home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!gossip.pyramid.com!olivea!isc-br!bunker!nuconvex!starpt!doiron
- From: doiron@starpt.UUCP (Glenn Doiron)
- Newsgroups: comp.sys.amiga.misc
- Subject: Re: Problems with the A4000
- Message-ID: <doiron.0cnc@starpt.UUCP>
- Date: 14 Sep 92 21:51:03 GMT
- Organization: 68K Software Development
- Lines: 197
- X-NewsSoftware: GRn 1.16e (7/4/92) by Mike Schwartz & Michael B. Smith
-
- In article <1992Sep12.083819.17967@news.iastate.edu> barrett@iastate.edu (Marc N Barrett) writes:
- > For about a week or so, I've been promising to post my detailed explanation
- > of exactly why Commodore's use of a super-cheap IDE HD interface in the A4000
- > has very likely pushed the initial list prices higher than what they would have
- > been had SCSI been used.
-
- Tune in next week for Marc's explanation of why Surface Mount Technology
- pushes price higher than conventional assembly methods.
-
- > The problem comes down to the architecture of the A4000 itself. As has been
- > rumored before, several good systems were canceled not long after DevCon '91,
- > a year ago. What has not been said is that these included all of the systems
- > that Dave Haynie designed that incorporated the AA chipset. Indeed, none of
- > the new Amiga systems -- including the A4000 -- were designed by him. They
- > were designed in a rush by junior engineers at Commodore.
-
- Whose names Marc will withhold because He Doesn't Know.
-
- > The whole design of the A4000 can be summed up in a fairly short sentence:
- > the A4000 is an A3000 with the CPU moved to a card, the AA chipset kludged in,
- > and with the SCSI subsystem replaced with IDE.
-
- Here it is folks- the great Marc Barrett facts.
-
- > I will get back to the issue
- > of SCSI & IDE in a minute, but I would first like to explain what other
- > problems this has.
-
- > Basically, the A3000 was not designed for the AA chipset.
-
- Well, one fact in an entire article is pretty good for -MB-.
-
- > The Ramsey and Buster chips in the A3000 were designed for use with the ECS,
- > in particular the 2M Agnus chip.
-
- Ramsey and Buster have nothing to do with ECS. Ramsey controls the fast
- ram arrays/SCSI address generation. Buster controls the Zorro3 bus.
- Neither Ramsey nor FAT Buster (which is the correct name for the chip in
- the A3000) have anything to do with ECS, indeed they are only connected to
- the local CPU bus (and Ramsey is, of course, connected to the motherboard
- fast RAM). Chip RAM arbitration is somewhat complicated, but I'll
- simplify:
-
- 1. Chip RAM is made up of 2 16-bit (512k or 1meg, depending on how much
- chip RAM you have) banks. Special "Bridge" circuitry electrically connects
- the two data busses together. When Agnus wants to access the "near" 16-bit
- bus the bridge circuits do nothing. When Agnus accesses the "far" 16-bit
- bus the bridge circuits kick in, moving the "far" 16-bit data onto the
- "near" bus for Agnus to get at. When the CPU accesses chip RAM, *BOTH*
- busses are accessed SIMULTANEOUSLY (bridge & cpu logic are provided via the
- morass of PAL's on the A3000's motherboard). So, now you can see, Ramsey
- and Buster have *not a thing in the world* in common the the ECS chips,
- except that they might be in the same machine.
-
- > The new Alice chip in the AA chipset -- the
- > replacement for Agnus -- can potentially address 4M of chip RAM.
-
- Sources please? Do you have Barrett (TM) X-RAY VISION that lets you see
- schematic diagrams thousands of miles away in Westchester, PA? Or perhaps
- you glanced a peek at the top layer of a multi-layer motherboard and
- deduced all the traces?
-
- > But since
- > the Alice chip has been kludged to work with other chips intended for the
- > 2M Agnus chip,
-
- Err.. yeah, whatever. You should check your "inside" sources at Commodore
- again.
-
- > none of the new Amigas can use more than 2M of chip RAM.
-
- This may or may not be the case of the Alice chip. Since I don't have any
- technical specs either, I'm not going to overtrump your BS with my BS.
- Suffice it to say that I as a developer have received *NOTHING* that states
- that Alice can address 4 megs of RAM.
