Scorpion CMOS

In RC 1.96, the current drive is saved in cell #10 CMOS
(in the ROM version), in which cell is it best to save
the panel configuration for RC 2.0?

I ask again, if no one answers it will be saved in #11.

* Crossposted in ZXNET.SOFT
* Crossposted in REAL.SPECCY

Bye!
Pa...@nm.ru realsoft.nm.ru [RC 2.0 47%] [ZX] [Real Software]

Original was from to .
PK> In RC 1.96, the current drive is saved in cell #10 CMOS
PK> (in the ROM version), in which cell is it best to save
PK> panel configuration for RC 2.0?
PK> I ask again, if no one answers it will be saved in #11.
^^^^^^^^^^ I answered, then you need to push it
bitwise from #11 to #18 ;)

[I ZX] !mPeR!0^PHG

PK> (in the ROM version), in which cell is it best to save
PK> the panel configuration for RC 2.0?
The free first one is #15th.

PK> I ask again, if no one answers it will be saved in #11.
No need, there is #aa for normal operation of the madrome, etc. Also #12,#13,#14.

Well, bye! Tigrr from -■Brainwave^X-Project■- tigr...@mail.ru

05 Mar 01, 13:22 "Pawel Kislyak" wrote to "All":

PK> In RC 1.96, the current drive is saved in cell #10 CMOS
PK> (in the ROM version), in which cell is best to save
PK> panel configuration for RC 2.0?
PK> I ask again, if no one answers it will be saved in #11.

Normal. My shell is using the last 5 bytes.

See you later.

AV> Normal. My shell uses the last 5 bytes.
What, if not a secret? I still suggest #15, not #11. Otherwise, I would rather not
use kmos in rts2.0 than constantly edit the cell.

IT> What, if not a secret? I still suggest #15, not #11. Otherwise, I
IT> would rather
IT> not use kmos in rts2.0 than constantly edit the cell.

To be more precise, it’s not the shell, but the ROM firmware. Or even more precisely - SOS128 with
TR-DOS.
There is such a feature - if “B” is written in #3f, then autostart of disk A: (type boot).
If “R” is written there, then the resident starts from the address
#3a - control byte from the start address, #3b - low byte of the address, #3c -
high byte of the address, #3d - what throw in the page port, #3e - high byte of the expansion port
(the low one is always #FD).
What is this glitch for?
I’ll clarify. For example, I installed the same RC, it started. They launched a program out of it, it got boring. We pressed reset and fell back into the commander. A very
convenient thing in some cases. It is advisable to have more than 128 kilos of memory for this purpose. And so I made myself a resident Real Commander, Consul
Commander, some kind of text editor and something else.

Another byte 10 (approximately from #30) will be used by tr-dos under the Kaevsky screw -
_I ASK YOU NOT TO TAKE IT._
By the way - does anyone besides me need this? I mean screw support.

And I would like to immediately throw up a topic for discussion.
In general, the following thing will be done. A free port is being invented that will switch the tr-dos ROM - regular tr-dos and screw-mounted tr-dos.Emulation
will be almost 100% and not only for #3d13! Is-dos will be used as a kind of boot - images of tr-dos disks will be stored there. There you can also link the
image to a real disk a-b-c-d. The only limitation is that the image must be
continuous. It's not that difficult to follow. There is one bummer - directly from the tr-board boot or directly from the command line it will not be possible to copy from the drive to the image and back. Only through is-dos. I hope everyone understands why.

And I want to say right away that such complexities at first glance, on the contrary, simplify
the work. There is no need to use RAM like in Scorp or Kay - only the memory and system
variables.

Now I hope to hear support from you, and not a discussion
of “it will work or not” - I know that it will definitely work.
It’s just a little difficult over time...:( And there are so many things I want to do...

See you.

AV> To be more precise, it’s not the shell, but the ROM firmware. Or even more precisely - SOS128 with
AV> TR-DOS.
Damn, again... ;))

AV> There is such a feature - if “B” is written in #3f, then autostart of disk A: (like
AV> boot).
Strange, of course. I usually don’t use autostarts like that - so that they are permanently on before I remember that they are on. IMHO this is not necessary. Is it possible to work
with kmos from your shell? For example, crash the resident, reset with
ignoring the resident, change the configuration from it yourself?

AV> If “R” is written there, then launch the resident from the address
You can stick to any single system of residents, otherwise there are already 3 of them -
MadROM, scorp, and yours. It is inconvenient to support everything. Support one - but what about
the rest?

