On Jan 31, 2022, at 5:26 PM, Thomas DeBellis <tommytimesharing@gmail.com> wrote:
You know, I think you might be right. Columbia did get a big discount from IBM and we had some stuff that was not commonly available (like the 360/91, of which three we made, I believe). I don't remember DEC giving us that kind of discount on our 20's except when they were desperately trying to save the account after they pulled the plug on 36 bits.
That point about CPU usage really was the point. I think I remember an IBM 3033 demonstration system with some absurd number of 3270's on it; I mean something like 11,000. No way you could that with full duplex; too many Start I/O's. Even if you did a Start I/O no-release, you just would never had the gas.
All things being equal, the deals we could get on a 3270 were about twice as expensive as what we could get for a HP2621 or something of that caliber. The wiring was more expensive to do. What I remember is our com guys complaining about the coax being difficult to work with while RS232 being nowhere as difficult and the wire itself costing less (not sure about plenum grade).
I think the IBM controllers may have cost more; at least a controller for 16 3270's cost more than a DH11, but I don't think that's a completely fair comparison because it didn't factor the cost of the front end (which you needed no matter what you did).
On 1/31/22 6:14 PM, dave.g4ugm@gmail.com wrote:
_______________________________________________Thomas, Dave, So Thomas, you say "expensive" well IBM wasn't always more expensive. You need to put the CPU cycles to handle the character editing somewhere. So if you were using IBM you distributed some of it in the 3274/3174 controllers. If you were DEC you simply sold more PDP11 or VAX which meant you needed more hardware and probably more software licences. It wasn't always cheaper. A very good friend once confided with me he was reluctantly rolling out All-In-One to senior staff in a big UK bank but was giving lower levels PROFS because it was cheaper. I don't know the details of the pricing, but I can't think why he would lie to me..... IBM also gave generous discounts to academia. Back in the day they would often loan a machine to a university for 12 months trial, with an option to buy after the trial, but at the used price, with academic discount which was typically 50%... These days I have both. A few VaxStations and MicroVax, plus an IBM P390 . I enjoy both, but when I fire up the VaxStation VLC I feel sad that Digital dropped the ball and that I can no longer legally use it with VMS. It must once have had a VMS licence, and I can't see that it was transferred any where ☹ What a different world it would be if we had all had one of those on our desks instead of IBM PC's... Dave-----Original Message----- From: Thomas DeBellis <tommytimesharing@gmail.com> Sent: 31 January 2022 22:28 To: hecnet@lists.dfupdate.se Subject: [HECnet] Re: native Dup sync line revisited --> revisited Dup test on pdp11 --> problem solved: Simh KG11 emulation probably defective Fair is an interesting word. What is abundantly true is that the full duplex and the half duplex folks NEVER agreed. I was one of the very few people who spent significant time on both (as long as the half duplex terminal was a 3270). And it's also true that the 3270 could blow a lot of things out of the water in terms of speed (when on a byte selector channel) and field definitions. If you could afford it. Everything about them had some serious technology, except the price. Yet you couldn't do WSIWYG, not real WSIWYG. And an attempted port of VisiCalc failed because of this. A full duplex terminal allowed chording and language understanding, which allowed faster movement to 'the right place'. It was easier to get where you wanted instead of having to arrow key because a 3270 has no idea where you want to go. You can certainly write REXX macros to do some of this and bind them to function keys. You can get some decent thing, but now you're hitting the CPU, which defeats the purpose of the buffering. I always picked EMAÇS over XEDIT, 100% of the time, but I didn't whine about using XEDIT, either. A similar fight happened with bringing EMACS up under Multics. An entire book got written about it. People blabbed about CPU usage and it's true unless it doesn't matter. And nobody agrees. I could bring an IBM programmer into my office and show him EMACS on a 9600 baud terminal. "Look how easy and it knows what you're typing and ... " Blah, Blah, Blah. They were utterly unmoved, so I gave up. For me, my official response was, "I like whatever I'm paid to use". Right now, that's Windows and a virtual 3270. Privately, it was just less fingering to do development. I am glad there is no more keyboard unlock any more. I can't remember how many times I pushed that thing. On 1/31/22 4:08 PM, Dave McGuire wrote:On 1/31/22 3:22 PM, Thomas DeBellis wrote:Then we are probably talking apples and pineapples. The product that I'm talking about is exclusively for the 36 bit line, requiring a DTE to get the data to the PDP-10.Well, Peter Allan did say he's interested in "2780 emulation for a PDP-11". And, Peter, so am I. We have several running mainframes at LSSM, across the room from a bunch of PDP-11s. One of the things we like to demonstrate is interoperability between disparate architectures; I'd love to get something like that working such that we could demonstrate RJE from a PDP-11 to a mainframe on the other side of the exhibit floor.I don't recall a 3271, I guess you mean those green screen terminals? What a beast... I do recall using 3270, 3276 and the like when I was hacking the bisync drivers on VM to talk to IBMSPL.The 3271 is a remote terminal cluster controller for 3270-protocol terminals.A number of us swore that IBMSPL was the only reasonable way to use an IBM mainframe as the 3270 was a half-duplex terminal that was all to ready to lock the keyboard if you even thought of typing ahead. The more modern emulators don't lock like that, but you are still effectively in a late 1960's half duplex paradigm. Still. Today. Really.Tom, I'm sorry but that's not a fair assessment at all. It's not so much a "half-duplex paradigm" as it is forms- and screen-based. You've used these systems; I know you know this. The world thought it was a good enough paradigm to reinvent it for the WWW, and it's still the most efficient way to get to a big machine, rather than peck-peck-peck and flood the box with interrupts. There's nothing "still today really" obsolete about it at all; it's just that people like you (and me) who are accustomed to character-by-character interaction with the machine see it as foreign. And of course it turns into a painful experience when you try to fake a character-by-character interface with it, with something like VM or TSO. Think about what's going on to make that happen; there's no way it won't be at least somewhat painful. And I'm saying this as a late-comer to the mainframe world; I cut my teeth on PDP-8 and PDP-11 systems. -Dave_______________________________________________ HECnet mailing list -- hecnet@lists.dfupdate.se To unsubscribe send an email to hecnet-leave@lists.dfupdate.se_______________________________________________ HECnet mailing list -- hecnet@lists.dfupdate.se To unsubscribe send an email to hecnet-leave@lists.dfupdate.se
HECnet mailing list -- hecnet@lists.dfupdate.se
To unsubscribe send an email to hecnet-leave@lists.dfupdate.se