Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!rutgers!lll-crg!styx!ames!ucbcad!ucbvax!CORY.BERKELEY.EDU!dillon
From: dillon@CORY.BERKELEY.EDU.UUCP
Newsgroups: comp.sys.amiga
Subject: Re: C-Ltd. (Disk Controllers Really)
Message-ID: <8611230958.AA07969@cory.Berkeley.EDU>
Date: Sun, 23-Nov-86 04:58:23 EST
Article-I.D.: cory.8611230958.AA07969
Posted: Sun Nov 23 04:58:23 1986
Date-Received: Mon, 24-Nov-86 00:54:43 EST
Sender: daemon@ucbvax.BERKELEY.EDU
Organization: University of California at Berkeley
Lines: 22

I beg to disagree.

DMA offloads the work from the CPU. Thus, disk transfers do not slow down
the computer in general (i.e. other processes doing other things)

When using DMA, you generally do not have disable interrupts or get into nasty
timing problems.  I have no desire to lose characters on the serial port
due to some idiot's idea of a hard-drive interface.

Lastly, the transfer time IS dependant on the disk->computer transfer rate.
Even assuming it must do a lot of tracking, there is still always that 
transfer time when the data is finally ready to be sent.  IBM HD's are MUCH
faster than Amiga HD's simply because they are DMA.  Besides, executables
are usually contiguous (if not, you can MAKE them contiguous).  I would 
rather have LC1/LC2 load in 4 seconds than 8.. or less.

Personally, I'm not going to even look at the HD market for the Amiga until
some good DMA drives (or SCSI controllers) come out.



				-Matt Dillon