More SCSI termination questions
Joanne Dow
sp...@drycas.club.cc.cmu.edu wrote:
>I have a A3000 with 1 internal hard drive which is terminated and a
>external CDROM which is also terminated. Now I want to add a 2nd
>internal hard drive. Could someone tell me the proper termination
>setup for this configuation? Do I put terminators on the new hard
>drive and/or take them off my old internal drive? Also at what unit
>number should I put the old drive. The drives are a Connor 170 meg
>SCSI (old drive) at unit number 6 and a Quantum 105 meg SCSI drive
>which is the new drive I want to install and a Toshiba CDROM in a
>external case at unit number 2. I also replaced the A3000 proto
>rev 04 SCSI chip with a rev 8 SCSI chip.
>
>
> Scott Corley
>
>
>sp...@drycas.club.cc.cmu.edu
>
Look at the cabling. Where are the ends of the cable if you figure the
external cable and the internal cable make one long cable? Of course the
external drive must still be terminated. On the inside terminate the drive at
the end of the internal cable be it "first or second" drive chronologically
and regardless of their SCSI IDs. I would, as a side issue, make the CDROM a
higher ID than the drives so that you can short circuit the drive search
before the CDROM drive, which can make booting materially faster on some
A3000s.
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
sp...@drycas.club.cc.cmu.edu
sp...@drycas.club.cc.cmu.edu
> In article <1995Aug2...@drycas.club.cc.cmu.edu>,
> sp...@drycas.club.cc.cmu.edu wrote:
>>external CDROM which is also terminated. Now I want to add a 2nd
>>internal hard drive. Could someone tell me the proper termination
>>setup for this configuation? Do I put terminators on the new hard
>
> external cable and the internal cable make one long cable? Of course the
> external drive must still be terminated. On the inside terminate the drive at
> the end of the internal cable be it "first or second" drive chronologically
> and regardless of their SCSI IDs. I would, as a side issue, make the CDROM a
>
Since my 1st internal drive which was already terminated was at the end of
the internal cable, I added the 2nd drive without terminators and it works
fine.
Thanks Joanne
Scott Corley
ChrisBig-KahunaRampson
: I have a A3000 with 1 internal hard drive which is terminated and a
: external CDROM which is also terminated. Now I want to add a 2nd
: internal hard drive. Could someone tell me the proper termination
: setup for this configuation? Do I put terminators on the new hard
: number should I put the old drive. The drives are a Connor 170 meg
: SCSI (old drive) at unit number 6 and a Quantum 105 meg SCSI drive
: which is the new drive I want to install and a Toshiba CDROM in a
: external case at unit number 2. I also replaced the A3000 proto
: rev 04 SCSI chip with a rev 8 SCSI chip.
: Scott Corley
: sp...@drycas.club.cc.cmu.edu
GVP says - If all are INTERNAL devices, terminate the last one. If any are
EXTERNAL devices, terminate that last one (none needed internally).
--
_/_/_/ _/ _/ Chris "Big Kahuna" Rampson
_/ _/_/_/ _/ _/_/ _/_/_/ cram...@ford.com
_/ _/ _/ _/_/ _/ _/_/ NO cutdowns,flames or insults
_/ _/ _/ _/ _/ _/_/ ONLY data as I interpret it
_/_/_/ _/ _/ _/ _/ _/_/_/ "NOT an OFFICIAL FORD spokesperson"
J_PL...@amtrash.comlink.de
(02.08.1995)
< [...] I would, as a side issue, make the CDROM a
> higher ID than the drives so that you can short circuit the drive search
> before the CDROM drive, which can make booting materially faster on some
> A3000s.
Use 3.1 (40.xx hardkicked) and A3000`s SCSI will be much easier...
/ J_Pl...@Amtrash.comlink.de \
/ Plew...@Informatik.fh-hamburg.de \
+-------------------------------------+
\ A3000/030/6/52 A3000T/040/18/512 /
Joanne Dow
>
>GVP says - If all are INTERNAL devices, terminate the last one. If any are
>EXTERNAL devices, terminate that last one (none needed internally).
>
You go ahead and believe that, Chris. If you have problems do not whine to us
here on the net. I have told people here repeatedly what the SCSI specs say,
every edition so far, and those who choose to ignore it are on their own. I help
once. Willfull ignorance I have little time for.
>
>--
>
> _/_/_/ _/ _/ Chris "Big Kahuna" Rampson
> _/ _/_/_/ _/ _/_/ _/_/_/ cram...@ford.com
> _/ _/ _/ _/_/ _/ _/_/ NO cutdowns,flames or insults
>_/ _/ _/ _/ _/ _/_/ ONLY data as I interpret it
> _/_/_/ _/ _/ _/ _/ _/_/_/ "NOT an OFFICIAL FORD spokesperson"
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
Kevin Allan Donald Carter
JD> You go ahead and believe that, Chris. If you have problems do not whine to
JD> us here on the net. I have told people here repeatedly what the SCSI specs
JD> say, every edition so far, and those who choose to ignore it are on their
JD> own. I help once. Willfull ignorance I have little time for.
Ha ha. Your oh so helpfull SCSI termination advice. It took Randell Jesup to
convince you A3000s didn't need an external terminator as they were already
terminated on the mother board and the over terminating you were telling
people they *MUST* do could potentially cause damage to the components. He did
convince you, I presume -- but then he replied here and not on BIX so perhaps
not. I only told you, what, three times? Just imagine, theory not living up to
practical application. Who'd of guessed? Now what's that you say about
willfull ignorance?
To Chris: if problems persist feel free to post. Some of us, particularly
those with actual practical experience, don't mind helping people who aren't
using a pay service.
Kevin
Warren Block
: On Thu 10-Aug-1995 7:54a, Joanne Dow wrote:
: JD> You go ahead and believe that, Chris. If you have problems do not whine to
: JD> us here on the net. I have told people here repeatedly what the SCSI specs
: JD> say, every edition so far, and those who choose to ignore it are on their
: JD> own. I help once. Willfull ignorance I have little time for.
: Ha ha. Your oh so helpfull SCSI termination advice. It took Randell Jesup to
: convince you A3000s didn't need an external terminator as they were already
: terminated on the mother board and the over terminating you were telling
: people they *MUST* do could potentially cause damage to the components. He did
You seem to be confused. If both internal and external SCSI devices are
to be used, the internal motherboard terminators should be removed. In
this case, you do need an external terminator. AFAIR (and I haven't paid
a lot of attention because it didn't make a lot of sense), it was
suggested that goofy (i.e., not SCSI-spec) termination schemes would work
on the A3000--and sometimes they do. That doesn't make it a good idea.
--
---------------------------------------------------------------------
| Warren R. Block * New EMail Address: wbl...@server.sdserv.org |
| Rapid City SD USA * Brought to you in majestic INEMASCOP! |
---------------------------------------------------------------------
Michael Greene
>In article <407pk8$f...@tbd120.tbd.ford.com>,
>>
>>GVP says - If all are INTERNAL devices, terminate the last one. If any are
>>EXTERNAL devices, terminate that last one (none needed internally).
>>
>You go ahead and believe that, Chris. If you have problems do not whine to us
>here on the net. I have told people here repeatedly what the SCSI specs say,
I missed it, what did you say about SCSI specs? Also, I have a GVP Series II
in my 2000 with no drives in the case. I run a 2 foot(?) cable to a mini tower
with the following devices: 52meg, 110meg, and 1.02gig hard drives; Tandberg
3820 tape drive; 44meg SyQuest removable HD. Here's the question, I got an old
NEC x1 speed External CDROM but the only cable I have to run between it and
the tower is a 6' cable. Isn't that a bit long?
// ---------------------------------------------------
// Michael K Greene PGP key:
// mgr...@exis.net finger mgr...@exis.net
\X/ ---------------------------------------------------
Thor2.01/GED3.1.3
Coffee - 2 sugars - cream - and aspirin.
Stefan Payment
to write something under the subject "Re: More SCSI termination questions"
this is what I think about it :
WB> You seem to be confused. If both internal and external SCSI devices are
WB> to be used, the internal motherboard terminators should be removed. In
WB> this case, you do need an external terminator. AFAIR (and I haven't paid
WB> a lot of attention because it didn't make a lot of sense), it was
WB> suggested that goofy (i.e., not SCSI-spec) termination schemes would work
WB> on the A3000--and sometimes they do. That doesn't make it a good idea.
jajaja
The SCSI-Bus has to be terminated on both (physicaly) ends.
There's just the thing that you can direct terminate the last HD (mostly
used internal) and so don't need an extra terminator.
and if you use external stuff you've to use an external termiator or
set the terminator on the last HD (or whatever is the last Item).But
then you MUST remove (if possible) the terminator on the motherboard
or Host Adaptor otherwise you'll have a third terminator in the chain
and that'll never work out.
--
Stefan
--------------------------------------------------------------------------
E-Mail:Les...@Fishtwn2.han.de|Realname:Stefan Payment|Voice:+49 4721 65077
runnin on AMIGA1200T/030/882/50/2C/6F/428+127HD/SCSI2 but who cares ? ;-)
--------------------------------------------------------------------------
Joanne Dow
ka...@julian.slip.uwo.ca (Kevin Allan Donald Carter) wrote:
>On Thu 10-Aug-1995 7:54a, Joanne Dow wrote:
>JD> us here on the net. I have told people here repeatedly what the SCSI specs
>JD> say, every edition so far, and those who choose to ignore it are on their
>JD> own. I help once. Willfull ignorance I have little time for.
>
>Ha ha. Your oh so helpfull SCSI termination advice.
