=== Does clipping affect me? ===At a rough estimate 99% of modern pop CDs are clipped. Without analysing each one it can be tricky to tell how bad the problem is.
Why should I spend my time trying to verify or falsify your marketing claims?
These days are confusing!First BS found we need MQA to know how it was meant to sound and now i learn everything is clipped and must be repaired in a way only CS knows.BZZOWNT
Quote from: Porcus on 09 January, 2017, 12:46:36 PMWhy should I spend my time trying to verify or falsify your marketing claims?Wrong premise, marketing claims are the capability of a product.
The main problem is the loudness war itself, which leads to overcompression in mastering, which affects the way the material sounds, even when there is no clipping problem. There are dynamic range databases for quite some time now, to allow people to make an educated guess before buying. They have their own serious problems, as we all should know by now.
I don't know why the reaction was as hostile as it was, but given that we have had discussions about this numerous times here, people may just have acquired short temper.
Firstly, the CD has long ceased to be regarded as a quality medium, and the fact that all attempts at establishing an effective copy protection scheme had led content producers to try to establish alternative media that allow them to control copying. I would venture to assert that at least some of them have tried quite deliberately to harm the CD's quality peception to help moving people over to a different medium. They gave the people overcompressed shit because that's what they wanted to do. It wasn't an accident. Much to their frustration, the other media (i.e. SACD, DVD, ...) didn't catch on. They now pin their hopes on streaming.Secondly, they haven't got enough money to do several different masterings of the same thing for different applications, and end up going for the broadest market. Kind of a lowest common denominator approach.
My own pet theory is that it could be killed dead by introducing a floating point distribution format that doesn't have a clipping point of any practical relevance. Removing the wall everyone is banging their head against should remove the damage to the heads. But alas, I seem to have difficulties convincing people of this way. I failed miserably here in this forum.
This discussion is trying to accomplish too many things at once.Substantive debate about declippingBack-and-forth about the loudness wars and statistics of distributed musicDebate about the proper ways to deal with overly broad statements on the wiki and other disputes thereQuestions about policies for allowing software authors and other experts to contribute to the wiki Questions about Greynol's behavior patterns as a moderator and a wiki editorQuestions about cutestudio's edits and whether he should have wiki edit privilegesAt least some of these discussions need to be separated somehow and addressed more generally and dispassionately.CuteStudio, if you think greynol has overreacted and acted irrationally or based on emotion/hot temper, the most important thing for you to do is to avoid overreacting, act rationally, and keep temper and emotion in check. Try to be more concise and dispassionate in your complaints.Here's my take on the first few of those topics; I may comment on the others later.1. In a couple of other threads as well, Greynol and others have been unjustly dismissive of declipping. For highly variable signals such as speech, where clipped regions are likely to be short, isolated, and very severe, good audio restoration algorithms frequently improve the SNR by more than 12dB. It is not at all difficult to hear the improvement either (esp. the reduction in boomy full-spectrum distortion during clipped vowels). For less variable signals (most music), if the severity of clipping is high enough to make an obvious audible difference, it's likely that so many samples are clipped that a restoration algorithm has insufficient information for a really good reconstruction. Improvements are likely to be much smaller, say 6dB or less, but may still be worth pursuing.This is not some kind of weird audiophile junk. This is a well-posed set of mathematical problems which have seen good theoretical and engineering work. Sadly the only open-source stuff I'm aware of are Audacity's clipfix, which is a very naive (cubic interpolation is not very appropriate for audio) simple hack Ben Schwartz (later of Xiph fame) did back as an undergraduate, and Monty's postfish, which does a reasonable job but is not as simple to work with (linux only, no distro includes binaries, source only available via svn, odd build dependencies, doesn't integrate into other kinds of toolchains). Audacity's "repair" effect, which does least squares autoregression, could probably be turned into a halfway decent declipper if combined with something that detected the clipped regions. Implementing the algorithms from any of the various recently published academic papers on the subject might be more competitive with the closed-source solutions.2. If someone puts an overbroad statement about clipping in pop songs on the wiki, the right way to deal with that is a  and then remove the offending statement if evidence is not provided in a timely way. We don't need to sit here and argue about the loudness wars and how they should have been addressed twenty years ago.3. I don't think mass deletion and reversion and blocking users at the first suspicion is sensible for dealing with a small closed-membership wiki.
