[comp.sys.dec] PDP 11/24 VS 11/44

jar@konkord.uucp (John Roberts) (09/26/90)

	PDP users of the world:

		Does anyone know which is the faster/better machine?
		(The 11/24 or the 11/44)?

	Is it possible to take out all the 11/44 boards and replace them
	with the 11/24 boards? Are the backplans compatable?

	What does it mean when the DC ON light blinks and its NOT the
	power suppy. Why does the DC ON light blink when grant cards
	are not there/or not in the slot?

	Sorry for all the questions...


				Thanks

-- 
John A. Roberts				(508)460-4646 X197
Concord Communications, Inc.			uunet!konkord!jar
753 Forest Street					or
Marlboro, MA 01752, USA				jar%konkord@uunet.uu.net

young@dynamo.ecn.purdue.edu (Mike Young) (09/26/90)

In article <1990Sep25.201504.5605@uunet!konkord> jar@konkord.uucp (John Roberts) writes:
>
>	PDP users of the world:
>
>		Does anyone know which is the faster/better machine?
>		(The 11/24 or the 11/44)?
>
>	Is it possible to take out all the 11/44 boards and replace them
>	with the 11/24 boards? Are the backplans compatable?
>
>	What does it mean when the DC ON light blinks and its NOT the
>	power suppy. Why does the DC ON light blink when grant cards
>	are not there/or not in the slot?
>

	First of all, sounds like you need to subscribe to the info-PDP11
mailing list - a mail-oriented newsgroup composed of those of us that still
have to deal with PDP11 stuff. Send a note to info-pdp11-request@transarc.com,
asking to be added to the mailing list. Re-posting your request to that
group will undoubtedly generate a useful reply! (The posting should be
mailed to info-pdp11@transarc.com.)

	I can give you some brief info here. The blinking DC ON light is indeed
power supply problems, but not necessarily +5. Could be +/- 12, or +/- 15
as well. The 11/24 and 11/44 do not share backplanes. They may look alike,
but they are wired completely differently. DO NOT try to put an 11/24 CPU
board into al 11/44 backplane, or vice versa. The 11/44 CPU is actually about
six hex boards; the 11/24 CPU is one board. 

	The DC ON light should not be affected by presence/absense of grant
cards.

	I have two 11/44s, and one 11/24. The '24 is in standby mode, in case
one of the 44s dies (11/44 is a lot faster than the 24, and can do split I/D).


>	Sorry for all the questions...

	No problem! 

>
>-- 
>John A. Roberts				(508)460-4646 X197
>Concord Communications, Inc.			uunet!konkord!jar
>753 Forest Street					or
>Marlboro, MA 01752, USA				jar%konkord@uunet.uu.net



--
    |  Mike Young KA9HZE                |       young@ecn.purdue.edu    |
    |  Purdue University EE Dept.       |       ...!pur-ee!young        |
    |  W. Lafayette, IN  47907          |                               |
    _____________________________________________________________________

kcwellsch@watmath.waterloo.edu (Ken Wellsch) (09/27/90)

>> What does it mean when the DC ON light blinks and its NOT the
>> power suppy. Why does the DC ON light blink when grant cards
>> are not there/or not in the slot?
>
> I can give you some brief info here. The blinking DC ON light is indeed
> power supply problems, but not necessarily +5. Could be +/- 12, or +/- 15
> as well. The 11/24 and 11/44 do not share backplanes. They may look alike,
> but they are wired completely differently. DO NOT try to put an 11/24 CPU
> board into al 11/44 backplane, or vice versa. The 11/44 CPU is actually about
> six hex boards; the 11/24 CPU is one board. 

I had a flashing DC-ON light as well.  I tracked it down to a jumper (okay,
zero ohm resistor) that was enabling the monitoring of the 12 volt supply.
From what I could trace down, the 12 volts was there for the old 256K memory
boards while the newer 1Mb boards do not need 12 volts.  In fact my +12/-12V
lines are at +15/-15V.  So I disabled 12 volt monitoring (as this is a
jumper option) and my supply is otherwise within spec.  Someone may correct
my interpretation.  The jumper is on the M7090 card (I think there are about
20 or so "jumpers" and they are not obvious, at least not on my M7090).

I've got a glossy "relative performance" histogram that shows the 11/44
as about 1/2 that of a VAX 11/780 while the 11/24 appears to be about
1/4 a 780.  In fact they show the 11/23+ and 11/24 together on that chart. 
I believe the 11/24 is just a UNIBUS based Micro/F-11 chip as in the 23+.

Most of the PDP Unibus models have uniquely designed back-planes.  The
44 and 24 as said may look similar but are radically different.

-- Ken Wellsch
   kcwellsch@watmath.waterloo.edu

terry@spcvxb.spc.edu (Terry Kennedy, Operations Mgr) (09/28/90)

In article <1990Sep25.201504.5605@uunet!konkord>, jar@konkord.uucp (John Roberts) writes:
> 		Does anyone know which is the faster/better machine?
> 		(The 11/24 or the 11/44)?

  The 11/44. It has:

  o '181-based CPU. The 24 uses a custom 40-pin microprocessor (2, actually)
  o Native addressing to 1920Kw. The 24 needs a KT24 option to go over 248 Kw.
  o Faster floating-point (FP11-F) and CIS (KE44) options
  o Split I/D space
  o More 11/70 compatibility for things like Unix
  o Always in a 10.5" box. Many 24's are in a horrible 5.25" box

> 	Is it possible to take out all the 11/44 boards and replace them
> 	with the 11/24 boards? Are the backplans compatable?

  Nope. The CPU backplanes are dissimilar (the 44 has a dedicated backplane,
the 24 goes in the first 2 slots of a (slightly-modified) DD11-DK. However,
expansion backplanes (DD11-DK, DD11-CK) are compatible, as are the boards
which inhabit them.

> 	What does it mean when the DC ON light blinks and its NOT the
> 	power suppy. Why does the DC ON light blink when grant cards
> 	are not there/or not in the slot?

  DC ON blinks if a power supply voltage is out of tolerance. The most com-
mon cause for this is someone installing a DD11 backplane and forgetting to
cut the battery backup jumpers on it 8-).

  If you have the environmental monitor option (it's a secret option) in the
44, you can also get a DC ON blink if all 3 of your fans are kaput. Watch
out for 44 fans - they are 35V 70Hz *square* *wave* drive and really hard to
get.

	Terry Kennedy		Operations Manager, Academic Computing
	terry@spcvxa.bitnet	St. Peter's College, US
	terry@spcvxa.spc.edu	(201) 915-9381