Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!microsoft!brianw From: brianw@microsoft.UUCP (Brian Willoughby) Newsgroups: comp.sys.apple Subject: Re: Apple's committment to the // line Message-ID: <8078@microsoft.UUCP> Date: 17 Oct 89 04:17:13 GMT References: <8910140920.AA29715@trout.nosc.mil> Reply-To: brianw@microsoft.UUCP (Brian Willoughby) Organization: Microsoft Corp., Redmond WA Lines: 77 In article <8910140920.AA29715@trout.nosc.mil> sschneider@pro-exchange.cts.com (The RainForest BBS) writes: > > >Why would =ANY= company (in its right mind) care to continue to develope >for an obsolete what it available today> computer such as the Apple // when they can spend the >same amount of money doing more with what is now state-of-the-art??? > >/steve Obsolete? The Compaq 20e on my desk here at work uses RAM caching because the CPU operates faster than normal DRAM. I know for a fact that the 65C8xx is capable of executing at speeds too fast for DRAM, but I don't know of any Apple II that employs cache technology. I think the reason that Apple's II line is slow is because they haven't taken the time to use their chip design teams to put the same kind of support circuitry in an Apple that is standard in high speed computers. The TransWarp GS has a cache (so I understand), but who would buy the Compaq 20e if only a third party provided the necessary hardware for high speed? On the subject of RAM slowing down the processor, this is one of the bottlenecks in the IIgs design. The Apple Mac SE/30 employs dual-port RAM to allow the 68030 and the video to BOTH access RAM without waiting. Apple has the technology for the Mac, where is it for the II series? The Apple II CPU is not obsolete, and you've probably read the other postings about how clock speeds offer no guideline for direct comparison between different CPUs. I'll bet that you can't show me a fast machine from anyone else that doesn't use at least one of the following: Cache, Virtual Memory, DMA. Granted, many other CPUs have these functions built in, but in ALL cases these support functions evolved as EXTERNAL chips. Only after they became accepted did Motorola and Intel begin building them in. Those two companies have enough demand to develop such integrated CPUs. But if you look closely at even some of today's IBM PC clones, you'll see that there was once a time when the 80x86 couldn't stand on its own without both an interrupt controller chip AND a clock chip - which means that it took a 3 chip CPU in the 4.77 MHz PC/XT to run as fast as a single chip 1 MHz 6502. In fact, it is these two chips (and a few others) which DEFINE a PC compatible. I should know, because we at Microsoft have to deal with the shortcoming of this architecture while designing OS/2 to multitask on this limited platform. The Catch-22 is that the Western Design Center cannot design these support chips alone without the revenues that Intel and Motorola enjoy. Without the support to make the chip more attractive, the revenues won't be there. Kinda sucks, don't it? Where does this leave the Apple II? Well, I wouldn't call it out of date unless you could compare EQUALLY EQUIPPED systems. It is not the CPU or clock speed (although Apple could increase thier standard speed to match AE's), but the lack of support hardware which strangles the line. There is the added fact that people are crazy enough to demo IIgs's at the computer stores WITHOUT A HARD DISK. When is the last time you WAITED for an IBM to RUN PROGRAM FROM A DISK-ONLY SETUP? At least the Apple II is faster in this area. My opinion? Apple has designed Dynamic RAM Controllers (the //e MMU), integrated all the I/O on one chip (IOU), and recently put the entire Apple II on a chip (MEGA II). If their very capable VLSI team, who have obviously had enough practice and experience, would turn to NEW CIRCUIT IDEAS instead of repackaging the original Apple I (sic) circuit - then perhaps we could see 65C8xx support chips like Virtual Memory and Cache Controllers from Apple. From what I understand of VLSI and ASIC, it must be incredibly cheap for Apple to do this considering the investment in on-site development equipment and the sheer volume of chips they have made so far. Hey, I tried to get into the Apple hardware group when I was fresh out of college, but received the 'try again after you've worked a little and have some experience' response. I'd love the opportunity to put some meat into my ideas :-) hint Brian Willoughby UUCP: ...!{tikal, sun, uunet, elwood}!microsoft!brianw InterNet: microsoft!brianw@uunet.UU.NET or: microsoft!brianw@Sun.COM Bitnet brianw@microsoft.UUCP