INTERACT FORUM
More => Old Versions => JRiver Media Center 25 for Windows => Topic started by: Matt on March 15, 2019, 01:19:17 pm
-
[Edit by JimH -- This change was discussed and then reversed.]
Hi everyone,
We're proposing turning off conversion from lossy formats to lossless formats.
Here's a screenshot:
(http://files.jriver.com/temp/Conversion.png)
Does this seem reasonable? We can't think of any reason to allow it, and can think of a few reasons it could be bad (pollution of our online music service for example).
Thanks.
-
Personally, I agree with this. But I'm sure there's some poor, misguided soul out there that actually wants to do this (WHY?!?!) and they'll probably end up complaining at some point. Maybe having an option to allow it hidden somewhere in the options would be a good compromise... even though I believe doing such things are blasphemous! Honestly, I'd just do it and see if anyone eventually complains about it on the forums so we can help guide the misguided soul. ;)
-
Good god people do this?? Yea, it's a terrible idea so I'm on board with turning that off.
-
Actually, what would be better is if MC warned users on Import/Analysing if you have such dodgy audio files. These guys have written some papers and even released an app that will scan your files (a work in progress). http://losslessaudiochecker.com/#workinprogress
Detection algorithms
Upscaling
Upsampling
AAC-SIN Transcoding: AAC tracks, encoded using sine windows, transcoded to a lossless format
Work in progress
AAC-KBD Transcoding: AAC tracks, encoded using KBD windows (e.g., Nero AAC), transcoded to a lossless format
MP3 Transcoding: MP3 tracks transcoded to a lossless format
-
Theres always going to be someone who claims that converting their MP3s to 256kHz 24bit or 4x DSD will improve the sound..
-
You can't create something that's not there. Good lord!
-
I am not going to be lead into commenting on their state of mind. Merely mentioning that this change will result in threadfuls of posts from such people, and similar responses as yours. Personally I wouldn’t go there..
-
Then perhaps its an opportunity to educate such people...
-
Yep, I have to agree with Andrew here.
Maybe instead of straight out preventing such conversions, it'll a) pop up a warning for the user informing them it's not a good idea to convert lossy to lossless (and give a brief reason why) with a "are you sure" type of button in case they wish to proceed and b) maybe add some tag to the converted file noting it's a lossy to lossless conversion? Though the sticking point there is probably adding a tag like that, since such users will likely try to remove it.
But... I also like Nathan's idea too regarding detecting files that are lossy to lossless conversions during import/analysis. I can see how such a thing could really come in handy.
-
Theres always going to be someone who claims that converting their MP3s to 256kHz 24bit or 4x DSD will improve the sound..
That's very true. The question is whether we should be their unwitting accomplice. I personally would like to just say we don't do that.
You can blame it on Inflateablemouse's question about quality:
https://yabb.jriver.com/interact/index.php/topic,119839.msg828519.html#msg828519
-
Then perhaps its an opportunity to educate such people...
Well, people like that (as misguided as they may be) can be pretty stubborn and just want to do it anyways. It's really a no-win situation, even though I support the idea.
-
So the choice is letting them live with their ignorance and belief that they are "improving" the quality vs irritating them by not "letting them" "increase quality" by disabling this functionality? Delightful.
-
Well they're idiots then. :) You can't create something that's not there. Good lord!
Actually you can, its called distortion. And some people think that makes it sound better.
-
Actually, what would be better is if MC warned users on Import/Analysing if you have such dodgy audio files. These guys have written some papers and even released an app that will scan your files (a work in progress). http://losslessaudiochecker.com/#workinprogress
Detection algorithms
Upscaling
Upsampling
AAC-SIN Transcoding: AAC tracks, encoded using sine windows, transcoded to a lossless format
Work in progress
AAC-KBD Transcoding: AAC tracks, encoded using KBD windows (e.g., Nero AAC), transcoded to a lossless format
MP3 Transcoding: MP3 tracks transcoded to a lossless format
I would love to see this in MC too.
-
Actually you can, its called distortion. And some people think that makes it sound better.
