maggior;535579 Wrote: > Has anybody compared the headphone out of the Touch to the SB3's > headphone out? I ended up using a headphone amp with the SB3 because I > could hear noise (probably from the CPU) when plugging directly into the > SB3. I recall others experiencing the same with the SB3. I'm wondering > if I'm going to have to do the same thing with the Touch, especially > since it will be put to headphone listening/controller duty. > > Regarding the "unknown reason for the SQ difference between FLAC and > WAV on the touch when using the internal server" issue - could it be > due to increaceed CPU usage to decode the FLAC files? The increased > CPU activity could presumably inject more noise into the audio section.
The electrical noise levels in the Touch (both on the board traces and EM fields in space) are much lower than for the SB3. The result is much lower noise in general on all outputs. The WAV/FLAC or TinySBS / external SBS is rather interesting. Both Phill and my tests have shown that there is no difference on the analog outs for these different configurations. Yet I can hear it via the headphones. What I HAVE measured is that there is an increase in jitter on the clock lines when the headphones are inserted, the jitter spectrum is based on 120HZ, NOT random noise or music correlated. My guess is that the headphone amp has to supply some significant current into the headphones and this extra current is slightly changing the behavior of the voltage regulators/power supply and letting some line frequency ripple into the circuits. My guess is that once this is happening and the power system has been pulled out of its best operating mode the slight differences due to cpu load/memory access patterns etc are showing up a little more on the supply lines. Note this is NOT proven, just conjecture it will take a lot more testing with headphones plugged in, out, maybe dummy loads of different resistances etc. There is a lot of work to do to find out what is happening. Btw my testing in general (not specifically on the Touch, but computers in general) seem to point to the issues NOT being CPU load itself that is the issue, but more along the lines of memory access patterns, certain patterns wind up with instructions being kept in cache which will cause fewer main memory accesses. So subtle changes in program or data can easily change cache usage which can have a big impact on memory bus usage even though average CPU utilization stays the same. John S. -- JohnSwenson ------------------------------------------------------------------------ JohnSwenson's Profile: http://forums.slimdevices.com/member.php?userid=5974 View this thread: http://forums.slimdevices.com/showthread.php?t=77158 _______________________________________________ Touch mailing list Touch@lists.slimdevices.com http://lists.slimdevices.com/mailman/listinfo/touch
