On 6/13/19 1:46 AM, JP wrote:
On 6/13/19 12:31 AM, Antti Palosaari wrote:
On 6/12/19 11:07 PM, Frantisek Rysanek wrote:
On 12 Jun 2019 at 1:28, Antti Palosaari wrote:
[...]
What is that T230C2 stick?
JP has already explained the details, how that name was arrived at.
As previously suggested, I can call it T230C v2 in the descriptive
texts. I'd suggest keeping T230C2 in the USB ID macro (or suggest
a more appropriate name for the macro).
Here in CZ, a company called Abacus imports and distributes consumer
electronics gadgets under a private brand "EvolveO" - and this is how
the "rebadged OEM Mygica" has reached me.
http://m.evolveo.com/cz/sigma-t2
This particular T2 dongle is "allover the place" around here, no
other dongle is this broadly available. (Well on our modest market.
We're a nation of 10M people.)
Naming sounds like a DVB-C2 capable, but I
found only T230C model from MyGica site.
The local brand's site only mentions DVB-T2.
The 2-page "brief datasheet" of the si2168 that's publically
available only mentions DVB-C, apart from T/T2.
And also patch should be split to two logical parts, first add
manual ts
frequency support to si2168 and then other patch which adds device
itself.
I'll try to find some time and massage that approach into the code.
I have read all the past attempts (example patches) and the
maintainer's polite objections.
And which are tuner and demod versions/revisions used for that device?
That's reported in dmesg if memory serves... I'll try to find the
answer.
Frank Rysanek
Yeah, all-in-all:
1) name it T230C v2
2) use manual ts clock speed
And according to old usb sniffs from pctv 292e [Si2168B] default
manual ts clock is set to 7.2MHz, which means 57.6Mbit/s datarate, it
should be quite optimal for DVB-T2 max. In theory it could be a
little higher only when 10MHz channel bandwidth and most less error
correction FEC in use. And currently driver is using some config that
uses dynamic ts clock which clocks only when there is data to feed.
For some reason, usb-ts-bridge does not understand that and manual
configuration is needed (ts valid or ts-sync connection?). If
possible use 7.2MHz, if not: set to 10MHz.
That's perfectly alright with me. I'm now testing that 7.2Mhz value.
Hold on.
The driver crashes with the 7.2Mhz value! That was totally not what I
ever expected.
Recompiled the whole kernel: crashes again. Then tried on debian kernel
4.19: same thing.
Food for thought?
And we were thinking in that same line of ts_sync connection.
BTW this is the link for where that value of 10Mhz stems from:
https://github.com/ljalves/linux_media/issues/164#issuecomment-335011689
JP