Well, ok, yes you can corrupt the data. If people think that makes it "sound better" there's very little hope for the world. LOL
-
Well, there is at least one situation where some people feel it is quite reasonable to convert a lossy file to a lossless format. Some people believe that at least some DSD DACs sound better than comparable PCM only DACs. And those people sometimes buy DSD only DACs. Therefore, they will want to convert not only PCM files to DSD, but also mp3 files to DSD. The issue is not creating something out of nothing. The issue is that the very process of playing DSD files is different than playing PCM files and some people find that sound more pleasing. So, a PCM file converted to DSD may sound better to some people - not because of the extra bits but because of the playback mechanism. It is a personal preference that some people have. If those people cannot convert to DSD, then they simply will not be able to play some of their music.
Lets not get into yet another discussion of what effect different playback mechanisms and playback filters have. Some people like the sound of DSD better than PCM. For those people, not being able to convert to DSD may be a detriment.
-
As to distortion, many people like the sounds of tubes over solid state and many believe that the odd order harmonics (distortion) that tubes produce is part of the reason. Well, maybe some people are wired to like that type of distortion. Does that make them wrong? Actually, the reason there are so many different types of electronics and speakers is that different people have different preferences for sound. There is no one standard - unless you want to tell people that they are wrong in enjoying the particular sound they like.
Objectionist versus subjectivist - there is no single answer to this question.
-
Also what about those that want to apply a whole heaps of DSP effects and then want to make a transcode of that file? Would't using a lossless container help preserve the additional detail that the DSP has created?
-
Hmmmm. Converting PCM to DSD or DSD to PCM would technically count as a lossy to lossless conversion as well but unlike other lossy to lossless conversions which are likely few and far between, there's plenty people who actually do that.
-
Actually, what would be better is if MC warned users on Import/Analysing if you have such dodgy audio files. These guys have written some papers and even released an app that will scan your files (a work in progress). http://losslessaudiochecker.com/#workinprogress
Detection algorithms
Upscaling
Upsampling
AAC-SIN Transcoding: AAC tracks, encoded using sine windows, transcoded to a lossless format
Work in progress
AAC-KBD Transcoding: AAC tracks, encoded using KBD windows (e.g., Nero AAC), transcoded to a lossless format
MP3 Transcoding: MP3 tracks transcoded to a lossless format
FYI - this tool has found a few of my FLAC files that have been up-sampled and with errors.
With his background in creating APE, I'm sure Matt could would be interested in their paper on how to detect these mods in lossless files
-
Are JRiver and its users here to evangelise only "pure" music, and prevent people from "doing the wrong thing" with their own music tracks? Are we here to change the minds of those with a cognitive dissonance regarding such lossy to lossless conversions?
I think not, even if the argument is appealing.
The bottom line is, if people want to do it, they will do it, just using a different tool.
I could also argue that there are indeed situations where one might want to do say an MP3 to FLAC conversion, even knowing the effect that will have. How about setting up for a funeral, there is a list of tracks to play, and the sound system or funeral company requires FLAC files. Are you going to stop and argue with them about MP3's being fine, while making arrangements for your parent's funeral, or just convert a couple of MP3s to FLAC to make them happy?
Same with a school play. There is a Playlist. Some tracks are only in AAC, but the teacher/director/whatever wants all tracks in FLAC. Okay, they should buy FLAC versions. Maybe they aren't available. Maybe they are expensive. It's a school play, the sound system isn't that good. Is anyone really going to notice the distortion introduced in the conversion, or are they going to be watching their six year old dancing around the stage?
You aren't going to educate all the funeral directors, teachers, directors, etc. on how to best handle music. In fact, you aren't going to want to once you start trying. Just look at all the discussion around MQA to see the issue.
I would suggest just a warning. Some people don't actually know it is a bad thing to do, so tell them, but still allow them to do it if they have their reasons.
For uploads to JRiver CloudPlay and so on, vet the files and reject them if they don't meet the standard. That is something that you can control and is a reasonable position to hold.
I would also really like to see a tool that can do what Nathan in suggesting, to identify tracks that have been poorly handled. That would give me the choice to re-rip them, or source better versions.
