[greenwoodwoo]

Hi everyone,
I'm currently running my MBP with iTunes and Amarra, with an Audio-GD FUN currently en route to my residence. Been burning in a new pair of AKG k701s this week (With pink noise, and silences in between so the tweeters can rest), and today i decided to take them for a test run. 
 
Playing Melody Gardot's "Our Love Is Easy" in lossless format, I discovered some serious distortion at higher frequencies that I figured were due to the file itself, since I tried the same file on my MBP and iPod, through different systems. Today I downloaded the demo version of Channel D's Pure Music player, and set my sample rate to 96khz as per the instructions in the guide. Once I enabled upsampling, it seemed as though all the distortion vanished!
 
Can someone explain if the sudden lack of distortion at higher frequencies is due to the upsampling or something else unbeknownst to this humble audio newb?

 

[silverxxx]

It could be caused by dithering. The channel D player appears to have dithering built in. You could verify this theory by playing the file in foobar and enabling dithering to see if the distortion dissapears. Or see if your program has an option to enable/disable dithering. Most tracks these days are dithered during mastering, but it is not impossible for one not to be.
 

 

[AtomikPi]

Quote:

It could be caused by dithering. The channel D player appears to have dithering built in. You could verify this theory by playing the file in foobar and enabling dithering to see if the distortion dissapears. Or see if your program has an option to enable/disable dithering. Most tracks these days are dithered during mastering, but it is not impossible for one not to be.
 

Dithering deals with bit depth, not frequency, and it is largely within the noise floor (<-85 dB or so) and is often transparent under DBT.
 
If the OP is taking a 16/44 file and upsampling to, say, 24/96, that wouldn't undo the loss of dynamic range caused by a lack of dithering in the original mastering process. Once you've lost the dynamic range, you can't get it back.
 
As for the OP's question, it seems very strange that upsampling would have that effect - even under non-blind conditions, most reviewers report only subtle changes from upsampling. However, I can't offer an alternative explanation.

 

[milosz]

Maybe a level somewhere was too high- either in digital or analog domain - and enabling the upsampling had the coincidental effect of normalizing the signal. 
 
You can over-level a digital stage, just as you can overdrive an analog stage.

 

[Currawong]

Are you sure the Sound Enhancer hasn't been left on in iTunes?