Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Me too, but I was able to do it around 1995-1996 :) Also remember Windows95 can boot with 4MB of RAM, and was decent with 12MB.


Windows95 was decent even with 8 MB, on a 66 MHz or 100 MHz 486 CPU.

With either 4 MB or only a 386 CPU, it was definitely crippled, making an upgrade not worthwhile.


Windows 95 on a 386 CPU with enough RAM was alright. Not fast but very useable.

https://youtu.be/Pw2610paPYM?t=72

But most 386 didn't have 8+ megabytes, and some 386 had a 286 like data bus, making it even slower. (386SX)


On paper a 386sx is slower than a 386dx, and certainly is in terms of RAM access. But in practice you'd need some expensive hardware to fully take advantage of that speed, like EISA cards and a motherboard that supported them (or, MCA cards on one of the higher end IBM PS/2 models). The typical ISA cards of the era were limited to 8 MHz and 16 bits no matter what processor or motherboard you used.

The 386dx could also use a full 32-bit address space, whereas the 386sx had 24 address lines like the 286. But again, having more than 16 MB would have been expensive at the time.


> Windows 95 on a 386 CPU with enough RAM was alright.

I benchmarked it for PC Pro Magazine when it came out.

We had to borrow a 4MB 386SX from a friend of the editor's, as we had nothing that low-end left in the labs.

In our standard benchmarks, which used MS Word, MS Excel, PowerPoint, Illustrator, Photoshop, WinZip, and a few other real apps, Win95 1.0, not 95A or OSR2, was measurable faster than Windows for Workgroups 3.11 on MS-DOS 6.22, hand-optimised.

When it needed the RAM, 95 could shrink the disk cache to essentially nothing. (Maybe 4 kB or something.) Win3 could not do that.

It was SLOW but under heavy load it was quicker than Win3 on the lowest-end supported hardware.

Under light load, Win3 was quicker, but Win95 scaled down very impressively indeed.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: