Re: [linux-dvb] Consistent kernel crash on latest hg repo with 2.6.24-rc6
Please advise on how to approach this problem. Since there's no kernel panic being printed to the console, I suppose using things like LKCD to obtain a crash dump won't do anything.. P.S. So far all tests have been done using 64 bit distro's and kernels. Erik Erik van Zijst wrote: Hi folks, Since I installed a TechnoTrend 1500 Budget (with optional budget-ci CAM board), my kernel crashes completely everytime I run a frequency scan using the w_scan utility. There's no kernel panic message or anything printed to the tty, just instant death. The server box (dual quadcore Xeon) ran perfectly stable until I installed the T-1500 last week, and still runs stable as long as I don't use the card. I started out using a stock 2.6.22-something kernel (vanilla Fedora 8 and Ubuntu 7.10 server), then switched to 2.6.24-rc6 (generated at 2007-12-21) and finally tried today's trunk of http://linuxtv.org/hg/v4l-dvb using hg, on top of my 2.6.24-rc6 kernel. Yet, every combination dies. The kernel usually dies the first time w_scan -x (http://wirbel.htpc-forum.de/w_scan/index2.html version 20071209) is run and then usually dies near the end when w_scan prints tune to:. Occasionally it takes more than one run before it crashes. The box is idle during the scan and there's no other PCI activity. I'm using firmware dvb-fe-tda10046.fw, installed using get_dvb_firmware. regards, Erik van Zijst ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] HVR-950 status
Hey Devin, I was able to get full ATSC support from the HVR-950 running the latest Ubuntu 7.10 with a std mythtv install. I was also able to finagle getting the other input devices (NTSC/S-Video/Composite) functioning by using their respective V4L names. There is a sox sound hack that's been floating around because analog sound support is not currently available. I have not tested it. Sorry to say it's not a problem specific to your environment. http://lunapark6.com/usb-hdtv-tuner-stick-for-windows-linux-hauppauge-wintv-hvr-950.html that's where I started. Let me know if you would like more detailed information about my setup. From what I have gathered, the HVR-950 does not do QAM decoding, so digital cable is not available (could someone clarify this as a hardware/software issue?), only OTA (VSB encoding). Good luck! -Kyle On Jan 1, 2008 11:05 PM, Devin Heitmueller [EMAIL PROTECTED] wrote: Hello, Could somebody please give me a quick status of where they believe the HVR-950 is at in terms of support in the mainline? Video appears to work for analog (once the xc3028 firmware is loaded), as well as channel tuning, although there is no audio at all. Also no ATSC support since we don't have multiproto support yet in the mainline (no /dev/dvb device appearing). If anyone else out there is working with the HVR-950, could you please confirm that audio is not working? I just want to be sure that this isn't something specific to my environment. Thanks, -- Devin J. Heitmueller http://www.devinheitmueller.com AIM: devinheitmueller ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
[linux-dvb] RE : New tuning file for eiffel tower DVB-T transmitter