-
I agree
-
The argument that a device may require FLAC doesn't hold water, in my opinion. If a device supports FLAC, it supports MP3. Apple supports MP3, but not FLAC.
-
why do you care if people convert lossy to lossless? does it cost you something?
-
Okay, I shouldn't have said that the sound system required FLAC. I should have just said that the Funeral Director required the tracks in FLAC format. Are you going to argue, or just provide them in that format?
-
Okay, I shouldn't have said that the sound system required FLAC. I should have just said that the Funeral Director required the tracks in FLAC format. Are you going to argue, or just provide them in that format?
I'll be dying to find out.
-
why do you care if people convert lossy to lossless? does it cost you something?
Because we have a plan, and it requires quality files.
It's also a very bad idea to do, and there are many other ways to do it if you insist on it.
Can anyone think of a realistic need for it?
-
Not my field of interest. I obviously wouldn't do such kind of conversion, but ...
Couldn't those "lossy to lossless" files be marked in some way, through metadata, through some kind of micro-tag/mark ?
Couldn't it be done, anyway, from "lossy to less lossy", like 192kbps mp3 to 256aac ? If user insists... Maybe an advanced option which should be unlocked to allow a "no-sense" conversion after specific confirmation/warning.
-
Because we have a plan, and it requires quality files.
It's also a very bad idea to do, and there are many other ways to do it if you insist on it.
Can anyone think of a realistic need for it?
Jim - I gave a realistic example based on a thread from a real MC user. It is not a common use, but it is based on a real user who your proposal would force to go to another tool. Maybe you are OK with that, but that is, of course, the issue.
Is your plan so critical that you cannot find a way to allow your users to do what they want to do? You can make it hard to do, but to eliminate the option takes a legitimate option away from real users.
By the way, what is to prevent users from moving the files to another tool, doing the conversion and moving them back to MC?
Rather than preventing your customers from doing what they want to do, maybe you should warn them about doing it, but let them do it if they want to.
Audio quality also brings up the problem of badly compressed files with very little dynamic range - the whole loudness war issue. If you want to prevent bad audio in your new program, are you also going to reject badly compressed files also?
-
I did not read the whole thread. But I just used mc to convert a YouTube video to wav. So I could edit it for myself. I would not want that to go away.
-
I did not read the whole thread. But I just used mc to convert a YouTube video to wav. So I could edit it for myself. I would not want that to go away.
You could convert it to another format.
-
The only time I convert lossy to lossless is with DVD rips and the occasional blu-ray rip that has DVD quality audio.
Multichannel audio on DVDs, and some blu-rays, uses Dolby Digital (AC3) and DTS which are both lossy. I rip them to flac which supports their 6 channels and high bitrate resulting from the 6 channels. I use another software for this but MC currently does it too: https://wiki.jriver.com/index.php/Convert_Format (https://wiki.jriver.com/index.php/Convert_Format).
As for converting mp3 to lossless, I say disable it. If it results in the end of the world, consider adding the ability back in with warnings.
-
Cannot think of a valid reason to have this functionality, and if enabled flies in the face of pure audio sensibility which is a cornerstone of this product.
Absolutely; remove it.
-
Because we have a plan, and it requires quality files.
It's also a very bad idea to do, and there are many other ways to do it if you insist on it.
Can anyone think of a realistic need for it?
I agree with jmone and RoderickGI.
Quality of content is a different matter. Your service that requires "quality" files should enforce that standard rather than arbitrarily removing some function in this app.
The main use I can think of is the previously mentioned use for DSP processing to extract some part of a track to use it elsewhere. A niche use case but a valid one. JRiver is v convenient for this as it can follow an MPLS playlist across many underlying tracks. Having said that, the underlying tracks in that case are lossless anyway so not relevant to this case.
To my mind it seems like a needless removal of customer choice and one which doesn't seem to do what you want (protect quality of some other service). However I can't say I do this myself so don't really have a strong opinion either way.
I do think the implementation seems a bit annoying though in that you give someone an option and then tell them no after trying to use it, I think it would be better to provide the available conversion formats for the selected content instead so the user can only choose from the actually available options.
