ftp://ftp.arm.linux.org.uk/pub/linux/arm/people/nico/vorbis/

Compiling (with the cross compiler) the source yields a decoder that will successfully decode ogg files on the empeg.

The player they include 'vplay' is non-functioning, but we've got ways around that ...

That decoder took just under two minutes to decode (to a file) a 5 minute ogg song. I think it was 117 bit. But there won't be so much disk overhead when it's not having to write out a file while reading another from the same filesystem. And never the less it was still over twice as fast as it had to be for writing out to the audio device.

I got the resulting pcm file (from the decoder) to play using:
'dd if=Hydrate.pcm of=/dev/audio bs=4608'

And it sounds great.

Greg

Very cool.

Wouldn't ``decoder < bitstream.ogg | dd of=/dev/audio obs=4608'' work?

In reply to:

---

a 5 minute ogg song. I think it was 117 bit

---

I'm impressed - if OGG encodes everything that small, I should be able to get my entire collection into RAM...

In reply to:

---

I'm impressed - if OGG encodes everything that small, I should be able to get my entire collection into RAM...

---

Yeah, and pgrzelak's empeg would be able to hold all music that has ever been recorded!

Greetings!

Haven't we had that debate somewhere before??? Still, a free integer OGG raises some interesting possibilities. Does anyone know of a good OGG to MP3 comparison table or study?

I'm sure my measurement and description was supposed to be pretty scientific to satisfy you guys, but looking at it in the light of day ... there's something to be said for going to bed early.



Greg

Well, something's not right then, cuz this produces no audio. Should that be bs=4608?

Greg

Maybe. I'm not up on how the empeg's audio device works, nor am I a dd expert.

I think he's just double checking that you meant to type "bs=4608", as opposed to "obs=4608".

Well, I meant ``obs'', but ``bs'' might work better. I don't know. I was suggesting he try it the other way.

Well, I meant ``obs'', but ``bs'' might work better.

You did mean "obs". Just setting "bs" doesn't do what you expect. The other thing to bear in mind is that you must have played audio from the player (since the last reboot) before doing this, so the player can unset the DSP mute, which is on by default.

Peter (BTDTGTTS)

The Empeg player has got to be upgraded to allow user-added decoders. If not, it's no better than a PhatBox or Neo.

It can be upgraded to allow user-added decoders.

If you replace the player software. Or you do something clever in the kernel too fool the player software. Both are possible. Would you like to be the first to implement it?

I can be upgraded to allow user-added decoders.

Aha!!! Proof!

We always knew you were a robot!

Ah, see, but in true Douglas Adams style, that's actually proof AGAINST my being a robot. Robots don't make typos.

(Corrected.)

The Empeg player has got to be upgraded to allow user-added decoders. If not, it's no better than a PhatBox or Neo.

Oh yeah right

This is what the problem was. "obs" and the rest of bitt's command line was correct. After playing audio from the player, this command worked, as did the decoder. That mute thing has bitten my butt a couple of times before, you'd think I would remember it.

Greg

'Ah, see, but in true Douglas Adams style, that's actually proof AGAINST my being a robot. Robots don't make typos.'

lol- probably my favorite part of any of his books BTW.

Ah, that was easy.

(/me gets killed at the next crosswalk)