Skip to main content
Recent Posts
2
Support - (fb2k) / Re: %CONTENT GROUP% mapping error
Last post by thomian -
@Temerald:
Thank you for bringing this topic into life. It makes handling of the grouping field with different files a lot easier!! Thank you very much! And thank you Peter for this gem of software.
3
Support - (fb2k) / Re: %CONTENT GROUP% mapping error
Last post by thomian -
This tag mapping is confusing now in Beta 17.
Yes, that's true.

The field was not mapped erroneously. I repeat: Vorbis comments has no standards. ONLY recommendations.

Yes, that's true. And it's even worse: Vorbis comments are not designed to be renamed or remapped at all. Vorbis comments are designed to show up in the application in the same way, as they are written to the file. If you want to have another name for this tag, just give it another name. This is different to other tag types, for example id3, where TIT1 has to be translated/mapped by the software to an understandable description. Vorbis comments have not to be mapped, they describe themselves.
The new mapping of foobar breaks the rules for Vorbis comments. It is a non-standard implementation. It's simply a mistake.
AFAIK there is no other Vorbis comment that is remapped to a different name in foobar.

Now there are four reasons, why this new mapping is a mistake:
1. foobars behaviour of tag writing and reading is confusing
2. CONTENT GROUP is a misleading term, in comparison to GROUPING
3. mapping of Vorbis Comments breaks the rules, as they should be displayed "as they are written"
   GROUPING should show up as <GROUPING> and not as <CONTENT GROUP>
4. In foobar no other Vorbis Comments are translated/remapped to a different field name

To solve these problems, I recommend to use the field %grouping% instead, and map all other tags to this field:
Vorbis: GROUPING
APE: Grouping
ID3tag: TIT1
iTunes MP4: ©grp
WMA: WM/ContentGroupDescription

Peter, would you please implement this in the next update.
Thanks
4
Scientific Discussion / Re: Help me understand why sound is one dimensional
Last post by saratoga -
Things like waveforms and time domain signals are one dimensional. Signal transformation through things like FFT is also one dimensional but returns a two-dimensional value.

I'm surprised how many people don't realize this, but the FFT returns the same number of dimensions as you put in, so a 1D function has a 1D FFT. It's a linear transform, so you get the same number of points and dimensions between domains. I suspect that the misconception you have hear (that frequency is somehow of higher dimensionality than time in spite of them having inverse units) is related to the general confusion most people have on this topic.
Well, a Fourier Transform (discrete or continuous) returns a complex-valued function, which real and imaginary components can be mapped onto a 2D plane. Complex numbers by definition extend the one-dimensional number line in \mathbb{R}, to the two-dimensional comlpex-plane \mathbb{C}. That's what I was referring to.

To try and make my point above clear, I'll reiterate that an FFT maps a (possibly complex valued) input to another (possibly complex) output of equal dimensionality. It is not correct that a (possibly complex) input of dimensionality 1 can map to an output of dimension anything but 1.

More importantly, being complex valued in this case is distinct from defining a plane as you are assuming. Recall that the FFT of a real valued function (such as a series of sampled points of a transducer in time) must be symmetric about zero. This means N unique values would seem to map to N/2 values - which would make the transform non invertible. Instead it actually maps to N/2 complex value pairs, or more simply just N values. Thus it's actually a mapping from N to N and therefore invertible.

Finally getting back to my original point, all of this math could be avoided by just defining dimensionality correctly.


People have such a limited ability to grasp what a 3D sound field is that most do not realize they even exist. The perception of sound is basically 1D with a bit of stereo, and this is the thing people are talking about. And they do mean to say 1D, they're just not sure what the words mean and are expressing themselves incorrectly.
Well, I'm not sure we're talking about perception of sound here. When Op started the thread, I thought we talk about sound as in the way signals are expressed.

Seems like a distinction without a difference given that they're expressed similarly 99.99999% of the time.
5
Opus / Re: Opus 1.3-rc
Last post by IgorC -
@jensend

I’ve tested a sample of German speech “FB vs SWB” at ~16.4kbps.  One sample just for now, as always no time.
The difference is audible. I could ABX and I prefer FB.  SWB was somewhat more muffled. I've repeated the test on two different open over-ear headphones. The results were same.

You can try it too by forcing --set-ctl-int 4004=1104 for SWB.
Also I had to increase bitrate for SWB just for 0.2 kbps because I guess this is exactly an amount of rate which is saved by coding SWB instead of FB. This way the bitrate was exactly same for both SWB and FB.