-
I would recommend blocking SACD to DSD conversions too, since most people are doing that to split their SACD to individual tracks without knowing about the DSD>PCM>DSD conversion that will happen.
It would be great if you could integrate sacd_extract (https://github.com/sacd-ripper/sacd-ripper/releases) for that, but I don't know what the license is for it.
Well, there is at least one situation where some people feel it is quite reasonable to convert a lossy file to a lossless format. Some people believe that at least some DSD DACs sound better than comparable PCM only DACs. And those people sometimes buy DSD only DACs. Therefore, they will want to convert not only PCM files to DSD, but also mp3 files to DSD. The issue is not creating something out of nothing. The issue is that the very process of playing DSD files is different than playing PCM files and some people find that sound more pleasing. So, a PCM file converted to DSD may sound better to some people - not because of the extra bits but because of the playback mechanism. It is a personal preference that some people have. If those people cannot convert to DSD, then they simply will not be able to play some of their music.
Lets not get into yet another discussion of what effect different playback mechanisms and playback filters have. Some people like the sound of DSD better than PCM. For those people, not being able to convert to DSD may be a detriment.
DSD is a lossy format, not lossless - so I guess it would not apply here?
Are JRiver and its users here to evangelise only "pure" music, and prevent people from "doing the wrong thing" with their own music tracks? Are we here to change the minds of those with a cognitive dissonance regarding such lossy to lossless conversions?
I think not, even if the argument is appealing.
The bottom line is, if people want to do it, they will do it, just using a different tool.
Lossy > Lossless is essentially an invalid conversion and I see no reason to support it.
-
I think that the ability to convert lossy>lossless should be an advanced option that is off by default. That way, users that do so to apply DSP or whatever other esoteric reason would be unaffected.
If the option is ever enabled, maybe it should permanently disallow uploads to CloudPlay.
-
I assume this wouldn't affect real-time conversion like for DLNA where lossy to lossless is useful (e.g. for volume leveling)?
-
Why not just make a standalone tool? I can honestly picture a lot of strange situations where for whatever reason someone wants their files in a particular format and there is no time or good reason to argue with them. You might be able to point the jriver user user in the right direction but it would still be up to said user to change the third party's mind.
-
Rather than putting any effort at all into removing an existing feature, I’d rather see that effort put into working on new features and fixing existing bugs. I agree that it’s generally silly to convert lossy to lossless, but if someone really wants to do it, they’re going to find a way; removing the feature from your software isn’t going to prevent someone from uploading low quality content to your servers.
-
Personally, I think at most it should warn people that its not recommended, just in case someone truely just doesn't know, with an option to dismiss the warning forever, but not block it. Blocking it will just cause complaints, and there really is no benefit to doing so. As others have said, they'll find a way if they really want to. We shouldn't try to "nanny" users like that. We can offer tools, and even recommendations if we want to, but trying to enforce rules on top of that is a very slippery slope - in my opinion.
If the entire purpose is to ensure quality uploads, then trying to do something like jmone suggested, and actually trying to analyze audio for its quality sounds like a great option - but I don't know how reliable something like that can be. Nothing is preventing someone from converting in another tool and uploading, or having low-quality recordings in the first place that someone once claimed to be "lossless". As an alternative, I also like the idea of writing a tag to identify the original format, if you just want to block simple cases of conversion in MC and then immediately uploading. Of course if someone is very eager, they might find and remove the tag, but in the same way they might also convert in another tool.
There is also some actually valid conversion scenarios, not to increase quality, but to preserve it. If I have audio in an uncommon or new lossy format, that some devices may not be able to play, and I want to convert it to a format they can play, what do I pick? Converting lossy to lossy will always further degrade quality. Always. So I would convert to FLAC. I'm under no illusions that it'll improve the quality, but it'll exactly preserve the quality I already had.
-
I like the idea of "Original Format" tag.
-
i actually have converted a small amount of lossy to FLAC in the past, for reasons exactly like Hendrik stated (archival purposes)
JRiver is all about choice and i'd hate to see it start going the other way
-
I like the tagging as well. I also thing a simple warning is reasnable. It's not something you should do unless you mean to.