Got your homonymns mixed up there I see. The advice is as helpful as people make
it. If they refuse to use it they lose.
> It took Randell Jesup to
>convince you A3000s didn't need an external terminator as they were already
>terminated on the mother board and the over terminating you were telling
>people they *MUST* do could potentially cause damage to the components. He did
>convince you, I presume -- but then he replied here and not on BIX so perhaps
>not. I only told you, what, three times? Just imagine, theory not living up to
>practical application. Who'd of guessed? Now what's that you say about
>willfull ignorance?
>
Randall rather concurs with me than disagrees if you read more carefully.
Regardless he has convinced me of no such thing as you are accusing.
I repeat - for a properly operating SCSI bus you normally want precisely two
terminators on the bus, one at each end. This is required for proper signal
propogation on the bus. Such is one of the fields of expertise I claim that Dave
Haynie may share with me but I doubt Randall, a software engineer, would. If you
have fast drives and a fast controller and a long unterminated external cable
you will experience problems as a general rule. The termination rules detailed
in both the SCSI-1 and SCSI-2 specifications state this clearly enough I daresay
even you could determine what they say.
Some A3000s were delivered with terminators on the motherboard. Based on my
experience here with machines tapped out for SCSI addresses and an amazing
variety of peripherals the termination rules are best followed. If you have two
drives inside you have enough cable to give you problems if you leave the
internal bus unterminated. If you can keep the cable run to under a foot you
might get away with the onboard terminator inside amd no other internal
terminator. *I* would go in and chop out the terminators on the motherboard and
properly terminate the last internal drive. Then I'd use proper 50 wire cables
to external peripherals and terminate the last of them. Using this technique I
have had three fast SCSI harddisks, a CD-ROM, a tape drive, and an unusual
scanner/plotter from Seiko that did 36" wide pages. Unless I had the termination
as described this lashup failed to work.
>To Chris: if problems persist feel free to post. Some of us, particularly
>those with actual practical experience, don't mind helping people who aren't
>using a pay service.
>
>
>
>Kevin
Feel free to post - just do not bitch if it does not work right. It might appear
to work. If it does "go for it." It sounds like you enjoy life on the edge. I
prefer data reliability. And with at least as much experience as it appears from
your postings you have been alive I think I may just know a little more about
how things work than you. (I've been working with transmission-line technology
for over 35 years now. What are YOUR qualifications, Sir.)
ChrisBig-KahunaRampson
2. I quoted the docs from GVP on termination.
3. I have NEVER seen an internal terminator. I work with UNIX workstations
all day, and NONE of them have internal terminators, although they all
use EXTERNAL terminators, even if there are no devices. May be just a
different design.
4. Please do not incinuate that I am incompetant! I have 7+ years of
experience in UNIX administration plus a Masters degree.
decu...@fn1.freenet.edmonton.ab.ca
: In article <407pk8$f...@tbd120.tbd.ford.com>,
: >
: >GVP says - If all are INTERNAL devices, terminate the last one. If any are
: >EXTERNAL devices, terminate that last one (none needed internally).
He appears to be un-aware that GVP solders their terminators to the
controller board, and makes no definitive statement about what needs to
be done when you have BOTH internal and external devices, other than to
state that ONLY the last device on each end of the chain needs to be
terminated. This unfortunately requires some logical thought to get to
the conclusion that if multiple devices are installed in both external
and internal chains, then the terminators need to be removed from the
controller.
: You go ahead and believe that, Chris. If you have problems do not whine to us
: here on the net. I have told people here repeatedly what the SCSI specs say,
: every edition so far, and those who choose to ignore it are on their own. I help
: once. Willfull ignorance I have little time for.
Well, you can lead 'em to water Joanne, but that's about it.
Cheers,
--
email: decu...@freenet.edmonton.ab.ca
+ _ ____________ tm Dale Currie ____ ___ _ +
| /.\ .. | __ \ / alias "ZZORRO", / __[___]__ T tm|
| /___\ /\/\ | | |_) | X Borg Systems Analyst / (o.o) | |
| / \/ ^^ \ | | | \ | / \ & AmiTrix / `-^-' | |
|/ - D E V E L O P M E N T - \ Technical Support /___ Z O R R O I N K !|
| --------------------- ---------------- |
+ ("Zorro lives!" -++- "He's been assimilated as part of my Amiga!") +
Scott Timberlake
JD> about how things work than you. (I've been working with transmission-
JD> line technology for over 35 years now. What are YOUR qualifications,
JD> Sir.)
Well, it looks like you are the perfect person to answer my question then ;-).
I have a 3000 with one internal HD and an external CD-ROM. I have never
altered the termination on anything, so I assume then that the HD and CD-ROM
are terminated. It would be hard to tell on the CD-ROM. It's a MediaVision
CDR-MV2 and any terminators would be inside the case.
The problem is that when I have the CD-ROM turned on, the SCSI bus ALWAYS
locks up. Sometimes it happens right at boottime, sometimes it waits until
the system has been on for an hour. But it ALWAYS eventually happens. And
it has nothing to do with whether or not I am accessing the CD in the drive.
When it locks up, only the SCSI bus is affected. All other system functions
remain unaffected. And if I wait for about 5-10 mintues, the bus will
"unlock", and everything is fine again(until the next lockup, of course).
My question is whether this is a SCSI controller chip problem(since I have
an older 3000 with the old SCSI chip) or a termination problem. I don't
want to spend $25 on a new chip only to find out it's termination.
Scott
* Q-Blue 2.0 [NR] *
Jim Pritchett
> He appears to be un-aware that GVP solders their terminators to the
> controller board, and makes no definitive statement about what needs to
> be done when you have BOTH internal and external devices, other than to
> state that ONLY the last device on each end of the chain needs to be
> terminated. This unfortunately requires some logical thought to get to
> the conclusion that if multiple devices are installed in both external
> and internal chains, then the terminators need to be removed from the
> controller.
Oh, it is worse than that. On my GVP controller, they don't tell you that the
terminators are there or even where they are. Further, they are non-standard
in location and in their appearance. Nice, huh?
Jim Pritchett
UUCP: j...@caleb.uucp "I am the way, the truth, and
rwsys.lonestar.org!caleb!jdp the life; no man cometh to the
caleb!j...@lerami.lerctr.org Father, but by me."
John 14:6
ChrisBig-KahunaRampson
: Joanne Dow (jd...@BIX.com) wrote:
: : In article <407pk8$f...@tbd120.tbd.ford.com>,
: : >
: : >GVP says - If all are INTERNAL devices, terminate the last one. If any are
: : >EXTERNAL devices, terminate that last one (none needed internally).
: : > ^^^^^^^^^^^^^^^^^^^^^^^^^
: He appears to be un-aware that GVP solders their terminators to the
: controller board, and makes no definitive statement about what needs to
: be done when you have BOTH internal and external devices, other than to
Hmm, I guess the manual that I read this stuff in was not valid . . .
: state that ONLY the last device on each end of the chain needs to be
: terminated. This unfortunately requires some logical thought to get to
: the conclusion that if multiple devices are installed in both external
: and internal chains, then the terminators need to be removed from the
: controller.
: : You go ahead and believe that, Chris. If you have problems do not whine to us
: : here on the net. I have told people here repeatedly what the SCSI specs say,
: : every edition so far, and those who choose to ignore it are on their own. I help
: : once. Willfull ignorance I have little time for.
: Well, you can lead 'em to water Joanne, but that's about it.
I guess YOU don't read these posts well enough! I was a respondent to the
original post. That person wanted to know about how to terminate SCSI. I
responded with what the GVP MANUAL SAYS. I personally believe that you
you should terminate if your machine needs it.
--
_/_/_/ _/ _/ Chris "Big Kahuna" Rampson
_/ _/_/_/ _/ _/_/ _/_/_/ cram...@ford.com
_/ _/ _/ _/_/ _/ _/_/ NO cutdowns,flames or insults
_/ _/ _/ _/ _/ _/_/ NO such thing as a stupid question!
Kevin Allan Donald Carter
JD> I repeat - for a properly operating SCSI bus you normally want precisely
JD> two terminators on the bus, one at each end. This is required for proper
JD> signal propogation on the bus. Such is one of the fields of expertise I
JD> claim.
You seem to enjoy avoiding the issue. No-one is denying you need to terminate
both ends of a SCSI chain (though you certainly like to repeat it over and
over and over...). We were discussing a specifica case of SCSI termination on
A3000s. You emphatically stated all A3000s "*MUST*" have an external DB25 SCSI
terminator. I corrected you. You ignored it and reposted including your
"experience" with SCSI manuals and helping people on BIX and all sort of other
stuff Amiga and otherwise. I corrected you again and included my first hand
experience with A3000s as well of that of three of my collegues. You made some
backhand comments (which is what I've come to expect from you when dealing
with non-BIX people). Randell Jessup corrected you. I post expecting some sort
of retraction on your part (actually, I expected you to pull a Bourdin and
avoid the issue) but here you go again replying but skirting the issue all the
same.
The simple fact of the matter is the external SCSI ports of A3000s are
terminated on the motherboard and in order to use external SCSI devices
reliably you need to remove those terminating resistors. This is the cause of
the famous A3000 SCSI lockup when adding external devices as I'm sure your
aware from all your years of being oh so helpful on BIX. You do not add
*additional* terminating resistors; one on each end is sufficient. It probably
says that in one of your SCSI manuals someplace...why don't you go look it up
just for fun. Each end of the SCSI chain needs to be terminated *once*. There,
I've repeated it for you. Double terminating the external port is pointless,
much as your ravings about the history of SCSI and BIX when we were dealing
with someone with a simple SCSI problem with their A3000.
