Am I processing my rips correctly?

[Wighty]

Hi, I have been using MakeMKV for a while now and have even purchased a key (it’s well worth it), and I thought I post my process to garner some additional advice from the experts out there.

My Process
1) First I rip the DVD/Blu-Ray (no 4K here) using MakeMKV (v1.18.2), with the following settings (changed from default):

• Video: Options: Minimum title length: <10 seconds> (Ensures I get all the extras!)
• General: Expert Mode: <ticked>
• Language: Preferred language: <eng: English> (Only need English tracks)
• Advanced: Default selection rule: <-sel:all,+sel:(favlang|nolang),-sel:mvcvideo,=100:all,-10:favlang,+sel:(attachment),+sel:(mvcvideo)> (Have added in options to auto grab both videos tracks for 3D and the attachment image (for Blu-Rays).)

This works perfectly and gives me a selection of .mkv file to process.

2) Once I’ve identified the files I wish to keep, and rename them accordingly, I then process them via HandBrake (cropping as necessary), using the Matroska H.265 MKV 1080p30 preset, to compress the Video track only (I remove the audio/subtitles).

As my preference is to have a compressed (but near-lossless quality file), it means that under Video I use the following options:

• Quality: Constant Quality: <22 RF>
• Encoder Preset: <<Slow>
• Encoder Profile: <<Main 10>
• Encoder Level: <Auto>

3) Once HandBrake has finished, I then use MKVToolNix to merge the original .mkv with the HandBrake output (basically swapping to the compressed video track, but keeping the original audio, subtitles and chapter tracks).

4) I then copy the file to my Media Server.

My Concern
However, I’m thinking I can most probably do better, especially with the audio tracks, and maybe the video tracks (perhaps some different Handbrake settings). Therefore, I have some questions:

My Questions
Video Tracks
Are my HandBrake settings appropriate to get the best out of the video track compression? Should I use a lower RF value or a slower speed?

Audio Tracks
I’ve been keeping the original audio tracks, as I wanted to keep the best audio quality, and assumed (wrongly) that compression would reduce quality. However, I’m wondering if I should now be processing them to make the files smaller?

Because I've been using the originally ripped audio tracks, all of my .mkv files utilise a selection of different audio tracks Codec’s:

• A_MS/ACM
• AAC
• AC-3 Dolby Surround EX
• AC-3
• DTS-ES
• DTS-HD Master Audio
• DTS
• MP2
• PCM
• TrueHD Atmos
• TrueHD

Whereas, if I was converting using FLAC, then that would simplify the codecs used.

I see that MakeMKV can compress the audio tracks, as I just come across the posting about on-the-fly audio conversion, should I be doing this? Selecting the FLAC profile, which I understand is lossless compression (so keeps quality)?

However, I noticed that only the first audio track is compressed, can MakeMKV be configured to compress all the audio tracks (so the additional audio commentary tracks are done as well), or would I have to use HandBrake to do this?

Finally, what are the AAC-stero and WDTV profiles for? The forum post (linked above) also mentions LPCM, but there is no profile for that, has the feature been removed, and is LPCM worth using?

Your feedback and advice is welcomed.

Thanks.
Jonathan

[DukeFleed]

2) Once I’ve identified the files I wish to keep, and rename them accordingly, I then process them via HandBrake (cropping as necessary), using the Matroska H.265 MKV 1080p30 preset, to compress the Video track only (I remove the audio/subtitles).

As my preference is to have a compressed (but near-lossless quality file), it means that under Video I use the following options:

• Quality: Constant Quality: <22 RF>
• Encoder Preset: <<Slow>
• Encoder Profile: <<Main 10>
• Encoder Level: <Auto>

3) Once HandBrake has finished, I then use MKVToolNix to merge the original .mkv with the HandBrake output (basically swapping to the compressed video track, but keeping the original audio, subtitles and chapter tracks).

