The last time it was released 16 years ago (when Windows 7 was in use), it was only provisionally compatible, but this time it has been significantly revised to be compatible with the latest Windows. There has been no particular improvement in functionality, but it has been modified in various ways to suit the latest PC environments. Also, some minor OS-dependent bugs have been fixed.
What's New
Since I've made changes to the fundamental parts, I've incremented the major version by 1, making it Revision 4.00.
The compiler has been changed to the latest Delphi 12.1 Community Edition, which is compatible with Windows 11.
As a result, the internal character code processing has been largely replaced with a Unicode-based system. As a result, even special characters such as emojis and Chinese characters are now displayed correctly. This was the most troublesome fix, so I left it for a while...I'm sure there are many free software programs that stop updating because of this.
ID3Tag related items have also been changed from Shift-JIS to Unicode. Previously, ID3V1 and ID3V2 tags were based on V1 and V2 was used as an extension, but now V2 is the base and V1 is downgraded to V2 data for reading and writing. This is troublesome. The default is now only V2 tags. Previously, tags were recorded in SHIFT-JIS, but now the base is V2.
It is now possible to enter Unicode tags (those that can only be in ASCII, such as genre and year, will remain the same as before). However, since there may be some players who have problems with this fixed setting, you can choose whether to keep it the same as before in the settings. The default is Unicode recording. Famous software such as foobar2000 also defaults to Unicode, so this should probably be fine for most users.
- When creating the track number, I changed it from track number/all track numbers to just the track number. Why did I do it like this...? (I've already forgotten)
- With the compiler update, the GUI, including common controls, has been changed to Windows 11-based. To make it look more reasonable, the font has been changed from MSP Gothic 9pt to Meiryo 10pt as the main font. The font has become a little larger, making it easier to see on a 4K display.
- File sizes and other file manager related data were processed in 32-bit format, but have now been changed to 64-bit format. This has eliminated digit overflow, and the size of large files and free disk space no longer appear as negative values. The display has also been updated to take into account terabyte-sized HDDs.
- Historically remaining hidden functions have been removed at the code level (such as the ability to communicate with Fake Haruna).
- The settings menu was also quite cluttered due to historical reasons, so I cleaned up things that were unlikely to be used. Basically, it has been modified so that it can be completed with just the output file tab.
- Since Windows 11 has become 64-bit OS only, the 16-bit code call function of MSCDEX, which can no longer be used, has been removed. Also, the compression function of TwinVQ, which is no longer used and is troublesome to test, has been removed (the setting menu is somewhat cleaner).
- I also considered removing the WinASPI32 support because Windows no longer properly supports SCSI, but I have left it for now. However, the process that tries to rewrite the registry internally when ASPI is not available seems likely to be caught by a virus check for zero-day vulnerabilities, so I removed the entire code. Therefore, the registry rewriting process is completely eliminated internally (the registry is touched for internationalization support of Delphi system libraries, so that's about it, if there is any at all).
- I don't have an environment where I can check the operation of ASPI at home, so I haven't checked the operation of related functions (such as burst read).
- While checking WMA, I found out that MP3 can now be created with the Media Foundation API, so I switched to the option to encode with this instead of ACM. I've changed the compression default to this. Also, to be honest, ACM encoding is confusing and probably not very useful, so I deleted it too. Who remembers RMP files anymore, right? So, the settings screen is now even cleaner.
- FLAC and AAC encoding can be done with the same API, but I don't support it because it's a hassle to implement tagging and other things. + I don't use it myself, so I'm not very motivated. Conversion is easy because you just need to change the parameters, but it's a hassle to build an integrated environment that combines tags.
- No one may be using WMA anymore, but I have changed the generation method to one that is supported by Windows 11, so it can now be generated normally, just like MP3, without having to add any DLLs.
- However, when using the built-in MP3 encoder, unknown data is added to the comments, but I don't know how to avoid it at the moment, so I'm leaving it alone. However, during the automatic CDDB-Rip-Encode-tagging sequence, these unknown comments are automatically overwritten and deleted, so I don't think it's usually a problem (I thought that something from Apple was wrong because there was an iTUN header, but it's the same on PCs that don't have anything like that installed, so I'm not sure).
