Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!husc6!hao!ames!ptsfa!ihnp4!chinet!nucsrl!ram From: ram@nucsrl.UUCP (Renu Raman) Newsgroups: comp.arch Subject: Re: Question: on-chip or off-chip MMU? Message-ID: <3810035@nucsrl.UUCP> Date: Fri, 22-May-87 15:02:07 EDT Article-I.D.: nucsrl.3810035 Posted: Fri May 22 15:02:07 1987 Date-Received: Sun, 24-May-87 03:19:25 EDT References: <579@gec-mi-at.co.uk> Organization: Northwestern U, Evanston IL, USA Lines: 24 Another nice thing about optical computing is the ability to represent and compute in multiple levels of logic. Imagine ternary or other higher radices used for computation. Now, whether this is possible or not, depends a lot on the scalabilty of detector technology. Frequency separation and Intensity level separation provides two dimensional representation. Fourier transforms will be simple mixing of two light signals (That was a guess, based on my rusty knowledge of signal processing. So be nice if you think it warrants flaming). It might ressurect analog computation. If optical computing becomes commonplace, what would be the equivalent of MIPS and FLOPS? MQPS (Million Quanta per second :-)) ------------------- Renu Raman UUCP:...ihnp4!nucsrl!ram 1410 Chicago Ave., #505 ARPA:ram@eecs.nwu.edu Evanston IL 60201 AT&T:(312)-869-4276 P.S.1: The most corrupted version and abuse of MIPS is B(illion)IPS. "The Connection Machine (II) is benchmarked at 2.5 BIPS"