AV> #3a - control byte from the start address, #3b - low byte of the address, #3c -
AV> high byte of the address, #3d - what to throw in the page port, #3e - high byte of the
AV> extension port (low byte is always #FD).
So. And if there is no space, nothing will work?

AV> Why is this glitch needed?
AV> Let me clarify. For example, I installed the same RC and it started up. We launched
AV> from
AV> the program and got tired of it. We pressed reset and fell back into the commander.
Mmm... What if I have a cache? As I understand it, it is not expected?

AV> A very convenient thing in some cases. Memory onlypreferably more than 128
Almost like MadROM, for the sake of something incomprehensible which I don’t want to change...

AV> Another byte 10 (approximately from #30) will be used by tr-dos under Kaev’s screw -
AV> _I ASK YOU NOT TO BORROW._
In which direction?

AV> By the way - does anyone besides me need this? I mean screw support.
Kaevskaya - I don’t know, but it will probably come in handy, since the circuit is simple and is assembled
in makeshift conditions.

AV> And I would like to immediately throw up a topic for discussion.
AV> In general, the following thing will be done. A free port is being invented, which
AV> will switch the tr-dos ROM - regular tr-dos and screw tr-dos. Emulation
AV> will be almost 100% and not only for #3d13! Is-dos will be used as
#cff7, make the bit customizable - since many people already have this port, if
anyone has done anything with #eff7. And set the bit when flashing the ROM itself
to make it convenient for everyone. And for prog, if necessary, it’s better to either make a subprogram
or store a mask somewhere.

AV> bummer - directly from the tr-boot or directly from the command line you can’t
AV> will be copied from the drive to the image and back. Only through is-dos.
AV> I hope
AV> everyone understood why.
And copying into an image will be slow as standard, like isdos itself? Those. Will I have time to drink tea 15 times before he copies something from me?

AV> And I want to say right away - these are the first oneson the contrary, complexity simplifies
AV> work. There is no need to use RAM like in Scorp or Kai - only memory and system
AV> variables.
Only one small difficulty - not everyone has memory, but more people have memory.

AV> Now I hope to hear support from you, and not a discussion
AV> “it will work or not” - I know that it will definitely work.
Still, nothing will work out... ;)))

08 Mar 01 22:45, Alexey Volyk wrote to Igor Turashev:

AV> Another byte 10 (approximately from #30) will use tr-dos under the Kaev
AV> screw - _PLEASE DO NOT TAKE IT._ By the way - besides me, this is for anyone
AV> necessary? I mean screw support.

I decided to upgrade the spec a little... I bought a custom power supply and a screw.
I soldered a 128kb flash drive into the spec. All that remains is to connect the screw and use the
flash drive
for its intended purpose - it is programmed on the go directly in the Spectrum. There is destined to be screw support in
my computer.

AV> And I would like to immediately throw in a topic for discussion.
AV> In general, the following thing will be done. A free
AV> port is being invented that will switch the tr-dos ROM - tr-dos regular and tr-dos under the
AV> screw. Emulation will be almost 100% and not only for #3d13!

I also want tr-dos on the screw. But for now I don’t know exactly how
the screw will be connected (currently it is connected only to the power supply). And I want
to be able to flash the modified TR-DOS firmware into the Kay ROM with a
Nemovsky
controller and everything should work there. Only in 64kb of ROM, 256kb of RAM and without
a soldering iron.
If this can be installed in Scorpio instead of a shadow service, then absolutely
wonderful.

AV> Is-dos will be used as a kind of boot - images of tr-dos will be stored there
AV>disks. There you can also link the image to
AV> a real disk a-b-c-d.

Probably a good option for the Spectrum. You just need to think about how the system
will
search for is-dos disks on the screw. It is difficult to make more than 16 disks using standard methods.

AV> The only limitation is that the image must be continuous. This is not so
AV> and difficult to comply with.

But I don’t like this. If the work with the isdos file system has already been done at
a level
sufficient for working with continuous images, then adding segmentation here
is not so
difficult. Of course, the disk image must be full size.

For full functionality of the isdos file system from under tr-dos, you can
provide
launching the is-dos kernel in who knows what jar. If you tear off the top two levels
it will fit
into the jar.

AV> There is one bummer - directly from the tr-boot or directly from the command
AV> line it will not be possible to copy from the drive to the image and back. Only
AV> via is-dos. I hope everyone understood why.

They didn’t understand. Did you understand what you wrote? Like you can't
work
with a TR-DOS disk image from TR-DOS...

AV> And I want to say right away that such difficulties at first glance, on the contrary
AV>, simplify the work. You don't need to use RAM like in Scorp or Kai - just
AV> memory and system variables.

