From: Philippe C. <la-...@or...> - 2012-11-25 11:44:06
|
Changes have been commited as revision 2254 and 2255. I just had a message for 192kHz; when you will have more information, we will modify. Best regards, Phil Le samedi 24 novembre 2012 à 14:12 +0100, Jano Svitok a écrit : > Hi Philippe, > > this is what I've got so far: > > https://github.com/janosvitok/ffado/commit/f72e116270eb367f482ec7e18acd27114b088504 > > 48 kHz should be ok, I'm not sure about the higher rates. I'm also not > sure about the mixer outputs - I'll check them later. > > I've created also http://subversion.ffado.org/wiki/Maudio > > Question: is 192 kHz supported at all? A comment in the code suggests > that it is not (I've not tested it yet). > > Do you want me to store test-dice-eap outputs somewhere? > > Jano > > > On Tue, Nov 20, 2012 at 8:23 PM, Philippe Carriere > <la-...@or...> wrote: > > Le mardi 20 novembre 2012 à 14:27 +0100, Jano Svitok a > écrit : > > > Hi all, > > > > Hi Jano, > > > > > > I'm owner of Maudio profire 2626. I'm thinking about adding proper > > mixer GUI for this card. > > Nice idea. I planned to extend the work I have yet done for > Focusrite devices and Presonus Firestudio Tube but help is > welcome (since I whatever have no such devices for testing) ! > > > > > > Before I dive deep in the sources, can you > > give me any hints how should I proceed? > > First have a look to changes revision 2251 and 2252 of the > trunk (involving Firestudio Tube). > You will create a "maudio" sub-directory in libffado/src/dice, > then create a cpp and a header file for customization. > You will have to follow the lines of the two yet existing > files for firestudio, for instance. > > > > Is there any howto? > > No. > So after the preliminaries, you will have to find by yourself > the correct configuration of the routing, that is essentially > find the registers corresponding to each source and each > destination. > > For this, you will use the information given in the output > (redirect this to file) of: > > $ test-dice-eap -c1 -a --port=fw1 > > and more particularly to the "showFullRouter output". All > destinations will be listed with their routed sources like > > 1348247999745837: (dice_eap.cpp)[ 843] showFullRouter: 128 > entries to read > 1348247999747426: (dice_eap.cpp)[ 854] showFullRouter: > Current Configuration: > 1348247999748913: (dice_eap.cpp)[ 861] showFullRouter: 60 > routes > 1348247999751026: (dice_eap.cpp)[ 872] showFullRouter: 0: > 0xb0 <- 0x50; > 1348247999751129: (dice_eap.cpp)[ 872] showFullRouter: 1: > 0xb1 <- 0x51; > 1348247999751200: (dice_eap.cpp)[ 872] showFullRouter: 2: > 0xb2 <- 0x52; > 1348247999751240: (dice_eap.cpp)[ 872] showFullRouter: 3: > 0xb3 <- 0x53; > 1348247999751291: (dice_eap.cpp)[ 872] showFullRouter: 4: > 0xb4 <- 0x54; > ... > > Not all sources are necessarily listed, so it would be better > first to set up your device under Windows or Mac, routing all > the "DAW" sources to some destinations so as to be able to > detect them in the output (either you would have to do this > several times). > What is routed is not important but by the way you will be > able to detect how are configured all destinations and all > sources. > Care that this is not necessarily straightforward. Also, it > might depend on the samplerate (you will see "low", "mid" and > "high" configuration, and just have a look to Pro 24 or 40 to > see that differences may exist). > > > > Any DICE/EAP documentation? > > https://dev.tctechnologies.tc/tcat/tags/release/public/latest/docs/firmware/DICE_Firmware_EAP_Specification.pdf > > > > > > I have seen that I have to write python class > > You will first use the generic EAP mixer, so you don't need to > change anything in the python support at first. > When the customization will work, well, that depends. I am far > from fond of the matrix mixer as it stands. > On the other hand, it is useful so as to have a general look > at a glance, so that it must be kept as it is presently. > What would be nice would be the possibility to switch, inside > the mixer, from the existing global view to a more per output > view with a more friendly interface (and vice versa). > > > > > > for the card, and I > > suppose also Dice EAP subclass in C++. > > > > Thanks. > > > > You can join me for any additional questions, when you will > have had a first look to this. > > Good cheers, > > Phil > > > |