u9050728@cs.uow.edu.au (Shane Kelvin Richards) (06/15/91)
Hello. I was wondering if anyone knows the "theory" behind how FTA and others have been able to write really fast disk readers (be it for copying or demos). Are they talking "directly" to the IWM? Why is it so much faster than the standard read? -- +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- Shane Richards u9050728@cs.uow.edu.au +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-
taob@micor.ocunix.on.ca (Brian Tao) (06/17/91)
u9050728@cs.uow.edu.au (Shane Kelvin Richards) writes: > > Hello. I was wondering if anyone knows the "theory" behind how FTA > and others have been able to write really fast disk readers (be it for > copying or demos). Are they talking "directly" to the IWM? Why is it so > much faster than the standard read? The FTA demos have all their code arranged so that you can simply do a sequential block read from the disk, and it will all magically fall into place in memory. Normal file operations (like under GS/OS) obviously cannot assume this, so when a group of files is read from disk, the OS has to check various other tags on the disk (directory entry, block pointers, etc.) You can't get the Apple 3.5" drive to read any faster than about 32K/sec or so. The specs say 64K/sec, but I have never seen it go _that_ fast (no matter what interleave or computer I use it on...)