Recording via optical-in

[sullen]

Why don't you donate for the sheer fact that Rockbox kicks ass and they took the time to port it to the iRiver to start with?  To me, that is enough reason in itself.

[dwonk]

back to the top 

Thom (preglow) threw something up just a little while ago on rockbox's feature request BB:

http://www.rockbox.org/tracker/task/3041#comment7848

Comment by Thom Johansen (preglow) - Sunday, 07 May 2006, 10:54PM

SPDIF recording is being developed (slowly) by me right now. A rudimentary patch is available, but there are some bugs I need to get rid of. As for formats, you'll currently have to be content with 96/16. 20 bits is the maximum possible on H1x0, so you'll pretty much have to forget anything greater. Even 20 bits isn't currently possible, since we'd need to create an interrupt based driver for the PCM playback code. Don't expect too much in the department of multichannel either. Any help on the coding would be appreciated.

[mborus]

SPDIF recording is being developed (slowly) by me right now. A rudimentary patch is available, but there are some bugs I need to get rid of.

For those of you that have not started testing:

Once you got the first digital recording running, it's bug-free until you shut down the unit.

I've recorded over 20GB worth of WAV files digitally testing this patch so far and all recordings are OK. (I'm using an external A-D converter and am recording a batch of 60 US radio shows on vinyl, if you are wondering)

So as long as you remember to:
- turn on digital optical OUT and
- to feed the digital in a signal before selecting digital input
you are safe...

[sullen]

Since you seem to have done the most recording with the optical patch, I have a question but you may not be able to answer it.  Is having the optical out, required to record at this point, draining the battery more?  I don't know the interworkings of the iRiver so I was wondering if this was added drain/strain on the battery.  About how much time are you getting per charge or do you record with the wall wort plugged in?

[whatboutbob]

Sullen,

Interesting question.  You have to enable optical out but it turns off when recording. mborus, have you noticed any significant impact on battery life?  i haven't (but I haven't been paying attention to it).

[sullen]

Sullen,

Interesting question.  You have to enable optical out but it turns off when recording. mborus, have you noticed any significant impact on battery life?  i haven't (but I haven't been paying attention to it).

Oh, I didn't realize that.  I had only performed a few tests and didn't pay attention.  When this works well, I may convert to the iRiver full time for optical, but for now I am sticking with my JB3 and MicroTrack for Digital.  iRiver will still be my main analog recording unit though, gotta love it for that!

[whatboutbob]

Woot!  Looks like amiconn may have solved the hanging (DMA error) issue.  I can't get recording to hang now anyways.  If there's a signal loss now (eg when the cable is pulled out mid-recording), the recording pauses (the time keeps ticking, but the peakmeter disappears and the size pauses) until the signal's found again.  This isn't obvious when it happens, but I imagine that's something they'll get to eventually. :-)

Here's a precompiled build w/ the new patch for anyone that wants to help test: http://download.yousendit.com/CF490F1E480279DC

Reminder: you'll need to enable spdif out, and its prolly safest to reset settings...not in that order. :-)

[mborus]

Is having the optical out, required to record at this point, draining the battery more?  I don't know the interworkings of the iRiver so I was wondering if this was added drain/strain on the battery.  About how much time are you getting per charge or do you record with the wall wort plugged in?

I've been recording with the power connector plugged in most of the time.

I would expect that the optical output power consumption is nothing compared to the many HD accesses needed to write the WAV file.

I've sucessfully recorded several hours on battery. Currently, I record aprox 3 GB and then connect the Iriver to the PC to transfer the data before I start recording the next 4LP set. So recordings larger than 4 hours are no problem...

[preglow]

mborus:
Could you try the build whataboutbob posted and tell me if you still get wrong sample rates on the WAVs that used to be tagged with wrong sample rates? Could you also check with the diagnostic screen and check to be 100% percent sure the sample rate is never correct there either?

[mborus]

mborus:
Could you try the build whataboutbob posted and tell me if you still get wrong sample rates on the WAVs that used to be tagged with wrong sample rates? Could you also check with the diagnostic screen and check to be 100% percent sure the sample rate is never correct there either?

I still get 44.1 kHz files if I record 48 kHz from satellite.
As before changing the playback speed to 48 kHz fixes
the problem.  Also the debug screen still fails to recognize
the 48 kHz signal.

The good news is that there are no more crashes when I start the recording with no signal present.

[Crumbo]

Woot!  Looks like amiconn may have solved the hanging (DMA error) issue.  I can't get recording to hang now anyways.  If there's a signal loss now (eg when the cable is pulled out mid-recording), the recording pauses (the time keeps ticking, but the peakmeter disappears and the size pauses) until the signal's found again.  This isn't obvious when it happens, but I imagine that's something they'll get to eventually. :-)

Here's a precompiled build w/ the new patch for anyone that wants to help test: http://download.yousendit.com/CF490F1E480279DC

Reminder: you'll need to enable spdif out, and its prolly safest to reset settings...not in that order. :-)

here's a perm link for whataboutbob's pre-compiled build:

whataboutbob's pre-compiled build 5-9-2006

[preglow]

OK, here we go with a new patch. This time the only problem I'm aware of is that the recording source doesn't seem to be saved properly. Also worth noting now that starting recording before a signal is applied isn't an error anymore, is that the sample rate is written to the WAV file at the very start of recording. Thus, if you have no valid signal applied at this point, an erronous sample rate will probably be written.
Please test the patch with all sample rates you can. It should work properly with all.

Patch: http://www.pvv.org/~thomj/rockbox/spdif_recording.patch
Pre-compiled build: http://www.pvv.org/~thomj/rockbox/rockbox.iriver

Note that the pre-compiled build can't load plugins and might have some other stuff included, please try to use the patch if possible.

[preglow]

Patch commited.

[whatboutbob]

\o/

[AGI]

Great news!

I have been silently following progress on s/pdif recording, reading IRC logs every day.
I want to thank very much preglow and amiconn for their wonderful work and whatboutbob for testing and gentle pressing ;-)

Thanks guys!