aduncan@rhea.trl.oz (Allan Duncan) (01/14/91)
From article <17197@cbmvax.commodore.com>, by daveh@cbmvax.commodore.com (Dave Haynie): > > That tends to be true. You do get good Zorro II bus designs for the Amiga, > but since that bus is so slow, it's no big deal. Same with the basic ISA > bus, I imagine. VME has been around for a long time, and the mere existence > of PCs has forced the performance of VME devices up the scale of things, > both in performance and cost. And PCs have only recently grown 32 bit multiple > mastered buses, while most VME systems have been doing that for some time. So > sure, VME (and Multibus, etc) are more mature. They pretty much have to be, > at those prices. Not to forget that on the full blown VME systems there is the VMX (or whatever it is designated) to handle 32 bit local inter-board stuff, so that the VME bus is only being used for global transfers and I/O. The processor(s) and local memory accesses are kept of it. A bit like the Amiga with its chip and fast bus. _And_ the specs were worked out by the prospective vendors before they went to market, so it is free of mindsets, and most boards are compatible. Allan Duncan ACSnet a.duncan@trl.oz (03) 541 6708 ARPA a.duncan%trl.oz.au@uunet.uu.net UUCP {uunet,hplabs,ukc}!munnari!trl.oz!a.duncan Telecom Research Labs, PO Box 249, Clayton, Victoria, 3168, Australia.
daveh@cbmvax.commodore.com (Dave Haynie) (01/16/91)
In article <2586@trlluna.trl.oz> aduncan@rhea.trl.oz (Allan Duncan) writes: >From article <17197@cbmvax.commodore.com>, by daveh@cbmvax.commodore.com (Dave Haynie): >> And PCs have only recently grown 32 bit multiple mastered buses, while >> most VME systems have been doing that for some time. So sure, VME (and >> Multibus, etc) are more mature. They pretty much have to be, at those >> prices. >Not to forget that on the full blown VME systems there is the VMX (or >whatever it is designated) to handle 32 bit local inter-board stuff, so >that the VME bus is only being used for global transfers and I/O. There are actually a fair number of different mezzanine buses around, to support low cost add ons to VME and Multibus main boards. These are a good idea, because they help lower traffic on the main bus and make it much cheaper to expand a VME system. Though a good part of the reason for the cost factors is that VME is expensive to implement, when compared to the mezzanine buses. In fact, lots of newer desktop machines out there, like the Sun 4s and DECStations, have gone to high speed buses (S-Bus and Turbochannel, in these cases) which are more in line with the mezzanine cost than something like VME or Multibus. Zorro II and Zorro III are, as well, much less complex than a VME bus; its relatively easy to implement the bus connection portion of the card with a few PALs and TTL chips. To leave room for real work on the [much larger] VME card, you have to use a dedicated VME controller chip these days. >_And_ the specs were worked out by the prospective vendors before they >went to market, so it is free of mindsets, and most boards are compatible. That's exactly what we did with Zorro III. I wrote the specification before any real hardware existed (I was developing the bus controller at the same time), and both C= and the 3rd parties have to conform to the specs. It's always better to write everything down ahead of time, in one place, rather than spread it all around (as we did for Zorro II, "well, you need this, and the Motorola book, and it helps if you have schematics"), or just plain don't say and let everyone guess, as in the case of IBM PC-AT bus (or its more standardized "ISA" incarnation). >Allan Duncan ACSnet a.duncan@trl.oz -- Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests" {uunet|pyramid|rutgers}!cbmvax!daveh PLINK: hazy BIX: hazy "Don't worry, 'bout a thing. 'Cause every little thing, gonna be alright" -Bob Marley