JD> If you have two drives inside you have enough cable to give you problems
JD> if you leave the internal bus unterminated.
Yes, but as no-one was suggesting to do such a thing why do you keep harping
on about it?
JD> If you can keep the cable run to under a foot you might get away with the
JD> onboard terminator inside amd no other internal terminator.
Again, this wasn't discussed here (except now for a third time by you).
JD> And with at least as much experience as it appears from
JD> your postings you have been alive I think I may just know a little more
JD> about how things work than you. (I've been working with transmission-line
JD> technology for over 35 years now. What are YOUR qualifications, Sir.)
Well, Chris has pointed out his qualifications (I'm not sure how he got drawn
into this as he had a different question) as have I. I'll assume you don't
need to see Randell's resume. Of course mine is strictly practical experience
selling, using, installing, and owning A3000s *first hand* PLUS information
from the Commodore technical manuals for good measure, as is that of my
collegues. That certainly doesn't compare to 35 years of phone installation
(transmission line technology), but then mine is *relevant* to the point being
discussed.
Flame away if you want, I can take whatever you can dish out, but lay off
putting down the people who are posting here sincerely asking for help. They
don't need your snide comments tacked on to the end of your "help". Just
because they aren't on BIX doesn't make them any less deserving of respect.
Kevin
Joanne Dow
ram...@ave050.ve.ford.com (Chris"Big-Kahuna"Rampson) wrote:
>
>I guess YOU don't read these posts well enough! I was a respondent to the
>original post. That person wanted to know about how to terminate SCSI. I
>responded with what the GVP MANUAL SAYS. I personally believe that you
>you should terminate if your machine needs it.
>
some drives the GVP way may work. For the general case the terminations at
the end of the COMPLETE bus is the way to go. (Suppose two drives inside and
two outside, thus:
drive 0 drive 1 controller drive 3 drive 4
\________/\________/\______________/\________/
In that wiring diagram you want one terminator on drive 0, one on drive 4,
and no other terminators in the system, including on the GVP card. This gives
your system the best chance of operating with any arbitrary set of drives.
>--
>
> _/_/_/ _/ _/ Chris "Big Kahuna" Rampson
> _/ _/_/_/ _/ _/_/ _/_/_/ cram...@ford.com
> _/ _/ _/ _/_/ _/ _/_/ NO cutdowns,flames or insults
>_/ _/ _/ _/ _/ _/_/ NO such thing as a stupid question!
> _/_/_/ _/ _/ _/ _/ _/_/_/ "NOT an OFFICIAL FORD spokesperson"
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
Joanne Dow
sti...@tbag.tscs.com (Scott Timberlake) wrote:
>JD> I think I may just know a little more
>JD> about how things work than you. (I've been working with transmission-
1) Never "assume" anything. "Assume == ass u + me". {^_-} (Do check that there
is a termination in the CDROM case.)
2) If the CDROM case has two of the 50 pin "Centronics" or such connectors on it
it is not likely terminated. Add one externally to the unused port.
3) If it cabled for "DB-25" connectors yank the drive, pull the built in cables,
install you own 50 pin ribbon cable with the connector for the drive in the
middle of the cable, reinstall the drive with whatever case work is needed. (OK
- it does not have to be in the middle so much as leave enough slack to get both
ends out of the case.) Get a *SHORT* DB-25 to 50 pin "Centronics" adapter that
is 1/2" thick or thicker. (They are most likely to have all 50 wires as proper
twisted pairs.) Get a pair female "Centronics" 50 pin connector to put on the
ends of the cable you used to connect to the drive. Place an appropriate
terminator on the short chunk of cable ourside the case and connect the other to
the adapter cable. If you have done everything right it should fire up and run.
And it is likely to run longer and more reliably.
4) Make sure it is not your CDROM filesystem software. (I have not heard of
either Commodore's or "Canadian Prototype Replicas'" driver failing this way,
though.)
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
(The "inventor" of that DB-25 and 25 wire cabling for the Mac should
experience some very special "treatment" at the hands of his
tormentors in the afterlife I am sure he will experience.)
decu...@fn1.freenet.edmonton.ab.ca
: In article <40pov1$c...@news.sas.ab.ca>, writes:
: > He appears to be un-aware that GVP solders their terminators to the
: > controller board, and makes no definitive statement about what needs to
: > be done when you have BOTH internal and external devices, other than to
: > terminated. This unfortunately requires some logical thought to get to
: > the conclusion that if multiple devices are installed in both external
: > and internal chains, then the terminators need to be removed from the
: > controller.
: Oh, it is worse than that. On my GVP controller, they don't tell you that the
: terminators are there or even where they are. Further, they are non-standard
: in location and in their appearance. Nice, huh?
Yes, wasn't it, NOT! Many of the manuals (mine included) do not show
them on the boards diagram, or mention that they are there. They also
don't mention a number of the jumpers that are on it, either. Makes it
very difficult for non-technical users. I removed mine and put sockets
in the board in case they had to be replaced.
Interestingly, they did socket the terminators on the Combo accelerator
with SCSI controller. Curious!
decu...@fn1.freenet.edmonton.ab.ca
: decu...@fn1.freenet.edmonton.ab.ca wrote:
: : In article <407pk8$f...@tbd120.tbd.ford.com>,
: : >
: : >GVP says - If all are INTERNAL devices, terminate the last one. If any are
: : >EXTERNAL devices, terminate that last one (none needed internally).
: : > ^^^^^^^^^^^^^^^^^^^^^^^^^
: : He appears to be un-aware that GVP solders their terminators to the
: : controller board, and makes no definitive statement about what needs to
: : be done when you have BOTH internal and external devices, other than to
: Hmm, I guess the manual that I read this stuff in was not valid . . .
Well, if it's like mine (and probably is), it's not that what they say is
not valid, it's that they don't mention all the possibilities. The manual
seems to assume that one will not connect multiple devices to both the
internal and external chains.
: : state that ONLY the last device on each end of the chain needs to be
: : terminated. This unfortunately requires some logical thought to get to
: : the conclusion that if multiple devices are installed in both external
: : and internal chains, then the terminators need to be removed from the
: : controller.
: I guess YOU don't read these posts well enough! I was a respondent to the
: original post. That person wanted to know about how to terminate SCSI. I
: responded with what the GVP MANUAL SAYS. I personally believe that you
: you should terminate if your machine needs it.
Quite true! I was not trying to flame you, Chris, but only point out
that there is a great deal of ambiguity in some of the GVP manuals. This
is born out by the vast amount of mail in the GVP echos on FIDO from many
users that had problems and were confused by the lack of clear information.
Scott Timberlake
KADC> the SCSI terminators from the A3000 motherboard in order to use
KADC> external SCSI devices reliably (if you can get them to work at all).
KADC> Check with your nearest Amiga dealer if you aren't keen on doing it
KADC> yourself. Check either the CD-ROM drive manual or with the dealer you
KADC> bought it from to see if it's terminated (it most likely is).
I was suprised, but the manual mentioned NOTHING about termination.
JD> 2) If the CDROM case has two of the 50 pin "Centronics" or such connectors
JD> on it it is not likely terminated. Add one externally to the unused port.
KADC> That's a big *IF*. Most external CD-ROM drives use DB-25 connectors
KADC> and are shipped terminated.
It has one DB25 connector.
JD> 3) If it cabled for "DB-25" connectors yank the drive, pull the built in
JD> cables, install you own 50 pin ribbon cable with the connector for the
JD> drive in the middle of the cable, reinstall the drive with whatever case
JD> work is needed. (OK - it does not have to be in the middle so much as
JD> leave enough slack to get both ends out of the case.) Get a *SHORT* DB-25
JD> to 50 pin "Centronics" adapter that is 1/2" thick or thicker. (They are
JD> most likely to have all 50 wires as proper twisted pairs.) Get a pair
JD> female "Centronics" 50 pin connector to put on the ends of the cable you
JD> used to connect to the drive. Place an appropriate terminator on the short
JD> chunk of cable ourside the case and connect the other to the adapter
JD> cable. If you have done everything right it should fire up and run.
JD> And it is likely to run longer and more reliably.
KADC> a) this still won't solve the problem
KADC> b) you'll void the warranty on the CD-ROM drive
KADC> c) it's time and money best placed in the "hack" category (not a
KADC> particularly
KADC> bad idea but hardly necessary)
I agree. That would be a last ditch effort. In fact, since I only use it
to access data disks occassionally, it probably isn't worth my time.
JD> 4) Make sure it is not your CDROM filesystem software. (I have not heard
JD> of either Commodore's or "Canadian Prototype Replicas'" driver failing
JD> this way, though.)
No easy way for me to test this idea, since Asim 2.0 is what was given to
me by the guy I bought the drive from. Anyways, I'm 99.99% sure that it's
not a software problem, as I'm sure everyone else is.
KADC> ps. I have seen good advice from Joanne Dow in the past... she just
KADC> seems to have this blind spot when it comes to A3000's and simple
KADC> solutions.
Yeah. I don't ever remember seeing anyone disagree with her until she started
giving A3000 SCSI advice. Then people came out of the woodwork disagreeing
with her. Then it got really weird when she started spouting off about how
many decades of telecommunications experience she has. Glad to see I wasn't
the only one who wondered what the f!?# that had to do with A3000 SCSI
knowledge/experience. Just proves that NO ONE knows EVERYTHING. Now someone
just needs to convince JD of this...
