FS#8807 - r16791 introduces playback skipping in ipod 5.5G (80GB)

Attached to Project: Rockbox
Opened by Bryan Childs (GodEater) - Wednesday, 26 March 2008, 09:58 GMT
Last edited by Michael Sevakis (MikeS) - Saturday, 29 March 2008, 23:25 GMT
Task Type Bugs
Category Music playback
Status Closed
Assigned To No-one
Operating System iPod 5G
Severity Medium
Priority Normal
Reported Version Daily build (which?)
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No


The changes introduced in r16791 have resulted in choppy playback on the iPod 80GB 5.5G.

This is at its worst when the player is first turned on, while the dircache is building - but also occurs every time the folder changes during playback.

Whilst the playback is suffering these glitches, the playback controls are also unresponsive, making it hard to pause / stop playback - although access to the menus seems unaffected.
This task depends upon

Closed by  Michael Sevakis (MikeS)
Saturday, 29 March 2008, 23:25 GMT
Reason for closing:  Works For Me
Additional comments about closing:  I suppose I'll close this as not 100% complete.
Comment by Michael Sevakis (MikeS) - Friday, 28 March 2008, 04:02 GMT
Argh. Let's try this nasty thing again.
Comment by Nicolas Pennequin (nicolas_p) - Friday, 28 March 2008, 12:47 GMT
This patch makes things much better for me... The unresponsiveness is gone and buffering is much faster :)
Comment by Michael Sevakis (MikeS) - Friday, 28 March 2008, 17:21 GMT
something in that port is being very bad >:-[
Comment by Michael Sevakis (MikeS) - Friday, 28 March 2008, 17:22 GMT
I guess I'll commit this awful thing but I will still find a way to investigate.
Comment by Sascha Wolf (Horscht) - Friday, 28 March 2008, 17:26 GMT
I am wondering, is there something wrong with the Ipod Video port?

every so often, a commit seems to break buffering on the device. Don't get me wrong, I am not complaining, I am realy just wondering :)
Comment by Michael Sevakis (MikeS) - Friday, 28 March 2008, 20:20 GMT
I am wondering as well since my test with an H10 running large sector access suffers no I'll effects - not quite as fast but no problems and supposedly disk reads aren't any slower on 5g. Do these problems happen if the CPU frequency is forced to 80MHz? Do they occur if a single core build is forced (see config.h to do that)? There's a huge list of possible checks but debugging remotely turns minutes into days :P.