For Paris downtown, the Eiffel Tower is the main transmitter. For the surburb, the transmitters are listed below with their list of UHF channels (frequency in MHz = 306 + 8 * UHF-channel) Paris - Tour Eiffel : 21 24 27 29 32 35 Paris Est - Chennevières : 35 51 54 57 60 63 Paris Nord - Sannois : 35 51 54 57 60 63 Paris Sud - Villebon : 35 51 56 57 60 63 I have no official information on other modulation parameters, although the following should be fine for most of them: freq:INVERSION_AUTO:BANDWIDTH_8_MHZ:FEC_2_3:FEC_NONE:QAM_64:TRANSMISSION_MODE_8K:GUARD_INTERVAL_1_32:HIERARCHY_NONE except for channel 35 (586 MHz) which is: 58600:INVERSION_AUTO:BANDWIDTH_8_MHZ:FEC_3_4:FEC_NONE:QAM_64:TRANSMISSION_MODE_8K:GUARD_INTERVAL_1_8:HIERARCHY_NONE Also, an offset of n * 166.7 kHz may be applied to the frequency with n in the range -1 to +3 in France. I do not know where to find the official list of offsets for France. Using either a Hauppauge Nova-T-500 or a Nova-T-Stick, using the central frequency without offset seems fine in the Paris area. I also noted that using wrong values for some modulation parameters such as FEC or guard interval has no impact. The two above mentioned Hauppauge tuners automatically switch to the right values (these values are transmitted in the TPS). I do not know if the correct values are read from the TPS and applied by the hardware or driver (any info on the subject would be appreciated). Note that one of the 6 frequencies is currently mute (the so-called R5 MUX). It has been used in 2006 for DVB-H tests and in 2007 for DVB-T/HD tests. It will be used for official DVB-T/HD broadcasting in Q1 or Q2 2008. The map of current DTTV transmitters in France can be found at http://www.csa.fr/pdf/carte_TNT.pdf The current list of transmitters and their corresponding UHF channels can be found at http://www.csa.fr/pdf/frequences_tnt_planifiees_alpha.pdf (document in French). -Thierry -Message d'origine- De : [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] De la part de Christoph Pfister Envoyé : dimanche 30 décembre 2007 13:48 À : Alexis de Lattre Cc : linux-dvb@linuxtv.org Objet : Re: [linux-dvb] New tuning file for eiffel tower DVB-T transmitter Am Montag 24 Dezember 2007 schrieb Alexis de Lattre: On 22.12.2007 14:01, Christoph Pfister wrote : - renamed file to fr-Paris_eiffel_tower as there are multiple DVB-T transmitters in the Paris area. Normally, all transmitters in big cities use the same frequencies (so called single-frequency networks), so this rename shouldn't be necessary. It is not the case in the Paris area unless for multiplex R1 : mux R1 is SFN but all the other muxes are MFN. That's why is it important to know the transmitter and not just the city name. Argh. How many transmitters are there? I think it would be quite convenient (for the user) to only have one fr-Paris, so we could put multiple transmitters into that file ... P.S. : I have enclosed the tuning file of another DVB-T transmitter of the Paris area called Paris Nord Sannois. Thanks, Christoph ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] AverTV Hybrid Volar HX
2008/1/1, Manu Abraham [EMAIL PROTECTED]: Zdenek Kabelac wrote: There is a OSS driver upcoming, supported by the chip manufacturer themselves (AFA Technologies.) I have been a bit involved in that process. Yep - looks nice - and do you have any idea about the release date? I don't have a date, but the chip manufacturer as well as everybody involved is under I mean - is it question for few weeks or couple months? Maybe I could help with testing and reporting bugs ? As I can see it would be probably useless to try hacking any code for this device myself. The