[mythtv-users] VIA EPIA LN10000EG
mikep at randomtraveller.org.uk
Thu Jul 31 15:23:33 UTC 2008
Udo van den Heuvel wrote:
> Nick Rout wrote:
>> On Thu, Jul 31, 2008 at 2:49 PM, Udo van den Heuvel <udovdh at xs4all.nl> wrote:
>>> Nick Rout wrote:
>>>>> I use my EN12000 as both.
>>>> No but the original poster asked about front end use, so stay on topic please.
>>> mythbackend runs most of the time.
>>> frontend runs less often on my box.
>>> because of the supposed leaks people see in the ebackend I would not be
>>> surprised if the frontedn suffers from the same issues.
>> To say that is just FUD. You have no basis for saying there is a leak
>> in the frontend.
> I don't have anything indeed.
> But assuming the same quality that is in the backend goes into the
> frontend... (why make a difference)
> See where I get? (there are several locations in the code where leaks
> are assumed; if we do a codesize comparison between frontend and backend
> we can distribute the leaks to that ratio)
> I don't use the frontend that often but of course I can run similar
> stats on the frontend as I do on the backend and give you the needed
> `proof` of assumed leaks.
> Sounds fair?
That's not the kind of remark that anyone who writes code will understand. Just
because two pieces of code have a certain size ratio implies *nothing* about the
performance of either. Even if they are written by the same people. Even if they
are parts of the same system.
The various parts of any system are designed for different purposes, and it may
be the code which handles a specific purpose, or device driver, or tuner, or
whatever, *or combination*, that is causing the problems you are apparently seeing.
Just saying that because two programs come out of the same codebase means they
will behave similarly is not useful. After all, Volvo makes both trucks and
sports cars, doesn't it?
More information about the mythtv-users