https://bugzilla.novell.com/show_bug.cgi?id=747391 https://bugzilla.novell.com/show_bug.cgi?id=747391#c6 --- Comment #6 from Dave Plater <davejplater@gmail.com> 2012-02-23 14:37:04 UTC --- (In reply to comment #4) Coolo split the ffado-mixer into a separate linked package a while ago to minimise rebuilds, the mixer has it's own spec file, if this is relevant. I will add the "-DENABLE_DICE=true" flag and notify you when (if) it has built for 12.1. Meanwhile I've found a possible patch which I'll first try on the release and I will work on getting the latest svn to build: + Euan de Kock <euan@dekock.net> 8th February 2011 12 13 + See Ticket #310 14 + 15 + This is a fix to force the Echo Devices to work at 16 + sample rates above 48000 where the ADAT digital channels 17 + get multiplexed to handle higher frequencies. 18 + At 88200 and 96000 the ADAT channels are reduced to 4, 19 + and at 192000 they should be reduced to 2. (I don't have a 20 + device capable of 192K to verify this) 21 + 22 + With the Echo devices (5.5 firmware at least), the device does 23 + not reduce its channel count when it multiplexes the ADAT data, 24 + but the actual packet does contain a reduced set of channels. 25 + 26 + The channels are referred to as Dimensions in the firewire parlance. 27 + With an Audiofire Pre8, we typically receive 17 channels of data at 28 + 44100 and 48000, being 8 Analog, 8 Digital and 1 MIDI. When we switch 29 + to 88200 or 96000 the ADAT (Digital) channel count is reduced to four, 30 + giving a total channel count (dimension) of 13. However Echo still 31 + reports the DBS parameter as being 17. 32 + 33 + To get around this, I don't bother using the DBS value at all. We know 34 + that the FDF is a reliable value for AM824 packets (type 0 to 7), so 35 + we can just use the FDF field to directly infer the size of each channel 36 + or dimension - 8, 16 or 32 Quadwords respectively for 41/48K, 88/96K and 192K. 37 + 38 + We set nevents directly to this for all AM824 packets directly, and do 39 + the standard calculation for any non-standard types. 40 + 41 + This is actually slightly less work than doing the calculation directly 42 + and should be portable across all devices that send AM824 data. 43 -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.