current state of testing is that, one needs to understand the chip internals and the testing happens in the Labs alone (lot of undocumented things in there, documentation being really crap), with some basic tests we have going on. As soon Well I wouldn't be afraid even from this kind of testing - I think I have enough skills for this. If you have some pieces of chip docs you could pass to me... I have redone the driver a few times by now, trying to handle the different aspects. There have been some discussions on this ML also, on the same a while back. Actually I've done some search through the list and I've found this: It looks like there are actually 3 different drivers for Afatech chip: (and look quite different) 1.) on from you: http://jusst.de/hg(last commit before 7 weeks) 2.) another from Antti Palosaari (crope) 3.) by afatech ?? http://af.zsolttech.com/ by Rick Huang Which leads me to the question - is it actually coordinated somehow? What is the plan - which driver will go to kernel ?? (IMHO the current situation looks a bit weird at this moment) And when you are say you are working on some driver - by this you mean only af901x ? (So I could try those other two if they will not work by some miracle) Or when you release your code my AverTV HX will work out-of-the-box? (with the support for SAA and TDA chips) Regards Zdenek ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Consistent kernel crash on latest hg repo with 2.6.24-rc6
Erik van Zijst wrote: So far all tests have been done using 64 bit distro's and kernels. I've just tried a fresh, vanilla 32-bit Ubuntu 7.10 install (kernel 2.6.22-14), but this still exhibits the same kernel freeze (albeit after a bit more time). Could this somehow be a firmware issue, or should a firmware or badly behaving card never crash the driver? Erik Erik van Zijst wrote: Hi folks, Since I installed a TechnoTrend 1500 Budget (with optional budget-ci CAM board), my kernel crashes completely everytime I run a frequency scan using the w_scan utility. There's no kernel panic message or anything printed to the tty, just instant death. The server box (dual quadcore Xeon) ran perfectly stable until I installed the T-1500 last week, and still runs stable as long as I don't use the card. I started out using a stock 2.6.22-something kernel (vanilla Fedora 8 and Ubuntu 7.10 server), then switched to 2.6.24-rc6 (generated at 2007-12-21) and finally tried today's trunk of http://linuxtv.org/hg/v4l-dvb using hg, on top of my 2.6.24-rc6 kernel. Yet, every combination dies. The kernel usually dies the first time w_scan -x (http://wirbel.htpc-forum.de/w_scan/index2.html version 20071209) is run and then usually dies near the end when w_scan prints tune to:. Occasionally it takes more than one run before it crashes. The box is idle during the scan and there's no other PCI activity. I'm using firmware dvb-fe-tda10046.fw, installed using get_dvb_firmware. regards, Erik van Zijst ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
[linux-dvb] Fwd: Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
Dwaine Garden [EMAIL PROTECTED] wrote: Date: Wed, 2 Jan 2008 08:42:23 -0500
(EST)
From: Dwaine Garden [EMAIL PROTECTED]
Subject: Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
To: [EMAIL PROTECTED]
I have a Digiwave 108G card, so I tried to add the card to the cx88 driver.
I can get the drivers to load, but the card will not scan or pick up any
