Diluting the mystery of Chris Long miracle MacIntel patches

Archive of historical development discussions
Discussions / Development has moved to GitHub
Forum rules
Please be aware we are now using GitHub for issue tracking and feature requests.
- This section of the forum is now closed to new topics.

Post Reply
Veteran User
Posts: 2888
Joined: Tue Apr 18, 2006 10:24 pm

Diluting the mystery of Chris Long miracle MacIntel patches

Post by rhester »


Before anyone gets their hopes up too much about the still-outstanding patches from Chris Long, please let me be quite clear on what they are (and aren't):

When Chris and I first agreed to try to pull HandBrake out of the ashes, x264 had come a long way from the revision bundled with 0.7.0. It made good sense at the time to take a look at the tip svn revision, figure out how (or if) it could be smoothly reintegrated with HandBrake, and use that as a basis for future development.

The good news is that libhb changes required to integrate the newer revision were minimal (but we still managed to bork them - thanks to you all for finding out flaws ;).

The bad news was that x264, at that time, was nowhere near compilable out-of-the-box on MacIntel.

Chris' work on patching was to achieve nothing more than a libx264 capable of compiling somewhat cleanly and executing without coredumping on MacOS X for Intel. Nothing more, nothing less. As I'm sure you've seen, work on that front has already been contributed in more recent days by Chris Long, titer, and others directly to the x264 project, and as such the value of the now-dated patches by Chris for HandBrake 0.7.1a1 is now quite diminished. I don't mean to downplay the great work he did, as it was invaluable at the time, but the reality is that libx264 now compiles quite effectively and runs quite stably on MacIntel with the patches we are currently using against revision 604.

So, the short version of the story is, we don't actually need the patches you guys are all waiting for from Chris, because they have since more-or-less been directly incorporated into x264 (which was only a matter of time, since it's clearly used in more MacIntel projects than just HandBrake...in fact, we come rather late to the game in that regard).

Hope this didn't disappoint anyone - but hopefully it encourages everyone, because there's no reason to wait any longer to continue full-blown development on HandBrake! =)

Posts: 28
Joined: Mon Dec 18, 2006 3:38 am

Post by mk2000 »

So can we interpret that statement to say that if we meet the pre-requisites (Jam, NASM) and compile HandBrake on Intel that it will come out correctly?

Thanks Rodney!

Veteran User
Posts: 3820
Joined: Thu Nov 02, 2006 8:19 pm

Post by dynaflash »

SndChsr was our tester for clean Intel Mac Compile. He reported a clean compile right out of the svn (Rev71) as far as I know. jbrjake and I got the clean compiles for PPC. I belive rhester was Linux.

We decided to go ahead with a rev71 without any BeOS testing (no one could think of anyone that uses it or even had it).
Post Reply