FS#12291 - Sansa Clip+ CPU/LCD noise while recording

Attached to Project: Rockbox
Opened by Samo Dadela (SamoDadela) - Saturday, 24 September 2011, 19:47 GMT
Task Type Bugs
Category Recording
Status Unconfirmed
Assigned To No-one
Operating System Sansa AMSv1
Severity Low
Priority Normal
Reported Version Release 3.9
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No


Recording noises on the Clip+

Probably a hardware problem though the original firmware somehow alleviates it.

When recording from:
1) Microphone
Noise is heard at the beginning of the recording while the LCD is still on, when it turns off
(goes to sleep) the noise goes away. Noise is then heard each time the CPU compresses
or writes the data. This can be alleviated by changing the rec. frequency from 44k to
22k. The noise at the beginning can also be heard using the original Clip+ firmware,
however the compress/write noise is not there (or is very low) - probably because the
original firmware writes at 22k (mono), too.
2) FM Radio
When recording FM using RB the CPU interferes with the FM receiver. This results in
the normally clean reception being broken by noises of 1-2 second duration happening
every 4-5 seconds (this can be longer if there is FM transmits silence - guessing
because the buffer fills slowly).
The original Sansa Clip+ firmware is also affected by this problem, however it is much
less noticeable. There the interference is heard like very tiny clicks of the period of 0.5
seconds - very silent but it is there.

As said its probably caused by the Clip+ not being shielded from interference (LCD, CPU
or something else) but the original firmware somehow works around the problem by not
working in bursts.

So... is it possible to process (record/encode/write) the audio in small pieces more
frequently instead of doing it in bulk each time the buffer fills? This could result in the
noise being less pronounced and thus not annoying. Maybe by making the buffers smaller?

This task depends upon

Comment by Samo Dadela (SamoDadela) - Tuesday, 27 September 2011, 18:20 GMT
More on the FM issue: it seems that the noise is dependent on the radio frequency we are tuned into - not much we can do here.
Checked again the CPU noise when recording from mic to exclude that the noise is induced only to the headphones - I can confirm the noise can be heard also on the recording.
Comment by Samo Dadela (SamoDadela) - Tuesday, 27 September 2011, 18:22 GMT
Correction: Hardware should be AMSv2.