-
Personally, I think at most it should warn people that its not recommended, just in case someone truely just doesn't know, with an option to dismiss the warning forever, but not block it. Blocking it will just cause complaints, and there really is no benefit to doing so.
I 100% agree with Hendrik. I also like the tagging as well.
-
There is also some actually valid conversion scenarios, not to increase quality, but to preserve it. If I have audio in an uncommon or new lossy format, that some devices may not be able to play, and I want to convert it to a format they can play, what do I pick?
MP3 high bitrate or VBR. You'll never hear the difference.
I think this issue is similar to the use of seat belts. I remember when cars had no seat belts. They were added, but they had no shoulder harnesses. Then a warning chime was added. So now, you wear a seat belt, even though you can easily buckle the belt behind you and never use it.
To the point of JRiver acting as a "nanny", I think part of our job is to promote best practices. Converting lossy to lossless is clearly not a best practice.
The issue came up because of Inflateablemouse's question on quality for the new Cloudplay, but when I thought about it, I decided that we should probably never allow lossy to lossless conversion. I've read all the posts above, and I still see no convincing argument in favor of allowing it. The best argument is that a few people won't like it. That's not enough. There are already a few things we do that someone doesn't like.
The idea of tagging files that have been converted is a good one, but it's easily defeated by a determined user.
Checking files to see if they have been converted might help, but it's a lot of work to do and may not be effective.
Again, if you have a good reason to allow lossy to lossless, please state it.
-
MP3 high bitrate or VBR. You'll never hear the difference.
With all due respect Jim, when you came out with your little MP3 versus flac tester, people did hear the difference. You cannot, but that does not mean others don't.
Again, if you have a good reason to allow lossy to lossless, please state it.
Actually, several people have.
-
MP3 high bitrate or VBR. You'll never hear the difference.
Thats not something I would accept, and just use another tool. File size is of no real concern for me with audio, since its all relatively small in comparison to my video archives, so any bit of quality I can preserve, I will preserve.
People going to another tool is the last thing you want, because if they do that often enough, they won't come back.
To the point of JRiver acting as a "nanny", I think part of our job is to promote best practices. Converting lossy to lossless is clearly not a best practice.
Promote and educate. Not enforce. If the reasons are good, and we can enlighten people, then they won't want to use it. If we force them, they'll just get bitter and any lesson is lost on them.
The idea of tagging files that have been converted is a good one, but it's easily defeated by a determined user.
Using another tool to convert to FLAC is also not hard, the web is full of them.
--
Ultimately, what I just don't get is what the advantage of such a change is. What does it do for the users of Media Center? Because I don't see anything positive there.
If all that is about is trying to prevent people from converting to lossless and uploading it, there must be other ways that don't leave a bad taste after you've been told by an app you bought that you cannot do something.
-
The key to what you just said is preserve. Not "manufacture." :)
-
Promoting best practices doesn’t have to be heavy handed. Surely there are more productive things to spend valuable development resources on than ripping out existing functionality where the justification for doing so is fairly week.
It sounds like the number one reason to do this is for your playlist upload feature. Making the proposed change isn’t going to even come close to ensuring that only high quality files get uploaded as people have a multitude of other ways to create poor quality files.
The overwhelming feedback here is don’t do it, but it sounds like you guys were expecting the user community to give you a rubber stamp. That’s something to think about as if you start to ignore your users, they’ll be inclined to look into other options over time...
-
If you do implement this, then I suggest it should be done properly by only allowing lossy to lossy filetypes to be selected in the conversion screens - there is nothing more annoying than going through a process to set up an action and then have a screen pop up saying "No you can't do that!".
All in all I would have thought that there are lots of positive things to do in MC25 rather than becoming Nanny MC. Just my opinion.
How would you stop someone burning a bunch of mp3 to an RW audio cd and then use mc to rip the disc to flac format?
-
There is one caveat though - users may reasonably want to transcode any file into a plain wav format, which then would be indistinguishable from a lossless file.
-
Thats not something I would accept, and just use another tool. File size is of no real concern for me with audio, since its all relatively small in comparison to my video archives, so any bit of quality I can preserve, I will preserve.
