Hacker News Reader

Comment by twoodfin | Hacker News Reader

It’s kind of wild that neither IBM (who got beat to the hardware punch by Compaq) nor Microsoft (who targeted the NT effort at everything except x86) initially grokked how revolutionary the 80386 was: A “decent enough” 32-bit architecture with a huge preexisting ecosystem that would be able to ride the rocket ship of commodity PC scale.

View on HN

On both counts, I think it's actually IBM who didn't get the importance. Both of the 386 to OS/2 and how important quick-to-market hardware was (even if just for brand prestige) versus Compaq.

Microsoft always got it, and I feel certain the first release of NT (3.1) sold many times as many copies for x86 as it did for other architectures; and it was targeted for it as much as for any other arch.

It was actually Microsoft that saw early that OS/2 needed to exploit the 386 but IBM dragged their feet on it. A strategy similar to Windows/286 vs. Windows/386 would've made a lot of sense IMO. And probably helped IBM sell more 32-bit Micro Channel hardware early on!

Microsoft kind of did though? Windows/386 was released in late 1987. It could run multiple DOS apps using the "virtual 8086" mode. That was pretty revolutionary at the time. I think I knew more people using DESQview back then, though.

Sure, but their mainstream OS didn’t support the bulk of the 80386’s capabilities until late 1995.

That’s kind of insane if you think about it.

True, though Windows 95 kinda ran like crap on the 386. We were well into the Pentium era by then.

Not just kinda, it was close to unusable without a 486.

Hah. I was being generous! My 386 box was collecting dust by that point. I did hear horror stories about people attempting to upgrade their 386 with 4 megs just to hit the "minimum" Win95 requirements and having their machines swap to death.

Reply on news.ycombinator.com