[mythtv] What other people tell me about CC...

Kevin Kuphal kkuphal at gmail.com
Tue Oct 21 13:45:36 UTC 2008


On Tue, Oct 21, 2008 at 5:28 AM, Linuxguy123 <linuxguy123 at gmail.com> wrote:

> Here is a sample of the emails I receive on channel changing.


I'll make a couple comments based on my observations (and I certainly don't
speak for all developers) but again, this is really more noise than signal
here because what I'm going to say probably isn't anything new if you look
back at the previous discussions

>
>
> ====================================================================
> For what it's worth, you will find MANY users that want faster live
> tv. I think that the majority of the developers are just bitching
> because they don't care about it and don't want to read about it.
> That's their problem. Stupid if you ask me. The kneejerk response
> becomes "myth wasn't designed for live tv. That's not what it's for"
> which is complete bullshit, because that is what a DVR does. Then they
> go into debates about channel flipping and how much time it wastes,
> and on and on. It is like they are trying to dictate how myth users
> watch their tv. I agree, myth does a very good job of scheduling and
> the devs have put a TON of work into making a stable system, but I
> don't think it's perfect in the least. I think there is likely room
> for tons of code optimization. I strongly encourage you to keep up the
> good fight, even if the devs are reluctant to help.
>
> ====================================================================


The reality is, none of the *active developers* have an overwhelming desire
to improve this feature.  I would venture to guess that few, if any of them,
use Myth in this way.  Isaac, whose project this is, most likely doesn't.
Until a developer, perhaps such as yourself, cares enough to improve it, it
likely won't.  Non-developers stating "I think there is likely room for tons
of code optimization" are not exactly the best judge of coding quality.

>
>
> Good job on not taking the stock original "this is as good as it gets"
> answer on this problem.  As you said, many people have asked about
> this - so if you can come up with a way to improve it, it would have a
> pretty big impact - and may greatly improve the MythWAF (wife approval
> factor) in a number of households.
>
> =====================================================================


It isn't that no one has come up with a way to improve it.  In fact, the way
has been probably tossed up every time I've seen this discussion:

Don't start recording immediately but instead just start streaming the video
and wait for the user to ask to begin recording

While on the surface this sounds great and may very well be how some DVR
products work, but Myth has some distinct differences that makes this
simplistic approach improbable.  The biggest being that Myth is a *multiple
tuner* system where you can't just lock the tuner and go about your
business.  You need to take the schedule into consideration and thus, there
is some additional processing that needs to be done.  I'm not saying that
this is the source of any delay, but it simply highlights that what may
appear to a user as "Why don't you just do it this way?" really ends up
being far more complicated then they think.

I doubt you'll find massive optimizations in the code itself that will
improve times.  The fact that it varies wildly between different users
points to something outside Myth as being a source of delay.

Good luck with your efforts, but again, there are reasons that this hasn't
been picked up before because it is generally a feature that just isn't used
by the development team and there are always other, more pressing features
that are desired.

Kevin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mythtv.org/pipermail/mythtv-dev/attachments/20081021/7a53fcde/attachment.htm 


More information about the mythtv-dev mailing list