In the process, you're creating a FLAC file that isn't lossless. Fine if it never gets out of your sight, I guess.
-
... when you came out with your little MP3 versus flac tester, people did hear the difference.
My memory of the results was that lossless and high bit rate MP3 weren't consistently distinguishable. It's still there if you want to prove me wrong.
Listening Test (https://yabb.jriver.com/interact/index.php/topic,107924.0.html)
-
Audio quality perception it's usually a very controversial subject. Too many factors enter in play. I don't want to derail this thread but...
I usually take in account this objective info: "The older you are, the less sensitive you are to higher frequencies". You can do a fast test. I did, and I'm unable to hear info beyond 15khz but my son and dougther, no problem in hearing up to 20khz (an easy way to test it is through some youtube videos about it or using sam sweep generator).
And there are a lot more... calibration, volume, noise... (yes, noise... it's also part of the music). In this test (http://archimago.blogspot.com/2013/02/high-bitrate-mp3-internet-blind-test_3422.html) the guys with higher quality gear did like more the mp3 than the flac. Two important things, the first one, they hear the difference, the second one, the "additional info" in the lossless track was perceived for most of them as "non-desirable sound", hence they "liked" more the mp3...
I think all opinions and user preferences should be respected, no matter if we think they could be wrong, or even if we could prove them that they are "scientifically" wrong, that wouldn't really make anyone change his mind if he/she firmly believes that he/she is right.
If someone is happy with his 192mp3 files ... why try to ruin the experience... ? If another one believes that a 100.000$ amplifier sounds waaaaaay better tthan a 10.000$ amplifier why argue ? Maybe under certain circumstances, equalization, room, volume, sources,... etc... we could share those opinions.
Cheers
-
There is zero gain to completely disable conversion from lossy to lossless. As others have already explained there are few use cases when it would be useful or justified. Warning about it is enough.
-
My memory of the results was that lossless and high bit rate MP3 weren't consistently distinguishable. It's still there if you want to prove me wrong.
There were people in that test who could distinguish between the various options, including at least one person who identified all the flacs. This is not about me, it is about the people who can distinguish between the mp3 and flac files. Your test brought forth people who could. Just because most people cannot tell the difference does not mean that nobody can.
-
I agree with disabling and being able to enable it in the options if desired.
It would be dumb to enable it but this would at least provide a barrier to a dumb action.
-
I convert lossy to lossless routinely. Why? To edit downloaded material. Rarely do I want to play something exactly the way I get it. I nearly always have to adjust the level, or remove silence, or filter hum, or... you get the idea. Frankly, I do't use MC for this anyway, but there is certainly a valid reason for doing so.
-
Again, if you have a good reason to allow lossy to lossless, please state it.
Yes. I buy a Music DVD. It is encoded in DD5.1. It is Lossy. I use MC's Convert function to create a FLAC 5.1 Audio Track version of each song for Audio Only Playback.
I'm with Hendrik on this. MC is not the FLAC police. I again suggest that if the aim is to prevent non lossless tracks being uploaded to the MC Cloud then by all means introduce some tech to identify and reject such tracks. Otherwise you will have non lossless tracks that users have "sourced" form the net being accepted just because they have a FLAC file type. Scan your library of music files with the util I linked and you may be surprised how many have not only be up-sampled but the number being reported with errors as well. If you want to build a quality lossless library in the cloud from user submitted tracks, you will need a process to verify these tracks are "good".
If you are really worried MC Users are going to use MC to covert MP3 to FLAC so they can upload it to the MC Cloud, then write a tag to the FLAC file and make that tag in MC non-editable. Have MC Cloud upload reject such files. Sure they could edit the tag out in a 3rd party app, but then again they could do the conversion in a 3rd party app and you would be none the wiser. Regardless you have the same issue. You will end up with poor files in the MC Cloud unless you verify the files regardless of the source..... or you live with the fact than some of the stuff upload is not correct.
-
After thinking a bit more about this, the main issue I see is the perception/trust we blindlessly put on everything tagged as "lossless". It works like high-quality-tag, no matter how has been cooked that file before reaching that format. We just pre judge those as better than anything else. Just take a look at some vynil 96khz/24bits rips...
