-- card: 22848 from stack: in
-- bmap block id: 23263
-- flags: 0000
-- background id: 21270
-- name: 


-- part 1 (field)
-- low flags: 00
-- high flags: 0000
-- rect: left=10 top=56 right=318 bottom=260
-- title width / last selected line: 0
-- icon id / first selected line: 0 / 0
-- text alignment: 0
-- font id: 2
-- text size: 10
-- style flags: 0
-- line height: 13
-- part name: 


-- part 2 (field)
-- low flags: 00
-- high flags: 0000
-- rect: left=262 top=56 right=331 bottom=512
-- title width / last selected line: 0
-- icon id / first selected line: 0 / 0
-- text alignment: 0
-- font id: 2
-- text size: 10
-- style flags: 0
-- line height: 13
-- part name: 


-- part contents for background part 1
----- text -----
47

-- part contents for card part 1
----- text -----
can be done with a small (2 K) INIT file called "SCSI Accelerator" (version 1.2 is the latest).  This INIT will apply a patch to the SCSI Dispatch call within the Macintosh system software which will increase the data transfer rate between the Macintosh and any SCSI hard disk (except drives that are already high performance, such as SuperMac's XP series drives).  Version 1.2 works with System versions 4.2 and 6.0 
(I did it with version 6.0.3).  No improvement will be experienced on a Mac SE or Mac II.  NOTE: before trying this you may wish to have a current backup of your hard disk drive.  I didn't, but then I'm a wild and crazy guy.
The author notes that to get the full potential you may need to reformat your hard disk drive with a different interleave 
setting.  Most Mac drives have a 3:1 

-- part contents for card part 2
----- text -----
interleave.  The author was able to use a 1:1 ratio on two different types of drives.  One had a RLL controller and the other a MFM controller.  You should experiment to determine the best interleave for your drive.  According to the author it could be anywhere from 1:1 to 3:1.
I was somewhat nervous about using the SCSI on my Quantum 80 MEG hard disk drive, but finally said what the heck.  I had no problems.  Using the utility DiskTimer II for 100, 24KB reads I recorded a read time of 103 without the patch and 57 with the patch.  
Write times were 100 and 56 respectively.  This is without changing the drive's 3:1 interleave ratio.  Seek times were essentially 
unchanged at
21 ms.Another 
utility called