Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site brl-tgr.ARPA Path: utzoo!linus!philabs!cmcl2!seismo!brl-tgr!tgr!GUBBINS@RADC-TOPS20.ARPA From: GUBBINS@RADC-TOPS20.ARPA (Gern) Newsgroups: net.micro Subject: Re: NEC V20 ---> 8088 Message-ID: <1603@brl-tgr.ARPA> Date: Thu, 19-Sep-85 09:51:39 EDT Article-I.D.: brl-tgr.1603 Posted: Thu Sep 19 09:51:39 1985 Date-Received: Sat, 21-Sep-85 11:41:38 EDT Sender: news@brl-tgr.ARPA Lines: 43 Answers to your questions: No, I do not have a NEC floppy disk controller. The Z-100 (a completely different, orginal design machine and far superior to the IBM/Clones) uses a Western Digital 1797. IBM liked the NEC because it was CHEAP. Yes, I have several friends that work at IBM that know these things as well as friends in the Reliablity Division here (we are a USAF research and development lab) that would have all sorts of nasty things to say about NEC chips, if only they were allowed to say such things. A just posted message has just confirmed (at least a rumor) that NEC did indeed pirate the Intel 8088, right down to the early design mistake in Intel's microcode (I did not think that the 8088 was a microcoded CPU?). Yes, I have a CMOS real-time clock calendar with auto-leap year in a device of my own design. Working/interfacing with CMOS is a pain DON'T TOUCH IT - IT DIES and the incompatible logic levels (pull-up, pull-up...) and the slow speeds. The very low power consumption makes CMOS ideal for battery backup/battery powered applications such as battery backup clock-calendar chips and the DG/One lap computer. Sticking a CMOS CPU in a mostly TTL machine, especially in a borderline spec designed machine such as the real IBM, may case line driving problems. Also the IBM design (again) is well known for its internal heat problems and CMOS ICs age very rapidly as heat increases, much more so than NMOS or TTL (in that order). SO it may work now, but maybe not a couple of years from now. CMOS ICs that are ALMOST TTL level/ TTL drive level/TTL speed level do so at a large increase in power consumption. Granded, not as much as TTL, but the future is not CMOS for high speed operations. The inherent properties of CMOS greatly limit its speed. And I still hope Intel wins... Cheers, Gern [The randoms in the message are my own opinions and usually that of my friends and other sane persons too and may or may not be those of the USAF, which won't tell] ------- Brought to you by Super Global Mega Corp .com