What made NeXT (and then OS X) so portable? NeXT never sold a whole lot, but I would like to see a breakdown of sales by CPU. And if certain platforms were missing things, etc.
I've heard many answers over the years:
Mach is a microkernel, they are easier to port
NeXT never made their own hardware so they designed it that way from the start
> NeXT never made their own hardware so they designed it that way from the start
When I think NeXT, I think the NeXTcube (1990-1993) [0], which apparently also had a predecessor in the NeXT Computer (1988-1991) [1]. NeXT made their own hardware for a solid chunk (nearly half) of their lifespan, and NeXTSTEP was originally released with the NeXT Computer [2]. It didn't even support CPU architectures other than the Motorola 68k series until 1993, around the demise of the NeXTcube! I think its ultimate portability probably has more to do with being a UNIX at its core (so a battle-tested design written in somewhat portable C) plus a user environment written in a higher-level, easier to port language (Objective-C).
The "NeXT Cube" was essentially the same thing as the "NeXT Computer" with an updated CPU. Basically, they renamed it after they released the "NeXTstation"[0] to avoid confusion. Otherwise, they'd have had two NeXT "Computers." Heh.
Though fine in 1988, NeXT's own hardware was rather under-powered for the early 90's. Their main competitor, Sun, was transitioning to their own RISC architecture (SPARC) after originally being on 68K. DEC was coming out with the Alpha. HP had moved to PA RISC. Though I digress, it's too bad NeXT never ran on the Alpha! That would've been a killer 90's setup.
I don't think running on a bunch of CPU architectures is really unique to NeXT. Windows NT ran/runs on x86, Arm, Alpha, MIPS, Itanium, and PowerPC, all with a desktop as a full-blown product. Forms of desktop Linux will run on all of those and more.
Super good question. I owned and used NeXT slabs, 1992 - 1999.
The Mach-as-microkernel thing couldn't have helped portability. The Mach task, port, thread and pager abstractions seem like they'd be difficult to port, but I could be wrong.
I read at the time that NeXTStep the GUI ran faster on Solaris than on Mach, because pipes and sockets were faster than Mach ports. So I'll believe the very little assembly.
I suspect that Mach-O file format had something to do with the portability, you can put arbitrary sections in it. Some NeXT games kept image and sound "files" in a single huge Mach-O executable.
"Mach 3 (Figure B.1) moves the BSD code outside
of the kernel, leaving a much smaller microkernel. This system implements
only basic Mach features in the kernel; all UNIX-specific code has been evicted
to run in user-mode servers. Excluding UNIX-specific code from the kernel
allows replacement of BSD with another operating system or the simultaneous
execution of multiple operating-system interfaces on top of the microkernel.
In addition to BSD, user-mode implementations have been developed for DOS,
the Macintosh operating system, and OSF/1. This approach has similarities to
the virtual-machine concept, but the virtual machine is defined by software
(the Mach kernel interface), rather than by hardware."
Early in that timeline Solaris was heavily into being a "pure" System V r4. The pipes and sockets implementations were based on System V Streams -- the in joke was that streams were much more like sewers and they weren't pleasant. Numbers wise, they were perhaps 2 orders of magnitude slower than on SunOS (BSD based). At one state university, where a very "beefy" SPARCcenter 2000 was purchased to support 500 simultaneous users, Solaris on this system could barely support 40 users (streams performance was the reason - telnet sessions pushed several streams modules). Sun ended up having to provide 10 SPARCstation 10 systems running SunOS which could easily support 70 simultaneous users each in order to fulfill the contractual obligations. The SPARCcenter 2000 was turned into an NFS file server.
After the early days, Solaris did improve, but I think that was because ideas / pieces of BSD (from SunOS) were put back into the OS.
I don't think Solaris pipes / sockets performed better than Mach.
Early Solaris was definitely not very good. I briefly ran Solaris 2.4 on a SparcStation 5, and reinstalled SunOS 4.1.x because Solaris 2.4 was so bad. 2.5 was decent though!
The impression I always got from those abstractions were that they were precisely there to support portability. Pagers, for example, are quite flexible about what can back them and what the geometry of that backing can be.
Yes, but it was explicitly a different kind of portability: running MSDOS or VMS executables unchanged, for example.
The task/port/thread/pagers are primitives that let Mach run some other OS executables, not made it easy to port Mach to other hardware. There were papers about getting MSDOS and VMS executables to run in Mach tasks. Some startup was making a system to run (pre-OSX) Mac executables.
« microkernels are easier to port » is not a thing.
NeXT did make their own hardware, it was 680x0-based and it was nicer to use than the x86 PC shitboxes it ran on when they stopped making hardware.
« very little assembly » is closest to the truth. NeXTstep and OpenStep were written almost entirely in C and Objective-C and compiled using the GNU toolchain. NeXT was always a hybrid of Unix userspace on top of a Mach microkernel. It’s just not that hard to port a Unixlike OS to another CPU architecture. It’s literally been done since the 1970’s.
However, it is much more difficult to port the OS and maintain binary compatibility with apps created for the previous CPU architecture, which has always been Apple’s special sauce since the original PowerPC Macs.
The hardware certainly looked nice (I have a "turbo" slab in my collection!), but the performance was "meh" at best. NeXTstep needed a lot of resources, and late 680x0 was no match for an early Pentium. It's no wonder NeXT ported to x86.
Business necessity. Something about your own hardware business failing makes you really eager to port to other hardware architectures. That and your processor manufacturer (Motorola) openly indicating that they're killing the ISA you built your platform on.
Obv writing in a portable language and on a portable microkernel helps.
I've heard many answers over the years:
Mach is a microkernel, they are easier to port
NeXT never made their own hardware so they designed it that way from the start
It contained 'very little assembly'