Cable Select vs. Master/slave settings

Mike McCarty mike.mccarty at sbcglobal.net
Tue Aug 9 19:18:25 UTC 2005


Craig White wrote:
> On Tue, 2005-08-09 at 12:03 -0500, Mike McCarty wrote:
> 
>>>There are standards, and there are standards, as you rightly point
>>
>>out, and 
>>
>>>much equipment is produced in complete violation of same - now why
>>
>>that 
>>
>>>particular instance occurred for me at that time, is lost to
>>
>>history, but it 
>>
>>>happened, and was formative in my reaching my current methods of
>>
>>operation. 
>>
>>Reinforeces my belief that C/S cables are for the birds.
> 
> ----
> the marketplace tends to absorb previous technology which incorporating
> new advances. Thus some backward compatibility. Everyone learns to cope
> in their own way...some by absorbing the changes, some by dismissing
> them.

Reverting hard disc address philosophy to the way floppies work with
twisted cables, after finding the mess that made, is not "incorporating
new advances". I have found, when moving things around, that the discs
often need to be in a particular position in the case, relative to other
discs. For example, a bay may be "closed", hence unusable for a CDROM.
This dictates where certain discs must go. If the cable does the select,
then the physical location of the disc forces it to be either master or
slave. Do you want the physical location of your disc in the case to
determine which disc is the master? Or would you prefer that the
function the disc performs determines whether it is the master?

This is a technical question, with a technical answer. CS cables, and
twisted floppy cables, and numbered slots all give the wrong answer.
They say that physical location, and not function, determine how the
device performs.

CS is not a new advance. It is a reversion to a previous method of
determining logical address from physical location; one which has
proven over time to be inferior to letting logical address follow
logical function.

This philosophy can be seen in many areas where
we gradually are abandoning geographical designations, for example
with local number portability and roaming with telephones. The
geographical addresses of telephones are gradually being replaced with
logical addresses. Which only makes sense.

CS is an anomaly in which, after abandoning the inferior earlier IBM
floppy disc method of letting the cable determine which device has which
address, and using a superior method of letting a hard disc device
know its own address, we reverted and call it something new.

You may set up your machines however you like. I never jumper discs
to be CS, but to be Master/Slave and recommend that others do likewise.

> Mixing a hard jumpered slave or master device with a device that is set
> to use cable select on a C/S cable seems to present issues (but then
> again, sometimes it works).

No disagreement there.

We all have to live with CS. We don't have to like it. Sanity
dictates that only one method should be used. The technical
superiority of having the logical function of a device be
dependent on its actual function in the machine, and
be independent of its physical location in the machine,
shows that Master/Slave setup is preferable to Cable Select.

If someone comes up with something which works better than
M/S setup, I'll gladly adopt it. Until then, M/S is the
only setup I use and recommend. Even after then, C/S is
for the birds, and should be abandoned.

And that's probably more than enough said!

:-)

Mike
-- 
p="p=%c%s%c;main(){printf(p,34,p,34);}";main(){printf(p,34,p,34);}
This message made from 100% recycled bits.
You have found the bank of Larn.
I can explain it for you, but I can't understand it for you.
I speak only for myself, and I am unanimous in that!




More information about the fedora-list mailing list