On Fri, 19 Mar 1999 [EMAIL PROTECTED] wrote:
> Marco Antonio Simon dal Poz <[EMAIL PROTECTED]> wrote:
>
> >> My project with MegaRAM uses a double-bank of mapped
> >> registers (16 bytes select ever 8kb block).
> >And how do you think you'll make the registers? Will you use the even
> >addresses to select LSB and odd addresses to select MSB?
>
> The great idea is more or less this:
> OUT (8Fh), A disable G-RAM (or enable ROM)
> OUT (8Eh), A enable LSB exp. address for bank switch
> OUT (8Ch), A enable MSB exp. address for bank switch
That's not a good idea, because for each block switch with distance higher
than FFh you'll need 2 OUTs and 2 LDs.
> >I prefer "Circuit Cellar",by Steve Ciarcia
>
> I collect copies of circuit cellar on BYTE small systems
> journal (USA/1979~1988). Thanks for remind !
I don't know Circuit Cellar publications on BYTE magazine. I was talking
about http://www.circuitcellar.com/
> >If you don't remember, Steve Ciarcia is the writter of the book called
> >"Construct your own Microcomputer using Z80", McGrawHill.
>
> I have a copy, too. :))))
> And plan to make a PCB for this schemes.
That will be very big!
> >>The most mysteriuos MegaRAM's feature is the mirror :)
> >>MSB of address bus (A15) is ignored by MegaRAM!
> >Yes, the creator of Megaram (Ademir Carchano) told me that he
> >ignored A15 to make the hardware much more simple,
> >and to fit in a standard cartridge box.
>
> Jeannie...(ops!) Ademir... is a genious!
I don't think so. He is a guy that had the chance of have a good
documentation and very recent equipments, and much more knowledge about
hardware than the most part of people in that time.
> (Blocks marked with "'" are mirrors of same block)
Right.
> >That can happen when you do a reset while Megaram is in "write enable
> >mode". But I don't know how much memory will be detected as normal RAM.
> >Does the BIOS verify if the mirror effect happens?
>
> Ok, the CHKRAM started on E000h, and decrease counter
> when search RAM... don't know how more is effective.
Does anybody know exactly how the system search for slots in which there
is RAM?
> >> MegaRAM are "RESET-insensitive". See technical informa
> >> tion and electrical diagram on CPU MSX Magazine nr 35.
> >Don't trust in that magazine! I'll analyse the contents of those
> >schematics to see if there aren't any bugs.
>
> The scheme do not use /RESET signal for clear MegaRAM
> registers. I'm based on CPU MSX scheme.
That's right. But I see that schematic doesn't use the BUSDIR signal. Does
anybody know how the BUSDIR signal should be used?
> >>> Simple and good method: select (in descending order) each possible
> >>> block, write its number in it in at some test address, and then
> >>> (starting with block 0) check up until which block the block number
> >>> matches what you find at the test address when selecting each block.
> >> Many faults for a test program...
> >Then show us which faults you are seeing!
>
> Checking for multiple-redundance nodes, and check for
> individual address bus lines minimize time on search
> size of available RAM.
What do you mean with multiple-redundance nodes? And checking for
individual address bus lines only works for Megarams (or Mappers) with
power of 2 kilobytes (starting with 16kb).
> Think how much time CPU need for measure the entire
> RAM connected at slots. This way is unreasonable!
Very few time! That's perfectly possible.
> >> I shop 4 ICs WD2793BL... but no have schematics for!!!
> >A friend of mine sended to me a copy of the port based disk interface, and
> >I saw that it really uses WD2793 FDC.
>
> The same friend send to me the copy of diagrams of this
> interface, too. But I don't know to make adjust.
I think that with the schematic, you can easily create an interface. What
don't you know?
> >>> But one of the difficulties is to understand how the all kinds
> >>>of MSX hardware work, because the variability is really big
> >>I think 1st priority detect all kinds of RAM.
> >That's the easy part.
>
> Not so easy. Several kinds of RAM is available:
> Cartridge 64kb RAM standart, MegaRAM, MegaRAM-Disk,
> Ext. Mapper, (new)G-RAM, 8/16/32kb SRAM modules...
I said that's the easy part because the others are much more difficult.
> Found size of RAM isn't all of test...
> I need mark the Wait cicles required for read/write on
> any kinds of RAM available on system.
That's not need and not possible! That's a hardware characteristic, and
can't be software controlled.
> I collect tricks for detect this kinds.
And I research tricks for that!
> >> Wait for G-RAM(TM) (MegaRAM with 16-bit registers)
> >> padronization. Will alocate 512Mbytes RAM/each slot.
> >Is the "G" from "G-RAM" a abbreviation of "gambiarra"? :-)
>
> <.br>
> "G" vem de GODZILLA, ficou bacana ?
Yes, but I think that "gambiarra" is much more funny!
> Mas nao espalha meu, senao posso me dar mal ! :))))
Mas voce ja' espalhou aqui!
> </.br>
Greetings from Brazil!
-----------------------------------------------------------------
Marco Antonio Simon Dal Poz http://www.lsi.usp.br/~mdalpoz
[EMAIL PROTECTED] "Apple" (c) Copyright 1767, Sir Isaac Newton
/"\
\ / CAMPANHA DA FITA ASCII - CONTRA MAIL HTML
X ASCII RIBBON CAMPAIGN - AGAINST HTML MAIL
/ \
****
MSX Mailinglist. To unsubscribe, send an email to [EMAIL PROTECTED] and put
in the body (not subject) "unsubscribe msx [EMAIL PROTECTED]" (without the
quotes :-) Problems? contact [EMAIL PROTECTED] (www.stack.nl/~wiebe/mailinglist/)
****