Maxwell Daymon
: Yeah. I don't ever remember seeing anyone disagree with her until she started
: giving A3000 SCSI advice. Then people came out of the woodwork disagreeing
: with her. Then it got really weird when she started spouting off about how
: many decades of telecommunications experience she has. Glad to see I wasn't
: the only one who wondered what the f!?# that had to do with A3000 SCSI
Just goes to show your ignorance in the matter (that you don't understand
the correlation).
: knowledge/experience. Just proves that NO ONE knows EVERYTHING. Now someone
: just needs to convince JD of this...
Joanne just needs to let some people hang themselves. Ignorance is the
only disease with a cure that nobody wants, but she keeps offering.
--
+------------------------------------+ you'll never believe me so,
| Maxx Daymon -+- mda...@rmii.com | why don't you find out for yourself
+------------------------------------+ then you'll see the glass
hidden in the grass... / Morrisey
Joanne Dow
>It has one DB25 connector.
>
headache brewing if you choose to go the DB25 route.
.. re replacing cable with proper 50 wire cable.
>I agree. That would be a last ditch effort. In fact, since I only use it
>to access data disks occassionally, it probably isn't worth my time.
Well - one either busts the warranty or puts up with the SCSI bus locks I
suppose. It depends on how often it happens and how irritating the locks are.
25 wire cables are a very clearly identified problem on a great number of
systems with large, hence fast, drives in place. (Large here being 500meg and
up.)
>
>Yeah. I don't ever remember seeing anyone disagree with her until she started
>giving A3000 SCSI advice. Then people came out of the woodwork disagreeing
>with her. Then it got really weird when she started spouting off about how
>many decades of telecommunications experience she has. Glad to see I wasn't
>the only one who wondered what the f!?# that had to do with A3000 SCSI
>knowledge/experience. Just proves that NO ONE knows EVERYTHING. Now someone
>just needs to convince JD of this...
Do remember that the SCSI bus behaves like a transmission line. This is
something well known to engineers in the telecommunications industry. I feel
quite qualified to discuss the proper way of terminating things from the
theoretical standpoint as well as the SCSI-1 and SCSI-2 specifications
standpoint. *BOTH* *REQUIRE* the terminations on the ends of the bus and nowhere
else barring 2" stubs and *BOTH* *REQUIRE* proper 50 wire cables and connectors.
These requirements come from standard transmissionline theory as well as
practice.
The committee generating the various SCSI-3 specifications (instead of one
document it is now broken up into several) are working to address the
termination issues more clearly as the existing standards are insufficient for
many applications. If you have a SCSI bus made up of a full load of *FAST* SCSI
devices you also have to address spacing and other issues on the bus as each
device is seen as a capacitive impedance "bump" on the transmission line. If too
many of these "bumps" are too close together the resulting apparent big "bump"
would also cause glitches on the bus.
If a less than perfectly setup bus works for you fine. Go with it. But if you
are having trouble getting the bus working you'll ease your load by doing things
properly. Maybe I am a dumb old fogey. But I do have a great number of SCSI
systems of varying types with very fast devices and very diverse devices to my
credit and all are very stable for the various customers, friends, and myself.
Feel free to ignore my advice. I recommend you abide by it if you want an easier
job setting up your systems.
> Scott
>
> * Q-Blue 2.0 [NR] *
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
James R Proffitt
: In article <stimbe...@tbag.tscs.com>,
: sti...@tbag.tscs.com (Scott Timberlake) wrote:
: >Yeah. I don't ever remember seeing anyone disagree with her until she started
: >giving A3000 SCSI advice. Then people came out of the woodwork disagreeing
: >with her. Then it got really weird when she started spouting off about how
: >many decades of telecommunications experience she has. Glad to see I wasn't
: >the only one who wondered what the f!?# that had to do with A3000 SCSI
: >knowledge/experience. Just proves that NO ONE knows EVERYTHING. Now someone
: >just needs to convince JD of this...
: Do remember that the SCSI bus behaves like a transmission line. This is
: something well known to engineers in the telecommunications industry. I feel
: quite qualified to discuss the proper way of terminating things from the
: theoretical standpoint as well as the SCSI-1 and SCSI-2 specifications
: standpoint. *BOTH* *REQUIRE* the terminations on the ends of the bus and nowhere
: else barring 2" stubs and *BOTH* *REQUIRE* proper 50 wire cables and connectors.
: These requirements come from standard transmissionline theory as well as
: practice.
: The committee generating the various SCSI-3 specifications (instead of one
: document it is now broken up into several) are working to address the
: termination issues more clearly as the existing standards are insufficient for
: many applications. If you have a SCSI bus made up of a full load of *FAST* SCSI
: devices you also have to address spacing and other issues on the bus as each
: device is seen as a capacitive impedance "bump" on the transmission line. If too
: many of these "bumps" are too close together the resulting apparent big "bump"
: would also cause glitches on the bus.
And believe me, she DOES know what she is talking about. (I have her in
my "search by author" function.) I read all I can find of what she posts.
--
Bob Proffitt, KH6C Pearl City, Hawaii b...@pixi.com
Kevin Allan Donald Carter
MD> Just goes to show your ignorance in the matter (that you don't understand
MD> the correlation).
Why would you say this when her "35 years of transmission experience and 9
years advising people on BIX" have lead her to repeatedly give incorrect
advice. In the very next post she once again ignores her own advice by
recommending termination on both ends *and* the middle by failing to recognize
that the A3000 terminates SCSI on the motherboard. With SO much experience
that should have been obvious to her.
MD> Joanne just needs to let some people hang themselves.
By following her advice. At least as far as A3000 SCSI problems go.
MD> Ignorance is the only disease with a cure that nobody wants, but she keeps
MD> offering.
The disease, in this case. She generally knows what she's talking about
otherwise -- I don't know what her problem is with A3000s but I wish she'd
check it out for herself (since she obviously isn't going to take anyone elses
advice) and stop giving people the wrong information.
Kevin
Kevin Allan Donald Carter
JD> Good luck using the DB-25 cables etc. If you have fast drives you have a
JD> major headache brewing if you choose to go the DB25 route.
Ah yes. Manufacturers so enjoy customer complaints and warranty returns that
they regularly provide the wrong connections and what not. Look, Joanne, we
are talking about a ---> CD-ROM drive <--- which you recommended taking apart
to connect a 50 pin cable. Since when did CD-ROMs become considered high-speed
data devices? A DB25 cable should easily be able to handle the minimal data
transfer provided by even the fastest consumer CD-ROM drive.
JD> Well - one either busts the warranty or puts up with the SCSI bus locks I
JD> suppose. It depends on how often it happens and how irritating the locks
JD> are.
That isn't the problem!!!! The SCSI chain is terminated improperly. As the
great SCSI termination evangelist you should know that until the terminator is
removed from the motherboard (ie. the *middle* of the chain) and properly
terminated on the *ends*, the problem will persist, regardless of any cable
hacking you might want to do.
JD> 25 wire cables are a very clearly identified problem on a great number of
JD> systems with large, hence fast, drives in place. (Large here being 500meg
JD> and up.)
How exactly do you figure large equals fast? This is a CD-ROM drive. Even a 6x
CD-ROM drive is hardly considered fast in terms of data transfer.
JD> Do remember that the SCSI bus behaves like a transmission line. This is
JD> something well known to engineers in the telecommunications industry. I
JD> feel quite qualified to discuss the proper way of terminating things from
JD> the theoretical standpoint as well as the SCSI-1 and SCSI-2 specifications
JD> standpoint.
How about from an A3000-SCSI-is-terminated-on-the-motherboard and
needs-to-be-removed standpoint?
JD> *BOTH* *REQUIRE* the terminations on the ends of the bus and
JD> nowhere else
Ding ding ding. I believe you're catching on. You can't have SCSI terminators
in the middle of the chain or you'll lock up, hence the problem at hand. Do
try and pay attention.
JD> barring 2" stubs and *BOTH* *REQUIRE* proper 50 wire cables and
JD> connectors. These requirements come from standard transmissionline theory
JD> as well as practice.
Which fails to explain why for decade (more or less) external SCSI devices
have worked just fine with DB25. We're talking about a CD-ROM drive, not a
"SCSI-2 FAST" hard drive.
JD> The committee generating the various SCSI-3 specifications (instead of one
JD> document it is now broken up into several) are working to address the
JD> termination issues more clearly as the existing standards are insufficient
JD> for many applications. If you have a SCSI bus made up of a full load of
JD> *FAST* SCSI devices you also have to address spacing and other issues on
JD> the bus as each device is seen as a capacitive impedance "bump" on the
JD> transmission line. If too many of these "bumps" are too close together the
JD> resulting apparent big "bump" would also cause glitches on the bus.
Thanks for the lesson but since we're dealing with SCSI-1, incorrect
termination, and an external CD-ROM drive perhaps you can bring yourself back
to the point at hand.
JD> If a less than perfectly setup bus works for you fine. Go with it.
He doesn't, that's the problem. Do you even read posts before you reply?
JD> But if you are having trouble getting the bus working you'll ease your
JD> load by doing things properly.
Yes, so please stop giving incorrect advice. The problem lies solely with
incorrect termination on the motherboard of is A3000 preventing the use of
external SCSI devices through the DB-25 SCSI port. Removing said terminators,
provided the CD-ROM is terminated, will solve the problem.
JD> Maybe I am a dumb old fogey. But I do have a great number of SCSI
JD> systems of varying types with very fast devices and very diverse devices
JD> to my credit and all are very stable for the various customers, friends,
JD> and myself.
I didn't think so before this started but you're starting to convince me. With
all that experience, why can't you recognize the problem? It's both a simple
one and a common one.
JD> Feel free to ignore my advice. I recommend you abide by it if you want an
JD> easier job setting up your systems.
