Talk:Ferranti Mark 1
Computing Start‑class High‑importance | |||||||||||||
|
It is requested that a photograph be included in this article to improve its quality.
The external tool WordPress Openverse may be able to locate suitable images on Flickr and other web sites. |
(untitled)
As an undergraduate, I spent one summer as a maintenance technician on the FERUT computer and learned programming from Christopher Strachey. My memory of the machine was that its Williams Tube memories had 64 64-bit words, arranged in blocks of 8 x 8 bits, with a parity bit row to the right and above the actual memory bits, making a 65 x 65 bit pattern. Strachey programmed the index register on the console (a similar Williams tube) to visually represent a checkers board, and programmed the machine to play against a human opponent who entered moves on the teletype. (I'm not putting this in the main page, because it relies on my memory rather than on documented evidence)Mmt (talk) 17:26, 8 February 2009 (UTC)
instruction represented by a single character?
There are some statements in the article, which, IMHO, are contradicting:
- The Mark 1 used a 20-bit word
- Instructions were stored in a single word (__how many?__)
- There were about fifty instructions in total
- the characters representing the values from 0–31 (five-bit numbers)
- Each instruction was represented by a single character
Now, if there were about 50 instructions in total, then how could they be represented by a single character? (ie. one of 32 possible values)?
It makes no sense.
Even if in a single word (20 bits) there would be two instructions stored, it would make 10 bits per instruction used, ie, __two__ 5-bit blocks or characters. — Preceding unsigned comment added by Klenot (talk • contribs) 06:22, 31 October 2016 (UTC)