- By the way, as usual, there is no API definition for the Media Foundation API in Delphi, so I created it myself, but it is huge and I have no energy to recreate it all, so I converted only the bare minimum of the API I will be using this time into Pascal.
- Since it is the bare minimum, it was helpful that it only took about 400 lines of code, but...
- However, when compressing MP3 or WMA files with the Media Foundation API, it seems like a 5-minute WAV file can be converted in an instant, but was it really that fast? Perhaps this is due to the power of the Ryzen 5 9600X.
- Deleted the help file. The OS has already removed the HLP file, so there's no point in leaving it on. It's a pain to recreate the chm file from there, so instead I made a PDF file based on the original RTF file of the help file (this was the easiest).
- I've made various changes to the display, so I've captured new screens and replaced them. I've also included notes for Windows 11.
- It is now possible to start the program even if there is no CD-ROM drive at all. It is now possible to add a drive after booting (this is supported because there are many cases these days where a drive is not usually attached). It is now possible to start the program only for encoding purposes (although I don't think anyone uses it for that purpose).
- The display of the splash window has been completely removed, and the graphics in the About window have been replaced with icon files. Since the URL of the web page has changed, it is troublesome to change the graphics, and the performance of PCs has improved so that the splash window is no longer necessary.
- Changed messages from resource strings to fixed strings, and removed some remaining resource switching processing (completely removed multilingual support: the main reason is that the community edition of Delphi 12.1 does not seem to support multilingual language.)
- CDDB service has been completely stopped (it really makes you feel the passage of time...). It seems that a volunteer is now providing a Japanese version, so I made it work (I think that in 3.xx, the CR/LF is split in the initial request, which makes it difficult to access). The default now points to this server (freedbtest.dyndns.org).
- However, I'm not sure if I can register new data correctly, so for now the registration function is blocked. As mentioned above, the character code has been completely changed this time, so I can't guarantee that it will work properly
- (If I register something strange, it will cause trouble for the server operator...)
- By the way, the compiler change caused Unicode to be partially supported, which affected me a lot, and immediately after the build went through, there were a lot of bugs around CDDB. For that reason, I disabled it for now.
- In 3.xx, the received data was converted from UTF-8 to Unicode and then finally converted to Shift-JIS, but now it is processed internally while remaining in Unicode, so unlike before, garbled characters that do not exist in ShiftJIS should be eliminated.
- This is also true for previous versions, but if you write EnableLog=Yes in the [CDDB] section of the cd2wav32n.ini configuration file, a communication log called CD2WAV32.Log will be created in the current folder where the software is located, so if you notice anything wrong with CDDB, please contact us with the log. The log is usually in UTF-8 text, so you can read it with a text editor that can read UTF-8 (as long as the cddb server supports UTF-8. It should normally support it, so we access it with Protocol Level 6).
- With overseas software, if you register Japanese data, the encoding may be incorrect and the data may be garbled. If the data is garbled at the log stage, there is a problem with the registered data itself. (The communication log records the data sent and received as is, so if there is something wrong at this log stage, it is a no-go.)
- The local DB related functions in the CDDB form have been deleted. They were only available on Windows 95/98.
- Installation is the same as before, just unzip the software to a suitable folder. All settings are written to and read from cd2wav32n.ini, so just delete the file to complete the uninstallation.
Other Changes
- The internal character code processing is now mostly Unicode.
- ID3Tag related items have also been changed from Shift-JIS to Unicode.
- The tool's GUI, including common controls, has been updated to the latest Windows 11 style.
- File size and other information related to the files is now processed in 64-bit format.
- The 16-bit code call function of MSCDEX, which can no longer be used, has been pruned, as have TwinVQ compression, and code to alert users of ASPI errors.
- Default compression is now provided by the Media Foundation API.
- The Help file has been removed and is replaced by a PDF.
- It is now possible to start this tool, even if you don't have a CD-ROM attached.
- And various changes have been implemented to get the CDDB service to run again so ripping can be done with meaningful file names, track titles, artist names, genres etc.