satellite/analog tv stations. I don't get how code for the two tuners. DVB
and the analog NTSC TV tuner.
I would like to get the card working and get a patch into the tree for this
card. Anybody want to help??
Here is the following chips on the card.
CX23882-19
CX24109-11
CX24123
The card does DVB-S and Analog NTSC/SECAM/PAL tuning. The card has multiple
inputs.
Card Inputs
1 LNB input
1 LNB output
1 AV Input
1 Analog TV Antenna Input
1 FM Antenna Input
Here is my entry to cx88-cards.c and cx88-dvb.c.
[CX88_BOARD_DIGIWAVE_108G] = {
.name = Digiwave 108G,
.tuner_type = TUNER_LG_TDVS_H06XF,
.radio_type = UNSET,
.tuner_addr = ADDR_UNSET,
.radio_addr = ADDR_UNSET,
.tda9887_conf = TDA9887_PRESENT,
.input = {{
.type = CX88_VMUX_TELEVISION,
.vmux = 0,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
}},
.radio = {
.type = CX88_RADIO,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
},
.dvb= 1,
}
},{
.subvendor = 0x14f1,
.subdevice = 0x2319,
.card = CX88_BOARD_DIGIWAVE_108G,
},
static int digiwave_108G_set_voltage(struct dvb_frontend *fe, fe_sec_voltage_t
voltage)
{
struct cx8802_dev *dev= fe-dvb-priv;
struct cx88_core *core = dev-core;
if (voltage == SEC_VOLTAGE_OFF) {
printk(KERN_INFO LNB Voltage OFF\n);
cx_write(MO_GP0_IO, 0xefff);
}
if (core-prev_set_voltage)
return core-prev_set_voltage(fe, voltage);
return 0;
}
static struct cx24123_config digiwave_108G_config = {
.demod_address = 0x55,
.set_ts_params = cx24123_set_ts_param,
};
Jan 2 06:30:47 mythtv kernel: cx2388x v4l2 driver version 0.0.6 loaded
Jan 2 06:30:47 mythtv kernel: CORE cx88[0]: subsystem: 14f1:2319, board:
Digiwave 108G [card=53,autodetected]
Jan 2 06:30:47 mythtv kernel: TV tuner 64 at 0x1fe, Radio tuner -1 at 0x1fe
Jan 2 06:30:47 mythtv kernel: cx2388x alsa driver version 0.0.6 loaded
Jan 2 06:30:47 mythtv kernel: cx2388x dvb driver version 0.0.6 loaded
Jan 2 06:30:47 mythtv kernel: cx88[0]/0: found at :00:09.0, rev: 5, irq:
18, latency: 32, mmio: 0xeb00
Jan 2 06:30:47 mythtv kernel: tuner 0-0061: chip found @ 0xc2 (cx88[0])
Jan 2 06:30:47 mythtv kernel: tuner 0-0061: type set to 64 (LG TDVS-H06xF)
Jan 2 06:30:47 mythtv kernel: cx88[0]/0: registered device video0 [v4l2]
Jan 2 06:30:47 mythtv kernel: cx88[0]/0: registered device vbi0
Jan 2 06:30:47 mythtv kernel: cx88[0]/0: registered device radio0
Jan 2 06:30:47 mythtv kernel: cx88_audio: probe of :00:09.1 failed with
error -2
Jan 2 06:30:47 mythtv kernel: cx88[0]/2: found at :00:09.2, rev: 5, irq:
18, latency: 32, mmio: 0xed00
Jan 2 06:30:47 mythtv kernel: cx88[0]/2: cx2388x based dvb card
Jan 2 06:30:47 mythtv kernel: DVB: registering new adapter (cx88[0]).
Jan 2 06:30:47 mythtv kernel: DVB: registering frontend 0 (Conexant
CX24123/CX24109)...
Jan 2 06:30:47 mythtv kernel: cx2388x blackbird driver version 0.0.6 loaded
I did get some information from Windows XP with regspy. Here is what I found.
CX88_MO_GP0_IO Always equals 00ff, Never changes at all.
CX88_MO_GP1_IO Changes when using DVB, TV Analog, Radio from f31d to
f39d. Meaning that f39d is ON.
CX88_MO_GP2_IO Changes constantly looks like the remote.
CX88_TS_INTMSK is at 000, when using DVB it changes to 001f1101
CX88_VIDEO_INPUT sits at 0781 for Radio, DVB and Analog TV. When scanning
it changes to 0181. When using Analog TV and changing inputs, it looks
like this.
TV Analog = 0781
S-Video (Not Present on Card) = 00018d91
Composite (Not Present on Card) = 4d81
Here is some information from the Windows driver inf file.
;
; Driver Configuration Registry Entries
;
;---
; AddReg sections for TS filter
;---
[CXTS.AddReg]
HKLM,System\CurrentControlSet\Control\MediaCategories\%TSPinID%,Display,1,
00,00,00,00
HKLM,System\CurrentControlSet\Control\MediaCategories\%TSPinID%,Name,,%TSPinName%
;---
;
[linux-dvb] Fwd: Re: Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
Dwaine Garden [EMAIL PROTECTED] wrote: Date: Wed, 2 Jan 2008 08:49:29 -0500 (EST) From: Dwaine Garden [EMAIL PROTECTED] Subject: Re: Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver To: [EMAIL PROTECTED] Here is a picture and information for the device. http://www.geniatech.com/pa/tvstar.htm Dwaine Garden [EMAIL PROTECTED] wrote: I have a Digiwave 108G card, so I tried to add the card to the cx88 driver. I can get the drivers to load, but the card will not scan or pick up any satellite/analog tv stations. I don't get how code for the two tuners. DVB and the analog NTSC TV tuner. I would like to get the card working and get a patch into the tree for this card. Anybody want to help?? Here is the following chips on the card. CX23882-19 CX24109-11 CX24123 ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Tuner drifts out of lock after tzap is exited
On Jan 2, 2008 12:14 PM, Roger James [EMAIL PROTECTED] wrote: If have been struggling for a couple of days trying to get decent NIT dumps off a local multiplex. Using tzap to tune a channel, exiting tzap and then running dvbsnoop, it seemed very variable whether I found the table before timing out. I noticed I had more success I kept retuning the frontend using tzap before each dvbsnoop run. But even then it would quite often produce nothing. I knew the driver and rf side setup was good because mtyhtv was using all the devices and multiplexes successfully. On checking the kernel logs I found I was getting a lot of cx8802_timeout messages on unsuccessful runs. So I tried a test running tzap, waiting till the frontend had lock, then exiting tzap and running femon. As I suspected, after a few seconds femon reported that the frontend had lost lock. So I a tried the dvbsnoop test with tzap left running in another session. Result, everything worked perfectly. Exactly as expected. I am running debian stable with a 2.5.19.2 kernel. Is this expected behaviour? Can anyone explain to me what is happening? Yes. The moment tzap is stopped, tuning ceases. Also, if you don't specify -r to tzap, you'll get nothing from the dvr device. In the past I had always assumed that could exit tzap after you had tuned a channel and the fe would stay locked as long as signal remained good. Nope. Always keep tzap running when performing this type of testing. Also, you sent this email to the video4linux mailing list, which only deals with analog video. cc added to linux-dvb, which is more appropriate for this topic. I hope this helps. Regards, Mike ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] HVR4000 DVB-S2 - What am I doing wrong ;-)
Maybe I can try with VDR again now as it seems like I can get a lock! Yes, I'm strictly recommend you to install the latest vdr 1.5.12 version with today's greatest h.264 + dvb-s2 patches from Reinhard Nissl http://linuxtv.org/pipermail/vdr/2008-January/014956.html @Reinhard thanks a lot of for your hardest h.264's implementation in VDR project Igor ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] AverTV Hybrid Volar HX
Zdenek Kabelac wrote: 2008/1/1, Manu Abraham [EMAIL PROTECTED]: Zdenek Kabelac wrote: There is a OSS driver upcoming, supported by the chip manufacturer themselves (AFA Technologies.) I have been a bit involved in that process. Yep - looks nice - and do you have any idea about the release date? I don't have a date, but the chip manufacturer as well as everybody involved is under I mean - is it question for few weeks or couple months? I guess it would be a few weeks. In fact working on a similar but different device as of now. Maybe I could help with testing and reporting bugs ? As I can see it would be probably useless to try hacking any code for this device myself. The current state of testing is that, one needs to understand the chip internals and the testing happens in the Labs alone (lot of undocumented things in there, documentation being really crap), with some basic tests we have going on. As soon Well I wouldn't be afraid even from this kind of testing - I think I have enough skills for this. If you have some pieces of chip docs you could pass to me... I would suggest that you request AFA for docs. One needs to be under NDA or so. (Though the docs are quite confusing) I have redone the driver a few times by now, trying to handle the different aspects. There have been some discussions on this ML also, on the same a while back. Actually I've done some search through the list and I've found this: It looks like there are actually 3 different drivers for Afatech chip: (and look quite different) 1.) on from you: http://jusst.de/hg(last commit before 7 weeks) 2.) another from Antti Palosaari (crope) 3.) by afatech ?? http://af.zsolttech.com/ by Rick Huang Which leads me to the question - is it actually coordinated somehow? What is the plan - which driver will go to kernel ?? (IMHO the current situation looks a bit weird at this moment) Nothing weird, the chip is a bit flexible, with that flexibility it brings in a lot of troubles. 