Hi. Sorry, I'm not sure I understand. Do you extract the entire disc, or just the movie, or just the episodes of a TV series with MakeMKV, then use Handbrake to compress just the video, removing the audio track and subtitles. Then, using Mkvtoolnix, you combine the Handbrake-compressed video file with the audio track you removed earlier. Do you do this to prevent the audio from being compressed when using Handbrake?

Why do you do these steps? Don't you know that Handbrake has a passthru option to prevent audio compression?
https://handbrake.fr/docs/en/latest/tec ... ed%20file.

[Wighty]

Hi. Sorry, I'm not sure I understand. Do you extract the entire disc, or just the movie, or just the episodes of a TV series with MakeMKV, then use Handbrake to compress just the video, removing the audio track and subtitles. Then, using Mkvtoolnix, you combine the Handbrake-compressed video file with the audio track you removed earlier. Do you do this to prevent the audio from being compressed when using Handbrake?

I extract the whole disk, as I keep all extras and episodes etc. Yes I was doing it this way to prevent audio compression.

Why do you do these steps? Don't you know that Handbrake has a passthru option to prevent audio compression?
https://handbrake.fr/docs/en/latest/tec ... ed%20file.

I was unaware of this.

This is why I posted, I have a "process", which I realised is most probably flawed, and thought it best to post and ask here.

I am also doing some more reading on audio compression to understand my options better.

J.

[Wighty]

Continuing on with my research, I have determined that some of the codecs in use are lossy (AC3, AAC, ACM, DTS, MP2), some are lossless (DTS-HD, True HD), and one is not compressed (PCM).

From what I understand, and I could be wrong, converting a lossy audio track into FLAC (lossless) will not gain any quality benefit, and in fact will always result in a large file size (so pointless to convert), and if you convert a lossy audio track to another lossy format there will be some quality loss (so again, pointless to convert).

I did some testing, and converting AC3 to FLAC does indeed increase the file size, and whilst converting AC3 to AAC did produce a smaller file and I didn't notice any sound differences (but I'm no audiophile!), I think I will stick with only converting the lossless tracks to FLAC.

Hopefully, I'm on the right path now.

J.

[dcoke22]

Have you considered keeping your original rips? In the years to come, h.265 compressed video might not be the best way to have your files. If you have your original rips, all you'd have to do is re-encode them. Without the original rips, you'll have to re-rip them first, assuming your optical drives and discs still work at all. Yes, keeping all that stuff will take up a lot of space, but big hard drives are pretty cheap on a $/TB basis and getting cheaper all the time.

Regarding FLAC… I don't think FLAC keeps any positional audio data like Dolby Atmos. (I might be wrong; I'm not an expert.) If you have an audio system in your home theater that can utilize that, you might not want to convert everything to FLAC.
Not everything can play back FLAC, although it seems relatively well supported. If you need or want highly compatible files, a basic AC-3 5.1 audio track can probably be played by almost anything.

[Wighty]

Have you considered keeping your original rips?

No, never thought to do that.

In the years to come, h.265 compressed video might not be the best way to have your files.

Why would you say that? Surely, it's the "new" standard?

I don't think FLAC keeps any positional audio data like Dolby Atmos.

I only have a basic home setup, so don't benefit from Atmos anyway.

Not everything can play back FLAC, although it seems relatively well supported. If you need or want highly compatible files, a basic AC-3 5.1 audio track can probably be played by almost anything.

Thankfully, everything I use does support FLAC, however I can see the benefit of sticking with AC3.

The main thing I have learnt on this journey, is that there is no right or wrong way to do this... Everyone has their preference, and each option has pros and cons!

Thanks for the reply though.

J.

[dcoke22]

In the years to come, h.265 compressed video might not be the best way to have your files.

Why would you say that? Surely, it's the "new" standard?

h.265 is fairly well supported these days. I would say the 'new' standard is AV1, which is not well supported yet. However, Netflix, YouTube, Amazon Prime, & Twitch are using AV1 for some stuff on some platforms. And hardware support for decoding AV1 is proliferating in the market.