Maybe the best way is letting users to qualify and share those files by comparing them hearing them and giving some kind of subjective rating. And I truly believe that some kind of "auto analyzer" could really help, at least to tag files which doesn't reach some minimum numbers (dynamic, frequencies, channel separation… ) when compared with more "trusted" version of the same tracks.
Even saving a hash of files (excluding tags/headers to reduce bad practices) could be stored in the database to "detect" some "too personally cooked files".
A slippery slope, but don't block features. Not a single advanced JRiver user wants to be said what to do with his files or his hobby. Most of us like to think we have golden ears or "special sensitivity". No matter if it's true or it isn't.
In any case be totally ready for users to upload sub-optimal files on purpose. No matter what you do. Some determined users will always find a way to do it.
-
I think people have made a good case for reasonable lossy > lossless conversions, so I don't agree with Media Center blocking the conversion any more.
I'd still include a warning which makes it clear that lossy > lossless conversions will not improve audio quality.
And Media Center should block SACD to DSD conversion.
Yes. I buy a Music DVD. It is encoded in DD5.1. It is Lossy. I use MC's Convert function to create a FLAC 5.1 Audio Track version of each song for Audio Only Playback.
If you rip to MKA files, the music will be stored in the original format and be approximately 1/4 the size of converting Dolby Digital (or DTS) to FLAC.
It's less portable outside of Media Center or other computer-based playback though.
-
There is no MKA Option in the MC "Convert" function (see pic).
I don't care if FLAC is "bigger" than MKA even if the option existed.
I agree that FLAC is more universal and would not want to use MKA as a result.
In this case, FLAC is just storing the decoded DD5.1 Audio Stream, where is the crime exactly?
...but most of all, I don't want my SW to tell me that I can or can't convert to any format I choose (regardless of my insanity to do so - :) ).
-
I guess you can't save people from themselves. I should know better.
We'll wait to release until April 1, and we'll just make everything FLAC.
-
;D Good to see mob rule wins over sanity!
Seriously, I do like Yaobings suggestion as a 1/2 way if coding a file checking is too time consuming. Make a Tag called "Original Format" and use that to store what the original format was during the conversion process. Like with the "Compression" field for video just make it not editable then test against that when you upload to the cloud for the minimum std you want. The side benefit is you could see what tracks saved in flac that were made from various sources (DVD, BD, CD, etc even 128kbs MP3 if you were nuts)... that would be great.
Else I like you April 1 idea, but I'm surprised it will not be APE!
-
There is no MKA Option in the MC "Convert" function (see pic).
I don't care if FLAC is "bigger" than MKA even if the option existed.
I agree that FLAC is more universal and would not want to use MKA as a result.
In this case, FLAC is just storing the decoded DD5.1 Audio Stream, where is the crime exactly?
...but most of all, I don't want my SW to tell me that I can or can't convert to any format I choose (regardless of my insanity to do so - :) ).
It's not a conversion that Media Center offers. I rip to MKV using MakeMKV, and convert to MKA using MKVToolNix.
Lossy formats like Dolby Digital are typically 32-bit floating-point. Lossless formats like FLAC are typically 16-bit or 24-bit integer.
So the AC3 bitstream is decoded to PCM and encoded to FLAC.
There was a time when most decoders would output 16-bit rather than 24-bit and the conversion may not have been dithered properly, for example.
So it inflated the file size and potentially reduced the audio quality.
It also means that you can no longer play it in 5.1 via S/PDIF unless you have a Dolby Digital encoder.
Offline conversion from AC3 to PCM/FLAC, and real-time encoding from FLAC to AC3 is going to degrade audio quality. Of course that particular scenario doesn't matter if you are using HDMI.
Rip to MKA, optionally splitting to individual tracks, and you still have the original AC3 bitstream in a much smaller file.
-
Seriously, I do like Yaobings suggestion as a 1/2 way if coding a file checking is too time consuming. Make a Tag called "Original Format" and use that to store what the original format was during the conversion process. Like with the "Compression" field for video just make it not editable then test against that when you upload to the cloud for the minimum std you want. The side benefit is you could see what tracks saved in flac that were made from various sources (DVD, BD, CD, etc even 128kbs MP3 if you were nuts)... that would be great.
