🚀 go-pugleaf

On Monday, August 13, 2018 at 5:23:26 PM UTC-4, John Levine wrote:
> In article <5b71ce88...@news.dslextreme.com> you write:
> >Your so-called "categories" are arbitrary and mostly meaningless.  Even before
> >the advent of the 360, they were always "general purpose" computers.  Meaning
> >they could be used for any application, given the proper programming.  The
> >salient differences were in CPU speed, storage capacity, and data transfer
> >rates. ...
> 
> Yes and no.  The architecture of the binary word addressed 704x/709x
> scientific machines was rather different from the decimal word 7070 or
> the character addressed 705/7080.  Sure, they were all Turing
> complete, but the 7080 was bad at floating point arithmetic and the
> 7090 was bad at moving character strings around.

True.  Plenty of customers did use binary machines for business
and character machines for sci/eng, but it was inefficient.  
Computer time was very costly, so it had to be used carefully.

Here, we discussed how using an offline sorter was more cost
efficient in the early days than sorting by the computer itself.




> We all think it's obvious to have byte/word addressing where the word
> addresses are byte addresses with the low bits zero, but it wasn't
> obvious at all before IBM invented it for the 360.  Once they had a
> memory architecture that could work either way, building the
> instruction set wasn't so hard.

As mentioned, memory was extremely expensive, so a bigger machine
had extra cost in having longer instructions to support a larger
address space.  Note that the low-end 1401 was only 1,400 characters,
and I believe the original S/360 was offered as low as 8k.

It's hard to believe.




> 
> > The big advantage of having a line of computers with a
> >single architecture would be to ease moving from one machine to another that was
> >faster and/or bigger.  It also helped IBM, assuming they could consolidate their
> >OS and system software development and support.
> 
> That too, although BOS, TOS, DOS, OS/PCP, OS/MFT, OS/MVT, CP/CMS, TSS,
> and PS44 suggest that consolidating the system software didn't work
> out so well.

Yes, the development and expectations for S/360 "OS" was a big
blunder.  I can't help but suspect IBM should have known better
that _both_ their performance goals and development schedule
were far too ambitious.   Didn't they have big software development
experience with SAGE (although contracted out), and SABRE?

In my humble opinion, I think IBM wasted very scarce resources
at the time developing PL/I and TSS.  COBOL and FORTRAN were 
adequate, they didn't need PL/I.  I can't help but suspect that
TSS was developed as a result of _perceived_ pressures, not
actual market demands--I don't think there was that much
demand for it in the early 1960s--most users were batch
oriented.

In other words, I'm pretty sure Watson wanted S/360 to be
a true universal computer for every possible customer,
when in fact there were some specialty uses best left
to others, an idea Watson couldn't stomach.

IMHO, IBM should've concentrated on getting the basic batch
versions of S/360 out the door first, then worked on enhancing
the software to exploit the hardware's advanced features.
In other words, walk before running.

IBM was correct that a unified product line would help them
on peripherals.  AFAIK, most major S/360 peripherals were
usable without modification throughout the product line.
I believe before S/360, each peripheral had to be customized
for a particular model of computer.  

I think most people know that IBM's 1959 technology--SMS circuit
cards--found even bigger use in S/360 peripherals.