The only way this guy (or anyone with the classic A3000 SCSI lockup problem)
is going to solve his problem IS to ignore your advice. Because it's simply
wrong. What would be easier is if you would just check into it for yourself
(as you obviously aren't interested in taking anyone elses advice). I
sincerely hope you'll be embarassed once you discover the truth as previous to
this latest A3000 SCSI termination problem your advice had generally been on
the money.
Kevin
Kevin Allan Donald Carter
ST> It has one DB25 connector.
Then there's every likelihood it is terminated since there's no passthrough.
Most external CD-ROM drives (well, most SCSI devices period) are shipped with
terminators in place.
ST> No easy way for me to test this idea, since Asim 2.0 is what was given to
ST> me by the guy I bought the drive from. Anyways, I'm 99.99% sure that it's
ST> not a software problem, as I'm sure everyone else is.
Asim is very popular (probably the most widely used, too) so I should think
you're in good shape in that area.
Do post once you get it working. I think there's a few people who need a
strong dose of reality here.
Kevin
Joanne Dow
>MD> Just goes to show your ignorance in the matter (that you don't understand
>MD> the correlation).
>
>Why would you say this when her "35 years of transmission experience and 9
>years advising people on BIX" have lead her to repeatedly give incorrect
>advice. In the very next post she once again ignores her own advice by
>recommending termination on both ends *and* the middle by failing to recognize
>that the A3000 terminates SCSI on the motherboard. With SO much experience
>that should have been obvious to her.
>
*VERY FEW* A3000s were delivered with terminations on the motherboard. If there
is a termination there it should be removed. (And FYI no, repeat no, A3000Ts
were ever delivered with onboard terminators.)
>MD> Joanne just needs to let some people hang themselves.
>
>By following her advice. At least as far as A3000 SCSI problems go.
>
>MD> Ignorance is the only disease with a cure that nobody wants, but she keeps
>MD> offering.
>
>The disease, in this case. She generally knows what she's talking about
>otherwise -- I don't know what her problem is with A3000s but I wish she'd
>check it out for herself (since she obviously isn't going to take anyone elses
>advice) and stop giving people the wrong information.
>
>
>
>Kevin
{^_^} Joanne Dow, Amiga Exchange Editor on BIX
jd...@bix.com The Wizardess
Joanne Dow
ka...@julian.slip.uwo.ca (Kevin Allan Donald Carter) wrote:
>JD> Good luck using the DB-25 cables etc. If you have fast drives you have a
>JD> major headache brewing if you choose to go the DB25 route.
>
>Ah yes. Manufacturers so enjoy customer complaints and warranty returns that
>they regularly provide the wrong connections and what not. Look, Joanne, we
>are talking about a ---> CD-ROM drive <--- which you recommended taking apart
>to connect a 50 pin cable. Since when did CD-ROMs become considered high-speed
>data devices? A DB25 cable should easily be able to handle the minimal data
>transfer provided by even the fastest consumer CD-ROM drive.
>
Hey, dweeb, the cdroms DO have high speed SCSI chips on them. That is all you
can buy from the likes of Cirrus Logic, NCR, etc. Further if you have *ANY* high
speed devices on the bus, and the WD chip qualifies, you are courting troubles.
>JD> Well - one either busts the warranty or puts up with the SCSI bus locks I
>JD> suppose. It depends on how often it happens and how irritating the locks
>JD> are.
>
>That isn't the problem!!!! The SCSI chain is terminated improperly. As the
>great SCSI termination evangelist you should know that until the terminator is
>removed from the motherboard (ie. the *middle* of the chain) and properly
>terminated on the *ends*, the problem will persist, regardless of any cable
>hacking you might want to do.
>
Hey, dweeb, how many A3000s have you seen delivered with onboard terminators? On
most machines they simply are not there.
>JD> 25 wire cables are a very clearly identified problem on a great number of
>JD> systems with large, hence fast, drives in place. (Large here being 500meg
>JD> and up.)
>
>How exactly do you figure large equals fast? This is a CD-ROM drive. Even a 6x
>CD-ROM drive is hardly considered fast in terms of data transfer.
>
Larger drives are more likely to be modern designs with modern SCSI chips on
them. These chips themseives are fast enough to respond to teh glitches inserted
on the bus from bad termiantors.
>JD> Do remember that the SCSI bus behaves like a transmission line. This is
>JD> something well known to engineers in the telecommunications industry. I
>JD> feel quite qualified to discuss the proper way of terminating things from
>JD> the theoretical standpoint as well as the SCSI-1 and SCSI-2 specifications
>JD> standpoint.
>
>How about from an A3000-SCSI-is-terminated-on-the-motherboard and
>needs-to-be-removed standpoint?
>
I have *NEVER* said to leave it there.
>JD> *BOTH* *REQUIRE* the terminations on the ends of the bus and
>JD> nowhere else
>
>Ding ding ding. I believe you're catching on. You can't have SCSI terminators
>in the middle of the chain or you'll lock up, hence the problem at hand. Do
>try and pay attention.
Catching on? This is what I have been preaching for over 9 years now on Amigas.
I can't help it if you cannot read.
>
>JD> barring 2" stubs and *BOTH* *REQUIRE* proper 50 wire cables and
>JD> connectors. These requirements come from standard transmissionline theory
>JD> as well as practice.
>
>Which fails to explain why for decade (more or less) external SCSI devices
>have worked just fine with DB25. We're talking about a CD-ROM drive, not a
>"SCSI-2 FAST" hard drive.
>
Older SCSI chips did not incorporate logic elements as fast as the logic
elements in modern SCSI chips. The modern ones are more sensitive to glitches.
>JD> The committee generating the various SCSI-3 specifications (instead of one
>JD> document it is now broken up into several) are working to address the
>JD> termination issues more clearly as the existing standards are insufficient
>JD> for many applications. If you have a SCSI bus made up of a full load of
>JD> *FAST* SCSI devices you also have to address spacing and other issues on
>JD> the bus as each device is seen as a capacitive impedance "bump" on the
>JD> transmission line. If too many of these "bumps" are too close together the
>JD> resulting apparent big "bump" would also cause glitches on the bus.
>
>Thanks for the lesson but since we're dealing with SCSI-1, incorrect
>termination, and an external CD-ROM drive perhaps you can bring yourself back
>to the point at hand.
>
What makes you think SCSI-1 is different in any salient manner from what exists
in SCSI-2 in regards termination or cabling? *NEITHER* SCSI-1, SCSI-2, nor any
SCSI-3 specification permits 25 wire cable. They *ALL* absolutely require
terminations at the two ends of the bus and nowhere else. Perhaps you should go
get copies of the specs and read, if you can.
>JD> If a less than perfectly setup bus works for you fine. Go with it.
>
>He doesn't, that's the problem. Do you even read posts before you reply?
>
I read 'em better than you.
>JD> But if you are having trouble getting the bus working you'll ease your
>JD> load by doing things properly.
>
>Yes, so please stop giving incorrect advice. The problem lies solely with
>incorrect termination on the motherboard of is A3000 preventing the use of
>external SCSI devices through the DB-25 SCSI port. Removing said terminators,
>provided the CD-ROM is terminated, will solve the problem.
>
Batguanno.
>JD> Maybe I am a dumb old fogey. But I do have a great number of SCSI
>JD> systems of varying types with very fast devices and very diverse devices
>JD> to my credit and all are very stable for the various customers, friends,
>JD> and myself.
>
>I didn't think so before this started but you're starting to convince me. With
>all that experience, why can't you recognize the problem? It's both a simple
>one and a common one.
>
>JD> Feel free to ignore my advice. I recommend you abide by it if you want an
>JD> easier job setting up your systems.
>
>The only way this guy (or anyone with the classic A3000 SCSI lockup problem)
>is going to solve his problem IS to ignore your advice. Because it's simply
>wrong. What would be easier is if you would just check into it for yourself
>(as you obviously aren't interested in taking anyone elses advice). I
>sincerely hope you'll be embarassed once you discover the truth as previous to
>this latest A3000 SCSI termination problem your advice had generally been on
>the money.
>
Kevin, I haven't the foggiest idea where you are getting your impressions of
what I have been telling people to do for quite some time now. All I can figure
is that you cannot read. Since your language towards me was so inflammitory I
have elected to answer you in kind. Maybe THAT is something you can understand.
A/B Roll
: advice. In the very next post she once again ignores her own advice by
: recommending termination on both ends *and* the middle by failing to recognize
: that the A3000 terminates SCSI on the motherboard. With SO much experience
: that should have been obvious to her.
The A3000 is *designed* to be terminated at the motherboard... what her
years of experience tells her is that most A3000's do not have
motherboard terminators installed because C= decided that they could get
away without them, right? Even with my limited (i.e. a B.Sc. in
Physics and 5 years in television) experience in transmission lines I can
tell you that you must terminate each end and not the middle which for
most A3000s means an external terminator and termination on the last
internal drive.
A/B
+----------------------------+------------------------------------------+
|Anthony Brown (g3...@unb.ca) | Make thirds, not war. - Fabian Stedman |
|Applied Optics Group | |
|Dept. of Physics | Campanalogio ergo sum: I ring, therefore |
|University of New Brunswick | I am |
+----------------------------+------------------------------------------+
Maxwell Daymon
: CD-ROM drive is hardly considered fast in terms of data transfer.
You are showing your ignorance _yet_again_
The SCSI *interface* is what controls the speed (yes, the buffer transfers
a hell of a lot faster than the CD-ROM drive) and the PARTS are what is
sensitive, regardless of the speed they are acutally being pushed to.
