HandBrake

Success! I just transcoded a large queue of items without an issue.

Thanks for the fix.

Thanks, guys. I'll give it a whirl when I get a chance. I really appreciate your effects.

No joy. HandBrake svn5912 (Nightly Build) - 64bit Version OS: Microsoft Windows NT 6.1.7601 Service Pack 1 CPU: Intel(R) Core(TM) i7-3820 CPU @ 3.60GHz Ram: 32708 MB, GPU Information: AMD Radeon HD 7900 Series - 9.12.0.0 Screen: 1920x1200 Temp Dir: C:\Users\Xenos\AppData\Local\Temp\ Install Dir: C:\...

There was a period when the setting was first added, that bicubic was default. The default change won't change the selection as it has no way of knowing if you chose that or it's an old default. Understood. I just wanted to make sure I wasn't doing something stupid. I'm not an A/V guy, and I try no...

If you turn on Lanczos vs Bicubic OpenCL does that make any difference ? OK, I've changed to Lanczos and am rerunning it now. I'll let you know the results. Interestingly, the "leaked bytes" is a lot smaller now (only 1K). BTW, I've noticed that the text in the options seems to imply that...

OK, this time it crashed right away (3rd file). HandBrake svn5912 (Nightly Build) - 64bit Version OS: Microsoft Windows NT 6.1.7601 Service Pack 1 CPU: Intel(R) Core(TM) i7-3820 CPU @ 3.60GHz Ram: 32708 MB, GPU Information: AMD Radeon HD 7900 Series - 9.12.0.0 Screen: 1920x1200 Temp Dir: C:\Users\Xe...

It isn't triggered by a particular file. That is, I can re-run it with the same file and it will work fine. I haven't seen any pattern yet. I will be happy to turn up the verbosity and provide another log. It will be forthcoming.

Sure, I get that. I guess I was just thinking out loud (at it were) that, even though it crashes, the file is OK and HandBrake continues on, I won't be stuck using a back-level version while I am waiting for a fix.

Thanks.

Starting to suspect it's some kind of finalization bug in the new muxers. John might have a better idea. Is the file playable? It looks like it may have crashed just after muxing. Thank you for responding. That's a good thought. Yes, the file is playable in it's entirety. Is is also the exact same ...

OK, I finally found time to play it the latest nightly (5912), and I still get a crash in HandBrakeCLI. I made a large queue and encode with 5878, and it went fine. I then attempted to encode an equivalent queue with 5912 and about an hour in, it crashed (although I think the amount of time may be r...

Yes, it seems to be the same as mine. Same exception number (access violation, which indicates a bad pointer, array index, or stack overrun, etc.), and same offset. If one of the developers could attach a debugger to the running process, the offset would tell you the exact point of the crash.

Ugh. It seems the answer is, "no." Even 35mm spherical undergoes a "pan & scan" treatment when presented as full screen.

http://www.widescreen.org/widescreen_matte.shtml

Thanks for replying. No, that would be trivial. I'm talking about a movie filmed with a negative 4:3 aspect ratio (35mm spherical), matted to 1.85:1 widescreen for theatrical release.

There is a movie that I want, but have been avoiding purchasing because it's never been released in its theatrical widescreen format (1.85:1). But it was filmed in 35mm Spherical and matted to widescreen. So, assuming the DVD is kept that way and not pan & scan, would I be able to get the widesc...

Yeah, I thought about that after I had shut it down, and reinstalled the older version (I had a lot of jobs I wanted to encode overnight). What I will do is wait for the next nightly, and if that crashes, I will be sure to save the encode log.

Thanks for the reply.

The CLI is crashing a lot for me in the latest nightly (5893). Previously, it would never never crash (and I would have it transcoding for days). The last version I had before installing 5893 was 5878. I checked the Windows event log and I see this (for each crash): Faulting application name: HandBr...

Thanks!

There is an issue with the Queue Window in the latest nightly (Windows GUI). If you expand the window horizontally, the start button disappears.

Thanks,

DrXenos

Here you go. Please read the notes.txt file for assumptions the tool makes for things it cannot glean from the old queue files. Let me know if you have problems. The tool is was quick-and-dirty and is very simplistic. As such, it will only process one file at a time. If you want to "batch"...

It's fixed! Thank you so much.

This is from "LibScan.cs" @ 277. The logic looks reversed: HandBrakeUtils.SetDvdNav(this.userSettingService.GetUserSetting<bool>(ASUserSettingConstants.DisableLibDvdNav)); SetDvdNav will use libDVDNav is the argument to SetDvdNav is true, but DisableLibDvdNav is true if the checkbox is set...

Maybe a byproduct of using libhb scanning, rather than the CLI? Edit: judging by your last post, it looks like the bug is dvdnav on/off has it's meaning reversed (your log with it set to off shows it's actually enabled). Thanks. I just going to post that. It does seem reversed. I don't know if it's...

I just tried 0.9.9 with libDVDNav disabled and it correctly found the 7 angles (of course the GUI lists 8 ). For 0.9.8, it's the opposite. libDVDNav finds the 7 angles, but libdvdread does not. I thought libDVDNav supported angles, but libdvdread did not. Am I missing someone, or is there something ...

OK, did some analysis and I figured out why 0.9.9 is always showing the # of angles as two. Comparing the output logs for 0.9.8 and 0.9.9 (latest nightly), the former correctly determines that title 15 has 7 angles. The later seems to go off in la-la land when it get to that point (It's hard to comp...

Has anyone been able to use the angle selection drop-down? It doesn't seem to be functional. It use to work with 0.9.8.