from the USB? I ask this because the MSII does have indicator lights. It stays glued on 96K when I set the computer output to 96K regardless of the source material.
 
c) Is the dac or the computer 'upsampling' - I suppose it would be nice to know. I am one of those guys who don't believe in creating something out of nothing - and would rather play music at the sampling rate it was recorded in.
 
d) Is it true that for the player to adjust its output based on the sampling rate and bit depth of the music file, you have to use another player that piggybacks on itunes? What happens if you dont? e.g. if I play a crappy MP3 through Itunes with the MIDI set at 96K/24bit? What does the mad do?

((a))  44,100 samples per second at 16 bit depth. I've verified this for iTunes, MOG and Spotify.  
 
((b)) Yes.
 
((c)) The computer is upsampling in this case.  I also prefer to avoid resampling: it can't make the sound better.
 
((d)) I abandoned iTunes as a player a year or so ago.  I've heard rumors it can handle 24 bit audio now.  I wouldn't rely on iTunes to play anything other than iTunes store material and mp3s.  Apple simply has no incentive to do more.
 
I like to play my music without up/down conversion.  I got tired of going into MIDI setup all the time so I bought Decibel.  It's maybe $15.  Decibel can completely bypass the OSX audio system (this is called "hog mode") and talk directly to the hardware.  Decibel will automagically switch your DAC to the correct sampling rate and bit depth.
 
But if all your material is MP3s and stuff from the iTunes store, you can set MIDI to 44.1 and use iTunes as a player and you're pretty much golden.  If you're system has a manual volume control (like on a headphone amp) then its a nice touch to set all the computer volume controls to 100%.
 
If your collection includes 90/24 material, I suggest buying a separate player (such as Decibel).  There's some free ones too: Cog comes to mind.
 
An earlier poster makes a good point: DVDs are at 48000 samples per second.  If you set the MIDI to 44.1., DVDs will be downsampled.  If you set the MIDI to 48000, then your music will be upsampled.  Adding DVDs to the mix, I'm not aware of a solution that will handling the sampling rate automatically.  The whole mindset of the operating system people (Mac and Windows; Linux too if you consider PulseAudio) is that there are multiple, simultaneous sound sources, each potentially at a different sampling rate and bit depth.  They're not thinking of matching the DACs rate to the music's sampling rate. The only way they can meet /their/ goals is to resample everything to a common rate (48000 or 96000, typically).  They go to a lot of trouble to make sure everything gets resampled:  resampling is a feature, a "good thing" from their point of view.
 
Much of the reason matching sampling rates to the DAC is hard is that we're fighting the design goals of the operating system.

 

[obobskivich]

An earlier poster makes a good point: DVDs are at 48000 samples per second.  If you set the MIDI to 44.1., DVDs will be downsampled.  If you set the MIDI to 48000, then your music will be upsampled.  Adding DVDs to the mix, I'm not aware of a solution that will handling the sampling rate automatically.  The whole mindset of the operating system people (Mac and Windows; Linux too if you consider PulseAudio) is that there are multiple, simultaneous sound sources, each potentially at a different sampling rate and bit depth.  They're not thinking of matching the DACs rate to the music's sampling rate. The only way they can meet /their/ goals is to resample everything to a common rate (48000 or 96000, typically).  They go to a lot of trouble to make sure everything gets resampled:  resampling is a feature, a "good thing" from their point of view.

You do understand that modern SRC implementations are generally operating with a lower noise floor than the best DtoA solutions (until you get into the *healthy* five figure range), right? As in - you will not hear them working.

SRC is not a "good" or "bad" thing - it's a "required" thing. If you want "perfect native playback" you need to be going from a single source/stream, like a CD, through a chain of associated equipment. Doesn't do a whole lot for SQ; isn't very convenient either.