From the present thread and from hdd size limits?
, I think I have enough material to create this summary of what we do know at present about the limits of the different programs related to HDD formating, partitioning and maintenance, so here it goes:
The limit for NDD32.EXE (up to v. 18.104.22.168, from NSW 2008) is somewhere between 7.8 and 7.9 million clusters, or somewhere between 61.0 and 61.7 thousand sectors per FAT (as I now believe it crashes when reading the FATs to a buffer in memory).
The limit for SCANDSKW.EXE (22.214.171.12400) is somewhere between 26.4 and 26.6 million clusters, or somewhere between 206.1 and 207.7 thousand sectors per FAT. It crashed with Marius '95 1 TB raid single partition (link
), although 98-Guy has reported it works up to 31.2 million clusters (follow the links inside this post
SCANDISK.EXE from Win ME works, at least, up to 1 TB, according to Marius '95, and to 31.2 million clusters, according to 98-Guy (both these limits are about the same, for a 1 TB partition, using 32 kiB clusters, has about 31 million clusters).
NDD.EXE for DOS (2002 ..10E) also is reported by the same users as having the same limits as those of SCANDISK.EXE, but that's now doubtful, because it crashes for wsxedcrfv with 22.9 million clusters. In any case, Marius '95, for whom it worked, said it was very slow, so maybe he just didn't wait enough time for it to crash...
FORMAT.EXE works up to, at least 1018 GiB, but above 1TiB a divide error occurs, according to RLoew, in the present thread.
And the limit of Petr's fixed FDISK
(based on the FDISK contained in this update: KB263044
, which has a numerical display bug) is 512 GB, according to Microsoft (KB280737
), and confirmed in the present thread. Suitable alternatives are The Ranish Partition Manager
, although it is not adequate to format the partitions it creates, because of defaulting to 16 kiB clusters, or the Free FDISK v. 1.2.1
, or Symantec's GDISK (not free), or RLoew's RFDISK