The point is, whatever the current best choice of codec is probably won't be the best choice in the future. That doesn't mean you won't be able to play your older encodes on future devices, you will. But without the original rips, you won't easily be able to re-encode with whatever the next best codec choice is.

In any case, I agree that there's no single best path through this stuff. Do whatever works best for you.

[SirYodaJedi]

I don't think FLAC keeps any positional audio data like Dolby Atmos.

I only have a basic home setup, so don't benefit from Atmos anyway.

I still keep the track in case I end up upgrading a few decades from now, or bring my rip to someone else's house with a fancier system.

Mono and stereo FLAC are supported very well (even my BD player can play MKVs with them), but 5.1 FLACs may be played back with the wrong channel layout (rears instead of sides, because that's how the FLAC spec is standardized), which either means it comes out of the wrong speakers on a 7.1 system, or sounds too quiet due to downmixing on a 5.1 or 2.0 system.

I will occasionally keep the original audio tracks for certain movies and tv series, but otherwise, since I can't tell the difference between lossless and high-bitrate lossy, I convert lossless tracks as so:
• Mono - Either FLAC (if small enough) or 160kbps AAC
• Stereo - 320kbps AAC
• Intentionally surround-compatible stereo: 448kbps AC3 with Dolby Surround flag enabled
• 2.1, 3.1 - Keep original track, and create stereo downmix in 320kbps AAC
• TrueHD 5.1: Keep lossy core, unless it's not 5.1
• non-TrueHD 5.1: 640kbps AC3 with downmix levels for center and surrounds set to -3dB
• 5.1 intended as matrixed 6.1: like 5.1, but add the Dolby Surround EX flag
• 6.1 DTS-HD: keep lossy core; create 5.1 EX downmix for compatibility fallback
• 6.1 and higher TrueHD: keep original and lossy core
• 7.1 and higher DTS-HD: keep original; convert extracted core to 448kbps AC3 for compatibility fallback

I'd love to keep the original rips, but I don't have the storage space. I convert to a 8-bit level 4.0 h.264 for compatibility (~2-12mbps, depending upon content and CRF), and keep a unconverted copy if I think that doesn't do it justice (mostly anime). I have yet to figure out good settings for similar quality high-bitrate h.265, but would like to in order to get hardware decoding for 10-bit (which greatly helps avoid introducing banding); all I've figured out is that I need to disable SAO to keep the encoder from blurring the heck out of things. I figure I just deal with whatever is "good enough that I won't complain if I lose the discs".

[dcoke22]

I have yet to figure out good settings for similar quality high-bitrate h.265, but would like to in order to get hardware decoding for 10-bit (which greatly helps avoid introducing banding); all I've figured out is that I need to disable SAO to keep the encoder from blurring the heck out of things. I figure I just deal with whatever is "good enough that I won't complain if I lose the discs".

I encode normal 1080p movie blu-rays to 10-bit h.265 for my Plex server. I don't really use the Handbrake GUI, I do it via the command line.

For video, I use the x265_10bit encoder and I set the bitrate to 4000. I do multi-pass encoding with a turbo first pass. The x265_10bit encoder will run on the CPU (mostly) so it'll take longer than doing hardware encoding depending on your computer. But I almost never have a problem with banding and I'm cursed with seemingly always noticing banding. However, every time I think I have a banding problem with my encode, I go back to my original rip and I find that the banding is also on the blu-ray itself.

For what its worth, as I approach the end of ripping and encoding all the movies on my shelf and I get a feel for how much space I'm going to use on my Plex server, I've been considering re-encoding all my blu-rays with a bitrate of 6000. The extra 50% for the video bit-rate can only improve the image quality at the cost of using more disk space but I think won't be an issue for me. Why watch movies at 'only' streaming quality if you don't have to?

YMMV of course.