[mythtv-users] Intel i965 video buffers errors on frontend - live TV

Daryl McDonald darylangela at gmail.com
Fri Jul 21 13:31:20 UTC 2017


On Fri, Jul 21, 2017 at 6:35 AM, evade. <evade at internode.on.net> wrote:

> On 21/07/17 20:17, Ian Campbell wrote:
>
>> On Thu, 2017-07-20 at 03:39 +0000, Mark Perkins wrote:
>>
>>> I don't think MythTV is any different to a standard PVR in that regard.
>>> In the ordinary course of events with modern MythTV (say 0.27 or above)
>>> I don't believe some sort of 10 sec pause is required (assuming
>>> sufficiently capable hardware, proper network config, file system config
>>> etc).
>>>
>>
>> I always had the (possibly ill-informed) impression that it was also
>> quite dependent on the type of tuner you had and the quality of the
>> driver for it. e.g. I wouldn't be surprised if a network based tuner
>> took a little longer to talk to and to begin streaming the new channel
>> than the onboard tuner on some dedicated h/w PVR. Likewise I'm sure
>> that the quality of PCI and USB tuners and their respective drivers
>> varies greatly wrt tuning times independently of any issue myth itself
>> might have. Whereas the h/w and drivers are something that h/w PVR
>> manufacturers may or may not have optimised (and I expect some are much
>> better than others). Plus with a PVR they are only supporting one
>> specific combination of hardware and an integrated driver+application
>> stack (quirks and all) whereas a myth system can host a wide variety of
>> different tuner hardware, all with different quirks.
>>
>> So even if there was someone willing to try and contribute to a
>> seamless live tv experience, it would very likely need to be a full
>> stack effort (so drivers as well as myth) targeting a particular subset
>> of tuners (of course some subset would be generic, but by no means
>> all). For it to be seamless across a wide range of tuner stacks you'd
>> need a wide range of people willing to contribute such things...
>>
>> Annecdata: My parent's Humax (I think?) PVR seemed to me to be fairly
>> clunky for channel surfing, worse than our merely slightly clunky
>> MythTV live experience (which is rarely used for live tv though).
>>
>> (rereading the quoted stuff above I'm not sure now I've picked the
>> right place in the thread to make this comment, oh well)
>>
>> Ian.
>>
>
> Thanks for the insights Ian.  I think you're right.
>
> By supporting a wide variety of hardware it makes sense that there's less
> optimisations of a particular card, it's firmware and drivers.
>
> I wish there was a recommended hardware list, because ideally I'd buy the
> same tuner a developer users!
> There's not a lot of choice in tuners for some countries though...
>
>
> This conversation started as I had naively hoped that by putting a new
> system together with a newer tuner that it could change channel faster.
>
> For reference my old system has two PCI "Technisat SkyStar2" DVB cards
> while my new one has a "DVBSky T982" PCIe card.
> https://linuxtv.org/wiki/index.php/DVBSky_T982  It's using the OpenELEC
> firmware.
>
> I spent some time researching DVB-T PCIe based cards and there wasn't a
> lot of choice.  As it is I had to buy the T982 directly from DVBSky in
> China via email and Paypal!
>
>
> evade.
>


> Is going to the guide, scrolling through the listed programs and deciding
>> on one an acceptable substitute for channel surfing?
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mythtv.org/pipermail/mythtv-users/attachments/20170721/040d72f3/attachment.html>


More information about the mythtv-users mailing list