RE: IBM SAGE 1955

["Frazee, Walt" <Walt@VistaControls.com>]

Dimiter,

<<
I am always impressed by what people were doing with 
so much less components off the shelf then. I wonder if you had
opamps - 702, 709? They probably were too slow for those
memory sense amps (but then I have no idea on the speeds there).
>>
I remember the uA709 coming out about then, but it was too expensive and not
good enough to use.  We even made gates and flip-flops out of discretes.
The first analog IC we found a use for was a multiplexer - PMOS, I think.  I
designed a Sample & Hold Amplifier using all discretes, with the input stage
transistors temperature controlled by sticking them, and a heater resistor,
and a thermistor into holes in an Aluminium block.  (Matched differential
pairs came out just a little later.)

<<
I guess most of the nice discretes we have today were available
then (I mean low power bipolar)?
>>
I remember using a lot of 2n2222's, 2n2907's, 2n3904's,2n3906's, and
1n914's, all of which are still available.

Regards,
Walt


-----Original Message-----
From: Dimiter Popoff [mailto:tgi_earth@yahoo.com]
Sent: Monday, December 18, 2000 3:02 PM
To: Frazee@vanilla.bulnet.com; Walt
Subject: RE: IBM SAGE 1955


Walt,

>... Being an Analog type, I
>designed Data Acquisition Systems, DACs, and such, for the most part, so
>wasn't directly involved with the computer and memory design (except I did
>design a core memory sense amp and a current mode line driver).

I am always impressed by what people were doing with 
so much less components off the shelf then. I wonder if you had
opamps - 702, 709? They probably were too slow for those
memory sense amps (but then I have no idea on the speeds there).
I guess most of the nice discretes we have today were available
then (I mean low power bipolar)? I remember I once saw an
instrumentation amp (a so called 'shaping amp' for nuclear
spectroscopy) built by a guy in Oak Ridge, TN . He had done
heroic things... Most of his opamps used an RCA transistor pair
(or perhaps a pair with a current source beneath it), and several
other transistors. Then there were some coils and a lot of stuff - which
worked fine, at all that precision and bandwidth.

I don't mean that we have a less challenging life today, of course,
we are still needed to design, after all...

Regards,
Dimiter