Le samedi 16 février 2013 à 15:26 +0100, Johannes Treß a écrit :
Yes sorry 1394/Out:03 -> Anlg/In:01 but as I wrote seconds ago, the preset seems stronger than my manual editings - sadly...

Modifications have been committed as revision 2265.

Best regards,

Phil



Am 16.02.2013 14:26, schrieb Philippe Carriere:

Le samedi 16 février 2013 à 13:38 +0100, Johannes Treß a écrit :
Hello Philippe,

so finally I found out the right setting for the Pro 24 (1394/Out:01 -> Anlg/In:03; 1394/Out:02 -> Anlg/In:04; 1394/Out:03 -> Anlg/In:02; 1393/Out:04 -> Anlg/In:02).
Do you mean 1394/Out:03 -> Anlg/In:01 ?
In which case, it is just some reverse between 03/04 and 01/02 ?

I think this was the reason for my not working LEDs. Thanks for you very kind help again!
Maybe you can help me out with another problem I'm trying to solve right now: You are not using Pure Data by any chance? I get constant A/D/A errors in its console and some devs told me that it has to be a ffado - Pure Data communication problem... Did you ever hear of something like this?
No, sorry, I don't use Pure Data.


Yours

Johannes


Regards,

Phil
Am 15.02.2013 19:39,systr schrieb Philippe Carriere:

Hi Johannes,

sorry, I had to delay my answer.

Le jeudi 14 février 2013 à 11:29 +0100, Johannes Treß a écrit :
Update:

It is definetly not a bug. I changed (try&error) the routing in the mixer and now at least the 3. and 4.
Yes, it is possibly not exactly a bug because routing is available.
No, maybe the default configuration is not fully correct.

Input are indicated as it should be in LED row 3 and 4.
It's obvious that I'm not understanding the principle of the Mixer...
Again the question if there is a diagram of the routing options?

Well, routing is as you want. Now, the 1394/Out:01 to 1394/Out:16 are not really intended to be changed (it is fairly useless).
So by running set-default-router-config-dice-eap ... as indicated in my homepage, it is assumed that the default routing enables the LEDs in the right way: this default configuration is simply

Anlg/in:01 -> 1394/Out:01
Anlg/in:02 -> 1394/Out:02
Anlg/in:03 -> 1394/Out:03
Anlg/in:04 -> 1394/Out:04
...

so, isn't this right ? You have the ability to see if a different order seems to be required. In which case, I would have to correct the code.

Thanks a lot!


Best regards,

Phil

Am 13.02.2013 21:19, schrieb Philippe Carriere:

Le mercredi 13 février 2013 à 20:25 +0100, Johannes Treß a écrit :
Hey there,
Hi Johaness,

I managed to get my Soundcard to work with your help a few Weeks ago.
Right now I'm struggling with the Level Indicator LEDs on the front 
panel of my device.
Philippe, you wrote on your Homepage, that the AnalogIns should be 
routed to 1394/Out:01 to 1394/Out:08. I already did this (it was the 
default setting) but theres only one LED row working (the one for 
channel 4). Despite this I recognized, that the channels itself are not 
in the same order as the input numbers at the device, they seem to be 
disarranged.
Had you a try with

$ set-default-router-config-dice-eap ...

at the corresponding samplerate (indeed, the ordering of the router is specific for focusrite's devices; LEDs answer according to the first register, intended to be 1394:01 to 1394:0x depending on the number of analogic inputs. If not so, answer is wrong).
If not working better, send me the output of

$ test-dice-eap -a ...

and if you have such a possibility, try to reset with the Windows/Mac OS software and send me the output of

$ test-dice-eap -a ...

again; since I have a Pro 40, obviously I am not able to test with Pro 24 directly and only encode in a blind way and what I know is only available from Pro 24 users.

To be honest, I'm not quite sure, if I understand the Description at 
Philippes HP, but I really tried it...
Maybe you can help?

Is my poor english writing involved :-) ?

Thanks a lot!

Yours Johannes

Best regards,

Phil

PS: I'm running Ubuntu 12.04 Dream Studio Distro, latest Jack Version 
and libffado 2.1.9999-2263 compiled as Philippe told me (It worked the 
second time,too!). 
svn 2263 is intended to work. So it looks like a possible bug, somewhere



--
Philippe Carriere <la-page-web-of-phil.contact@orange.fr>