That was not my suggestion. I just aired my support for it.
My take on this is that if people want to do it, they will do it with another application and import the files back into MC. We lose control that way. I would rather let them do it inside MC and we tag it. We still maintain some control.
-
We'll wait to release until April 1, and we'll just make everything FLAC.
April 1 ;D ;D ;D
If I understand correct the impact for audio files is:
- mp3 input -> DSD Output will then forbidden (I have only 10 mp3 files out of ~14000 audio files in DSD, DFD, AIFF, FLAC, WAV. So it will it only have a very small/(no) impact on me?
- if you have a DSD DAC like the T+A DAC 8 DSD you can not convert all input files to DSD anymore on the fly?
For video files I have a lot of DVD ripped to mkv only with 5.1 lossy sound formats (no stereo track available). I let MC mix it down to stereo. What will be the impact on it?
-
We settled on just showing a warning if you're doing lossy to lossless conversion. You can still continue.
Thanks for all the feedback everyone.
-
We settled on just showing a warning if you're doing lossy to lossless conversion. You can still continue.
Thanks for all the feedback everyone.
Please actually *block* SACD to DSD conversions.
The only time anyone would ever do that is if they mistakenly think that Media Center is extracting DSD tracks from the ISO, rather than the DSD>PCM>DSD conversion it actually does.
-
Actually, switching MC to 'extract' DFF/DSF files from a SACD ISO (extracting the raw SACD ISO track data and putting it into a DFF/DSF container method sacd_extract uses) would probably be better than just disallowing SACD to DSD conversions. In addition, allowing DFF > DSF or DSF > DFF 'conversions' by repackaging and not re-encoding (so there's no actual DSD > PCM > DSD conversion) would also be nice.
MC could do the following...
1) If a SACD ISO is imported into a library, give the user a specialized tool to extract the SACD/DSD data into a individual track (by extracting the track's raw data, then packaging it into a DFF or DSF container), giving the choice of either DFF or DSF.
2) If the convert format option is being used to convert a SACD ISO to individual DFF/DSF tracks, either simply point it to the specialized tool mentioned above or do the conversion the same as above by extracting the raw track data and packaging it without doing any kind of re-encoding (and thus avoiding the DSD > PCM > DSD conversion).
3) If MC is being used to convert DFF to DSF or DSF to DFF, instead of converting (and going through the DSD > PCM > DSD conversion), simply repackage the container from DFF to DSF or DSF to DFF without re-encoding the actual data of the original file.
That way, the DSD > PCM > DSD conversion can largely be avoided. There are apps that can do both, so it's possible (sacd_extract (https://github.com/sacd-ripper/sacd-ripper) 'extracts' and dff2dsf (https://2manyrobots.com/plaster/) 'converts DFF to DSF by repackaging but it's Mac-only).
At the very least, having MC repackage DFF or DSF files without re-encoding would likely be MUCH appreciated. Probably would be interesting to experiment with seeing if you can take a DFF file and repackage it to a DSF file (then vice versa) without doing any re-encoding to see how easy it is.
If you want I can start a new topic regarding both requests. EDIT: New topic is here (https://yabb.jriver.com/interact/index.php/topic,119883.0.html).
-
PS I am assuming you must still allow conversion from lossy to lossless when pushing “Always Convert” to a DLNA renderer (at least to L16 and WAV). Because otherwise you risk breaking existing stuff on some people’s renderers, for some lossy media types.
.. and BTW, if you do continue to support that type of conversion, then there’s nothing to stop people raiding their conversion cache for “illicit” up-conversions.. ;)
-
1.) There are tools that accept wave or flac as input only, so I need to convert aac to one of these.
2.) Mediacenter used to be a swiss army knife for music management. It won't be after such change.
3.) I should not be up to jriver to decide what is proper and what is not.
-
We settled on just showing a warning if you're doing lossy to lossless conversion. You can still continue.
Thanks for all the feedback everyone.