http://blog.nullspace.io/apple-2-lisp-part-1.html
4min video
I am amused that the narrator describes an odd (to him) bit of
functionality of the Apple ][: it contains built-in hardware to turn
an incoming audio stream into data in memory, which they use to
transmit their Lisp into its RAM, as it has insufficient memory to
develop code on the machine itself. :?)
--
Liam Proven ? Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk ? GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com ? Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 ? Cell: +44 7939-087884
Hi
As you may already know, at the N8VEM project we are working on an S-100
VDP. This board has a V9939 video display processor and a AY-3-8910 sound
generator which should allow us to have some MSX2 compatibility when
combined with a backplane, S-100 Z80 CPU board, and a memory board.
There are more IO and memory features necessary for full MSX2 compatibility
required though. I am considering a designing a board to provide the
balance of MSX2 functions so that the combination of an S-100 backplane,
S-100 Z80 CPU, and S-100 VDP.
I envision such a board to be composed of SRAM memory, a couple of 8255
PPIs, a card slot plus whatever extra IO and memory we need. If anyone has
detailed MSX2 experience and would like to help out on designing an S-100
MSX2 compatibility board to complement the S-100 VDP please contact me.
S-100 and MSX seem like a neat combination but no one has tried it AFAIK.
Thanks and have a nice day!
Andrew Lynch
Hi,
I need a replacement Am26LS29 for a CNC machine that has one that is
broken. Our normal suppliers don't have them anymore. Any advice where I
might get one?
Regards,
Bert
I've started reverse-engineering the disk format for Model II
TRSDOS-II 4.2. (I'd still like to get images of other TRSDOS
versions.) I haven't actually started disassembling the code yet.
I've gotten to the point of being able to extract unfragmented text
files and BASIC programs from a floppy image, and that was the primary
objective, so it's possible that I might not have to pursue the
reverse-engineering any further.
I was pretty sure before I started that they had switched to 16
512-byte sectors per track (vs. 26 256-byte in Model II TRSDOS 1.2 and
2.0). That is the case, but they still logically deal with it as
256-byte blocks, so they must have put in code to do blocking and
deblocking. They do allocation in units of the 256-byte blocks,
rather than in granules.
The most interesting part is that they use a hierarchical directory
structure. I'm not sure what versions of TRSDOS-II 4.x are covered by
the manual I have, but the manual doesn't say anything about
subdirectories, so perhaps they aren't made available to the end user.
The root directory on the disk image I'm examining has room for four
entries, of which only two are used, for SYSTEM/DIR and USER001/DIR.
As one might guess, all of the system files are in the former, and
user files go in the latter.
The filename portion of a directory entry is no longer rigidly 8+3,
but is now a 24-character field with the slash separating the filename
and extension explicitly included. If TRSDOS-II 4.2 limits the
filename to 8+3, that isn't due to the on-disk filesystem structure,
but might be enforced for compatibility with earlier versions of
TRSDOS.
A directory entry now contains a linear block number of an index block
for the file. Actually it contains two linear block numbers, for the
index blocks of a primary and a backup copy of the file, but I've only
seen this used in the root directory for the index blocks for the
primary and backup SYSTEM/DIR and USER001/DIR directories. The
superblock contains two linear block numbers for the index blocks of
the two copies of the root directory.
As far as I can tell, they have abandoned using a HIT (Hash Index
Table) to speed directory searches, as was done in the earlier
versions of Model II TRSDOS, as well as TRSDOS and compatible
operating systems for the Model I/III/4.
The index block apparently contains a list of extents, each of which
is specified by a starting linear block number and a block count. The
floppy image I'm examining has no fragmented files, so I've only seen
a single extent in each index block, and am only assuming that
additional extents pointers would be consecutively stored after the
first. I'm not sure whether they can support a file needing more
extents than can be listed in a single index block.
The linear block numbers are probably three or four bytes, to support
hard drives, but since I've only examined a floppy disk, I've only
seen two bytes used, with the following two bytes being zero and
presumed to be the high order part of the block number. Oddly enough,
the low 16 bits of the block number is stored in big-endian form,
which is unnatural on the 8080/Z80 processors. However, since a 24 or
32 bit block number would have the 8 or 16 bits stored after the low
16 bits, it is a mixed-endian format.
The format of SYSTEM/SYS, the system overlay file, has changed again.
They now use smaller overlays, most of which are only 256 or 512
bytes, but there are a lot more of them than there were in 1.2 and
2.0.
Eric
Hi,
Does anyone have any details on how write precompensation is implemented
on MFM-formatted floppy disks?
I've been reading through the description in various disk controller
datasheets, but they're as clear as mud - to the point where most only
state that an EARLY or LATE signal is generated, not how those signals
are generated.
Reason I ask is, I'm trying to write a disc with a DiscFerret, but
without precompensation they're unreadable... :(
Thanks,
--
Phil.
classiccmp at philpem.me.uk
http://www.philpem.me.uk/
It does not seem like I'll be able to find media for this gadget, so I've
move to Plan B: Reverse engineer the 5 existing diskettes I have.