1) a second attempt at getting it right (got many things right, but some went wrong. Works with just one device alone) 2) a RE driver for the AF9015 chip based, for a MASTER device based configuration. 3) an AF9015 driver which was done by AFA for one customer, who violated the NDA, which is almost the source for 2 but very much specifically written for one tuner in mind. According to AFA Tech, None of these they would like to have in kernel, since they wouldn't be able to support newer configurations based on them. And when you are say you are working on some driver - by this you mean only af901x ? (So I could try those other two if they will not work by some miracle) Neither 3 will work. Lot of differences, what's common is very small. Or when you release your code my AverTV HX will work out-of-the-box? At least digital should work for you (when done), though there will be the need for a small stub additionally. Regards, Manu ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] VP1032 + Cryptoworks
David Santinoli wrote: On Thu, Dec 27, 2007 at 12:46:38AM +0400, Manu Abraham wrote: David Santinoli wrote: On Wed, Dec 19, 2007 at 11:49:42PM +0400, Manu Abraham wrote: en50221_app_ai_parse_app_info: Received short data This requires quite a bit as to what messages are sent to the driver. Also possible is a bug in the driver, which can't be ruled out either. Is there any other relevant info I can post to the list to help debugging? I could even allow root access to the machine in question. Do you think the CAM could be ruled out, or should I try a Philips Cryptoworks module instead of the SCM one? I don't think it's an issue with the CAM. Looks like a bug on second thoughts :-( Any chance of a fix in the near future? Alternatively, could you suggest another CI card with a less problematic driver? Nothing immediate is expected. It will take time to diagnose what the problem is itself. If you need to look at another card, something based on the SAA7146 might work out. Regards, Manu ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
[linux-dvb] repost: Tuner drifts out of lock after tzap is exited
On Jan 2, 2008 12:14 PM, Roger James roger at beardandsandals.co.uk wrote: I have been struggling for a couple of days trying to get decent NIT dumps off a local multiplex. Using tzap to tune a channel, exiting tzap and then running dvbsnoop, it seemed very variable whether I found the table before timing out. I noticed I had more success I kept retuning the frontend using tzap before each dvbsnoop run. But even then it would quite often produce nothing. I knew the driver and rf side setup was good because mtyhtv was using all the devices and multiplexes successfully. On checking the kernel logs I found I was getting a lot of cx8802_timeout messages on unsuccessful runs. So I tried a test running tzap, waiting till the frontend had lock, then exiting tzap and running femon. As I suspected, after a few seconds femon reported that the frontend had lost lock. So I a tried the dvbsnoop test with tzap left running in another session. Result, everything worked perfectly. Exactly as expected. I am running debian stable with a 2.5.19.2 kernel. Is this expected behaviour? Can anyone explain to me what is happening? Yes. The moment tzap is stopped, tuning ceases. Also, if you don't specify -r to tzap, you'll get nothing from the dvr device. In the past I had always assumed that could exit tzap after you had tuned a channel and the fe would stay locked as long as signal remained good. Nope. Always keep tzap running when performing this type of testing. Also, you sent this email to the video4linux mailing list, which only deals with analog video. cc added to linux-dvb, which is more appropriate for this topic. I hope this helps. Regards, Mike Mike Sorry about the intial post to the wrong list. Tnanks for the heads up. The doumentation I have read does not make the fact that you have to leave tzap running clear. It does seem a bit counterintuitive for an app that is named afer a zapper/remote, you don't expect to keep having to point your zapper at the TV to keep the channel tuned :-). Can anyone point me at somewhere were this is documented/explained. Is there some active element in the frontend tuning process that is disabled once tzap exits and closes its interfaces to the dvb driver? Thanks, Roger ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
I have added the following code to cx88-cards.c and it seemed to activate the
card. I can now here the regulator for the LNB switch on. Also, I'm getting
both signal and SNR values that look better than before. I also get locks on
radio channels when scanning.
Is there a way to confirm the actual analog TV tuner for the device? I think
this is the code that is missing.
