home *** CD-ROM | disk | FTP | other *** search
-
- Many people have inquired whether programs designed to "speed up"
- floppy disk access (SPEED41.COM, SP40.COM) are safe to run. There has been a
- concern that the use of this type of a program would increase the wear and tear
- on the disk drives and shorten their life. This question was raised once again
- by one of the callers on my BBS and his call galvanized me to investigate
- and determine just what the effect of these programs is.
-
- The two programs mentioned above are identical. Somewhere along the
- line someone apparently renamed it. They are both available on various BBS's.
- The program looks at Interrupt 30 (located at 0000:0078) to determine the
- segment and offset of the disk parameter block. In DOS 1.0 this is located
- in the ROM BIOS, but in DOS 1.1 and subsequent it is a part of DOS located at
- 0000:0522. The first byte (out of eleven) of the Disk Parameter contains the
- hexidecimal value CF in DOS 1.0 and DF in DOS 1.1 and subsequent. The
- speedup program changes that value to EF.
-
- Now let me quote from Peter Norton's book "Inside the IBM PC", page
- 138:
-
- "IBM made only two changes in the disk base (sic. parameter)
- table from DOS release 1.00 to 1.10, but they resulted in
- much faster use of the diskettes. First they made a minor
- increase in the time allowed for the diskette drive to move
- from track to track. In the ROM version of the table, used
- by DOS version 1.00, the time was 24 milliseconds; the new
- time, starting with DOS version 1.10, is 26 milliseconds.
-
- On the face of it, this change should have slightly increased
- the operating time, but it apparently was aimed at reducing
- the number of times that a diskette operation failed and had
- to be re-tried - and that speeds up access. If you are looking
- this item up in the disk base table, it is the first half - or
- first nybble, as it is sometimes called - of the first byte
- of the table. The original value was 12, hex C, and the DOS
- 1.10 value was 13, or hex D. This parameter is counted in
- units of two milliseconds, so the increase was 2 milliseconds,
- and the new value is 26 milliseconds."
-
- The SPPED411.COM or SP40.COM programs therefore further increase this
- value to 28 milliseconds, causing a SLOWER operating disk drive! However, it
- apparently also further reduces the required number of retries for there is a
- noticeable improvement in the speed with which the drives access data. It
- would be my surmise that these programs would place LESS stress on the disk
- drives, not more, yet provide improved performance. It is well know that IBM
- designs conservatively, but this is ridiculous.
-
- An additional benefit is that these programs produce a MUCH quieter
- running disk drive.
-
- John McDermott
- (D) 703-321-7441
- ooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooooo