[mythtv] Re:Next Scheduler Patch
mechou at hotmail.com
Tue Mar 2 04:42:09 EST 2004
David and Bruce,
Well, I don't recall the exact time frame, but in my case even though I
had two tuners (one on each machine), the "jarring changes" I'm
referring to were the deferment of (weekslot) recordings even though a
tuner was available. (In other words myth no longer tried to use all
tuners at the earliest opportunity, but had instead changed "strategy"
of using only one tuner as much as possible, in the case of weekslot
records). This had the unfortunate side-effect of pinning virtually all
recordings to the master backend, and rendering the slave backend
unused. The bottom line was there were consequences in terms of storage
allocated to each machine, and the change took me by surprise,
unannounced and unplanned. I suppose these changes could conceivably
have happened prior to myth .12, but it wasn't until myth .13 until the
"user impact" became obvious.
In your prior descriptions of the (new) scheduler, my *perception* was
that the user was now responsible for setting priorities on recordings,
so that myth would "do the right thing." Perhaps this is a
misconception, but nothing you've stated so far has led me to believe
otherwise. To me, even myth .14 got a bit onerous because I now have to
manually override deferred records via the "fix schedules" screen,
virtually all the time. And now as a user of the new scheduler, I've
got the added burden of dealing with program priorities? In some ways I
can't but feel nostalgic for the scheduler of myth .10 :). Please
correct me if this is a misconception.
As a programmer in a former life, I understand the value of prototyping,
which is probably the approach you decided to take. But please keep in
mind, like I said before, some have come to rely on mythtv as an
appliance, and therefore would want to understand the implications or
changes before (fully) adopting them. That's why in standards bodies
and such there is a "request for comments" period where a preliminary
specification drafted up and (hopefully) some meaningful & coherent
discussion takes place.
I certainly don't want to come across unappreciative or ungrateful.
Like I said before, it's clear to me that you and Bruce have given this
matter much thought and work. In this particular case, I'm simply
providing some personal feedback on why I personally would be very leery
of even trying any changes to the scheduler, (especially as there was no
comprehensive explanation or specification such that one could ascertain
the "user impact" prior to this thread) having been bitten by myth .13
scheduler. Perhaps this is one of the reasons, coupled with the fact
that relatively few people have spare myth boxen, why you have gotten
little feedback on the patch itself.
David Engle wrote:
> >As a myth user who uses/gets cvs updates at least once every two weeks,
> >quite frankly the change in schedule conflict resolution between myth
> >.12 and myth .13 was jarring, to say the least. This (unexplained)
> >change just deepens the mystery, without allowing the user to evaluate
> >the potential trade offs a priori.
I have to take some exception with this. There weren't any scheduling
changes made (at least of substance) during that time. What did
happen is that the improved status reporting had the (some would say
beneficial) side-effect of showing what the scheduler was actually
going to do if you didn't intervene.
More information about the mythtv-dev