[linux-dvb] ves1820 FEC off? EN 300 468 -> cable:has inner fec, on implementing

[thomas schorpp]

Robert Schlabbach wrote:
> From: "thomas schorpp" <t.schorpp at gmx.de>
> 
>>in 429 the framing shows only outer r-s fec.
>>the latest 468 standard says we HAVE inner fec(!)
> 
> 
> No, it just defines a way to indicate an inner FEC method through the DVB
> Service Information. But since there is no inner FEC method defined for the
> DVB-C transmission standard, the only possible value you can use is:
> 
> 
>>Table 35: Inner FEC scheme
>>1111 no conv. coding

hmm..

>>
>>i'll implement this in the driver until weekend
> 
> 
> What do you want to implement? I suggest you start by examining the DVB
> Service Information you receive (try e.g. dvbsnoop). I'm 99.99% sure you'll
> find "no conv. coding" indicated in your cable delivery system descriptors.

will do, thx.

> 
> 
>>-if- someone gets me programming registers and values or a full
>>datasheet, theres no on the web, or i try using the ves1x93 data...
> 
> 
> The VES1820 is a demodulator for the DVB-C standard and thus has no inner
> FEC (Viterbi) decoder. The circuits for this are simply not there, period.
> Stop thinking about inner FEC, it is definitely _not_ the problem you are
> facing.

thx for clarifying that.

i was wondering about this in ves1820.c:

static int ves1820_read_status(struct dvb_frontend* fe, fe_status_t* status)
{
	struct ves1820_state* state = fe->demodulator_priv;
	int sync;

	*status = 0;
	sync = ves1820_readreg(state, 0x11);

	if (sync & 1)
		*status |= FE_HAS_SIGNAL;

	if (sync & 2)
		*status |= FE_HAS_CARRIER;

	if (sync & 2)	/* XXX FIXME! */ <----???
		*status |= FE_HAS_VITERBI;

but ok.

many thx to all.


> 
> Regards,
> --
> Robert Schlabbach
> e-mail: robert_s at gmx.net
> Berlin, Germany
> 

y
tom