Further, very VERY few Amiga 3000 systems actually have the terminators
installed on the motherboard, and NO recent models came with the
terminators on the motherboard. Hence, it is NOT likely that the
termination problem is occuring there.
: How about from an A3000-SCSI-is-terminated-on-the-motherboard and
: needs-to-be-removed standpoint?
See above.
Jim Pritchett
> Interestingly, they did socket the terminators on the Combo accelerator
> with SCSI controller. Curious!
Not on mine...
Phil Wright
: strong dose of reality here.
Why? You seem to be doing so well off in your fantasy world.
What a dweeb.
Kevin Allan Donald Carter
JD> *VERY FEW* A3000s were delivered with terminations on the motherboard. If
JD> there is a termination there it should be removed. (And FYI no, repeat no,
JD> A3000Ts were ever delivered with onboard terminators.)
If "*VERY FEW*" equals most then you're right. ;) At least you finally
acknowledge the problem. It only took you a month and a dozen posts.
(And FYI I know about A3000Ts... after all, it was I who pointed it out to
*you* in the first place! Sheesh, anything to try and save face, eh?)
Kevin
Kevin Allan Donald Carter
AR> The A3000 is *designed* to be terminated at the motherboard... what her
AR> years of experience tells her is that most A3000's do not have
AR> motherboard terminators installed because C= decided that they could get
AR> away without them, right? Even with my limited (i.e. a B.Sc. in
AR> Physics and 5 years in television) experience in transmission lines I can
AR> tell you that you must terminate each end and not the middle which for
AR> most A3000s means an external terminator and termination on the last
AR> internal drive.
You hardly need to tell me what I've now repeated over a dozen times.
Regardless of the number of A3000s which *might* not have the terminators on
the motherboard, in the posts under discussion this is clearly not the case
and they *do* have terminators which must be *removed* so as to provide proper
SCSI termination on the *ends* of the SCSI chain *only*.
Kevin
Kevin Allan Donald Carter
MD> You are showing your ignorance _yet_again_
No, Maxwell, as usualy I leave ignorance to you since you're SO good at it.
MD> The SCSI *interface* is what controls the speed (yes, the buffer transfers
MD> a hell of a lot faster than the CD-ROM drive) and the PARTS are what is
MD> sensitive, regardless of the speed they are acutally being pushed to.
Funny then how hundreds of thousands of users and manufacturers continue to
use DB25 SCSI connections without fail.
MD> Further, very VERY few Amiga 3000 systems actually have the terminators
MD> installed on the motherboard, and NO recent models came with the
MD> terminators on the motherboard. Hence, it is NOT likely that the
MD> termination problem is occuring there.
Since these people are clearly reporting classic A3000 SCSI termination
problems (ie. they are "correctly" terminated (if it weren't for the
motherboard terminators) at both ends of the chain, have no conflicting ID
numbers, are correctly cabled, yet still experience SCSI lockup) they clearly
*DO* have SCSI terminators on the motherboard. Hence it *IS* likely that the
termination problem is occuring there.
Kevin
Kevin Allan Donald Carter
JD> you can buy from the likes of Cirrus Logic, NCR, etc. Further if you have
JD> *ANY* high speed devices on the bus, and the WD chip qualifies, you are
JD> courting troubles.
Oh no, Joanne hurt my feelings. :( Ha! Grow up, grandma (since we're
exchanging pleasantries). Yes they have SCSI 2 and even SCSI 2 Fast -- they
still has to deal with SCSI 1 transfer rates as they're dealing with a stock
A3000 SCSI controller. Nevertheless, if using *industry standard* DB25 SCSI
connections is sooooo bad, that'll come as BIG news to litterally hundreds of
thousands of people who have used it successfully for nearly a decade and are
using it still on *new* drives which are currently *manufactured* to use DB25
connectors.
JD> Hey, dweeb, how many A3000s have you seen delivered with onboard
JD> terminators? On most machines they simply are not there.
Hey, grandma, how about all of them? Most A3000's had terminators on the
motherboard as far as I've personally seen (or know of second hand). The only
exception being A3000Ts. I've heard rumours of some A3000s not having
terminators but having seen so many with them, I've written that off as people
getting lucky with their external devices. (How do they function with multiple
drives internally and for so long without proper termination, as you so often
point out?) At least you admit that, at the very least, some A3000s did ship
with terminators on the motherboard.
JD> Larger drives are more likely to be modern designs with modern SCSI chips
JD> on them. These chips themseives are fast enough to respond to teh glitches
JD> inserted on the bus from bad termiantors.
Indeed, which is why it is even more important to get the termination correct,
which I what *I've* been trying to do.
JD> I have *NEVER* said to leave it there.
You have said, repeatedly, to terminate the external SCSI port on stock A3000s
never once mentioning the need to remove the terminators on the motherboard
(at LEAST on certain A3000s, and *certainly* in these cases). If it were a
simple omission on your part that wouldn't be a problem, but you kept
repeating the exact same advice over and over and over and over. That goes
beyond simple negligence. Most importantly, in the cases we have been
discussing, regardless of if some A3000s had shipped without terminators on
the motherboard, the problem has clearly been the classic A3000 SCSI lockup
problem when there *is* motherboard termination that needs to be removed.
KADC> Ding ding ding. I believe you're catching on. You can't have SCSI
KADC> terminators in the middle of the chain or you'll lock up, hence the
KADC> problem at hand. Do try and pay attention.
JD> Catching on? This is what I have been preaching for over 9 years now on
JD> Amigas. I can't help it if you cannot read.
Um, put on your spectacles, grandma. You're the one who's been advising people
on incorrect termination. I've been trying to straighten out the half-complete
(and hense bad) advice you've been giving.
JD> Older SCSI chips did not incorporate logic elements as fast as the logic
JD> elements in modern SCSI chips. The modern ones are more sensitive to
JD> glitches.
No one is debating that, nor are you the first to point it out. It does,
however, make correct termination even more of an issue. Can't you at least
admit that you may have "overlooked" the fact that A3000s came with SCSI
termination on the motherboard that needs to be removed in order to use
external SCSI devices?
JD> What makes you think SCSI-1 is different in any salient manner from what
JD> exists in SCSI-2 in regards termination or cabling? *NEITHER* SCSI-1,
JD> SCSI-2, nor any SCSI-3 specification permits 25 wire cable. They *ALL*
JD> absolutely require terminations at the two ends of the bus and nowhere
JD> else. Perhaps you should go get copies of the specs and read, if you can.
Hello? It wasn't me giving people incorrect termination advice. *I've* been
busy pointing out that you have to remove the terminator now in the *middle*
of the SCSI chain so that there is terminators *only at both the ends of the
chain*. Turn up your hearing aid, gradma, and I'll read the specs to you the
next time I visit the old age home.
JD> I read 'em better than you.
You should. You've posted the same bogus advice enough times to have it
memorized by now.
JD> Batguanno.
We'll know soon enough.
JD> Kevin, I haven't the foggiest idea where you are getting your impressions
JD> of what I have been telling people to do for quite some time now. All I
JD> can figure is that you cannot read. Since your language towards me was so
JD> inflammitory I have elected to answer you in kind. Maybe THAT is something
JD> you can understand.
Well, Joanne, after reading reply after reply from you I can certainly see
where you might be confused. I've certainly come to the same conclusions with
regard to your comprehension skills. Similarly, you memory seems to be fading
if you believe I have responded to you in anything other than the tone *you*
set in your multiple backhanded comments about me which started this whole
mess.
This "debate" (though clearly I'm the only one addressing the issue) revolves
around one simple fact; that being your refusal to acknowledge (until now,
though you went on and on afterwards in an attempt to further confuse the
issue) that A3000s (at the very least the A3000s in question) have SCSI
terminators on the motherboard which must be removed in order to provide
proper termination when adding external SCSI devices through the DB25 port.
Your desire to void warranties rather than use the provided DB25 SCSI cables
is a point we shall apparently never resolve as the theory does not match
practical application.
Furthermore, as long as you continue to put down every non-BIX user, even
those not asking advice, this "wasted, worthless netizen" as you have called
us, shall contiue to hold a similar opinion of you.
Sincerely,
Kevin A. D. Carter
Kevin Allan Donald Carter
PW> Kevin Allan Donald Carter (ka...@julian.slip.uwo.ca) wrote:
PW> : Do post once you get it working. I think there's a few people who need a
PW> : strong dose of reality here.
PW> Why? You seem to be doing so well off in your fantasy world.
Ew! Brilliant rebuttal! Of course, why should the original poster bother
telling us what the real problem turned out to be? No, far better to let the
issue hang indefinately.
Now, since you weren't involved in the "debate", I fail to see why you would
be afraid to discover the truth of the matter. I am fully prepaired to be
proven wrong in my analysis of the problem. At least my solution causes no
damage to components, voids no warranties, and insures proper termination.
PW> What a dweeb.
Wow! Another snappy retort! Not exactly original but hey, a guy like you only
has so many brain cells to go around. Back to the playground, junior. Come
back when you have something original to say.
Kevin
Joanne Dow
Kevin, I have not seen a single A3000 with stock internal terminators on the
motherboard other than possibly one I saw at the prerelease mini-devcon at which
the A3000 was revealed to the developer community. This was brought up on BIX
and my understanding from people at Commodore who should know manufacturing was
instructed to leave off those terminators after the first manufacturing run.