Is it simpler? Doubtful. But it's definitely worse.In the case of using RAM
of course
it is possible that some programs will not work (they will crash that area of ​​RAM and that’s it).
But the percentage
of such programs is small. And on top of that, you can use the large amount of RAM on top of
megabytes where
normal programs do not climb. The only problems are at the Pentagon. Can you come up with something?

I hope it is clear that it will be very difficult to work without RAM and you will have to tie everything
to
CMOS memory, which is not available in every computer? And not all operations will
be easy to perform
with virtually no RAM. IMHO, for normal operation you need a jar or two.

AV> Now I hope to hear support from you, and not a discussion
AV> “it will work or not” - I know that it will definitely work.

I will do something for my spec. Doing something like I wrote above -
on a kai with one controller and nothing else should work. But first
I have to connect the screw. So far I've settled on three options:

1. Using a bus width converter 8->16, connect a multi-tool with a
controller.
2. Nemov's circuit or something similar.
3. Perhaps the 8-bit version will work on my screw. But will I be able to
connect the CD later?

Probably the first option will be the most useful.

P.S.: Miha Ulanov wrote a program for the pisyuk to pull out img files
files.
You can use it (raw)probably use it (working with IS-DOS disks).

P.P.S.: Flash connection diagram:

(I haven’t figured out the bugs yet)

A0..A13 Spectrum -> A0..A13 flash \
A14'ROM Spectrum -> A14 flash | everything is the same as with ROM
ROMDOS Spectrum -> A15 flash /

RD -> OE#
WR [OR] locking -> WE#
/RESET -> RP#

+5V -> VCC
+12V -> VPP

reading/writing in ROM -> CS#

didn’t come up with that -> A16

Chip type 28F001BX-T. It plugs into the socket under the ROM (4 contacts in the air - it has 32 legs). There is an 8kb boot block in size from the very top
of the memory which according to the given diagram
is not programmable (you need +12 volts on OE or RP). In the 28F001BX-B
chip the boot block is located
at the bottom of the memory and it turns out that it cannot be connected in the Spectrum using this scheme. But in
pisyuks they install 28F001BX-T
or similar ones.

In general, there are more modern microcircuits that are only programmable from +5 volts
and do not have any problems with the boot block. For this microcircuit it is written that
is guaranteed
100,000 reprogramming cycles, for new ones 1,000,000 cycles and more.

AV>> SOS128 with TR-DOS.
IT> Damn, again... ;))

The first version of this case was written in 1997. All of us use it
Specrumists use it - everyone is happy!

IT> Strange, of course. I don’t usually use autostarts like that - so that they are

Convenient in general.

IT> it is necessary. Is it possible to work with kmos from your shell? For example, a resident
IT> crash, reset, ignoring the resident, change the configuration from it
IT> yourself?

It turns on and off naturally. If you hold down "R" during the reset, then all
resident autostarts are switched off. Autostart of the disk is also removed by a brick
during the launch of tr-dos and returnom.

IT> It may stick to some single system of residents, otherwise there are already
IT> things
IT> 3 - MadROM, scorp, and yours. It is inconvenient to support everything. Support
IT> alone - but what about the rest?

Yes, I’m not asking you to support my system as a whole - I’ll redo the program I need
in half an hour. And we gave everyone who needed the source codes of the installers
residents with a bunch of comments.

IT> So. And if there is no space, nothing will work?

Prize for the studio!!! Naturally. But the presence of space is determined automatically.

IT> Mmm... What if I have a cache? He, as I understand it, is not expected?

I have work with the "reset" button. The cache works with "magik".

IT> Almost like MadROM, for the sake of claritywhich I don’t want to change
IT...

No one is forcing me.

09 Mar 01, 22:21 "Kirill Frolov" wrote to "Alexey Volyk":

KF> food). And I want it to be possible to flash the modified TR-DOS
KF> firmware into the Kay ROM with the Nemov controller and everything should work there.
KF> Only in 64kb ROM, 256kb RAM and without a soldering iron. If this can
KF> be installed in Scorpio instead of a shadow service, then that’s absolutely wonderful.

It won’t work instead of a shadow service. And for KAI directly too - you will have to install a separate ROM
- there, on port #7ffd you can only switch to a third-party ROM.

KF> Probably a good option for the Spectrum. You just need to think about how
KF> the system will search for is-dos disks on the screw. Using standard methods
KF> it is difficult to make more than 16 disks.

It will not search for anything - in the is-dos itself, we bind
images to drives using a small utility. Well, as in emuls.

