This is the bug/patch tracker for Rockbox. Click here for more information.
Quick links: Bugs · Patches · Rockbox frontpage
FS#9394 - IRiver H10 sample rate change
Attached to Project:
Rockbox
Opened by Christian Lees (cdlees) - Thursday, 11 September 2008, 08:37 GMT+2
Last edited by Peter D'Hoye (petur) - Monday, 29 September 2008, 01:26 GMT+2
Opened by Christian Lees (cdlees) - Thursday, 11 September 2008, 08:37 GMT+2
Last edited by Peter D'Hoye (petur) - Monday, 29 September 2008, 01:26 GMT+2
|
DetailsAllows you to change the recording/playback sample rate on the IRiver H10. It was locked at 44.1kHz, now you can select 8, 32, 44.1, 48, 88.2 and 96kHz. Checked on r18325 (I don't have svn access, firewall). The patch was checked by recording 15 seconds of a 1kHz sine wave on the line in input at each sample rate. Tested by playing the samples back on the H10 as well as looking at the files in Cooledit.
Has only been tested on the H10 20GB. |
This task depends upon
Closed by Peter D'Hoye (petur)
Monday, 29 September 2008, 01:26 GMT+2
Reason for closing: Accepted
Additional comments about closing: Thanks & welcome to the Rockbox credits file ;)
Monday, 29 September 2008, 01:26 GMT+2
Reason for closing: Accepted
Additional comments about closing: Thanks & welcome to the Rockbox credits file ;)
This should follow the convention where pcm_set_frequency doesn't have immediate effect on the hardware.
pcm_apply_settings should apply all parameter changes done by other calls - this is done to have on-the-fly changes without having several settings take effect in succession.
pcm_apply_settings should synchronize with the interrupt properly by masking FIQ.
Any changes should be applied when starting playback (via pcm_apply_settings semantics): stopped->play, stopped->paused, paused->play
If any ports have implemented the interface outside spec they should be synced to proper behavior or else software development can't rely on a consistent interface.
The internal sampling frequency of the codec is changed. I have tested the bandwidth of the codec by recording a white noise source then processing the data in matlab to get the spectral response at 44.1kHz and 96kHz to verify that it works.
This patch should work for any device using the PP5020 and a WM8731, if there are any others. The config_XXX.h will just have to be updated.
Would be nice if you could create patches against SVN and not against another tree on your disk. (svn diff)
ipod 3gen could also use it if it has recording....