It was realized that external expansion would be added and the motherboard
termination would have to be removed. This is not an easy operation. They
discovered that they could "survive" with one terminator if the bus inside was
short enough. They relied on the external drive's termination for the second
terminator, as is proper from that standpoint. I understand that the very last
A3000s delivered had the DB-25 SCSI terminators included with them just as the
a3000T comes with that terminator. I have never seen such so I cannot declare
that as a fact. Do note that the machines I know are US delivered and configured
machines. In Europe the situation may have been different. (If so I'd like to
know this for a fact.)
In re your remarks about the 25 wire cables and DB-25 connector - if it works
for you do not fix it. If you have configuration problems look to it as a
potential (likely) source of your problems. If you think it is not a likely or
possible source of problems because it is "industry standard" (note it is *NOT*
a standard by the formal ANSI/EIA SCSI specifications at any level) I firmly
believe you have blinders on. That's your problem. If you try to get naive
people on the net here to adopt your "wisdom" that is the problem of the rest of
us on the net with more experience. This is why I have been argueing this with
you. As it stands my points have been made. You can continue to bray shrilly in
the breezes. Anyone crazy enough to adopt your philosophy rather than a
conservative approach via the SCSI specifications diserves their problems. They
have other sources of better information they chose to ignore. That ain't my
fault or responsibility.
realwombat on BIX
>If "*VERY FEW*" equals most then you're right. ;) At least you finally
>acknowledge the problem. It only took you a month and a dozen posts.
>(And FYI I know about A3000Ts... after all, it was I who pointed it out to
>*you* in the first place! Sheesh, anything to try and save face, eh?)
Sheesh, Kevin. Everything she's ever posted on SCSI
and Amigas has proven out true. The first post I ever saw
from you (here) had obvious false information and inconsistencies,
aside from you arguing by claiming she said the opposite of
what she actually said. <shrug>
realwombat on BIX
Hmmm. By listening to everything I've ever heard (seen)
Joanne Dow post about SCSI buses/drives and Amiga Computers AND
following her advice, I've been able to setup three Amigas with
eight different SCSI devices (internal and external) with a
grand total of ZERO problems. Can't go wrong there. :)
Alva B. Powell III
Just so everyone can know/find out.
I recently (about 6 months ago) had many scsi problems. I had just added a
new external SCSI device. That gave my system a total of four devices (2
internal, 2 external). Then I started having major boot problems.
Prior to that I had 2 quantums internal and a syquest 105 external, the last
quantum was terminated and the syquest was terminated. System ran fine for
about 18 months (Not on lock up!).
Well I was reading posts in c.s.a.h one day an saw a post that mentioned the
terminators on the mother board (after I started having all the problems).
I thought (OH DUMB ME!! Of course the mother board has terminators how else
could the internal drive work. I didn't know you could get away with one
terminator on a short cable). So I opened the manual (got to find the silly
things before one can remove them), not one reference to them... OK open
A3000 box (once again), looked around for something that resembled them. I
found something that I thought might be the terminator on the motherboard
but alas they were soldered (sp) in. Well OK, pull all internal devices and
move them external. All problem vanish (just about, I then started getting
this little pop up box about every 30 minutes telling me I had a read or
write error, no big deal I figured it was my 14 foot SCSI cable
db25 -> 50 centronics (6ft) -> 8ft ribbon cable in tower winding beween
devices).
No another boot up problem in several months. Then I try to install Linux
68K. OOPS!!! boot after boot failure... Someone suggest removing the syquest
still got problems!! OK, maybe some how its my 14 foot scsi cable, maybe it
is a termination problem.
I sent e-mail to Joanne Dow, to find out where the motherboard terminators
would be, if I had them. Sure enough right where I thought they were
originally, just behind the scsi connector on the mother board (THANK YOU,
THANK YOU, THANK YOU JOANNE!!!!) They seem to be in rp802 and rp804 (only
two on my box Joanne).
So I my termination is ok for linux. Since I had my drive tower ripped
apart at this time I checked all the drive parity, one quantum had PE jumper
set. Remove that rebuild all the hardware. Everything is GREAT! Linux 68K
boots. No more little reminder boxes jumping up about a read error on the
disk... Lovely!!!
So the bottom line is:
1) my A3000 (circa 12/90, yeow! I didn't think it that old!) has SCSI
terminators on the motherboard.
2) I ran fine with three terminators for about 18 months (2 internal
quantum drives 50 and 40, 1 external syquest 105). 40 MB quantum
had the PE jumper set the whole time.
3) I ran fine with 4 external devices 25DB->50Centronics (6ft) ->8 ft
of ribbon in a tower case (some minor errors because of the PE jumper).
4) Now I have a 50pin cable plugged into the mother board->50Centronics
woven cable -> 8ft ribbon in tower case (not a problem PE jumpers
removed on all drives.) Linux 68K and Amiga Dos work find.
I suggest to all with hook up problems - Check you mother board for
terminators, it only takes about 30 minutes and a screw driver (use
parental supervision ;-).
I still cannot get my maxtor MXT-540S (made for IBM) to run. When ever I
place it on the scsi chain the whole system hangs up as soon as it tries
to ID the drive.
OH, well it works fine in an RS-6000....
Still not really sure about the parity on the Amiga SCSI (I guess just leave
the jumpers off...)
In Christ,
Alva B. Powell III al...@free.org
A mind is a terrible thing al...@rogue.com
to waste... Think when alv...@netcom.com
you read the Bible... II Tim 3:16&17
Scott Timberlake
MD> installed on the motherboard, and NO recent models came with the
MD> terminators on the motherboard. Hence, it is NOT likely that the
MD> termination problem is occuring there.
Okay everybody, I see a lot has been discussed in my absence. As usual,
the BBS I have to use was down for 5 days or so, that's why this message
has taken so long. I ripped apart my 3000 to take off the MB terminators,
only to discover that it wasn't quite as obvious as I thought it would
be as to where they are! Turns out there are several things that look
like they might be it, so I didn't touch anything. Just for the hell of
it though, since I was in there, I removed the terminators from the HD.
As I figured, it didn't help, so I put them back.
My best guess is that the terminators are just below the SCSI chip.
There is what looks like a socket for a terminator, since it has holes
in it. Just under that is something that looks like a terminator. I'm
guessing that the terminator is not installed, which would not be good
since that is what I was hoping was the problem.
I went ahead and sent money to Bob Krushinski for a SCSI chip. I hope
to hell that it does the trick, because when my Zip drive finally gets
here at the end of the month I know that it will just make things worse
and not better. I'll post what happens when I get the chip. But in the
meantime, could someone tell me exactly where the MB terminator is?
Joern Plewka
(02.09.1995)
> In article <19950831.7...@julian.slip.uwo.ca>,
> ka...@julian.slip.uwo.ca (Kevin Allan Donald Carter) wrote:
> >On Mon 28-Aug-1995 10:22a, Joanne Dow wrote:
> >JD> *VERY FEW* A3000s were delivered with terminations on the motherboard. If
> >JD> there is a termination there it should be removed. (And FYI no, repeat no,
> >JD> A3000Ts were ever delivered with onboard terminators.)
> >
> >acknowledge the problem. It only took you a month and a dozen posts.
> >(And FYI I know about A3000Ts... after all, it was I who pointed it out to
> >*you* in the first place! Sheesh, anything to try and save face, eh?)
> >
> >
> >
>
> I understand that the very last
> A3000s delivered had the DB-25 SCSI terminators included with them just as the
> a3000T comes with that terminator.
these arrays placed in the sockets!
The A3000T doesn`t work without the terminator...it`s a little different.
(None of the 3 ones I saw worked without the terminators.)
Everybody (with problems) should check his terminators;-)
/ J_Pl...@Amtrash.comlink.de \
| J_Pl...@People-S.people.de |
/ Plew...@Informatik.fh-hamburg.de \
+-------------------------------------+
\ A3000/030/6/52 A3000T/040/18/800 /
Joern Plewka
(31.08.1995)
> On Mon 28-Aug-1995 10:22a, Joanne Dow wrote:
> JD> Hey, dweeb, the cdroms DO have high speed SCSI chips on them. That is all
> JD> you can buy from the likes of Cirrus Logic, NCR, etc. Further if you have
> JD> *ANY* high speed devices on the bus, and the WD chip qualifies, you are
> JD> courting troubles.
>
> still has to deal with SCSI 1 transfer rates as they're dealing with a stock
> A3000 SCSI controller.
6 meters with 10Mhz...
> Nevertheless, if using *industry standard* DB25 SCSI
> connections is sooooo bad, that'll come as BIG news to litterally hundreds of
> thousands of people who have used it successfully for nearly a decade and are
> using it still on *new* drives which are currently *manufactured* to use DB25
> connectors.
And it`s not a good choice if companies are trying not to waste material;-)
Latest cables I`ve seen (and destroyed due malfunktion) didn`t use
25 pairs (twisted? no way!). They are using a f*cking printercable with
SCSI plug. I`ve got a ScanjetIIcx and tried to make it work with any cheap
non-HP cable. It really was impossible. OK, the scanner worked but
wangdat and Syquest disappeard from the INTERNAL bus or replied defective
characters while using SCSI tools!!!
Even aktive termination wasn`t able to fix it. We put a nice oscilloscope to
the bus, replaced the WD to 93C33B-04 and increased it`s clock to 20Mhz.
Shiiiittt.
Now the bus of our machines is "extremly clean" (with real cable) but much
work and money...
If you want to get good 25pin cable, do it on your own!
OK, Syquest270 Rev1/2.xx is a very bad reference but it works with real cable.