IT>> more people have memory.
AV> In KMS you can store data about the screw. Is this possible in memory?

Data about the screw should only be stored on the screw itself. Otherwise, if this
screw is inserted into another computer or if the computer is killed, you will not find
where any disks were.

KF>> Flash TR-DOS into the Kay ROM with the Nemov controller and everything there
KF>> should work. Only in 64kb ROM, 256kb RAM and without a soldering iron.
KF>> If this can be installed in Scorpio instead of the shadow
KF>> service, then that’s absolutely wonderful.
AV> It won’t work instead of the shadow service.

Why? Where will it go...

AV> And KAI directly too - you will have to install a separate ROM

WITHOUT A SOLDERING IRON. EVERYTHING!

AV> - there, via port #7ffd, you can switch only the third-party ROM.
^^^^^^^^
Did you mix up anything?

There, via #1FFD, you can force enable the third-party half of the ROM.
The shadow part will be installed when writing #00 to #7FFD.

Only on the board this seems to be the case (not sure) can be prohibited
by jumpers.

AV> Turns on and off easily. If you hold down "R" during the reset, then all
AV> resident autostarts are switched off. The autostart of the disk is also removed by a brick during the
AV> time of launching tr-dos and returnom.
And set the clock, change the desired cell, alarm clock, configuration of software
mics (for example, port #cff7)?

[...]
KF> WR [OR] blocking -> WE#
It’s better to run it through the KP so that you can programmatically shut it down.

KF> /RESET -> RP#
??? Why?

[...]
KF> didn’t come up with that -> A16
Cache sampling - you will have the largest cache of all :) Or attaching a software microphone
is also not bad.

KF> Microcircuit type 28F001BX-T. It is plugged into the socket under the ROM (4 contacts in
We need to look for something newer.

KF>> contact in
IT> We need to look again.

I don’t need it. There is no point. Moreover, it’s already soldered.
And those who will do it again should probably look towards similar
chips from AMD. Or make the computer start from address 0x1C000
- there is one of the “parameter blocks” that can be erased separately. Only 3
inverters
to be installed at the addresses. I'll probably do that later.

KF> Data about the screw should only be stored on the screw itself. Otherwise, if this
KF> screw is inserted into another computer or if the kmos is killed, you won’t find
KF> where what disks were there.

But this, IMHO, is correct.

For simm all... Squeak!
Nikolay G. Arhipoff

KF> Data about the screw should only be stored on the screw itself. Otherwise, if
KF> this screw is inserted into another computer or if the kmos is killed not
KF> you will find where what disks were there.

Z-a-g-r-u-z-a-e-sh-s-ya from the backup is-dos disk. :))
In general, after the reset, it won’t take long to read the screw and register the data in kmos.

See you later.

KF> ^^^^^^^^
KF> Didn’t you mix anything up?

There is a tr-dos-sos128 switch on the 1st leg and if you want to start the second half of
tr-dos, then you need to throw 0 in #7ffd. And what program will work in this
case? In my opinion, no one will switch this case.

KF> There, using #1FFD, you can force enable the third half of the ROM.

I looked at the diagram - it’s possible.

KF> The shadow part will be installed when writing #00 to #7FFD.

KF> Only on the board this thing seems (not sure) can be prohibited
KF> by jumpers.

Nope - there Using jumpers you can “rearrange” the halves and switch in inversion
from port #1ffd.

KF> And how in ZXEMU do we exit the emulator every time (that is, load
KF> is-dos) to change the disk? I understand. You are not expected to work with the isdosa
KF> file system at all. Simply the coordinates of the file on the hard drive
KF> are written in kmos. So?

It’s easier.

KF> What can I get? To work with segmentation, you need approximately 300
KF> bytes
KF> of data memory and two times less than program memory. EVEN IN YOUR
KF> CASE (even without data memory). You remember the coordinates of the block list
KF> of the segmented file (its length is 256 bytes) and the coordinates of the beginning and the
KF> disk on the hard drive.

Yes, I don’t want to work with “clean” files at all.is-dosom.

KF> Let’s not waste our brains here. Copying is two operations:
KF> reading and writing. Is it possible to read from a soft disk? Is writing to a
KF> virtual disk possible? So what's the problem??? :-/ Or images
KF> READONLY? :-( ) Why the hell is this necessary?

In the image itself everything will be like a regular disk. You can simply create an image from scratch (a blank disk) or with a program sv_trd.com - that’s what it’s called.

KF> Tell me, for example, how Are you planning to WRITE ONE SECTOR
KF> of 256 bytes onto the disk? There are 512 byte sectors inside the screw.