My challenge was for Chibisteven to prove his assertion that '99% of all modern CDs are clipped ... bullshit '
I suspect it's a taboo subject. HA's 'holy cow'.
... most have given up or stopped caring in the face of an intransigent music industry.
I had a read of that thread - thanks for posting it. At reply #9 you were basically fighting off a hostile audience picking irrelevant holes in the idea. Floating point would be a very good idea IMO, with the standard mandating a min/max extent number in the header - to remove the +/- 1.0 convention in current WAVs.
Since the CD many viable file formats have been created, including ODF which is rather more complex than an array of IEEE floats and a trivial header.
Perhaps Denial is at play here - people I guess don't want to know the truth about the music they are playing.
Because it's easy to make a loud CD without introducing any clipping. I've done it with a few recorded MIDIs using the free version of Stereo Tool and that's an application designed for being as loud as other radio stations and even louder if one wants to.
I can imagine the methods and stuff used by more experienced professionals who do this for a living can do a 100 times better job at it than me and have access to tons of expensive tools at their disposal.
I've bought CDs in the last few years of recent albums and found no signs of clipping but they were loud nonetheless.
It's also easy to transfer music to 16bit without detectable compression or clipping like they did up to the 1990s, but however much people bleat about 16bits being good enough on HA it doesn't cut it when you have dynamic source material like this:(Saturn The Bringer Of Old Age)There's parts on here that average -48dB which on a 16bit format is using 8 bits. Is 8 bits still HiFi? I suspect this is one subtle driver of the loudness war (A war we've lost BTW).
Quote from: CuteStudio on 10 January, 2017, 05:39:52 AMQuote from: pelmazo on 09 January, 2017, 02:09:45 PMThe main problem is the loudness war itself, which leads to overcompression in mastering, which affects the way the material sounds, even when there is no clipping problem. There are dynamic range databases for quite some time now, to allow people to make an educated guess before buying. They have their own serious problems, as we all should know by now.Agreed 100%.You agree, yet seem to miss my point entirely. I was trying to tell you that your tool and your general approach doesn't solve the problem. You can't agree with me and in the next sentence carry on advocating your tool. There's a large cognitive dissonance here.
Quote from: pelmazo on 09 January, 2017, 02:09:45 PMThe main problem is the loudness war itself, which leads to overcompression in mastering, which affects the way the material sounds, even when there is no clipping problem. There are dynamic range databases for quite some time now, to allow people to make an educated guess before buying. They have their own serious problems, as we all should know by now.Agreed 100%.
QuoteI suspect it's a taboo subject. HA's 'holy cow'.That's almost certainly the wrong suspicion. Particularly when you are linking it with gear price and cable sound. Elsewhere perhaps, but not here.
Quote... most have given up or stopped caring in the face of an intransigent music industry.Any solution to this problem must take the business realities of the music business into account, or it will fail.
QuoteI had a read of that thread - thanks for posting it. At reply #9 you were basically fighting off a hostile audience picking irrelevant holes in the idea. Floating point would be a very good idea IMO, with the standard mandating a min/max extent number in the header - to remove the +/- 1.0 convention in current WAVs.Thanks for taking the effort to read it. I'm not yet sure you have understood the point I was trying to make there. You are mentioning a standard, implying that something is to be mandated. Which standard are you referring to, what exactly should it mandate and why, who would be in a position to mandate someting like this, and how would one ensure it is being complied with?
QuotePerhaps Denial is at play here - people I guess don't want to know the truth about the music they are playing.That may be the case with many people, but the people involved in the discussion here are much more likely to have known the sad truth for a long time. They don't question the problem, they question your approach to solve it.
