HandBrake

rollin_eng wrote: ↑Tue Jan 26, 2021 6:54 pm It’s more: You’ve asked for this/similar before, it probably ain’t gonna happen if you don’t do it yourself.

You're trolling... naughty, naughty.

https://docs.github.com/en/github/colla ... l-requests

What am I supposed to conclude from that? Is that a *F*YOU*: If you're not a coder you have no place here? Is that it?

Oh god it's the 5-5 telecine guy again. Yup! Have you tried it? -filter_complex "telecine=pattern=55, split[1][2], [1]select=not(eq(mod(n\,5)\,2))[3], [2]select=eq(mod(n\,5)\,2), pp=linblenddeint[4], [3][4]interleave" I've spent the intervening time learning ffmpeg to a pretty deep depth....

May I give you an example of what Handbrake currently does?

Suppose I represent 24000/1001FPS frames via my notation, thusly:
[A.a][B.b][C.c][D.d]
Setting CFR=60000/1001 in Handbrake produces this:
[A.a][A.a][B.b][B.b][C.c][C.c][C.c][D.d][D.d][D.d]
It's offal. The judder makes it almost unwatchable.

Since Handbrake cannot possibly support the vast number of ffmpeg filters and combinations, help with filters inserted here is best obtained directly from ffmpeg (ffmpeg.org). Sorry. I revised this several times prior to posting. While revising, I lost track of the sentence subject, so I failed to ...

The issue may be that enabling these ffmpeg options may cause other issues within HB. Obviously they may not, but that’s for the developers to decide as they have to program it and support it. Indeed. But there's a way out of supporting inserted ffmpeg filter graphs. Simply include a note in the in...

Obviously the Handbrake team have integrated a wide range of tools; one need look no further than how subtitltes are handled. That's what makes Handbrake great. The Handbrake team can justly feel proud. Here's a use case that applies to over 90% of DVDs and to all Blu-rays. -filter_complex "tel...

Howdy, If you need "exact trimming", you REALLY need to be using an editor. You cannot guarantee that a key frame happens at a specific location to make a clean cut. Are you saying that the time stamps & frame numbers returned by ffprobe aren't correct? I successfully trim via MKVTool...

Two enhancement requests: 1, In order to do exact trimming, would you expand PTS selection (i.e. 'Range: Seconds') from hh:mm:ss to hh:mm:ss.sss? Alternatively, trimming to exact frame number, N=0,1,2..., would be very welcome and could be used to bypass potential PTS errors. 2, Would you provide a ...

Bad news but not unexpected. Simple operations like frame double & telecine would make transcoding a lot easier and give better results. What does HB do if out_rate exactly equals in_rate * 2 (e.g. 24000/1001 to 48000/1001)? What does HB do if out_rate exactly equals in_rate * 1.25 (e.g. 48000/1...

Description of problem or question: Here's what I need to do: 23.976_FPS ==> [_frame_double_] ==> [_23-telecine_] ==> 59.940_FPS I need expert advice how to do it. This: 23.976_FPS ==> 47.952_FPS ==> [_save_file_] 47.952_FPS ==> 59.940_FPS will not do for 3 reasons: 1, 23.976_FPS ==> 47.952_FPS (wh...

https://youtu.be/5eu_KjKsnpM You might find this video entertaining. For my terms interlaced systems vs progressive system. We will go with 60hrz cause it’s what we are taking about. Both “systems” (Maybe a poor choice of word, but it’s what I’m gong with. The concept is what matters) are a 60FPS m...

It’s universally rejected because it has no real world use outside your use. The math doesn’t add up and Combing is bad in a progressive system. What math doesn't add up? When telecining p24 to p30 or p60, combing is the result. How else can you add the additional frames? Would you add blank frames...

So far, 55-telecine has been universally rejected and I've been ridiculed for proposing it. Frankly, I'm amazed. I think it's rejection is caused by abstraction of video metrics to numbers instead of pictures and reasoning. Well, I have some numbers, too. 23-telecine produces 30FPS video having 12Hz...

No, I really meant field. Try to draw a picture of that. I think you'll see the problem with it. A combed/interlaced image has 2 fields making one frame. Each field a different moment in time. You are left with interlaced/combed frames problem in a progressive system. First, all frames are progress...

I understand the combed frames. But what to do with them in a progressive system. I can A. Leave as is and have combing artifacts few times a second. Yes, that's right. Every 1/12th second, there's a 1/60th second combed frame. It's so brief, it's almost invisible. B. Blur them. Yes, you can do tha...

You have the essential idea but I don't know why you write "not a true representation". How not? Because there is no mathematical way to represent 24 frames into 60 evenly without juter through 6:4 pulldown. I draw pictures (timing diagrams, actually) as I drew in the previous postings. L...

As M2TS files dont contain chapters why would you process one in HB and expect the correct chapters? I don't... Ah, I see the source of confusion, and it's my fault. I've got about 6 balls in the air and things have gotten a little confused. At some particular experimental set up in the past, I ext...

PS: I have done 55-telecine via ffmpeg. ffmpeg -i IN -filter_complex "telecine=pattern=5, split[A][B], [A]select='not(eq(mod((n+1)\,5)\,3))'[C], [B]select='eq(mod((n+1)\,5)\,3)',pp=linblenddeint[D], [C][D]interleave" -map 0 -codec:v libx264 -crf 18 -codec:a copy -codec:s copy -dn OUT The r...

From what i'm reading as I look what the OP wants to do is basically create a fake frame that's a combination of 2 frames for 1/60th every 2 frames/10. Basically create your own interpolated frame by using combing and blend interlacing. Ingenious, but is not a true representation of the original fi...

You could also use a hex editor and read the 1's and 0's from the HD platters as thats where the data is, its still not a good idea. The playlists exist for a reason. I dont know of any program that gets chapters from an m2ts, do you? M2TS files don't contain chapter indexes. HandBrake gets chapter...

While you seem determined to mess about with m2ts and vob files there is not much more I can help you with. I understand. Thank you for your help; very generous. I feel your frustration and share it. Yes, I mess about with M2TSs & VOBs. They are where the streams are. I don't know of any altern...

Ok, so you are still watching every m2ts/vob to see what they are? Yes. That will be your problem then, you need to find a way to open the root folder or use makemkv to rip your disc. I will try 'eac3to D: -demux' again. I will also try makemkv, though I've already ripped to ISO. I've included the ...

rollin_eng wrote: ↑Mon May 04, 2020 7:15 pm Try mkvextract to get chapters then.

mkvextract apparently only accepts MKV containers, not M2TSs and not VOBs.

In fact your workflow could be: Extract everything with mkvextract/eac3to. mkvextract will not extract chapters. eac3to extracts solely the main feature (longest?) unless I supply a title #. In a script, supplying title #s is problematic. Encode video with ffmpeg, are timestamp issues in container ...