Half an operating system: The triumph and tragedy of OS/2
- Thread starter JournalBot
- Start date
Nice read, thanks for that.
Couple of points for clarity. The PC was a quick move to market that was specifically Open in the beginning. That was because nobody had the capacity to manufacture all those funky serial port cards and the myriad of other items that extended the PC to be more than a terminal. In fact, nobody knew what was going to be needed or what the market would actually want.
When the PC/AT was released IBM was receiving ~90% of all 286 productions and many journalists were astounded that IBM displayed stacks of PC/AT ready to ship that, to their knowledge, were more than all the 286s that existed - in several different locations.
Personal anecdote, running a multi-tasking solutions was interesting on both systems. I built a small system that drove a specialized hardware card. We would fit several in the larger machine. Running on DOS it ran in X time but was unable to run two. Windows ran two in 2X time, OS/2 ran two in 1.06X and eventually NT ran in ~1.25X. In many of my cases OS/2 was a clear winner.
I have long thought that this legacy position in the mainframe business tainted the method for dealing with PCs. Trying to put the genie back in the bottle was vastly more difficult when the genie was released by IBM's PC division and the bottle was theoretically held by the mainframe group. Trying to close an open design was never going to work.
Couple of points for clarity. The PC was a quick move to market that was specifically Open in the beginning. That was because nobody had the capacity to manufacture all those funky serial port cards and the myriad of other items that extended the PC to be more than a terminal. In fact, nobody knew what was going to be needed or what the market would actually want.
When the PC/AT was released IBM was receiving ~90% of all 286 productions and many journalists were astounded that IBM displayed stacks of PC/AT ready to ship that, to their knowledge, were more than all the 286s that existed - in several different locations.
Personal anecdote, running a multi-tasking solutions was interesting on both systems. I built a small system that drove a specialized hardware card. We would fit several in the larger machine. Running on DOS it ran in X time but was unable to run two. Windows ran two in 2X time, OS/2 ran two in 1.06X and eventually NT ran in ~1.25X. In many of my cases OS/2 was a clear winner.
I have long thought that this legacy position in the mainframe business tainted the method for dealing with PCs. Trying to put the genie back in the bottle was vastly more difficult when the genie was released by IBM's PC division and the bottle was theoretically held by the mainframe group. Trying to close an open design was never going to work.
Upvote
16
(16
/
0)
- Status