Re: DolphinDOS information needed (addendum)

On Tue, 30 May 2006 17:56:57 +0200, silverdr
<silverdr@xxxxxxxxxxxxxxxxx> wrote:

There was a product, which utilised a different approach than most of
the speeders, including DD. It had "much more" RAM and used it as a
buffer for the whole disk, rather than a single track. That buffer was
filled with the disk contents upon inserting every diskette. All later
transfers were done from the RAM cache, which was very fast as it
exposed neither mechanics related latencies nor was limited by the
diskette's bitrates. The drawback was that it required to re-buffer
every diskette upon insertion and this process took some 15 seconds or so.

I read up on that, but have never actually used it myself. It seems
to me that the best way to handle something like that is to start
reading the disk into RAM when the disk is inserted. If the disk is
accessed, it pauses the read-to-RAM, reads what is being asked for
(storing it in the appropriate location in RAM at the same time,) then
go back to where it left off. From what you wrote it doesn't seem to
do this.

Also,
since I just cannot find the original for sale (new or used,)
anywhere, the only option I have left is to find someone willing to
send me images of the ROM.

I have the ROM contents for DD2 and the drive side of DD3. I lack the
KERNAL of DD3 but DD3 drive works with DD2 KERNAL.

I already have the DD2 C64 & 1541 ROMs. A friend supplied me with
the DD3 C64 ROM, and a partial of the DD3 that should have been
enough. Couldn't get it to work in Vice128 though. The only thing I
don't have access too is the DD3 1541 & DD3 1571 ROMs. I'd also
dearly like to have DD2 ROMs for those as well. I don't know if they
exist though. From what I've found on the web, it's "possible" that
DD2 came out for the 128 & 1571 prior to being upgraded to DD3. DD3
came out right after the 128DCR came out, and my DD2 manual has
instructions for installing into a 128 (2 kernal chips,) if that is
the version I bought.

I managed to track down the distributor
and was basically told to knock myself out, as nobody there knows
anything about it. Do you know a way to mount the mc6821 where DD3
can use it from the expansion board?

Nicolas AFAIK didn't do it and was actually interested in having it too.
And the reason why I still didn't do it either is that I used a lot of
effort to put the DD3 board into one of the 1541-II drives with good
amount of attention to the details how it's positioned, mounted etc. and
am now somewhat lazy to disconnect all the stuff, remove the drive and
free the board again, invalidating the previous work ;-)

I don't blame you :) The only reason I'm willing to open my 1541
and pull the DD2 expansion is because somehow the parallel cable was
destroyed in storage. No shorts, it seems to have been cut fairly
cleanly right after it comes out of the drive. I cannot even find the
User Port plug end of the cable :( One of the reasons I was looking
into the RAMROM card. Get the RAM, use Joe Forster's instructions to
connect up my own parallel port, and away we go :) The only problem
with that is that the original DD2 user port plug had pass through to
allow multiple drives to be attached, and still attach something else
to the port. Talking it over with others, I gave up the idea of just
daisy chaining the parallel connection, so I'm now looking into
replicating the way DD2 did it (to protect the VIA chip.) I just
don't know how.... Someone mentioned a User Port expander card (allows
3 or 4 devices, much like the cartridge expander,) but haven't found
any yet.

I've managed to track down 3 stock kernal sets now (for both types of
kernel setups - the 2 16k chips and 1 32k chips.) One is the US
version, 1 just says DIN, the other is Swedish. The only difference
between the US & DIN version is a linkage table that just FF filed in
the US version. From what I've been able to read (and not allot,) it
was an International version, and is what can be found in most German
sold systems. It's possible that that chipset (in it's entirety could
be used in my system.) Since the link table is for the chargen chip,
I'd have to replace that one as well.

Have you checked this? I'd check various ROM images to see if they work
correctly for me. If not with real hardware then at least with VICE's
128 emulation for a good start.

I've tested the DIN Kernal in VICE and it seems to be exactly the same
as when I use the US Kernal. Replacing the chargen though makes a
difference - some of the characters a different, very minor. Maybe
just using the DIN version of the Kernal won't effect my system at all
:). I downloaded ROM images for the US & DIN models and did a binary
file compare. There are a 26 bytes changed in the low area in 8
separate groups, then a many, many bytes in the upper area are
different (almost completely in sequence.) That area in the US chips
is just filled with $FF, no real data or information there at all, as
if it isn't used. I've pasted a complete FC output below.

ROMs won't fry any chips nor boards as long as they are installed
properly. They may not work properly though ;-)

Ok, I'm not worried about the ROM's frying. But I do know that in
some cases (not many,) if you place a chip designed around a PAL clock
into an NTSC clocked system, something will get damaged. Had a friend
put a PAL VIC-II chip into his 64 years ago. After about an hour
total running time (all of which was spent trying to figure out why
the video didn't work right,) his system froze up and would never boot
again, even after he figured out the PAL VIC-II and replaced it with
an NTSC VIC-II chip. Now I don't personally see how the ROM chips
could be configured specifically for PAL or NTSC, but I'd sort of like
some reassurance in that area :) All things considered, I don't see
how the "image" of the ROM in an EPROM could possibly tell the
difference HW wise. So, I don't see a problem putting a DD3 Kernal
image (on an EPROM,) as my Kernal ROM.

As well as on 1541. It was quite an update to the original design. The
addition of the extra IO port made it work with multiple drive models
without any extra considerations.

I'm just wondering how that parallel port would work with say the
XAP1541 adaptor. If push comes to shove, and nothing against
Nicolas's work (he did an awesome job IMHO,) I'd much rather install
original DD3 HW into my computer & drives. But I just cannot find
them :( I don't see how I could possibly need more than 8k extra RAM
in my floppy (extra RAM in my 128 is another story - but that's not
the topic here.) I do sort of like the idea of installing multiple
ROMs and then flipping some switches to chose between them, but other
than DD, there isn't really any that I'm interested in.
.

Relevant Pages

  • Re: CP/M-85 ??
    ... The basic answer still stands, 8085 runs CP/M, yes. ... M10x have 32k of rom at 0000. ... All the extended mapped ram in the world makes only for a ramdisk. ... disk and housekeeping ...
    (comp.os.cpm)
  • Re: CP/M-85 ??
    ... ReMem allows for an all RAM mode to be switched in, ... ReMem also has an IO based "ram disk" referred to as a RAMPAC. ... Any help in getting to a working CP/M for this hardware would be ... Disk in mine is mapped in Rom and RAM both. ...
    (comp.os.cpm)
  • Re: CP/M-85 ??
    ... the rom, map in a boot rom then map in ram and uses CF as disk ... I have put CP/M on the M100 and a few other 8085 powered ... found on the 'net and store them on local disk so long as you have ...
    (comp.os.cpm)
  • Re: Altair Questions
    ... unless its around 75 F. It might be the disk or ROM, ... ram or 2SIO. ... "French Luser" ...
    (comp.os.cpm)
  • ColdFire M5271 use 52Kb SRAM for C-stack to test DRAM at OS boot ?
    ... allocate C-stack there to test DRAM at the boot time? ... allocating stack on DRAM and jumping over it during DRAM test - anybody ... I need to run the RAM test before the bootrom copies itself into RAM, ... At this point the system is still executing instructions out of ROM. ...
    (comp.arch.embedded)