The 2 second gap at the beginning is there because it was felt that this is what is mandated by the Philips Red Book Standard or IEC 960 or IEC 60908. I've not seen either of these documents as these are not free. It was my understanding that (probably older) CD players also required this. See http://www.gnu.org/software/libcdio/libcdio.html#SEC25
Occasionally folks want to violate what the standards allows. The most common case is where people want an ISO 9660 filesystem to be larger than what is allowed by ECMA 119. In some cases we allow for deviation from the standard. However there should be justification, sufficient reason and a understanding that this occurs often enough to be useful by a majority of people to outweigh the negative effects of deviation from standards. In your situation, it sounds like you are writing DTS-encoded information using Redbook formatting. I don't doubt for what you are doing this is convenient for you. But I'm not sure this is common practice, either. If there some document you can point to that indicates this common practice or is part of some standard, that would make a more compelling case for change. As for the mailing list and libcdio, it's always been a volunteer organization. People come and go as they feel the need, have time or the interest. Best as I can tell, the library is used more now than it's ever been. Robert William Fuller writes: > Robert William Fuller wrote: > > The quirk results in a serious usability problem with the library. As > > I'm sure you know, the library assumes a two second gap at the beginning > > of all audio tracks. Put simply, more often than not, this is an > > incorrect assumption. Hence, when you rip tracks with libcdio-paranoia, > > often times the beginning of the songs are clipped. It is especially a > > Umm ok so I was partially wrong on this. libcdio-paranoia apparently > uses either LSN's or LBA's because it ignores the two second gap that > the library reports for MSF's. > > Nevertheless, it would be nice if applications that used MSF's, such as > those that create cue or toc files, did not have to subtract two seconds > from the value returned by the library for audio tracks! Hence, there is > still a usability problem, but it is within the library, and > libcdio-paranoia does not suffer from it. > > > Now I realize that cdparanoia does not currently detect gaps and that is > > the reason that an assumption is made. I also saw somewhere a > > suggestion for a GSoC project to add gap detection. However, in the > > interim, I would suggest that the library make a better assumption, that > > audio tracks do NOT start with a two second gap. > > Still stands... > > > Also, I have not seen any traffic on this list since I signed up, and I > > notice that aside from configuration code, little has changed in the CVS > > for many months. Is this project still active? > > Still wondering... > > > > > Thank you. > > > > Regards, > > > > Rob > > > > > >
