Very slow booting and running and braindead OS's
No, I counted one user with multiple sign-ons as one user. If I would have been counting jobs...
Faint memories, I hope I'm not slandering any systems too badly, but can anybody verify and expand this list of doggish OS's?:
All kinds of things. The OS-2 desktop itself uses something like 35 concurrent threads, and I tend to run a fair number of things in the background. A semi-complete list of...
1130-CAI A multi-terminal Computer Aided Instruction OS for the 1130, using some huge swapping disk, which also held and refreshed the video contents of the terminals. Was usable with ONE terminal, unbearably slow with two. The main CAI language was so brain-dead, to put text up on the terminal screen, one had to punch countless cards with counted hollerith fields, i.e.: 43H--WELCOME TO THE ARITHMETIC TUTOR-- 47H (C) 1970 U. OF DISMAL SEEPAGE, OHIO
TSS-360 I seem to have a faint memory that TSS-360 was a bit of a dog. 1200 programmers were thrown into the coding frenzy and apparently little restraint was used. Many wonderful advanced features led to really poor performance-- like it could only support two users if they had a lot of patience.
Here in alt.folklore.computers, A lot also depends on how you count "users", on the types of activities those users are performing, and (on some systems) on the clbuttification and nature of "users" that you're talking...
UNIVAC-1100 OS A very sophisticated OS by the looks of the manuals! But a short FORTRAN compile with 12 light users could take 30 minutes. No security system, as it would try to reserve for you that many files, of that many tracks of that many blocks on across that many disks. Which exceeded the total disk storage on the planet. But the OS would bravely try, issuing every minute a message: "GATHERING FACILITY....". You could literally request that the OS finish running your job by course was not only impossible physically, but violated the applicable theoretical NP "predict ending time" theorem. But the system would try, issuing to your terminal every minute "gathering CPU facilities..." messages. And oh, the most brain-dead feature: any way around it. Which made tallying user-chaddable survey cards a real exciting job.
Of course not (heck, we supported thousands of concurrent TIP users on much lesser 1100- and 2200-series mainframe boxes at the airline I used to work for). The "40-50" number is simply...
HP-2116-DOS I still can't beleive what the sysop told me-- you had to allocate file space by hand, specifying actual track and sector numbers? Could this be true?
HP-3000 OS Another rather sophisticated system for its time. Except everything was dynamically linked, so it could take 20 minutes for the OS to boot up and link itself enough to present the first login prompt. Not fun when you'd paid several hundred thousand dollars for the privilidge.
UCSD Pascal, version V Previous versions would boot in about 3 seconds from a 400K floppy, but had slightly obtrusive need for the user declaring overlays if a program grew larger than 64K (which was a LOT of memory back then, good for about 100 lines of Pascal). But in version V they added full dynamic linking and segment swapping, which made coding large programs somewhat easier. The downside-- It could take 20 minutes of disk clicking for the system prompt to appear.
Hoare's second Algol system-
The first one had a very clever one-pbutt, fast Algol compiler, with a few minor limitations. The second version of the OS-Compiler was a full implementation, but the overlay swappng limited speed compile speed to a very few ccharacters per second. Unusable.
Very slow booting and running and braindead OS's 4523
Alt Folklore Computers from Newsgroups