[CX88_BOARD_DIGIWAVE_108G] = {
.name = Digiwave 108G,
.tuner_type = TUNER_PHILIPS_FQ1216ME,
.radio_type = UNSET,
.tuner_addr = ADDR_UNSET,
.radio_addr = ADDR_UNSET,
.tda9887_conf = TDA9887_PRESENT,
.input = {{
.type = CX88_VMUX_TELEVISION,
.vmux = 0,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
}},
.radio = {
.type = CX88_RADIO,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
},
.dvb= 1,
}
},{
.subvendor = 0x14f1,
.subdevice = 0x2319,
.card = CX88_BOARD_DIGIWAVE_108G,
},
void cx88_card_setup(struct cx88_core *core)
{
static u8 eeprom[256];
if (0 == core-i2c_rc) {
core-i2c_client.addr = 0xa0 1;
tveeprom_read(core-i2c_client,eeprom,sizeof(eeprom));
}
switch (core-board) {
case CX88_BOARD_DIGIWAVE_108G:
cx_write(MO_GP0_IO, 0x00ff);
cx_write(MO_GP1_IO, 0xf39d);
break;
___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
Hi Dwaine,
Am Mittwoch, den 02.01.2008, 18:03 -0500 schrieb Dwaine Garden:
I have added the following code to cx88-cards.c and it seemed to activate the
card. I can now here the regulator for the LNB switch on. Also, I'm getting
both signal and SNR values that look better than before. I also get locks on
radio channels when scanning.
Is there a way to confirm the actual analog TV tuner for the device? I think
this is the code that is missing.
the FQ1216ME with tda9887 is the most doubtfully analog tuner in it's
current shape I know about. There are some old multi standard PAL/SECAM
tuners in a similar uncertain condition, but at least they don't claim
suddenly to be compatible with more recent MK3/4/5 tuners too.
For the two physically existing devices I imagine, none of them has a
tda9887. For the more recent FQ1216ME H-3 (MK3) it is a tda9886, means
no analog radio support on that analog TV tuner. It would need an extra
radio tuner chip for that then.
Those are all still huge tin/can tuners, but from the fuzzy picture I
saw, you definitely seem to have some sort of silicon behind the analog
antenna connectors.
Likely it is worth to have a closer look at these chips again.
For what is confused currently with FQ1216ME, you should be either far
away from it or you are close to shed some light on it, how it could
come into this stage.
Interesting enough, that you seem to get no errors on analog tuning
attempts.
Cheers,
Hermann
[CX88_BOARD_DIGIWAVE_108G] = {
.name = Digiwave 108G,
.tuner_type = TUNER_PHILIPS_FQ1216ME,
.radio_type = UNSET,
.tuner_addr = ADDR_UNSET,
.radio_addr = ADDR_UNSET,
.tda9887_conf = TDA9887_PRESENT,
.input = {{
.type = CX88_VMUX_TELEVISION,
.vmux = 0,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
}},
.radio = {
.type = CX88_RADIO,
.gpio0 = 0x00ff,
.gpio1 = 0xf39d,
},
.dvb= 1,
}
},{
.subvendor = 0x14f1,
.subdevice = 0x2319,
.card = CX88_BOARD_DIGIWAVE_108G,
},
void cx88_card_setup(struct cx88_core *core)
{
static u8 eeprom[256];
if (0 == core-i2c_rc) {
core-i2c_client.addr = 0xa0 1;
tveeprom_read(core-i2c_client,eeprom,sizeof(eeprom));
}
switch (core-board) {
case CX88_BOARD_DIGIWAVE_108G:
cx_write(MO_GP0_IO, 0x00ff);
cx_write(MO_GP1_IO, 0xf39d);
break;
___
linux-dvb mailing list
linux-dvb@linuxtv.org
http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
[linux-dvb] Theater Pro 650 PCIe
I should have an MSI Theater 650 Pro PCIe hybrid TV tuner in the mail soon. I don't have any experience with TV tuners under Linux. What can I do to get this card working under Linux? ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
