Taylor Jacob wrote: >I have been very busy with work today, and would have posted this earlier.. >Sorry. > >Here is the "cleaned up" patch for adding support for the Air2PC/Airstar2 HD5000 >(LGDT3303 Frontend). > >http://www.digitalregime.com/patches/hd5000-cleanup.patch > >Changes to note from the last one I submitted are: > >- I have tried to impliment the SNR for the LGDT3303 since it appears to be the >same formula for the 3302 and 3303.. > >- I have also tried to impliment the signal level based on information I was >provided by BBTI.. I am not sure if this is correct for the 3302, and it is a >unitless measure since I have no idea how it was calcualted.. I would need to >spend some time digging it spec sheets for both and see if I can come up with >something, but I only have the 3303 right now.. > >- The serial_mpeg value needed to be set to 0x04 for the demux to return data.. >I am not sure as to why but the suggested value of 0x40 didn't return any data >from teh demux. (Could this have something to do with the wiring of the >frontend to the demux?) > >- The frontend reset delay added to flexcop.c was reduced from 100 or 200ms to a >reasonable 1ms which gives adequate time for the 3303 to reset.. > >Also to note: > >I have also put together a few liner patch to make the older version of the >DVICO FusionHDTV Gold3 work.. I have one that has a LGDT3302 and has a card id >of 18ac:d800 instead of 18ac:d820 (CX88_BOARD_DVICO_FUSIONHDTV_3_GOLD_T), but >appears to be the same.. I will validate this and post this seperately.. > >Taylor > > Taylor- Too many changes in one patch... Please regenerate into multiple patches in the following manner: Patch #1 should be a trivial patch, similar to your first one, that adds AirStar2 HD5000 support to the flexcop driver, without changing anything else, unnecessarily. This first patch can most likely be applied immediately. Patch #2 should contain the changes that you want to make to lgdt330x frontend module. I'd like Mac Michaels to approve your second patch before it gets applied to cvs, as he is the original author of this driver. Also, please see my previous email containing comments from Mac about this SNR issue. http://linuxtv.org/pipermail/linux-dvb/2005-September/005001.html Patch #3 should contain any changes to the AirStar2 HD5000 card support that depend on Patch #2. (maybe it's okay to combine patch #2 and patch #3, but it's best to have them separate.) Please follow my advice on this. I have no problem applying your card support patch, as it has already been proven that it works by some testers that have posted to the list. However, I have three other lgdt330x cards that I would prefer to test first, before applying your lgdt330x changes to cvs. When making so many changes like this, it is ALWAYS preferred to send these patches separately, so that they can be reviewed, tested and applied separately. This way, if there is a problem, we can apply the safe patches, and only reject the problem patches. Mac has made it clear that the SNR issue is not as simple as you seem to think. It is very possible that you've looked at the situation in different light, and you may have discovered a better method to do this. Nonetheless, Mac should take a look at your changes to lgdt330x before we apply them to cvs. About the FusionHDTV3 Gold card, Mac programmed the Gold-Q model (18ac:d810), and I have programmed the Gold-T model (18ac:d820) ... I'd be happy to add a definition for your other model... Please either forward all the info that you have on that card, or send me a patch against video4linux cvs. If the card is programmed exactly like one of the other models, namely GPIO, then we can just add a subsystem ID reference pointing to the other model in cx88-cards.c ... Please verify the GPIO programming using the RegSpy program included in the dscaler project. You can send your FusionHDTV3 patch directly to me, and I will apply it to video4linux cvs for you. Cheers, -- Michael Krufky