Since the originals are read with a WD FDC chip, I thought this would be a
matter of letting ImageDisk have at it to determine the format.
Unfortunately, IMD cannot make heads or tails out of what's on the
diskettes. They do seem like they have 80-tracks, probably at 500KB/sec
data rate - at least that's what IMD tends to converge on in between
cascades of low-level data errors. Even at full analysis setting it's not
able to decide what it's looking at.
These diskettes read fine when I connect the unit to an Apple 2, so
whatever is on them is intact. Anyone have an idea how to decipher it?
I have a suspicion that they deliberately offset the tracks from a
"normal" position. At one point there was a special fluid available to
reveal magnetic domains under a magnifier, but this seems exotic and
not something I would know where to obtain.
Steve
--
Hi, I've recently got around to try get and my MINC-11 working and not
getting very far! So I'm after advice.
Right now I have a minimal system consisting of
M7270 LSI11/2 CPU
MSV11 RAM
DLV11/J serial/console
PC PSU with reset circuit to generate BPOK and BDCOK
Here's a rough guide to what I've diagnosed so far:
1. Jumpered the CPU to boot into ODT and checked the jumpers on the
serial board for console address
2. No activity on the console port on power on or reset
3. Double checked the timing of BDCOK & BPOK and they're in spec.
4. Debugging on the DLV11 reveals the MATCH signal isn't being generated
>from the address decode logic
5. The bus looks partly correct. SYNC, BDIN & BDOUT look fine. The mux'd
data/address lines are active in the BDIN/BDOUT cycles but within those
cycles the signals "oscillate" (they cycle 20-30 times within a single
BDIN/BDOUT pulse, which doesn't look good!)
6. Probing around the CPU board now. All the signals look correct, with
the exception of the 16 data/address lines generated by the data chip
(1611). Power and all 4 clock phases look good as does the internal
data/control/microm WMIB bus and the WAIT signal from the control chip.
7. OK, so looks like the 1611 is faulty. I wanted to check it wasn't the
1611-Qbus interface circuitry so I "modified" a junk 40 pin socket to
sit between the 1611 and the on-board socket and removed pins 3-18
(hence isolating the WDAL0-15 signals of the 1611 from the rest of the
CPU board). This made no difference - the signals on those pins are very
weak (2V peak) and "spike" for very short periods at the beginning of
data read/write cycles.
So I'm stuck and I have no spare 1611 or other LSI-11 CPU module to try.
Does anyone have a spare 1611 or complete LSI11/2 they'd be willing to
lend/sell? I'm in the UK. Alternatively, any advice for other things to
check? I can try and grab some screen shots of the signals if that would
help.
Thanks for any help!
James
I replaced my noise side panel fan on my HP 16700A Logic Analyzer on
Friday with some much quieter ones.
I documented my process and wrote about it on my blog, available here
http://techtravels.org/?p=902
I cross-posted this to the HP Agilent Equipment yahoo groups, so sorry
for the dupe if you also belong there.
Thanks
Keith
Hi
I'd like to make a production order for S-100 6502 CPU V2 boards. ?These are similar to the S-100 6502 CPU V1 boards we released earlier but with some minor fixes and also improvements for TMAx circuitry. ?Now the board behaves well in a multiple CPU S-100 system as Master or Slave CPU board.
There has to be firm demand for 20 boards to make a production order viable. ?So far there are 15 boards on order for builders. ?If we could just get a few more then I will make the order. ?There will still be some left over we can sell the usual way and if there are still leftovers I will sell them on VCGM and eBay.
It is the usual arrangement except this time I am asking the builders to pay in advance for a good portion of the production order. ?That's a bit different than before but I can't "carry" the inventory like I used to be able to do. ?Each PCB is $20.
Shipping in the US is $3 for a single PCB and $2 for each additional PCB. ?Shipping internationally is $10 for a single PCB and $3 for each additional PCB. ?This is for the bare basics USPS first class postage with no tracking or insurance. ?The builder assumes all risk of delivery as per usual arrangement.
If you would like one or more S-100 6502 CPU V2 PCBs please send a PayPal to LYNCHAJ at YAHOO.COM
Andrew Lynch
PS, there are a number of S-100 boards backed up in the queue I'd like to distribute in this manner. ?Probably the S-100 80286 CPU board is next but there are several others John and I are working on as well.
Howdy!
I lovingly restore the good 'ol mechanicals and I have a very nice gently
used 1392366 with box. Rivets are in great shape and aside from a small
cosmetic issue here and there, it's about as good as you can find outside
of being new.
I thought I'd ask here first before the keyboard enthusiast forums. I know
it's a rather elusive keyboard so if it could be paired up again with an RT
I'd be thrilled.
Kind regards,
-Maxx
