The Perfect Computer - 36 bits?
|
From: Rich Alderson on 1 Apr 2007 16:16 Brian Inglis <Brian.Inglis(a)SystematicSW.Invalid> writes: > How many FE machines could you attach to a PDP-10? On the KL-10 (without a KA/KI compatible I/O bus), you can install 4 DTE-20 interfaces, to connect a Unibus 11. One is taken up by the console (which can handle terminals and a lineprinter); the other three can be used for terminal concentration, HASP RJE, ANF-10 networking (IIRC), DECnet, or whatever else you might need from an 11. The DC-10 that BAH mentions is an I/O bus device with which I have no experience. The DEC NI ("Network Interface") is an Ethernet(TM) interface that is part of internal I/O workings of the KL-10, inferior in many ways to the Stanford (and later cisco) MEIS which was available several years earlier. -- Rich Alderson | /"\ ASCII ribbon | news(a)alderson.users.panix.com | \ / campaign against | "You get what anybody gets. You get a lifetime." | x HTML mail and | --Death, of the Endless | / \ postings |
From: Patrick de Zeester on 1 Apr 2007 16:17 Andrew Swallow wrote: > krw wrote: >>> VAX/VMS the reliable alternative to Windows PCs. >> >> You're dreaming. That bus left a long time ago. > > Probably, it is Linux that will have to clean up the mess Microsoft > makes. Linux is a rather big mess itself. For some unknown reasons many distributions have chosen to go the Microsoft way; horribly bloated, resource hungry and unstable.
From: krw on 1 Apr 2007 19:16 In article <4610138f$0$24164$e4fe514c(a)dreader14.news.xs4all.nl>, invalid(a)invalid.invalid says... > Andrew Swallow wrote: > > krw wrote: > >>> VAX/VMS the reliable alternative to Windows PCs. > >> > >> You're dreaming. That bus left a long time ago. > > > > Probably, it is Linux that will have to clean up the mess Microsoft > > makes. > > Linux is a rather big mess itself. For some unknown reasons many > distributions have chosen to go the Microsoft way; horribly bloated, > resource hungry and unstable. > Just a guess, but perhaps that's what people really want. The ones who don't are generally savvy enough to strip it down (or build it up) to their liking. I know it's heretical, but I believe Linux' problem is much simpler; compatibility. -- Keith
From: Brian Inglis on 2 Apr 2007 02:32 fOn Sat, 31 Mar 07 12:29:26 GMT in alt.folklore.computers, jmfbahciv(a)aol.com wrote: >In article <mddejn6wjk5.fsf(a)panix5.panix.com>, > Rich Alderson <news(a)alderson.users.panix.com> wrote: >>jmfbahciv(a)aol.com writes: >> >>> In article <mddd52rx59j.fsf(a)panix5.panix.com>, >>> Rich Alderson <news(a)alderson.users.panix.com> wrote: >> >>>> They wanted 8-bit bytes and power-of-2 words, and nothing was going to >>>> change their minds about that. >> >>> Then customers would continue to buy -11s and move the boring grunt work to >>> the -10s and their secretaries. >> >>I see that the point was completely missed, and >>there's no use in trying to fix it. > >It was not missed by me. Those people you talked to will continue >to have their minis bought for them. However, the infrastruture >that surrounded them would have computing needs beyond mini >capabilities. Thus, the _bosses_ of those people you talked to >would buy a -10 because 1. they were already doing business >with DEC and were pleased with the services; 2. it was easier >to stay with one supplier than introduce a brand new computer >culture 3. the -10 was "compatible" with the gear that those >people to whom you talked insisted on using. > >Now do you see? In the PDP-11 era Marketing seemed to consider 10s(/20s) suitable only for educational institutions, not for businesses. 1. Digital made no promises of better (hours of) service from the branch for 10s than 11s. 2. The OS and third party software products were from different cultures, different products would have to be chosen, code would have to be rewritten, and data converted, regardless of which dissimilar architecture or vendor was chosen. 3. Going from 8 bit characters to 7 or 6 bit was seen as a step backwards, even in an environment where only 7 bits mattered. How your group saw business was one thing, how local field service and marketing saw business was another, and how businesses saw Digital thru their local reps was yet another. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis(a)CSi.com (Brian[dot]Inglis{at}SystematicSW[dot]ab[dot]ca) fake address use address above to reply
From: Brian Inglis on 2 Apr 2007 02:41
fOn 29 Mar 2007 22:10:41 -0700 in alt.folklore.computers, "Tarkin" <Tarkin000(a)gmail.com> wrote: >On Mar 30, 3:46 am, CBFalconer <cbfalco...(a)yahoo.com> wrote: >> Nick Maclaren wrote: >> >> ... snip ... >> >> > No, but nor could the Z80 compete on industry-quality functionality >> > and reliability. I know quite a few people who used Z80s for that, >> > and they never really cut the mustard for mission-critical tasks >> > (despite being a factor of 10 or more cheaper). >> >> Nonsense. I had 8080 based communications systems that ran >> continuously (no restart) for 2 to 3 years, until brought down by a >> mains power failure. >8080 != Z80. ISTR reading from a few different >places that early Z80's were 'twitchy'; that's >also why there are 'undocumented' opcodes- >those opcodes did not work reliably until the kinks >were worked out of the (wafer production [?]) >process. Nonsense. Undocumented opcodes tend to be a side effect of a particular implementation of an architecture: that implementation does this if you set those bits in an instruction. That's why they're undocumented: if they change the implementation, the side effects of setting those bits may produce a different result. Instruction decoders nowadays have transistor budgets that allow them to raise illegal instruction traps on any bit combos the implementor chooses to disallow. Early micros did not have those transistor budgets, so they left non-architectural features in the implementation and just chose not to document them. -- Thanks. Take care, Brian Inglis Calgary, Alberta, Canada Brian.Inglis(a)CSi.com (Brian[dot]Inglis{at}SystematicSW[dot]ab[dot]ca) fake address use address above to reply |