P.S. I couldn't find the way to try SWB at 12 kbps as I suspect it can perform better than WB (for speech).
6
Support - (fb2k) / Re: %CONTENT GROUP% mapping error
Last post by thomian -
CONTENT GROUP is much more logical, than GROUPING. ...
I respect only logics. For me logical thinking is the only "authoritative".

@infinci
I try to give you a logical explanation, for you to understand, how this grouping field is meant, and why it might be indeed more correct to name it GROUPING than CONTENT GROUP:

The term "content group" was invented by the id3tag consortium for the documentation of id3tags (http://id3.org/id3v2.4.0-frames).
It is the first of three tags holding "TITLE information" for the content of a track:
TIT1 = 'Content group description'
TIT2 = 'Title/Songname/Content description'
TIT3 = 'Subtitle/Description refinement'
Unfortunately the term "content group" is a bit misleading and the description in the documentation is not very well comprehensible, too... The TIT1 tag should be used, if the track belongs to a group of consecutive tracks, that form together a section of music. The TIT1 tag holds the title of this group of tracks. With other words: TIT1 groups tracks.

In classical music, a composer's "work" in most cases consists of more than one "movements". Normaly each movement is recorded to a separate track. For example, Chopin's piano concerto splits into three tracks:

Chopin: Piano Concerto No. 1, op. 11
  1. Allegro maestoso                   (=title/TIT2 of first track)
  2. Romance: Larghetto                 (=title/TIT2 of second track)
  3. Rondo: Vivace                      (=title/TIT2 of third track)

Then, the group of these three tracks is named "Piano Concerto No. 1, op. 11" (=content group/TIT1)

As this field is mostly used for the "works" of classical music, it would have been easier and much more comprehensive to call this field "work" instead of "content group" or "grouping". But the term "work" is constrained to classical music, and in some cases you might also want to group some tracks of non-classical music, for example the "first part" and "second part" of a rock concert. Therefore the id3tag consortium did not want to restrict this field to classical music, and they looked for a more universal term and chose "content group description".

As you can see, the grouping field really groups consecutive tracks to a group: Each grouped track has a different "track title", and all grouped tracks share a common "grouping title". (This is a bit similar to the ALBUM ARTIST field that groups different ARTISTS under one common ALBUM.)
This field has nothing to do with genre, style, music category, or content type... it really is meant to tie tracks together.

And logically now: CONTENT GROUP is much more misleading than GROUPING. The term "content group" even misled you, to assume that it describes some form of "content type", but it instead groups tracks together.

Vorbis comments should be named with a term as clear as possible. This is why most software developers mapped the id3tag TIT1 to the Vorbis comment GROUPING and not to CONTENT GROUP. And this is why the mapping proposal of wiki.hydrogenaudio.io recommends GROUPING for Vorbis comments and for APE files. It does not recommend "content group".

Also, iTunes developers chose "@grp" for the tagging of m4a files and not "@cnt" or "@cgr", because "grouping" is much more clear than "content group".

AFAIK there is no software in this world that writes a tag named CONTENT GROUP into files. In the linked tag-mapping-proposal of hydrogenaudio, only WMA files should be tagged with "WM/ContentGroupDescription". But the Windows MediaPlayer does not even support this field. It is likely that this tag name has not been designed by Microsoft. Presumably it has been designed by other developers with the need to map TIT1 to an apropriate field in WMA files. And the term WM/ContentGroupDescription seems to be simply copied from the id3tag documentation.

Ergo, to my opinion and (hopefully) fully logical: GROUPING is the better term than CONTENT GROUP.
7
Support - (fb2k) / Re: %CONTENT GROUP% mapping error
Last post by thomian -
If you will kindly follow the link I provided, hydrogenaud.io's Tag Mapping Page you would learn I'm not trying to change the standard; I am trying to bring foobar2000 up to standard.
These changes are not persional, or based on a singular opinion; I'm simply pointing out erroneous field mappings in foobar2000 based on  hydrogenaud.io's Tag Mapping Page.

