Inertia wrote:Buffer underrun protection is useful for preventing inadvertant buffer underruns, but I would not choose to use it as a method to purposely suspend a burn with a technology that produces inaudible "gaps" as most BU technologies still do. In other words, I accept it when it happens, but I would try to avoid it when possible.
<snip>
Given the choice of an interrupted burn with artifacts (although inaudible) or a burn without interruption or artifacts I would choose the latter.
Oh, I agree. In fact, I went so far as to buy a Yamaha CRW3200EZ because it used P-CAV technology when most of its peers were using Z-CLV, and I wanted to avoid links on general principle. But I really feel that I was being overly paranoid, and recognize that P-CAV and CAV have their own potential issues (we won't get into that here
![:wink:](http://www.cdrlabs.com/forums/images/smilies/icon_wink.gif)
). There are millions of Z-CLV drives out there producing links as a normal part of their operation without causing any problems at all. I think the only time it might really be a concern is when producing a master for pressed CD duplication, and I'm not even sure about that.
Inertia wrote:DAO burning may not be an issue with buffer underrun protection if you don't mind stopping a burn while changing CD's. It is a major issue for the many people who are still using older burners without buffer underrun protection. Caching definitely helps but as burning speed increases it won't prevent problems without buffer underrun protection. Caching has been around much longer than buffer underrun protection, and it hasn't been reliable for preventing coasters.
I sure know caching predates buffer underrun prevention.
![:)](http://www.cdrlabs.com/forums/images/smilies/icon_smile.gif)
I burned without a net for years using a Memorex 1622, relying on caching only since the drive had no buffer underrun protection. I never made a coaster, even though I burned all my compilations directly from their constituent files and the system was only a 400 MHz Pentium II. Of course, the drive only wrote at 2x, so keeping the pipe full probably wasn't such a challenge even for that old system...
If all the files are already on the hard drive and none need to be converted, caching is sufficient to prevent buffer underruns even if the burner doesn't have buffer underrun protection, at least in my experience. When burning more complicated compilations that require file conversions or source material from CD-ROM, then I would agree with you that making an image on the hard drive first is advisable for those whose burners don't have buffer underrun protection.
cfitz