Some senseless babbling from Craig Swanson to Mike Ruskai
on 29 Jan 96  13:19:22 about Cluster algorithm...

 MR> This is the algorithm I'm using to determine a drive's
 MR> cluster size based on
 MR> its size:
 
 MR> Rough cluster size is drive size divided by 65536

 CS> I think this should be 65528.  My recollection is the top 8 numbers
 CS> (i.e., 0xFFF8 through 0xFFFF) are reserved for flagging free
 CS> clusters,bad clusters, and some other uses.

I'm not doing anything serious - just wrote a REXX program to scan a drive 
and present the wasted space in the current file system, and what it would 
be in the alternative.  So the way I did it (already uploaded it to 
Hobbes), there would be a progressive margin of error starting with 16k 
heading towards 128MB, all the way up to 256k when approaching 2GB.  Since 
FDISK only increments 1MB at a time, I don't believe it would ever cause a 
problem.  Thanks for the info.

Mike [Team OS/2]
mruskai{at}microfone.net


... Are there any lawyers here?  Any more?
~~~ ReneWave v1.01+
--- RG 10-5 Exp/GE 1.02+