Hi,
I hardcoded the tuner to 56, and at least for me it seems to use it.
Jun 14 19:13:14 linux kernel: cx2388x v4l2 driver version 0.0.5 loaded
Jun 14 19:13:15 linux kernel: ACPI: PCI Interrupt 0000:00:0b.0[A] -Link
[LNKB] -GSI 10 (level, low) -IRQ 10
Jun 14 19:13:15 linux kernel: CRE cx88[0]: subsystem: 0070:3401, board:
Hauppauge WinTV 34xxx models [card=1,autodetected]
Jun 14 19:13:15 linux kernel: TV tuner 37 at 0x1fe, Radio tuner -1 at 0x1fe
Jun 14 19:13:15 linux kernel: cx88[0]: cx88_reset
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: (0) scl=1, sda=1
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: (1) scl=1, sda=0
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: (2) scl=1, sda=1
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: (3) scl=0, sda=1
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: (4) scl=1, sda=1
Jun 14 19:13:15 linux kernel: i2c-algo-bit.o: cx88[0] passed test.
Jun 14 19:13:15 linux kernel: tveeprom 1-0050: Hauppauge model 34704, rev
J1B3, serial# 8804439
Jun 14 19:13:15 linux kernel: tveeprom 1-0050: tuner model is TCL M2523_3DB_E
(idx 113, type 4)
Jun 14 19:13:15 linux kernel: tveeprom 1-0050: TV standards PAL(B/G)
PAL(D/D1/K) (eeprom 0x44)
Jun 14 19:13:15 linux kernel: tveeprom 1-0050: audio processor is CX881 (idx
31)
Jun 14 19:13:15 linux kernel: tveeprom 1-0050: has radio
Jun 14 19:13:15 linux kernel: cx88[0]: warning: unknown hauppauge model #34704
Jun 14 19:13:15 linux kernel: cx88[0]: hauppauge eeprom: model=34704
Jun 14 19:13:15 linux kernel: input: cx88 IR (Hauppauge WinTV 34xxx
as /class/input/input5
Jun 14 19:13:15 linux kernel: cx88[0]/0: found at 0000:00:0b.0, rev: 5, irq:
10, latency: 32, mmio: 0xe2000000
Jun 14 19:13:15 linux kernel: cx88[0]/0: registered device video0 [v4l2]
Jun 14 19:13:15 linux kernel: cx88[0]/0: registered device vbi0
Jun 14 19:13:15 linux kernel: cx88[0]/0: registered device radio0
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: "NTSC-M" fsc8=28636360
adc=28636363 vdec=28636360 db/dr=28636360/28636360
Jun 14 19:13:15 linux kernel: cx88[0]: set_pll: MPLL_REG 0x00fffffe
[old=0x00fffffe,freq=28636360]
Jun 14 19:13:15 linux kernel: cx88[0]: pll locked [pre=2,ofreq=28636360]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MINPUT_FRMAT 0x00000001
[old=0x00000007]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MUTPUT_FRMAT 0x181f0008
[old=0x181f0008]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MSCNV_REG 0x00020000
[old=0x00020000]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MSUB_STEP 0x00400000
[old=0x00400000]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MSUB_STEP_DR 0x00400000
[old=0x00400000]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MAGC_BURST 0x00007270
[old=0x00007270,bdelay=114,agcdelay=112]
Jun 14 19:13:15 linux kernel: cx88[0]: set_tvnorm: MHTTAL 0x0000138e
[old=0x0000138e,htotal=910]
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: 320x240 [TB,NTSC-M]
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: hdelay 0x0038
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: hscale 0x15b3
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: hactive 0x0140
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: vdelay 0x0018
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: vscale 0x1e00
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: vactive 0x01e0
Jun 14 19:13:15 linux kernel: cx88[0]: set_scale: filter 0x80009
Jun 14 19:13:15 linux kernel: set_control id=0x980900 reg=0x310110 val=0x00
(mask 0xff)
Jun 14 19:13:15 linux kernel: set_control id=0x980901 reg=0x310110 val=0x3f00
(mask 0xff00)
Jun 14 19:13:15 linux kernel: set_control id=0x980903 reg=0x310118 val=0x00
(mask 0xff)
Jun 14 19:13:15 linux kernel: set_control id=0x980902 reg=0x310114 val=0x5a7f
(mask 0xffff)
Jun 14 19:13:15 linux kernel: set_control id=0x980909 reg=0x320594 val=0x40
(mask 0x40) [shadowed]
Jun 14 19:13:15 linux kernel: set_control id=0x980905 reg=0x320594 val=0x20
(mask 0x3f) [shadowed]
Jun 14 19:13:15 linux kernel: set_control id=0x980906 reg=0x320598 val=0x40
(mask 0x7f) [shadowed]
Jun 14 19:13:15 linux kernel: cx88[0]/0: video_mux: 0
[vmux=0,gpio=0xff00,0x0,0x0,0x0]
Jun 14 19:13:22 linux kernel: cx88[0]/0: open minor=0 radio=0 type=video-cap
I have a channel list, created with an other card. When I switch the channels
in the TV application, this is the output in /var/log/messages
Jun 14 19:26:51 linux kernel: get_control id=0x980900 reg=0x310110 val=0x80
(mask 0xff)
Jun 14 19:26:51 linux kernel: get_control id=0x980901 reg=0x310110 val=0x3f
(mask 0xff00)
Jun 14 19:26:51 linux kernel: get_control id=0x980902 reg=0x310114 val=0x7f
(mask 0xff)
Jun 14 19:26:51 linux kernel: get_control id=0x980903 reg=0x310118 val=0x80
(mask 0xff)
Jun 14 19:26:51 linux kernel: get_control id=0x980906 reg=0x320598 val=0x00
(mask 0x7f) [shadowed]
Jun 14 19:26:51 linux kernel: get_control id=0x980905 reg=0x320594 val=0x1f
(mask 0x3f) [shadowed]
Jun 14 19:26:51 linux kernel: get_control id=0x980909 reg=0x320594 val=0x01
(mask 0x40) [shadowed]
but the image is still black on all channels, also a channel scan does not
work.
any hints what I can try else?
kind regards
Sebastian
Linux and Kernel Video <video4linux-list (AT) redhat (DOT) comwrote:
Wed, 14 Jun 2006, hermann pitton wrote:
If tuner=56 is tested to be fully compatible, you can replace the
TUNER_ABSENT on idx 113 there for TCL M2523_3DB_E with tuner=56
TUNER_PHILIPS_FQ1216AME_MK4 from tuner.h.
I'm quite sure 56 is the one, by experimenting. Not by opening the
thin-can.
Since tuner=56 is a MultEurope MK3 tuner with tda9886, but without
radio, which would need a tda9887 and tuner=38 to work, further details
in case anything goes wrong or is missing are wanted. For example you
could enable debug=2 for the tda9887 module and see if it responds on tv
standard switching on card=1, else it is some more restricted
conventional tuner using only MK3 API on the pll chip. If all is fine
after some testing, we could add it hopefully such easy then, since I
believe no data sheet is available for that one.
I don't have time (or actually the hardware freedom) to test it. Because
it is 24/7 used to record. But your tips are a start :)
I created once a patch with a dummy hauppauge device and disabled
hauppauge eeprom detection to test on such always appearing new tuners
and suggested several times to allow tuner types forced by user insmod
options
I did it by hand, one by one ;) I recall 38 did give tuning result,
like many others. The problem was for them they only did upper band over
the complete bandwidth.
Stefan