Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!bonnie.concordia.ca!uunet!elroy.jpl.nasa.gov!usc!jarthur!nntp-server.caltech.edu!toddpw From: toddpw@nntp-server.caltech.edu (Todd P. Whitesel) Newsgroups: comp.sys.apple2 Subject: Re: My decision for the LC Message-ID: <1991Feb14.111611.26037@nntp-server.caltech.edu> Date: 14 Feb 91 11:16:11 GMT References: <9102132324.AA00480@apple.com> Organization: California Institute of Technology, Pasadena Lines: 112 jpenne@ee.ualberta.ca (Jerry Penner) writes: >In article <9102132324.AA00480@apple.com> MQUINN@UTCVM.BITNET writes: >the Mac. There is a future in the 68000 architecture because it was >designed with the future in mind. The 65816 is a horrible kludge. I disagree. It is the most elegant kludge I have ever seen. It is however far better suited to small and tight controller applications as opposed to general purpose programming (for which the 680x0 is indeed superior). The Super Nintendo uses a 65816 and its specs indicate that it will pound the Genesis and TurboGraphix (both 68000 boxes) into the dirt when 16 bit game machines finally oust the original Nintendo (which is 65c02 based). GS programmers, take heart! Get a job programming the Super Nintendo. 65816 at 7 mhz, kick-butt coprocessors for everything at your command... it's what the Amiga should have been but without the keyboard! Apple and NeXT will soon be the only companies whose flag product uses the 680x0 series. Apple is moving to Moto's 88xxx series and probably the ARM chipset. Both will leave the 680x0 in the dust (high end, and low end). >it had been designed more with a separate 16-bit mode (ie. no more >6502 instructions, just similar) it might have been worth taking a lot >further. If you've read the article Randy Hyde posted about what the >65816 could or should have been, you know what I mean. If you haven't >he basically outlined a completely new instruction set for the 16-bit >mode with emulation of a 6502. Randy Hyde's article (frankly) disgusted me. The bus architecture of the 65816 prefers a small and tight instruction set, whereas the bus scheme of the 680x0 prefers a larger and more powerful instruction set. Mixing the two (large instructions on a small bus & register set, or few instructions -- not RISC but way too few instructions -- with a wide, slow bus) does not easily produce efficient results. (I know there must be exceptions to this general statement). I claim that good 65816 compilers are possible (the Apple guy who worked on HyperCard IIgs posted recently and mentioned that MPW IIgs pascal had an awesome new code generator), and I have some ideas for a C code generator that I believe will produce excellent code -- but that doesn't count because I don't have time to try it and I don't have a job at ByteWorks. >But I digress... What I am saying is if you were a manufacturer who >was trying to put power in people's hands, you wouldn't want to make a >choice that was going to limit you in the future. I think Apple has >done that. Given Bill Mensch's track record I must agree to this. It comes down to an "I wish things had happened differently" but they didn't. What I blame Apple for is for not taking charge. They had the capability to do what the ASIC 65816 guys claim to be doing years ago -- I am not joking -- but did not consider the Apple II worth the effort, and I disagree. > Apple gives you the video overlay card, the high >speed DMA SCSI card, (all Apple IIs) and if you have a GS, new system >software, Hypercard, and ongoing technical support even to those who >don't pay for it. What more should they really do? Think carefully >about what you would do if you were Apple. What can you do with the >GS that's not a bunch of new hardware kludges? I can do a lot, Apple's hardware team could do a lot, given the ASIC resources alotted to two Mac CPUs and not much else. It'd take a lot more time for me to do it alone, of course, and I'd need access to Apple Internal specs for the Sound GLU, the SWIM, and the RTC interface section of the VGC. (Actually, I believe the Mac Hardware Reference describes enough of the RTC interface to design an interface for it.) The other custom chips (FPI, Mega II, Slotmaker, VGC, and the ADB/Keyboard GLU) could be totally redesigned and combined in order to simplify the logic board, add features, reduce chip count and overall logic complexity, and so on. While I cannot guarantee perfect hardware compatibility, I believe I can do it where it counts without violating the goal of the design effort -- an elegant reimplementation and improvement. > The architecture just doesn't lend itself to upgrading. I'm sorry, but that can't be true. Look at the IBM PC. They did what I'd love to see done to the IIgs and gee Mr. Wizard it worked. >Yes, their prices are high, and have always been that way. I don't >like their idea of telling me what to buy by the pricing strategy but >if I didn't own a computer now and wanted to buy a new one, I sure >wouldn't buy an Apple II. Neither would I, but I wouldn't buy a Mac either. I'd get a NeXT pizza box or a Sun. At least they run unix properly. >Since this discussion was about LC/GS's, I have to agree. I think the >LC is a crippled Macintosh. (maybe like the crippled GS was in '86) I'll agree with that. If I did have to buy a Mac, the si would be minimum, preferably a ci because the si has no FPU. >You really think that's why they created Apple? More like Woz liked >giving people something fun to play with and do cool things on and >Jobs wanted to make piles of cash. Apple's have always cost lots. I agree with this. Read Steven Levy's book _Hackers_. The original reason the paddle inputs were added was so he could play pong on it... > Sticking to the II line just because I like a lot of things about it means >I'm missing a lot of nice things about other systems. If you are primarily a user, sure. I however can't believe the amount of crap in the Mac O/S. While the IIgs still lacks many practical features of the Mac O/S, its overall design is infinitely cleaner and (ironically) more powerful in the long run. Of course there were mistakes, and it's too late to completely fix most of them. But most of these (like the Mac's) can be worked around or avoided. As far as the mass market is concerned, it is too late. I believe however that a ROM 04 done right could allow the Apple II a dignified death and give the schools the LC they really need. Todd Whitesel toddpw @ tybalt.caltech.edu