I’m still thinking... I have thoughts - I’ll have to try.

AV>> all Speckrumists use it - everyone is happy!
IT> I’ve never even heard of this one. :). No offense intended, of course, mr.
Glukу.

IT> And set the clock, change the desired cell, alarm clock, configuration
IT> of software microphones (for example, port #cff7)?

Glukovsky seemed to be a kmos commander - do what you want with the kmos.

IT> Then why is it stitched at all?

It’s stitched there sts6.2, hdd boot under the Kaevsky screw (it boots directly from the start menu with the screw), the computer is ticking, launches are processed. There is also a memory test
when resetting, which will give the number of the broken bit :) - it does not slow down when resetting. There is also
boot with a mouse.

IT> You need a prize - it has narrowed down the circle of potential users quite well.

Without the computer, but there will be STS and hdd boot.

IT> And if I want to get from the shell and into the cache, for example, I have
IT> something
IT> important sitting there, like RC? And I haven’t converted the magick for cache -
IT> there is no need.

So how do you get there with standard firmware?

IT> But I also want a screw, but just how many conveniences will I lose at once...
IT> It’s scarybecome...

AV> I actually wanted to use 24LC16 for this task, but I don’t know yet how best to do it
AV>.
Why? Should we still look for her? For example, I definitely don’t have it now.

Well, bye! Tigrr/Brainwave tigr...@mail.ru http://www.brainwave.dax.ru

12 March 01, 01:20 "Igor Turashev" wrote to "Alexey Volyk":

IT> Missed, mr.Gluk wrote Gluk reset service, and madrom -
IT> compowelcome.
IT> Not that - what exactly? There is everything for convenient work on a computer with
memory >> 128k, kmos, cache.

Well, not mr.Gluk - I don’t remember exactly.

[extra skip]
I didn’t even offer the firmware to anyone, I just wrote what’s there.
So if you don’t need it, don’t pour water.

IT> I don’t have the addresses at hand, but it looks like it’s possible in that diagram #aff7.

This port has an expansion of up to 4 meters at the Pentagon.

IT> Well, thanks for the sts... This is probably how they clog up the “unnecessary” place in the
IT> ROM... It would be better if they plugged in a text viewer and a simple shell, otherwise it
IT> after all eats up half a ROM!

But that’s wrong! STS is stitched not just as a load, but integrated into the system!
How do you like the fact that at any time in a running program I can click on
Magic, go to STS, trace, and then continue executing the program? IMHO, the shadowbagger on the scorpion is not up to the STS level? And at the same time, STS can
be thrown into any page (even >128K)!

AV> This port has an expansion of up to 4 meters at the Pentagon.
Then... Well #9ff7.

Bye, Alexey.
Tigrr/Brainwave/X-project tigr...@mail.ru http://www.brainwave.dax.ru

IT>> in ROM... It would be better if a text viewer and a simple shell were plugged in,
IT>> otherwise it, after all, eats up half a ROM!

NA> But he’s wrong! STS is stitched not just as a load, but integrated into the
NA> system! How do you like the fact that I can at any time in
NA> a running program click on Magic, go to STS, trace, and then
NA> continue executing the program? IMHO, the shadowbagger on the scorpion doesn't reach the STS level? And at the same time, STS can be thrown into any page
(even >> 128K)!

Yes, by the way, this is the coolest part of this version. Moreover, the "R" register is completely preserved. I tried to press magic while performing a cool XOR and then
launch again - everything is OK!!!

In general, this thing helps “newcomers” understand the code at least a little.

With respect, Volyk Alexey
from Phantom Family.

14 Mar 01 20:29, Alexey Volyk wrote to Igor Turashev:

AV> In general, this thing helps “newcomers” understand at least a little
AV> in the code.

If there was at least one serious book for Spectrum, you wouldn’t
have to figure it out.
And of course you can get to everything with your mind, but very slowly it turns out.

March 16, 2001 00:43, Kirill Frolov wrote to Alexey Volyk:

AV>> In general, this thing helps “newcomers” understand at least
AV>> a little bit of the code.
KF> If only there was at least one serious book for the Spectrum to understand
KF> I wouldn't have to. And of course you can figure out everything with your own mind, but it’s very slow.

ZX-Spectrum and TR-Dos for the user and programmer + the Inforcom trilogy
“Programming in machine codes and assembly language” - why not serious
books?

For this I say goodbye, with respect Alexey Kravchenko AKA Kurleson^HS^CPU np: Era -
Sentence