[Theora-dev] Directshow filters 0.64.7878

illiminable ogg at illiminable.com
Sun Oct 3 07:26:38 PDT 2004

----- Original Message ----- 
From: "Robert Huitl" <theora-dev at huitl.de>
To: <theora-dev at xiph.org>
Sent: Sunday, October 03, 2004 9:49 PM
Subject: Re: [Theora-dev] Directshow filters 0.64.7878
>> It should do, because the filters use that interface... however what it
>> most likely is, is because the encoder settings get  finalised at the 
>> time
>> the filter connects, if the filter is already connected, those settings
>> won't be loaded into the actual encoder. I'll do something about that 
>> when
>> i get the chance. In the meantime, if you set the settings on the encoder
>> before you connect it up, it should work.
>Well... no ;)  I set the properties on the unconnected filter:

Actually i meant before it is connected to the downstream filter... but is 
connected to the upstream one (sorry wasn't very specific)... i'll really 
have to take another look at it... i haven't looked at that code for a 

I'm quite busy the start of this week, so it may be a few days before i get 
the chance... but i'll definately get on it ASAP.

Also here http://www.illiminable.com/ogg/enc_theora_graphedit.html is a 
description of how/when to do it in graphedit, and maybe some other info... 
if you build your graph similarly it will hopefully  work for you.

Here's the code that implements the interface, it maps directly onto a 
theora-info structure. (bottom 5 or 6 methods)

I'm unsure why it wouldn't return exactly what you put in... AFAIR the only 
times that structure is changed is....

a) When the upstream filter connects... it gets reset and the info from the 
source is added (like frame size/rate etc).
b) When you change it on the interface.

Sorry about that, i'll get onto it as soon as i can... and at some point 
very soon there will even be nice pretty documentation... which will make 
everyones life easier :)

Thanks for taking the time to look into it and report it.



More information about the Theora-dev mailing list