brianw@microsoft.UUCP (Brian Willoughby) (09/30/89)
In article <CMM.0.88.622959499.cwilson@naurs2.NISC.SRI.COM> 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
crash@pro-colony.cts.com (John Stephen III) (10/02/89)
Network Comment: to #10579 by microsoft!brianw@uunet.uu.net Bill Mensch has been able to produce SOME 10MHz '816s been he has never been able to produce them in quantity. Our TransWarp GS board is capable of running at 10MHz+ speeds and, as many TWGS owners know, we have a hard time just getting chips that'll run at 7MHz! We've had some samples of 10MHz parts and we demonstrated a TWGS running at 10MHz at the Boston Applefest. Even these parts, which WDC rated as 10MHz, would not really run as specified. The power supply voltage needed to be boosted to 5.5v and small amounts of heat would caused them to crash. The problem with the faster chips is not in the timing between the TWGS and processor -- the problem is that certain instruction combinations will not work at the rated speed (this is true all the way down to the 6MHz (or whatever) parts that really are available) and the TWGS board "watches" for these combinations and slows down for them. In short (in long?), faster parts are in the future but we have not seen them. I wish Mr. Mensch the best of luck in producing them as well (I'd certainly like to have my //gs running at 10-20MHz!). One other interesting note - another company (I forget the name) is working on their own 816 which they are estimating will run in the 15-20MHz range. Should be interesting... John Stephen "These opinions are my own, my employer doesn't endorse them, blah blah blah" nnnnn | / . . \ | UUCP: [ ihnp4 sdcsvax nosc ] !crash!pnet01!pro-sol!pro-simasd! (| u |) | !pro-colony!crash \ \_/ / | ARPA: crash!pnet01!pro-sol!pro-simasd!pro-colony!crash@nosc.mil \ / | ProLine: crash@pro-colony /-->=<--\ |
lbotez@pnet02.gryphon.com (Lynda Botez) (10/03/89)
John Stephen writes: >One other interesting note - another company (I forgot the name) is working >on their own 816 which they are estimating will run in the 15-20 mhz range. >Should be interesting... Sure should be interesting. I heard the same thing... I think I remember hearing that this company will be working with Zip Technologies. Anyone care to elaborate? Also, in regard to the faster chips on the TWGS; it's true, when the chip runs at 9 or 10 mhz, some things crash... it would be advantageous to be able to set it to run at 7 mhz OR 10 mhz. Lynda UUCP: {ames!elroy, <routing site>}!gryphon!pnet02!lbotez INET: lbotez@pnet02.gryphon.com