HandBrake

Haha, I'm not listening to the AC3 track for DRC ... that would be silly as I am indeed passing it through. I'm listening to the AAC track for DRC. The encode is from an AC3 Track and I'm monitoring the DRC on the AAC track on output. Sorry if that wasn't clear.

My problem is I don't know what to expect. Thanks mduell for answering. Here's an encode log for the DRC 1.0 version. I can dig out the DRC 4.0 version when I get home. I don't know how helpful this will be tho ... I don't see anything useful in the logs in the way of audio or DRC related statistics...

I phrased this question to be general case. This is not a support request, it's a how does it work question. At no point do you need a log to relate your experience with this setting. If it's clear there is something wrong after I understand how it works, then I'll post logs. Random sarcasm is alway...

I'm varying the DRC level between 1.0 and 4.0 on an AC3 track where i have problems with dialogue being too soft, etc. Can't really tell any difference at all between the two encodes ... and dialogue is still not any louder relative to the explosions, etc. Is it supposed to be that subtle or does th...

What character is the "proper" character? I have trouble with the Apple TV chopping off subtitles on the second line, but I'm not sure if it is due to whatever is being described here.

Basically, here's what I'm seeing. I have two files of the same source movie ... one 720p, the other 1080p. Both sources were liberal with the bitrate. I am encoding both to 720p using the presets I specified in Devices/Apple TV presets thread ... essentially the high profile preset. Now, If I encod...

Maybe you should post the encodes somewhere so we can compare? I certainly don't see anything unusual. I've seen the occasional unreproducible hitch watching a long movie which I attribute to background caching or something.

I tested the Monsters vs Aliens scene that you are talking about and did not observe any frame skipping. It's possible you have better eyes than me tho. For reference, I simply encoded to 720P with default high profile settings and vbv-maxrate=14000:vbv-bufsize=14000 added to --x264opts. Frame dropp...

Limit the peak bitrate with vbv params to x264. And have you checked to see if your original source is stutter free?

Reports from macrumors that the ATV is decoding 1080p but only outputting 720p. Also curious if the 1080p content being played by folks is playable or syncable to the iPad or iPhone 4 ... I feel like if the A4 was capable of this it would have been caught by now, so the new Apple TV being able to d...

It may not be the bitrate spikes per se but just simply tough scenes to decode. Basing a 1080p encode off the universal or original ATV profile would be worth testing just to narrow down what's happening.

I get mine today so will be playing around later.

Reports from macrumors that the ATV is decoding 1080p but only outputting 720p. Also curious if the 1080p content being played by folks is playable or syncable to the iPad or iPhone 4 ... I feel like if the A4 was capable of this it would have been caught by now, so the new Apple TV being able to do...

1080p would be a huge win. I would bump the vbv to the next highest level (4) ... that would be vbv-maxrate and vbv-bufsize at 20000. Then see what happens.

cmdArgs.add( "--x264opts" ); cmdArgs.add( "--encoder" ); cmdArgs.add( "x264" ); // default high profile settings ... LGTM! cmdArgs.add( "b-adapt=2:rc-lookahead=50:vbv-maxrate=14000:vbv-bufsize=14000" ); // really slow setting ... probably too aggressive. I co...

Well, here's the code snippet in java that I'm using to generate my Apple TV (2010) encodes: In plain English, I run a script before hand that selects the proper source DTS track and converts it to AC3 using the following linked mkvdts2ac3 script. http://forums.macrumors.com/showthread.php?t=836043 ...

What if my source is 60 fps? This isn't really a practical situation for me, I'm just curious. I set --pfr --rate 30 assuming 30 would be a valid setting, but it wasn't when I reviewed the logs.

I noticed in the CLI that "30" is not a valid option for the "--rate" option. I also get a warning about this when using the CLI. Is there a particular reason for this? I believe plenty of true 30 fps (and not just 29.97) videos are out there.

Wow ... if you went that high then it seems the margin for safety is very large. I'm just a bit worried at the thought of doing some mass encoding and ending up with something incompatible. The Apple TV is also streaming so that might put a more serious strain on the parameters if Apple isn't carefu...

I've seen it stated in a number of places while researching x264 options for the A4. I think it might just be implied by the Main Profile lvl 3.1 support.

Have we settled on encoding settings for the new Apple TV yet? I've basically gone with high profile defaults while adding vbv-maxrate=13000:vbv-bufsize=13000 to the x264 options since the stated limit is 14 mbps. In testing with a couple of files this works OK on my iPad so the Apple TV is likely t...

I've been doing the math and mac mini's don't end up being much more efficient than a big 8 core mac pro due to the speed of the i7. The high speed 8 core or 12 core mac pro's may even come out ahead on a KWh/TB basis. If the cluster is idling at all, then the mac mini's do come out ahead by quite a...

I have an xgrid cluster of mac minis uniformly encoding to 720p with all computers going to and from an NFS mount. This is with a distributed job queue, not single file parallelization. Works quite well. You need to get comfortable with the CLI tho and xgrid is a little quirky.

Are you trying to encode a single video quickly or several in parallel using the command line? If the former, get the 6 - core ... if the latter, then the 12-core will have more total throughput.

OK, simple parameters, really. I'm curious what the most energy efficient setup is to encode via Handbrake en masse. Assume multiple computers will be processing multiple files in parallel for long periods of time. So energy overhead per computer counts. Individual speed per computer is not importan...