On Sat, Sep 8, 2012 at 6:45 PM, Dan Dennedy wrote:
> I found a regression related to fixes for using mlt_cache for frames
> and the way position is set/get on frames. The proper fix for this
> requires a big cleanup with respect to how nearly every producer today
> must set two positions (one gen
Branch: refs/heads/master
Home: https://github.com/mltframework/mlt
Commit: 0aa230434b3a44495484e389176cb6eeeb9a
https://github.com/mltframework/mlt/commit/0aa230434b3a44495484e389176cb6eeeb9a
Author: Dan Dennedy
Date: 2012-09-08 (Sat, 08 Sep 2012)
Changed paths:
Branch: refs/heads/master
Home: https://github.com/mltframework/mlt
Commit: 6a5e8781dade4bb685207c91ddbe91e3391487aa
https://github.com/mltframework/mlt/commit/6a5e8781dade4bb685207c91ddbe91e3391487aa
Author: Dan Dennedy
Date: 2012-09-08 (Sat, 08 Sep 2012)
Changed paths:
On Mon, Sep 3, 2012 at 5:02 PM, Alberto Villa wrote:
> Hi Dan and list!
>
> I'm attaching some patches to improve FreeBSD support.
thank you for the contribution
>
> patch-src__framework__mlt_property.h:
> xlocale was backported to 9-STABLE.
ok
> patch-src__modules__kdenlive__Makefile:
> it is
On a separate note, please be aware that doing a transition between 2
parts of the same producer is going to perform worse than 2 producers
due to having to seek back and forth within the same producer (and the
performance penalty of seeking that is especially bad with some
formats). Separate produ
I found a regression related to fixes for using mlt_cache for frames
and the way position is set/get on frames. The proper fix for this
requires a big cleanup with respect to how nearly every producer today
must set two positions (one general, one custom) on the frame. So, I
am going to take some
Hello Dan,
Following Kdenlive issue #2668 (1), I have a question for you.
Some time ago, I had to change Kdenlive's handling of avformat clips so that
when a clip is used on several tracks, each track uses a different producer.
I clearly remember that I introduced this because otherwise we had a
