When the System/360 was designed, nearly every new computer was different and incompatible with previous ones. Computer architectures came and went, few of them lasting very long. It is remarkable that the basic architecture of the System/360 has lasted for well over 5 decades while continuing to evolve.
Amdahl, Blaauw, and Brooks, the principle designers of the System/360, wrote a paper in the IBM Journal of Research and Development in 1964, titled "Architecture of the IBM System/360." In it they describe the trade-offs that they considered and the reasons for the decisions that they made. I think of it as a 15 page precursor of the Principles of Operation. For anyone interested in the reasons why S/360 was designed the way it was, I highly recommend it. Here are a couple of places where it can be found. http://www.ece.ucdavis.edu/~vojin/CLASSES/EEC272/S2005/Papers/IBM360-Amdahl_april64.pdf http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.72.7974&rep=rep1&type=pdf -- Tom Marchant On Mon, 4 Dec 2017 23:17:40 +0000, Swarbrick, Frank wrote: >I think someone else said it was choosing EBCDIC over ASCII. Personally I >think that one's worth, because apparently it will never be "fixed". > >-----Original Message----- >From: IBM Mainframe Assembler List [mailto:[email protected]] On >Behalf Of Charles Mills >Sent: Monday, December 4, 2017 3:55 PM >To: [email protected] >Subject: [EXTERNAL] Re: Access registers > >I believe someone (Harlan Mills? Fred Brooks?) said that he felt the only (or >most significant?) *error* in the System 360 design was the 24- >rather than >31- or 32-bit addressing.
