Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!uunet!microsoft!brianw
From: brianw@microsoft.UUCP (Brian Willoughby)
Newsgroups: comp.sys.apple
Subject: 10 MHz 65C816 chips (was Re: Applefestive)
Summary: Also, tricks with the TransWarp
Keywords: AE TransWarp
Message-ID: <7907@microsoft.UUCP>
Date: 30 Sep 89 01:55:37 GMT
References: 
Reply-To: brianw@microsoft.UUCP (Brian Willoughby)
Organization: Microsoft Corp., Redmond WA
Lines: 74

In article  cwilson@NISC.SRI.COM (Chan Wilson) writes:
>]>Someone from Western Digital (I'm pretty sure it was Bill Mench (sp?)
>]>asked him why Apple hasn't had more commitment towards faster 65816's.
>]
>]It was indeed Bill Mensch, and the name of his company is Western Design
>]Center.
>
>Does anybody know the truth behind the claims he made that he has
>lots of 6-10mhz 65816 chips? And is this true?
>
>] --Dave Lyons, Apple Computer, Inc.          |   DAL Systems

I placed a phone order to Western Design Center on July 14.  At the time,
65C816's were not to be had, I was told that they sell out of every batch
they get.  I can't remember if a 10 MHz '816 existed at that time, but a
10 MHz W65C802 was so hot off the presses that the lady didn't even have
pricing information on them.  Lucky me, she just charged me 8 MHz prices,
which was $24.84 for the cheaper plastic chip packaging.  I needed to
fill a $50 minimum order anyway.

I don't know, is three months enough time for manufacturing to ramp up?
I have a shadow of doubt that they have 'lots' now, since Apple and AE
aren't the only companies buying them.

Michael Steele mentioned that AE claims chips stamped 12 MHz are only
running at 10 or 8 MHz, but my 10 MHz '802 arrived with a little sticker
on it identifying it as 10 MHz.  It seems that the LAST step in their
production is changing the numbers stamped on the chip.  If AE is having
trouble with 12 MHz chips only running at 8, then their circuitry must
rely on extra critical timing.  WDC publishes all of the timing
measurements at each processor speed, and I'm confident that their 12 MHz
chips match their timing specs (although I have no way of verifying
this).  That doesn't guarantee that a 12 MHz '816 will work in every
circuit clocked at 12 MHz.  There might be something about the chips that
are used in interfacing to the processor which aren't cooperating.

While I'm on that subject (and hopefully have the attention of interested
hardware folks), would anyone care to hazard a guess as to whether or not
I can get this hot little 10 MHz '802 running a little faster than the
current 3.58 MHz that my TransWarp (II Plus) is doing?  My first thoughts
were that the 3.58 MHz speed used implies that the CPU must still be
easily synched up to the 1 MHz clock for video and Apple II bus
transfers.  If you recall the ancient II circuitry, the 3.58 MHz signal
on the Apple bus is derived from the same clock that the 1 MHz clock is.
That means that the edges of the 3.58 MHz clock always line up with the
1 MHz clock, and I assume that that fact makes it easier to switch back
and forth between clocks as the TransWarp does fast RAM accesses
interspersed with slow Apple I/O.  My next logical assumption was that
the 7.16 MHz signal would also satisfy this criteria, and double my speed
(although I probably would still lust after a full 10 MHz).  I talked to
Matthew Stier about this idea for a short while, and he said I should
give it a try.  Matthew said that there is probably a clock divider
anyway, but I still haven't checked out the circuit.

I would have asked AE these questions, but as I've been told their
TransWarp engineer (what, only one?) is out on assignment.  I found this
out when trying to determine how I could access ALL 256K on the
TransWarp.  128K is available in the standard, published Apple //e
interface (even on my Plus!), and another 12K is used to shadow the old
Apple bus-speed ROM chips in fast RAM, but the remaining 116K is just
sitting there unused.  I *have* found a few 'secret' unpublished
softswitches for the TransWarp - do any of you programmers out there want
to share discoveries?  Since I have both Integer and Applesoft Basic in
ROM (through a switch added to my keyboard), I was happy to find out what
soft-switches to select to re-load the shadowed ROM without powering
down.  Have any of you hackers figured out how to make that extra 116K
into a RAM disk?  Seems that AE is not providing the information
or documentation for full utilization of the hardware I paid for :-)

Brian Willoughby
UUCP:           ...!{tikal, sun, uunet, elwood}!microsoft!brianw
InterNet:       microsoft!brianw@uunet.UU.NET
  or:           microsoft!brianw@Sun.COM
Bitnet          brianw@microsoft.UUCP