Nick Andrew schrieb: > Attached is the firmware for the "LifeView TV Walker Twin" (01fd:0514) > > 7119b154e7975a7fc2d5cb7c4bed5b28 dvb-usb-lifeview-tv-walker-twin-0.95-20051010.bin > > ... as reconstructed from a dump of the usb transactions sent by > windows with the M9207BDA.sys driver loaded. However it's possible > that M9207BDA.sys contains more than one firmware. > > This is the diff between the firmware reported by Pierre Willenbrock > and mine: > > 5,8c5,8 > < 0000100 8e 83 ed f0 22 00 f6 40 c3 00 40 00 12 01 00 02 >...."..@..@.....< > < 0000120 00 00 00 40 fd 10 13 15 95 00 00 00 00 01 00 00 >...@............< > < 0000140 00 ff ff ff ff ff 08 4c 69 66 65 56 69 65 77 00 >.......LifeView.< > < 0000160 00 00 00 00 00 00 08 44 56 42 54 20 20 20 20 00 >.......DVBT .< > --- > > 0000100 8e 83 ed f0 22 00 91 40 c3 00 40 00 12 01 00 02 >...."..@..@.....< > > 0000120 00 00 00 40 fd 10 13 05 95 00 00 00 00 01 00 00 >...@............< > > 0000140 01 ff ff ff ff ff 08 4c 69 66 65 56 69 65 77 00 >.......LifeView.< > > 0000160 00 00 00 00 00 00 05 44 56 42 2d 54 00 00 00 00 >.......DVB-T....< > > The differences are: > > byte 0106 f6 -> 91 (load address??) > byte 0127 15 -> 05 (this is the MSB of the PID after firmware download) > byte 0140 00 -> 01 (no idea) > byte 0166 08 -> 05 (length of the following device name string) > bytes 0167 - 0176 (different device name) > > What are the first two bytes of every chunk in the firmware file > (the "value")? My guess is that the lsb is a checksum over the packet. I have not tried to reproduce it, yet. If you look for the firmware in the windows device driver, you will probably find that the fields for pid and name are not filled. Regards, Pierre _______________________________________________ linux-dvb mailing list linux-dvb@xxxxxxxxxxx http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
