OK nerds, here's what I have found out.
I did a little testing tonight because I was curious. I made a file, 44.1/16 with a -6dbfs 1K tone in one channel, and digital silence in the other. I encoded it to various bit rate MP3 files using various encoders. Although I found some interesting artifacts at the end of the files on the "blank" channel, no "cross talk" was created in
any of the files by the lossy encoding. Perfect -96db silence. So that little part of the question is answered.
I then took one of the 128K fixed rate MP3s of the file, and played it back through various pieces of hardware into my DAW, careful to calibrate levels the same for comparison. The outputs were:
1- RCA Lyra MP3Player Headphone Out
2- HP Pavilion Laptop Headphone Out
3- Palm Treo 650 Headphone Out
4- M-Audio USB Mobile Pre Line Out
Results were all over the place. The RCA player had channel separation of around 60 dB. Although it seems like a lot, the tone was clearly audible at high levels in the "silent" channel.
The Laptop Headphone out fared the worst of all three. Only about 40 dB of separation. The tone was even higher above the noise floor than with the RCA Lyra.
My Palm Treo was the first of the "headphone" connections that provided enough channel separation that I couldn't hear the tone above the noise floor of the "silent" channel.
As I expected the only proper line connection, the Mobile Pre, provided complete channel separation between the tone and blank channel as well.
So, of the four playback methods I tried, we're batting .500 for having enough channel separation for use in a high level environment. I wish I could try an iPod, it being so poplar and all, but I don't have one in the house.
I'd say this proves that the MP3 player itself
can be the cause of crosstalk between track, and "click" channels when used the way that we're finding more and more these days. If you're using a setup like this, best be careful how you're rig is put together, and take the time to check it out before showing up to an important gig through a large rig, to find out your player's headphone jack output isn't going to keep your click track out of the FOH mix.
warrbeat wrote:Act2 had a tube like Sony MP3 player (with a very small screen) that had a recently mixed track loaded on it that day. 1/8" - 2x1/4" lines. In this case the mixer supplied was given to him "pre connected" completely including: red 1/4" to balanced input on "Behringer like" (I can't remember what kind it was) mixer. The yellow to a prewired Radial Design DI. The only difference was there was a second 1/4" cable from the link back to the mixer. The thought was to be able to mix his own level of the playback into his cans. I unplugged the XLR from the DI from Act1 and plugged it into the RDI supplied. He used the identical DI that the previous band used for his Roland Handsonic. We achieved signal for both and moved on.... We later found the click through the play back line throughout monitors and FOH.
The FOH operator first blamed the patch, then the connections then didn't have anymore ideas. Since there was only a mono signal coming from the DI, having both the click and the playback on the same line means it had to be from one or a combination of places including: a faulty splitter cable, crosstalk inside the mixer which came back through the link creating a loop, the MP3 player could have been set to mono playback or the track had been improperly mixed down although the latter is improbable but not impossible as it was recorded and tested in the studio but not with the player.
After what I've found in my little experiment, and by what you describe there, it seems likely the player itself caused the crosstalk we heard. Like we said, assuming the files were good, hardware somewhere is to blame.