Talk:IBM 700/7000 series

From Wikipedia, the free encyclopedia

This article is part of WikiProject Technology, an attempt to better organize information in articles related to technology. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and/or contribute to the discussion.

[edit] 7070 as 650 replacement?

I'd be interested in seeing some backup for the claim "The IBM 7070, IBM 7072, and IBM 7074 were designed to provide a "transistorized IBM 650" upgrade path. They replaced the drum memory with core memory, but were not instruction set compatible with the 650 (so a simulator was needed to run old programs)." The 7070 did have a 10-digit decimal word length, same as the 650, but that is the end of the similarity. I attended one talk in the 1959-1960 time frame where someone from IBM presented the 1620 as the replacement for the 650. I have no doubt that IBM sold some 7070's to large 650 shops that wanted more capacity, but the 7070 was a much more expensive machine. And I never heard of anyone simulating the 650 for production purposes on any machine. Computer time was considered more valuable than programmer time back then, and my recollection is that 650 programs that were still needed were rewritten for whatever machine was leased as a replacement. Anyone have a different recollection or documents?--agr 22:14, 28 May 2006 (UTC)

I have an article that was written by Bob Bemer, and was posted on his website until he died and some casino took over the URL, that explains how the 7070 was originally planned as a unified upgrade path to replace both the 650 and 705. Because it was incompatible with both it was not that well liked (several incompatibilities resulted from engineering to management communication problems with managment making the decision to go with a more 650 like machine when the programmers were recommending a more 705 like machine). Ultimately IBM resolved the issue by also building the 7080, which was compatible with the 705.

As Bob Bemer's website is no longer available and the casino site that took the URL has blocked archiving, it is hard to get copies of this article (titled BIRTH OF AN UNWANTED IBM COMPUTER). Let me know if you want a copy of the HTML file. -- RTC 18:15, 7 June 2006 (UTC)

Yes, I'd like to see it. You can follow the link from my user page for my e-mail address. This sort of thing needs to be preserved. I wonder if the computer history museum would host it?--agr 19:09, 7 June 2006 (UTC)

[edit] The 700/7000 series is not a series

Is the definition of a series all the machines built in some time period using the technology of that time period and the naming conventions of that time period? No, series, or families, share architectural features and you can say useful things that apply to all of the machines in a series. Like the IBM 1400 series.

Saying that all these machines were built during the same time period and that they were obsoleted by the machines that followed them is a tautology. Other than that, the article is divided by architectures and it is those architectures that define families, that define series.

This article should be deleted, replaced by new articles, one for each architecture listed in this article. New articles like IBM 702/705/7080 series. 69.106.254.246 02:13, 27 September 2006 (UTC)

"Series" is the term generally used for these machines, see for example, http://www.piercefuller.com/library/ibm7000.html. Your definition is a more modern one and results in no small measure from IBM's problems developing software for so many different architectures. One reason to keep the article intact is to make clear to the reader how different these machines were from a programming point of view and how they each influenced the next-generation System/360 architecture.--agr 17:50, 28 September 2006 (UTC)