[MPlayer-dev-eng] Adding threaded functionality to mplayer

[Ed Wildgoose]

>lol, do you realy believe that PCI-Express will solve the I/O issue ?
>  
>

My mistake, I meant PCI-Express, not PCI-X.

Any, yes, I think it will solve a huge amount of bandwidth limitations 
on the current architecture.  As you point out, network cards in the 
gigabit range are suddenly feasible, etc

>MPlayer is mostly about two things: performance and stability.
>And if you think that threading will not have any drawback on
>performance then you should walk out of your university and have a look
>at the real world. 
>

I think I wrote a message a few clicks back where I have retracted the 
suggestion to thread.  I only claimed that it might be cleaner, not 
faster (although with multiple CPU's it might be).

I see the resistence, and I have no desire to run a whole fork myself.  
Please don't jump down my throat I was only trying to ask a general 
design question since I was thinking quite seriously about having a bash 
at cleaning up the main() loop.  Please try to offer me some support and 
not drive me a way?

I will bow to your experience about the overhead of multi-threading 
apps.  Personally I have little experience, but I do some work on MythTv 
which is a heavily threaded design, and also nowhere near as efficient 
as mplayer either.


>A nice top design with a hackish implementation isnt better than a
>hackish implementation with no design.
>  
>

You are putting words into my mouth.  At no point did I start a hackish 
design.  I only asked some questions before starting on some work - this 
sounds sensible to me - yes?


Sorry for putting people's backs up.  We all have to learn somewhere.  I 
am still interested in streamlining main() though if I can

Good luck all

Ed W