PS: None of the 3 A3000 in my room had had resistors...none of the other
5 machines had had terminators. Revisions 7.x up to 9.x;-)
John Crookshank
> In article <19950831.7...@julian.slip.uwo.ca>,
> ka...@julian.slip.uwo.ca (Kevin Allan Donald Carter) wrote:
>>On Mon 28-Aug-1995 10:22a, Joanne Dow wrote:
>>JD> *VERY FEW* A3000s were delivered with terminations on the
>>motherboard. If JD> there is a termination there it should be removed.
>>(And FYI no, repeat no, JD> A3000Ts were ever delivered with onboard
>>terminators.)
>>
>>If "*VERY FEW*" equals most then you're right. ;) At least you finally
>>acknowledge the problem. It only took you a month and a dozen posts.
>>(And FYI I know about A3000Ts... after all, it was I who pointed it out
>>to
>>*you* in the first place! Sheesh, anything to try and save face, eh?)
No, I can back up Joanne here. I work at our area's largest service center,
and I have seen just a mere handful of A3000's that had terminators
installed on the motherboard, and those were only during the very first
production run. And I've never seen a 3000T with motherboard terminators.
Every one of them shipped with an external terminating plug.
As an aside, to get around this internal/external termination dilemma, GVP
shipped EVERY single one of their "card+drive" packages with terminators
installed on the card, and none on the internal drive, which had a 2-inch
cable. This way, you could usually just plug in an external drive and "get
away" with it, even if it was not "standard". The problems would always
arise then the user added a second or third internal drive, plus wanted to
have an external drive. Then we would have to go and de-solder the
terminators from the card. They wised up later and put the host adapter
terminators in sockets so you could just pull 'em when needed.
Personally, I wish more manufacturers would have done it like the Commodore
A4091 - with a DIP switch on the back panel of the card, so you could just
flip the DIP switch when you wanted to add an external device, and flip it
back when you didn't. No opening the case, no "bending" of the SCSI rules,
no external plugs needed, etc.
______________________________________________________________
/ John Crookshank | Raptor3 DEC Alpha Workstations \
| MicroTech Solutions, Inc. | Lightwave 3D, Toaster/Flyer |
| Desktop Video Systems Dealer | JVC Professional Video Products |
|------------------------------|---------------------------------|
| jo...@mcs.net http://www.mt-inc.com/MT.html |
\______________________________________________________________/
Michael van Elst
>A3000 SCSI controller. Nevertheless, if using *industry standard* DB25 SCSI
>connections is sooooo bad, that'll come as BIG news to litterally hundreds of
>thousands of people who have used it successfully for nearly a decade
They might successfully use it for a few limited configurations. Anyone
who uses DB25 SCSI connections for a larger setup does run into
problems.
>and are
>using it still on *new* drives which are currently *manufactured* to use DB25
>connectors.
Really no "drive" uses a DB25 connector. You might get a DB25 connector
for an external box though. The large majority of external SCSI boxes
however uses Centronics-style 50pin connectors or the SCSI-2 high
density connectors. And of course cables with 50 wires.
>Hey, grandma, how about all of them? Most A3000's had terminators on the
>motherboard as far as I've personally seen (or know of second hand).
You probably haven't seen A3000s that were produced later. The first
A3000s (including mine) had termination resistor packs installed. A3000s
sold later had the terminators removed.
>The only
>exception being A3000Ts.
A3000Ts didn't have terminators on the motherboard but an external
terminator for the DB25 connector.
>I've heard rumours of some A3000s not having
>terminators but having seen so many with them, I've written that off as people
>getting lucky with their external devices.
And I have seen more A3000s that came without terminators than those
that came with terminators on the motherboard. So much for personal
statistics.
>(How do they function with multiple
>drives internally and for so long without proper termination, as you so often
>point out?)
Termination resistors are necessary for two reasons: they provide
default levels to all signals (and one terminator is sufficient there)
and they reduce signal reflections (and with a very short cable this
isn't needed).
>the motherboard, the problem has clearly been the classic A3000 SCSI lockup
>problem when there *is* motherboard termination that needs to be removed.
The "classic" A3000 SCSI lockup problem is a software problem.
>Hello? It wasn't me giving people incorrect termination advice. *I've* been
>busy pointing out that you have to remove the terminator now in the *middle*
>of the SCSI chain so that there is terminators *only at both the ends of the
>chain*.
You should also consider the situation where the motherboard terminators
are the _only_ termination (which is equivalent to an internal drive
being terminator and no motherboard terminators). If the internal cable
is short enough you can leave the termination as is.
>Your desire to void warranties rather than use the provided DB25 SCSI cables
>is a point we shall apparently never resolve as the theory does not match
>practical application.
I wonder why you are so concerned about termination "to the specs" when
you don't refuse to use DB25 cables that aren't "to the specs" either.
Regards,
--
Michael van Elst
Internet: mle...@serpens.rhein.de
"A potential Snark may lurk in every tree."
decu...@fn1.freenet.edmonton.ab.ca
: On Mon 28-Aug-1995 10:22a, Joanne Dow wrote:
: JD> *VERY FEW* A3000s were delivered with terminations on the motherboard. If
: JD> there is a termination there it should be removed. (And FYI no, repeat no,
: JD> A3000Ts were ever delivered with onboard terminators.)
: If "*VERY FEW*" equals most then you're right. ;) At least you finally
: acknowledge the problem. It only took you a month and a dozen posts.
: (And FYI I know about A3000Ts... after all, it was I who pointed it out to
: *you* in the first place! Sheesh, anything to try and save face, eh?)
: Kevin
Well, considering that the 3000 desktop service manual does not even list
the terminator packs in the parts list, and the 3000T manual specifically
states that they are UNSTUFFED (unlike yourself, who seems to be completely
"full of it"), I would say that *VERY FEW* is quite accurate! I have yet
to see one that has them installed, but of course, it never hurts to
check just in case the dealer or someone else put them in.
Did you ever consider that the reason everyone's disagreeing with you
might be because you're (perish the thought), WRONG!
Cheers,
--
email: decu...@freenet.edmonton.ab.ca
+ _ ____________ tm Dale Currie ____ ___ _ +
| /.\ .. | __ \ / alias "ZZORRO", / __[___]__ T tm|
| /___\ /\/\ | | |_) | X Borg Systems Analyst / (o.o) | |
| / \/ ^^ \ | | | \ | / \ & AmiTrix / `-^-' | |
|/ - D E V E L O P M E N T - \ Technical Support /___ Z O R R O I N K !|
| --------------------- ---------------- |
+ ("Zorro lives!" -++- "He's been assimilated as part of my Amiga!") +
decu...@fn1.freenet.edmonton.ab.ca
: Funny then how hundreds of thousands of users and manufacturers continue to
: use DB25 SCSI connections without fail.
And a lot of them that bought the cheaper 25 wire cables DO have problems!
I have 3 of them hanging here that are useless because they are not
reliable, whereas the heavier 50 wire versions do work most of the time,
but even they are not imune to problems when multiple &/or high speed drives
are connected to them.
: MD> Further, very VERY few Amiga 3000 systems actually have the terminators
: MD> installed on the motherboard, and NO recent models came with the
: MD> terminators on the motherboard. Hence, it is NOT likely that the
: MD> termination problem is occuring there.
: Since these people are clearly reporting classic A3000 SCSI termination
: problems (ie. they are "correctly" terminated (if it weren't for the
: motherboard terminators) at both ends of the chain, have no conflicting ID
: numbers, are correctly cabled, yet still experience SCSI lockup) they clearly
: *DO* have SCSI terminators on the motherboard. Hence it *IS* likely that the
: termination problem is occuring there.
Possible, but not necessarily so! "Correctly terminated" is at the whim
of the devices connected, their physical order on the bus, and the
quality of the components, cables, and connectors; NOT the eye of the
beholder!
Jim Pritchett
> Really no "drive" uses a DB25 connector. You might get a DB25 connector
The Zip drive has two DB25 connectors, unfortunately.
Jim Pritchett
UUCP: j...@caleb.uucp "I am the way, the truth, and
rwsys.lonestar.org!caleb!jdp the life; no man cometh to the
caleb!j...@lerami.lerctr.org Father, but by me."
Jesus Christ
decu...@fn1.freenet.edmonton.ab.ca
: Okay everybody, I see a lot has been discussed in my absence. As usual,
: the BBS I have to use was down for 5 days or so, that's why this message
: has taken so long. I ripped apart my 3000 to take off the MB terminators,
: only to discover that it wasn't quite as obvious as I thought it would
: be as to where they are! Turns out there are several things that look
: like they might be it, so I didn't touch anything. Just for the hell of
: it though, since I was in there, I removed the terminators from the HD.
: As I figured, it didn't help, so I put them back.
: My best guess is that the terminators are just below the SCSI chip.
: There is what looks like a socket for a terminator, since it has holes
: in it. Just under that is something that looks like a terminator. I'm
: guessing that the terminator is not installed, which would not be good
: since that is what I was hoping was the problem.
: I went ahead and sent money to Bob Krushinski for a SCSI chip. I hope
: to hell that it does the trick, because when my Zip drive finally gets
: here at the end of the month I know that it will just make things worse
: and not better. I'll post what happens when I get the chip. But in the
: meantime, could someone tell me exactly where the MB terminator is?
: Scott
They are: RP802 (just behind the DMAC), RP803 & RP804 (just behind the
internal SCSI header), the only socketed 8-pin SIPPs on the board, and
probably empty as most have speculated! (but we wait with baited breath
for the final verdict) ;-)
Scott Timberlake
De> internal SCSI header), the only socketed 8-pin SIPPs on the board, and
De> probably empty as most have speculated! (but we wait with baited breath
De> for the final verdict) ;-)
ALL of them have to be removed, right? As I recall, when I was looking
at the MB, there was only ONE missing. Strange, huh?