@Temerald:
You are wrong assuming, that the list on wiki you linked is a "standard" of mapping. There is no standard of correct mapping of the tags of different file formats.
In fact, every music software does its own mapping. That is: WinMediaPlayer maps in a different way than iTunes does. iTunes maps differently compared to WinAmp. WinAmp maps in a different way than MediaMonkey. And MediaMonkey maps differently than your Android phone...
This is a big problem for interoperability of tagged music. It's a mess and it happened, because there is NO standard.
The widely used tagging software MP3tag was initially designed for iTunes users, and it mainly uses the iTunes mapping scheme. And your linked mapping scheme on wiki is based on this software. But again, this mapping scheme produces errors when used with other software.
The only way to achieve "correct" tagging is:
1. decide the software you want to use
2. decide the file format to use
3. read the documentation of your software's mapping scheme for your file format or check it out by yourself
4. put your information in the appropriate tags, so that it will be read correctly by the software you want to use
This means unfortunately: If you want to read your tagged music files with another software, you first need to "translate" your tags.

Mediamonkey found a solution for this. They do double and triple tagging for "ambiguos" tag names: for example, an album artist like "Pink Floyd" is written into a flac file using three Vorbis comments simultaneously: "ALBUMARTIST=Pink Floyd", "ALBUM ARTIST=Pink Floyd" and "BAND=Pink Floyd". As a result, the album artist will be read correctly by a broad variety of software. But that's off topic.
8
Scientific Discussion / Re: Help me understand why sound is one dimensional
Last post by polemon -
Things like waveforms and time domain signals are one dimensional. Signal transformation through things like FFT is also one dimensional but returns a two-dimensional value.

I'm surprised how many people don't realize this, but the FFT returns the same number of dimensions as you put in, so a 1D function has a 1D FFT. It's a linear transform, so you get the same number of points and dimensions between domains. I suspect that the misconception you have hear (that frequency is somehow of higher dimensionality than time in spite of them having inverse units) is related to the general confusion most people have on this topic.
Well, a Fourier Transform (discrete or continuous) returns a complex-valued function, which real and imaginary components can be mapped onto a 2D plane. Complex numbers by definition extend the one-dimensional number line in \mathbb{R}, to the two-dimensional comlpex-plane \mathbb{C}. That's what I was referring to. Then again, judging by how things were discussed here, I doubt most participants have grasped the concept of complex numbers...

People have such a limited ability to grasp what a 3D sound field is that most do not realize they even exist. The perception of sound is basically 1D with a bit of stereo, and this is the thing people are talking about. And they do mean to say 1D, they're just not sure what the words mean and are expressing themselves incorrectly.
Well, I'm not sure we're talking about perception of sound here. When Op started the thread, I thought we talk about sound as in the way signals are expressed. Hence I tried to clarify in that direction. Perhaps I should've given this a pass.
9
Support - (fb2k) / Re: foobar2000 v1.4b19 - visualizations on toolbar don't have native borders anymore
Last post by j7n -
The visualization boxes indeed do not have borders since the interface was "flattened" for Windows 10 in the current version. The boxes had a "sunken" style like an embedded LCD display. When the toolbars are locked, the visualization touches the left side of its parent toolbar. They were never correctly centered, but now it is more obvious. Some UI panels and splitters have a new option Border > Standard, which makes them look closer to the 1.3 version, but nothing like that for the toolbar boxes. I guess you have to roll with the times and accept the metro style.
10
Scientific Discussion / Have some wonderful demos (after decoding only -- sorry.)
Last post by jsdyson -
I uploaded some wonderful demos -- fairly hard for a real DolbyA to decode nearly as clearly.  I grabbed material which has lots of opportunity for intermodulation, possible stereo image smearing/shift, mirrored (quick echo) voices, and just vocal reproduction in general.  A lot of this older material seems to have significant vocal emphasis -- but I think that you might agree that the sound is pretty good.  In recent days the decoder has gotten more correct LF behavior (so true LF like Scarborough Fair is much better), and also the distortion sidebands as created by a changing gain slew are mostly cancelled now.

I cannot claim that the results are impressive -- but they are darned good.  The decoder really does well considering that some people say that it is impossible. 

The only frustrating thing is that the site that I use limits the mp3 to 200kbits/sec and any mp3 causes a significant loss of the subtle detail.

The full feature and quality decoder will be available sometime in the future.  Right now, the currently available 'free' version still does okay for casual use, but the very enhanced and completed version is very significantly better.  Hopefully, there will be something released as a plugin.

DEMOS:
https://spaces.hightail.com/space/z3H68lAgmJ

John

 
SimplePortal 1.0.0 RC1 © 2008-2018