Talk:Data General
From Wikipedia, the free encyclopedia
 |
This page has been cited as a source by a media organization. See the 2005 press source article for details.
The citation is in: Andrew Orlowski. "Wikipedia: magic, monkeys and typewriters", The Register, October 24, 2005. |
Just a couple of observations. The FHP architecture was a potential world beater. However, it had some fatal floors at ground level. In fact, the turf war with the Eagle didn't last long as FHP died, causing a very big write off in investment.
The other thing was that the very first laptop came from Data General - the Data General/One. It's design concept was the progenitor of every laptop we see today - fold away LCD screen, miniaturised keyboard etc. The company made two strategic errors: commercially, they attempted to sell it under their OEM contract with it's bill backs; the retail world certainly wasn't having any of that. Technically, the DG/One was almost, but - critically - not quite MS-DOS compatible. So when Toshiba came oalong with their fisrt laptop a few months later it was the kiss of death for the DG/One.
- The article has a longish section on the Data General-One. (I Googled a number of sites and couldn't find any uniformity with regard to the punctuation and spelling of that name, by the way... ). But you might consider adding your remarks. I felt that the premature use of 3-1/2" diskettes, the low-contrast mirror-like screen, and the very high price also played a role in its near-failure. I hadn't realized there were MS-DOS compatibility issues. My company never got beyond looking at it (and seeing our faces reflected back at us) and wincing at sticker shock, so we never got one. (We got bitten by "almost MS-DOS compatibility" with the Digital Rainbow 100...) Dpbsmith 11:44, 24 Jun 2004 (UTC)
The story that I was taught in school was that the basic design of the Nova was developed by de Castro while he was at Digital. When the general register PDP-11 was chosen as the 16 bit successor to the PDP-8, it was selected over the single accumulator design whose team then left and founded DG.
Contents |
[edit] Intel 8088
The 8088 was not the "true commodity processor" in the early 90's; it was years obsolete for any sort of workstation or server application by that time. (Pointym5 12:59, 31 August 2005 (UTC))
[edit] PDP-8
The second para of the history section says that many PDP-8s still operate today. Is this still true? It might be safer to change this from "where many still operate today, decades later" to "where many continued to operate for decades".-gadfium 21:10, 26 December 2005 (UTC)
[edit] "infamous" ?
IMHO the wise-crack (and use the term deliberately) about "the now infamous PowerPC" is inapropriate and should be removed.
It seems to me that a strong opionated statement like that should be supported if it is to be included, and as such support is beyond the scope of the article, the phrase "now infamous" should just go away.
This sort of thing casts doubt on the overall reliability of what otherwise seems to be a very good article.
[BTW: IMHO, recent negetive events connected with this this processor architecture probably reflect more on the business relationships between the firms involved with it than on the actual hardware. AFAIK PPCs generally have worked pretty well and continue to do so (even if intial production batches shipped late, or whatever), which a good thing for many of us, as they have found their way into enough embedded applications that many of us probably use one or more every day without knowing it. "Many of us" may include the author of the article.]
- ef
- I agree, and have removed the description.-gadfium 00:01, 9 April 2006 (UTC)
[edit] Multi-processing
2006-08-02_01:03
Should this read: "supported multi-processor usage" instead of "supported multi-processing"?
Quote:
To compensate, the AViiON machines supported multi-processing, later evolving into NUMA-based systems, allowing the machines to scale upwards in performance by adding additional processors.
- I don't see anything wrong with the way the article is worded, although I'm not at all an expert on the subject. Our article on Multiprocessing seems relevant. Could you explain in more detail why a wording change would be beneficial?
-
- i think i see what this question is driving at, though i also wonder if the proposed remedy works.
-
- the term "multi-processing" can refer to a number of things, including the use of multiple physical processors and methods of running multiple processes irrespective of the number of processors. it's ambiguous.
-
- "multi-processor usage" is awkward and doesn't necessarily clear things up. "multi-processor configurations" might be more effective. (...assuming, of course, that i've understood the question correctly. have i ?) - ef
-
- Ericfluger 22:48, 15 October 2006 (UTC)
"were multiprocessors" would probably be the most economical locution. I don't really consider it ambiguous in that context, although one could also say something like "multiple processor configurations" or something like that, although it starts getting to be an awkward way to express what is pretty well-accepted and understood terminology. The more serious problem with the paragraph is the assertion that AViiONs were MP because the 88K was slow. This isn't really true. In fact, the 88K was one of the faster processors around when it first appeared. AViiONs were MP because all high-end servers were MP regardless of the architecture. (Over time, Motorola walked away from the 88K and it fell behind, but that's not the reason that AViiONs were MP.)
Ghaff 12:42, 16 October 2006 (UTC)
[edit] Virtual Memory
I'd like to question the use of the term "virtual memory", with regards to the 16-bit Eclipses. While the majority of knowledgeable people agree that "virtual memory", on most systems, refers to any type of hardware address translation, that's not the case in the context of Data General. When DG talked about "virtual memory" they specifically meant the ability to use paging space on the hard drive to allow the machine to appear that it had more memory than it actually did. The 16-bit Eclipses could only run AOS (Advanced Operating System) and did not have this ability. If I remember correctly, each process only had a 128k address space. The 32-bit Eclipses could run AOS/VS, where the "VS" stood for "virtual storage". On AOS/VS, I could have a process that used 8MB of memory, even if the machine itself only had 4MB.
--Dadoo3002 02:30, 26 March 2007 (UTC)