-
- > The
- > use of the A3000 architecture also has other problems, namely the
- > restriction of 16M of fast RAM on the motherboard.
-
- Yes, but how that suddenly became a restriction or why you can't use the
- ProRAM-3000 boards in the A4000, only you know...
-
- > This was not a problem
- > several years ago with the A3000 was designed, but 16M is not all that
- > much anymore, with RAM prices at $34/megabyte for 60nS static-column memories.
-
- Duh, like it would be cheaper to redesign Ramsey to address more RAM. Few
- people even have the whole 16megs that they can in their machine.
-
- > OK, I've established that the A4000 is an A3000 with the AA chipset kludged
- > in.
-
- No, you've established that you have absolutely not a (ahem) CLUE about
- what you're talking about. But, I digress. You've already proven that
- many times.
-
- > Now it makes sense to me that it would have cost less in development costs
- > to simply leave the A3000's SCSI design there instead of removing it and
- > replacing it with something else.
-
- Pretty ridiculous. So your whole bullshit argument revolves around the
- bullshit which you've tried to prove with bullshit. Eeew, what did I just
- step in?
-
- > The costs of including SCSI in the A4000,
- > then, would have been practically zero in development costs.
-
- Just about as much as including IDE, maybe more, considering what IDE
- really is. All the s/w work was done by Randell Jesup for the A600 anyways.
-
- > In addition to the costs involved in removing the SCSI subsystem from the
- > A4000's design,
-
- ...this should read: "...involved in not putting SCSI in there in the first
- place"
-
- > though, we also have to consider the costs involved in
- > designing the IDE design. These costs would have been practically zero if
- > the A4000 had simply used the same IDE design as the A600.
-
- Hardware-wise it's trivial. Software wise it's trivial too- all the work
- was already done on the A600.
-
- > The problem is
- > that it does not -- the IDE design used in the A4000 is a whole new design.
-
- Wooooooooweeeeee. Big farts. Even if it is a "new design", it still costs
- less to put IDE on than SCSI, although I don't agree with the decision on a
- design level myself.
-
- > The new design was created by Randall Jessup, and includes extra hardware to
- > make the IDE interface and HD look somewhat like a SCSI interface and HD to
- > Amiga software.
-
- All that "new design" is in fact "old design" because it was already done
- on the A600 which was also announced at the show.
-
- > This is a reasonable effort to make the most of a bad
- > situation, IMO,
-
- No, my reply is a reasonable (IMHO) effort to make the most of a bad
- situation (more -MB- facts).
-
- > but the simple fact is that a design like this must have cost
- > quite a lot to develop,
-
- New machines do have a tendency to be that way.
-
- > and would not have been necessary had the A3000's
- > existing SCSI design simply been left in the design for the A4000.
-
- Since your point is moot (as pointed up above), this statement simply does
- not apply.
-
- > As Dave Haynie and Randall Jessup have already pointed out, IDE -- even the
- > new one in the A4000 -- costs next to nothing in terms of manufacturing costs.
-
- Therefore it will cost less to produce each unit.
-
- > But the development costs involved in creating this IDE design must have been
- > very substantial compared to the costs that would have been involved in
- > simply carrying over the A3000's existing SCSI design to the A4000.
-
- Development on SCSI was already done. Development on IDE was already done.
- Since carrying over SCSI would mean carrying over extra chips like the
- WD33C93A/SuperDMAC/glue logic, it would be more expensive than the itty
- bitty piece of space on a chip they use for IDE logic.
-
- > These
- > development costs are still costs, and Commodore is going to want to recoup
- > these costs as quickly as possible.
-
- Yep. Expensive first, cheaper later.
-
- > This means boosting the list prices
- > to pay the development costs, and then lowering the list prices later as
- > soon as these development costs have been payed for.
-
- Duh.. was Economics one of those courses you weren't sleeping through?
-
- > In the short-term, then,
- > the prices on the A4000 and other new systems will be higher than if they
- > had used a SCSI design.
-
- Yeah, sure. Whatever.
-
- > ---
- > | Marc Barrett -MB- | email: barrett@iastate.edu
- > --------------------------------------------------
-
- Glenn Doiron
- --
- Amiga UUCP+
- Origin: uunet!starpt!doiron (Organization:68K Software Development)
- BIX: gdoiron
- ** Not enough memory to perform requested operation. Add 4 megs and retry.
-