I ripped a track and manipulated it by several means:original: original file.-6db: simple -6dB volume decrease.cassette: record the original to a cassette and re-digitize it.phase -90: phase shift -90 degrees with normalization to avoid processing clip.phase +10: phase shift +10 degrees with normalization to avoid processing clip.The free version offers a tool to analyse clipping and dynamics with a rating. See the screenshots attached.In a short summary:-6db: Faircassette: Aceoriginal: Poorphase +10: Goodphase -90: Ace
So In CuteStudio's opinion, the best declippers should be a cassette deck or a phase shifter
Some analysis are disabled, maybe a limitation of the free version. If you are going to tell me the full version has a more accurate analyser then sorry, it means the free version is a scam to fool and scare users in order to encourage them to buy the full version.
Also I wonder how accurate your "Hall of fame/shame" in your homepage is since apparently the analysis is based on your own algorithm.
I agree you should be banned to edit because of your dishonesty.
How about this? Some other members put an entry of SeeDeclip but warn about the inaccurate nature of the quality analyser by linking to this thread? It is quite predictable that someone interested in studying about clipping is also interested to know how to declip, adding this part into the wiki can prevent people being fooled by scamwares.
No cognitive dissonance, I just don't think you've understood what SeeDeClip4 is about which is probably my fault. Could you please read the last part of the post #17, starting "Reply #17 – Today at 05:16:37 PM".Here - and on the main webpage (http://www.cutestudio.net/ at point 1 in the first numbered list) - it's quite clear that the main purpose of SeeDeClip4 over the previous V3 'declip only' software is to identify the quality of the the recordings so you can select and play the better mastered ones.If you still disagree you'll have to define the phrase 'solve the problem' first, it could mean:To avoid playing damaged audio tracksTo repair damaged audio tracksTo buy every music company and force them to master music properlyYour dissonance appears to originate from 2 and 3, whereas I am clearly pointing to 1), with a little 2).
There are several hostile replies on this very thread where people are asserting their faith that their waveforms are not clipped, and then refusing to even name the album/track - much less post up the waveform. This has all the hallmarks of denial and taboo.
What we see is falling CD sales - a 24bit/96k DVD format could have made downloading and MP3s laborious and buying silver discs worthwhile. SACD was the window for this but the music sabotaged it. This hasn't helped them one bit and now it's too late because download speeds and disk space have grown to make these readily pirateable too.They also engineered the dip in mastering quality that made easily copied MP3 as good as the CD and thus wrecked their revenue.
"the standard" was referring to your proposal: A standard based on 32bit floats so there's no magnitude limit.I assume you'd already thought about the other questions - I'm not sure why I'd know the answers, it wasn't my proposal.
You're the only one saying that, no one else has admitted to any clipping at all, their recent pop albums are all clip free. So clip free that no album name or track has been mentioned in case I analyse it for them and tell them how damaged it is .
I ripped a track and manipulated it by several means:original: original file.-6db: simple -6dB volume decrease.cassette: record the original to a cassette and re-digitize it.phase -90: phase shift -90 degrees with normalization to avoid processing clip.phase +10: phase shift +10 degrees with normalization to avoid processing clip.The free version offers a tool to analyse clipping and dynamics with a rating. See the screenshots attached.
I mean, who will on purpose compress in order to boost volume up to the maximum only then to reduce by 6 dB?
Also, I am not so sure that it should be considered a failure to overlook signals that do not reach the full digital volume. The algorithm should be able to find clipping even if you bump down a little bit, but the application could very well just scan near the 0 dB mark - I mean, who will on purpose compress in order to boost volume up to the maximum only then to reduce by 6 dB?
The phase shifted files are indeed from the same source, you can try it yourself, it is totally unsurprising.
Quote from: bennetng on 12 January, 2017, 06:48:10 AMThe phase shifted files are indeed from the same source, you can try it yourself, it is totally unsurprising.I do not dispute that they originate from the same file, but they do not look like the same time - the legend suggests they are thirty or seventy seconds apart, isn't it so?
The irony of all this is that the LP now has a second spring because of their complete failure to kill the CD off. The LP became the audiophile medium again, purely because it was the minority format that could be targeted to quality-conscious consumers, and because the mastering for the LP doesn't easily tolerate the crimes that can be committed on a CD. This is not a testament to the inherent qualities of the LP, rather it is the testament of what a complete cockup the music business has become as a result of those failed CD replacement strategies.