l i n u x - u s e r s - g r o u p - o f - d a v i s
L U G O D
 
Next Meeting:
October 20: Web Application Hacking: How to Make and Break Security on the Web
Next Installfest:
TBD
Latest News:
Oct. 10: LUGOD Installfests coming again soon
Page last updated:
2001 Dec 30 17:09

The following is an archive of a post made to our 'vox-tech mailing list' by one of its subscribers.

Report this post as spam:

(Enter your email address)
Re: [vox-tech] brain dead hardware question: power pc vs x86 scsi
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [vox-tech] brain dead hardware question: power pc vs x86 scsi



> On Sun, 9 Sep 2001, ME wrote:
> > Those 25-pion SCSI connectors are not part of the SCSI standard. Sure
> > Apple used them on lots of their boxes and IOMega used them on the SCSI
> > and ZIP+ drives, but they are cheating when they do. Whet they do is
> > combine several independednt ground lines into only a few so that each
> > previously independent ground now is shared. Though SCSI is rather robust
> > in many cases and can even "work" with risk/loss in performance without
> > ermination, and seem just fine with the 25-pin connectors, they are not
> > part of the official pinout standard AFAIK.

On Sun, 9 Sep 2001, Jeff Newmiller wrote:
> Be careful... I don't think you should suggest that a scsi system can EVER
> be set up without exactly two (not zero, one or three) termination points,
> located at the ends.  Termination is not the same as sharing grounds.  
> Most SCSI interface cards come with a termination option... in some cases
> it amounts to these ten-pin resistor packs that you can insert or removed
> as needed.

These two were not confused. The point was that todays SCSI adapters are
mord robust than earlier controllers. Though you *could* choose to not
terminate one end of some busses in use today and still see data move, it
is not advised. (Demonstration of robustness being greater by
demonstration of being more tolerant of people breaking standards such as
poor terminators that may be passive and have insufficient reflection
dampening as opposed to active termination. If you had a poor termination
at one end of earlier SCSI busses with older controllers, the devices
would often just not work. Now, a bad terminator or lack of a
terminator on many moden buses can still allow the devices to "work" with
risk of data corruption, drops in spead and degredation in
performance. The issue with the ground is yet another place where a
compnay may not follow the standard and newer controllers do a better job
of dealing with 25-pin SCSI than older controllers. (Also plays to
robustness.)

Also, the issue of required termination was covered in the later portion
of my first e-mail that you chopped off of this in your response.
(re-included it here:)

> On Sun, 9 Sep 2001, ME wrote:
> > Make sure both ends of the SCSI bus are "terminated"  Don't
> > terinate devices in the middle of the bus, only the ends are
> > terminated.

> Also I won't depend on the "termination" switch on a zip drive... ever
> since I lost all the data on all my hard disks due do unreliable
> termination by the zip.  I use a 25-pin termination block.

Yes, on the ZIP SCSI (non plus) I think they used passive termination and
have had some that were not "perfect" (bad resistors, or other reasons
that may include more shortcuts to save money.) I have had some ZIP SCSI
drives that had such poor termination with their switch that they would
not even work on the older cards until I placed a good terminator on the
end of the bus.

Of course, IO Mega's Zip drive still uses that 25 pin SCSI connector which
is also a cheaper shortcut that is not in the SCSI standards, so it is
possible they may have found other shortcuts with passive termination
that is not part of the standards of SCSI.

-ME
-----BEGIN GEEK CODE BLOCK-----
Version: 3.12
GCS/CM$/IT$/LS$/S/O$ !d--(++) !s !a+++(-----) C++$(++++) U++++$(+$) P+$>+++ 
L+++$(++) E W+++$(+) N+ o K w+$>++>+++ O-@ M+$ V-$>- !PS !PE Y+ !PGP
t@-(++) 5+@ X@ R- tv- b++ DI+++ D+ G--@ e+>++>++++ h(++)>+ r*>? z?
------END GEEK CODE BLOCK------
decode: http://www.ebb.org/ungeek/ about: http://www.geekcode.com/geek.html
     Systems Department Operating Systems Analyst for the SSU Library



LinkedIn
LUGOD Group on LinkedIn
Sign up for LUGOD event announcements
Your email address:
facebook
LUGOD Group on Facebook
'Like' LUGOD on Facebook:

Hosting provided by:
Sunset Systems
Sunset Systems offers preconfigured Linux systems, remote system administration and custom software development.

LUGOD: Linux Users' Group of Davis
PO Box 2082, Davis, CA 95617
Contact Us

LUGOD is a 501(c)7 non-profit organization
based in Davis, California
and serving the Sacramento area.
"Linux" is a trademark of Linus Torvalds.

Sponsored in part by:
O'Reilly and Associates
For numerous book donations.