Looking at the chip which is right beside the Analog TV input. It says TBA2028E3. Is that the Xceive 2028 tuner chip? Dwaine hermann pitton [EMAIL PROTECTED] wrote: Hi Dwaine, Am Mittwoch, den 02.01.2008, 18:03 -0500 schrieb Dwaine Garden: I have added the following code to cx88-cards.c and it seemed to activate the card. I can now here the regulator for the LNB switch on. Also, I'm getting both signal and SNR values that look better than before. I also get locks on radio channels when scanning. Is there a way to confirm the actual analog TV tuner for the device? I think this is the code that is missing. the FQ1216ME with tda9887 is the most doubtfully analog tuner in it's current shape I know about. There are some old multi standard PAL/SECAM tuners in a similar uncertain condition, but at least they don't claim suddenly to be compatible with more recent MK3/4/5 tuners too. For the two physically existing devices I imagine, none of them has a tda9887. For the more recent FQ1216ME H-3 (MK3) it is a tda9886, means no analog radio support on that analog TV tuner. It would need an extra radio tuner chip for that then. Those are all still huge tin/can tuners, but from the fuzzy picture I saw, you definitely seem to have some sort of silicon behind the analog antenna connectors. Likely it is worth to have a closer look at these chips again. For what is confused currently with FQ1216ME, you should be either far away from it or you are close to shed some light on it, how it could come into this stage. Interesting enough, that you seem to get no errors on analog tuning attempts. Cheers, Hermann ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
[linux-dvb] initial scan file for Australia - Tamworth
Attached, Scan file for Tamworth, NSW, Australia, Mt Soma Transmitter. Note that as at January 3rd 2008, channel NBN are yet to commence broadcasting in Digital. au-Tamworth-Soma-Mt Description: Binary data ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
Re: [linux-dvb] Digiwave 108G/GeniaTech TVStar Card - Adding it to the CX88 Driver
Am Mittwoch, den 02.01.2008, 23:21 -0500 schrieb Dwaine Garden: Looking at the chip which is right beside the Analog TV input. It says TBA2028E3. Is that the Xceive 2028 tuner chip? Dwaine 2028 looks suspicious, but as far as I remember, they have been marked differently previously. I'll try to come back to it ASAP. There is so much out in the wilderness, others already seem to care for ;) Likely, we'll have already people knowing better ... Hermann hermann pitton [EMAIL PROTECTED] wrote: Hi Dwaine, Am Mittwoch, den 02.01.2008, 18:03 -0500 schrieb Dwaine Garden: I have added the following code to cx88-cards.c and it seemed to activate the card. I can now here the regulator for the LNB switch on. Also, I'm getting both signal and SNR values that look better than before. I also get locks on radio channels when scanning. Is there a way to confirm the actual analog TV tuner for the device? I think this is the code that is missing. the FQ1216ME with tda9887 is the most doubtfully analog tuner in it's current shape I know about. There are some old multi standard PAL/SECAM tuners in a similar uncertain condition, but at least they don't claim suddenly to be compatible with more recent MK3/4/5 tuners too. For the two physically existing devices I imagine, none of them has a tda9887. For the more recent FQ1216ME H-3 (MK3) it is a tda9886, means no analog radio support on that analog TV tuner. It would need an extra radio tuner chip for that then. Those are all still huge tin/can tuners, but from the fuzzy picture I saw, you definitely seem to have some sort of silicon behind the analog antenna connectors. Likely it is worth to have a closer look at these chips again. For what is confused currently with FQ1216ME, you should be either far away from it or you are close to shed some light on it, how it could come into this stage. Interesting enough, that you seem to get no errors on analog tuning attempts. Cheers, Hermann ___ linux-dvb mailing list linux-dvb@linuxtv.org http://www.linuxtv.org/cgi-bin/mailman/listinfo/linux-dvb
