OT: Yet Another Unhappy Customer for Vista

he

MassivelyWrong, once again.

There was no such rule, Dimbulb. There were *many* TSRs that would multitask under DOS. They weren't too difficult to write either; hook the timer interrupt and to a terminate (INT2h, IIRC) with stay resident.

Wrong again, Dimbulb. TSRs came along long before DesqView. The TSR hook was in there pretty much from the beginning of DOS.

They did, but not very well.

Preemptive multi-tasking

Wrong again, Dimmy. There were several TSRs that allowed preemptive multi-tasking.

Ever hear of a printer driver? Dumbass!

on

"switching" he

Oops... s/driver/spooler/

he

It SWITCHED between two tasks. I doubt very seriously that both had CPU cycles at the same time. The background task was likely getting none, AND DOS apps back then spent 90% of their time waiting for keyboard input. Hardly what one would call a live process.

Fuck you, KiethKeithStain.

It seems that you ALSO don't know the meaning of terms used in Computer Science.

A TSR is NOT a separate task.

You can worm and squirm all you want, KiethKeithStain, but you won't get anywhere until you actually learn about what you are spewing about.

WRONG.

Multi-Threaded Multi-tasking, which started with OS/2.

Everything else was a glorified SWITCHER.

Get a clue.

The discussion is about APPLICATIONS, not some side routine, dumbass.

You're repeating yourself again, Dimbulb.

It certainly is, if it's also hooked the timer.

MassivelyWrong no matter what moniker. Not surprising.

It is an embedded task, idiot. It is ALSO NOT any type task we are talking about here in this thread. The thread was and still is about running more than one APPLICATION at a time where both get processed at all times, and THAT did NOT happen.

You're an idiot.

Irrelevant.

Wrong again, Dimmy.

Wow! I scored a SIX! Pretty good.

You're still wrong. It's about multi-tasking. Print spoolers multi- tasked. Both the spooler and whatever other application DOS was running both got time slices. Print spoolers used TSRs.

Dimbulb is still MassivelyWrong.

he

The tricky bit of those sorts of programs is doing the needed disk I/O etc. Anything that had to deal with real hardware needed special care. To do a true multitasker under DOS you had to take over the DOS and BIOS disk operations vectors.

Some programs like the DOS print program attempted to do it by just saving the state of things like the disk tranfer area. They had to pause the printing if they happened to land in the wrong time.

Enough programs bypassed the DOS disk operations and did BIOS reads that you had to check for that.

There is no difference, Dimmy.

MassivelyWrong, once again, Dimbulb.

he

Then, in that respect DesqView was inferior to the prog I had,

No. it was multitasking, the background task was transmitting and receiving data, reading and writing files, etc... If I didn't tell it to use BIOS calls for its display output it'd crap all over the display of the foreground task.

it was a nasty hack that had other disadvantages (like having to partition the 640K into chunks for each application) but it did do multitasking.

You're misinformed.

that gave timeslices to all active apps

yeah until 3.x's 386 enhanced mode there was no dos multitasking in windows.

Bye. Jasen

yeah this one was like doubledos, I wish I remeber what It was called, I got it from a BBS in the late 80s.

Bye. Jasen

I'm sure that some lifer in prison would take a shine.

I think you are getting caught up in the semantics of the words "Multitasking" and "Task Switching". Until you had multiple cores or execution units in the same box, you were in fact task switching. A CPU that has only one execution unit could only execute one instruction at a time and part of the OS would switch between tasks so fast that the user feels as if the system is multitasking. Just because the system could preempt a task in progress, did not mean the system was multitasking, in fact the very fact that the OS had to preempt a task meant that it was getting ready to do a context switch, not multitasking. Not TRUE multitasking anyway. Too bad our engineering lingo can be so imprecise at times.

Jim

No, "task switching", as used here, only allows one application to run at a time (human time). "Multi-tasking" allows more than one application to operate "at a time". We understand the terminology being used.

Right. The point is that both task are getting CPU time. A "task switcher" suspends applications that don't have the "focus".

It's not. We all understood exactly what was being said. Dimbulb was simply MassivelyWrong, again.