This is NeXT-vs-Mac-IIfx-Innovation-argument in view mode; [Up]
Date: Sun 04-Apr-1990 22:58:33 From: Unknown Subject: NeXT vs Mac IIfx Innovation argument Folks, thought you'd like to see a discussion I had with an Apple felow about the Mac IIfx vs NeXT innovation-wise I mean. Interesting! Larry, You wrote: 6502's are convenient to use (there are standard cells available). It's always best to use the appropriate tool for the job, and a 6502 work well as a low speed I/O processor. Certainly, and for thinking of a low-cost answer that's a plus fro Apple. BUT, innovation? A REAL move would have been similar to what NeXT did with those two (admittedly proprietary) DMA-I/O chips...trying to move more power (Steve calls it mainframe power....well, not quite, but a nice try) into a machine that NEVER dreamed (030, 25Mhz) about that kind of power was a real gutsy move! You already had the 6502 installed and running ina pretty debigged machine! By that argument, there's not a lot of innovation in the NeXT. A large part of the system consists of other technologies that they put together (Mach, Display Postscript, the optical drive). The use of object-oriented programming wasn't new, and the interface builder is descended from work done on the Mac. Nope, completely disagree (which is nice isn't it?) NO ONE at CMU EVER thought Mach's first shot was a desktop unit. More like a massively paralleled monster like the BBN Butterfly (64 CPUs). The OD wasn't (probably still isn't) ready for prime time at all when Steve first started design work, and DPS had NO Applications to take advantage of. OOP wasn't available in REAL form except for SmallTalk which was more of an Academic exersise at PARC, and IB (while it might have had seeds from work started at Apple, I don't know. If it does, where's Apple's product, they would have had lots of time by now to get it out. MPW certainly ain't it!!) is just on another plane as far as development environments go. The Fluggleman people thought so too! As for the performance, based on my use of a NeXT a IIfx would be faster than just 10-20%. (Of course the NeXT we have here only has 8Mb and conventional wisdom is that you need 16Mb for top performance. Also, the stuff you were seeing is still beta software; how much improvement was NeXT 1.0 over 0.9?) You're right, lots. And 16MB is sorta required. But I was talking about looking at GUI UNIX on the IIfx vs NeXTStep'd UNIX. Sorry, I just thought that at 40Mhz, the Mac should have been MUCH faster, but it wasn't. > Uhm...nope. Apple didn't put a GREAT deal of thought into this box. That's not true. A lot of thought went into squeezing out the last bit of performance of the CPU. The basic hardware is comparable to the NeXT in terms of its attention to I/O and performance. Well, again, I disagree. Innovation, REAL innovation would have been something like a RISC (88k) chip and a GOOD cross compiler to show some of the SAME MS Apps (Excel) run under RISCified A/UX. That's innovation. The IIfx is just more of the same...with cheap I/O processors from the IIe! I'd say the same if Steve NEW NEXT was an 030 at 40MHz. At LEAST 040 and maybe even a MIPS RISC would be better. > My Mac can do 24 bit color, but so what, it looks rather cartoony because > QuickDraw just isn't in the same class. NeXT will do 32-bit Renderman I think this confuses the graphics system with what applications do with it. PIXAR is shipping a Mac version of Renderman that uses 32-bit QuickDraw. If you use 24-bit color to color windows, then you aren't buying much. In that sense, NeXT has the potential to do something different. Yup! > So what do y'all think! A low-end NeXT! ....reality, what a concept! NeXT has a similar problem to Apple with respect to the low-end. They have raised the base level so much that to call a machine a NeXT it needs a lot of stuff in it. Yeah, hoisted on there own petard!
These are the contents of the former NiCE NeXT User Group NeXTSTEP/OpenStep software archive, currently hosted by Marcel Waldvogel and Netfuture.ch.