Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!seismo!gatech!hao!ames!ucbcad!ucbvax!unisoft!lll-lcc!well!spg From: spg@well.UUCP (Stephen P. Grant) Newsgroups: comp.sys.amiga Subject: Re: flakey memory boards/CHIP contention Message-ID: <3117@well.UUCP> Date: Fri, 22-May-87 22:55:12 EDT Article-I.D.: well.3117 Posted: Fri May 22 22:55:12 1987 Date-Received: Sat, 23-May-87 19:13:34 EDT References: <8705110100.AA27374@cogsci.berkeley.edu> <3077@well.UUCP> <1776@te Lines: 52 kigm2.TEK.COM> Sender: Reply-To: spg@well.UUCP (Stephen P. Grant) Followup-To: Distribution: Organization: Whole Earth 'Lectronic Link, Sausalito, CA Keywords: In article <1776@tekigm2.TEK.COM> phils@tekigm2.TEK.COM (Philip E Staub) writes: >It's suggestion/comment/question time. Does anyone think it feasible or >desirable to develop an expansion connector plug in board which passes a >buffered, self-powered (i.e., this board would have it's own power supply or >a connector for an input from an external power supply) version of the >expansion bus out the other side? > >Yes, yes, I know the standard answer: buy an expansion box. But I can't help >but feel that there have to be people out there who bought (86 pin) bus >expansion boards, thinking they would "never" have a need for any more >than one or two plug in boards, then find themselves in a bind if they >need three or four and can't get enough drive on the bus pins to make all >the boards work at the same time. The standard answer is an expansion chassis OR a properly designed buss passthru. It is certainly feasible to design and build a "buffer" card complete with power supply, but it makes little sense economically. This buffer card is essentially a "Zorro" interface without the slots. It would have to buffer ALL of the signals on the 86 pin connector with transceivers (74F245s). It would need to provide buss steering logic. It should have a power supply that provides +5 volts and probably +12 and -12 volts. Add in a good printed circuit board, a few connectors, capacitors, a case for the whole thing, and FCC approval and what you have is an expansion device with no expansion that will have to sell at the retail level for $100 to $200 or more. Would you be willing to pay for this? I submit that few people will. >Or maybe there's already such a product. (If so,...........NEVER MIND 8-) ). > >Phil Earlier this year, Cltd and Microbotics where supposed to be working on this type of product to fix the problems that they created by not buffering the buss on their implementations of a passthru. Since then, nothing seems to have come of it. Why? Because they probably found that it was impossible to do economically especially when you consider that there are at most 150K Amiga 1000s in the world and little hope for more. Stephen P. Grant Engineering Manager Access Associates