Rockbox

This is the bug/patch tracker for Rockbox. Click here for more information.

Quick links: Bugs · Patches · Rockbox frontpage

Tasklist

FS#10816 - Simulator returns distorted PCM data

Attached to Project: Rockbox
Opened by Delyan Kratunov (archivator) - Thursday, 26 November 2009, 23:34 GMT+2
Last edited by Andree Buschmann (Buschel) - Saturday, 20 November 2010, 00:43 GMT+2
Task Type Bugs
Category Simulator
Status Closed
Assigned To No-one
Player Type All players
Severity Low
Priority Normal
Reported Version Daily build (which?)
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Private No

Details

I started working on  FS#10065  again and immediately found a bug most likely in the sim. The attached screenshots/screendumps show obvious distortion in the data received from pcm_get_peak_buffer, while everything is fine on the actual target (iPod 5.5G, if that matters).

Last confirmed on r23717.
   1 kHz dev.png (65.9 KiB)
   1 kHz sim.png (188.1 KiB)
   sweep dev.png (72.1 KiB)
   sweep sim.png (222.4 KiB)
This task depends upon

Closed by  Andree Buschmann (Buschel)
Saturday, 20 November 2010, 00:43 GMT+2
Reason for closing:  Works For Me
Additional comments about closing:  r28115 on sim for iPod 5.5G and nano 2G
Comment by Nils Wallménius (nls) - Friday, 04 December 2009, 21:32 GMT+2
Hi i tried looking into this but the patch in  FS#10065  doesn't build anymore, maybe you could post an updated patch to use when trying to track this down?
Comment by Delyan Kratunov (archivator) - Monday, 07 December 2009, 19:19 GMT+2
The attached patch is a snapshot of my current working tree. It's a bit larger than usual as it also includes the iPod Video graphics needed for the color and spectrogram modes.
   10816-fft.patch (172.2 KiB) 
 b/apps/plugin.c                       |    1 
 b/apps/plugin.h                       |    1 
 b/apps/plugins/CATEGORIES             |    1 
 b/apps/plugins/SUBDIRS                |    3 
 b/apps/plugins/bitmaps/native/SOURCES |   20 
 b/apps/plugins/fft/SOURCES            |    3 
 b/apps/plugins/fft/_kiss_fft_guts.h   |  145 ++++
 b/apps/plugins/fft/const.h            |    4 
 b/apps/plugins/fft/fft.c              | 1167 ++++++++++++++++++++++++++++++++++
 b/apps/plugins/fft/fft.make           |   27 
 b/apps/plugins/fft/kiss_fft.c         |  428 ++++++++++++
 b/apps/plugins/fft/kiss_fft.h         |  119 +++
 b/apps/plugins/fft/math.c             |   13 
 b/apps/plugins/fft/math.h             |   28 
 b/firmware/export/pcm.h               |    1 
 b/firmware/pcm.c                      |    5 
 16 files changed, 1966 insertions(+)

Comment by Delyan Kratunov (archivator) - Wednesday, 28 April 2010, 18:00 GMT+2
Using r25695, I get the proper graph on device but the sim still returns incorrect data.
   dev.png (24.6 KiB)
   sim.png (147.3 KiB)
Comment by Andree Buschmann (Buschel) - Monday, 31 May 2010, 06:59 GMT+2
Hmm, at a first glance it looks like the spectrum that you show in the sim samples a signal with 1/32 of the targets sample rate.
Comment by Andree Buschmann (Buschel) - Wednesday, 17 November 2010, 20:55 GMT+2
The spectogram looks fine on simulation when using real music. Is this bug fxied or still occuring for you?

Loading...