Question for knowledgeable people — ZXNet «real.speccy»

From Roman Petrov To All 21 August 1998

Hello! Dear All!

People, explain to me this situation:
Calling TR-DOS (via #3D2F) with ROM 128 BASIC on does not work
(does not switch), but with ROM 48 BASIC on, everything works great.
Knowledgeable people, Tell me - is this an emulator glitch or a Spectrum feature?
Tested on 2 emulators. I just don’t know, should I change the emulator
to write a BIOS or should I change the functions of working with the disk?

With respect, Roman aka Megus of Brainwave Tracks.

[Team ZX-Spectrum] [Team Iron Maiden] [Team Die Krupps] [Team Guitar]

From Roman Milukov To All 26 August 1998

Hello, Roman!

21 Aug 98 Roman Petrov, hoping that Everyone will read it, wrote:

RP> People, explain this situation to me:
RP> Calling TR-DOS (via #3D2F) with ROM 128 on
RP> BASIC does not work
RP> (not switches), and with ROM 48 BASIC turned on, everything
RP> works great.
RP> Knowledgeable people, tell me - is this an emulator glitch or
RP> a feature of the Spectrum?

This is a feature of the Spectrum! In those computers where a ROM with the 128th BASIC
is installed (and some, when upgrading to 128K, expand the memory, but not the ROM, leaving
the old one), it is mandatory to prohibit the inclusion of TR-DOS when a ROM
page with BASIC-128 is connected, otherwise this BASIC will not work it works!
You can remove this in the Pentagon (as, indeed, in other computers) by cutting one
track: free leg 11 D75 (LA2) and apply “1” to it.

All the best.
Roman.

From Igor Zelenin To All 26 August 1998

Greetings, dear Roman!

Friday August 21 1998 13:29, Roman Petrov for some reason wrote All:

RP> People, explain this situation to me:
RP> Calling TR-DOS (via #3D2F) with ROM 128 BASIC enabled not
RP> works (does not switch), but with the 48 BASIC ROM turned on, everything
RP> works great. Knowledgeable people, tell me - is this an emulator glitch or
RP> a Spectruma feature?
Spectrum feature.

With respect, Phantom Ltd/PoS/PCCC.

From Konstantin Samoylov To All 26 August 1998

Of course, the feature of the Spectrum and the emulators have nothing to do with it. :)
Standardly in 512 ROM, as I remember, the entries are located in the following
order:
Profi
CP/M \ this is one half of the chip 0-\
TR-DOS / /-1 - this so
BASIC 128K \ and this is another and this is so - 0-/
BASIC 48K / \-1

Basic 128k is selected 0, and 0 is common for 2 halves. So, when we transition from it to Tr-Dos, we end up in the CP/M firmware. These are the pies.

With respect, SP/PoS.

From Denis Sotchenko To All 27 August 1998

Once on 21 Aug 98 Roman Petrov wrote the following for All:

RP> explain this situation to me: Calling TR-DOS (via #3D2F)
RP> with the 128 BASIC ROM turned on does not work (does not switch), and
RP> with the 48 BASIC ROM turned on everything it works great.

Just think about it. Calling TR-DOS is an access to a memory area that does not contain code in BASIC-48 (the font is there, it seems). And in BASIC-128 there is a code. To prevent BASIC-128 from accidentally falling into TR-DOS during
work, a hardware ban is made on calling TR-DOS.

__
__/ / Powered [pepsi inside]
\_\/ by MOTOROLA [smoking suxx]

From Dmitry Lomov To All 27 August 1998

RP> Calling TR-DOS (via #3D2F) with the 128 BASIC ROM turned on does not
work
RP> (does not switch), but with the 48 BASIC ROM turned on everything works fine.

This is how it should be. this is a feature of the Spectrum.
the fact is that from address #3D00 in the 48th ROM there is a font that in no way
cannot be executable code, and this feature is used
to work in TR-DOS - if there has already been a transition to #3D.., then this is
obviously for a reason, this is a challenge.
in the 128th BASIC at addresses #3D.. there is an executable code
which, when attempted, would crash into TR-DOS, which is unacceptable.

therefore, in TR-DOS controllers, crashing into DOS is prohibited by hardware
by the sampling signal of the 128th ROM (bit 4 #7FFD).

All the best.

Dmitry aka -=LD=- / X-TRADE GROUP.

From Aleksandr Majorov To All 27 August 1998

Khayushki, Roman!

Once in morning 21-08-98 14:30:29 somebody Roman Petrov has
written to All about 'Question for knowledgeable people'.

RP> People, explain this situation to me:
RP> Calling TR-DOS (via #3D2F) with ROM 128 on BASIC does not
work
RP> (does not switch), and with ROM 48 BASIC turned on, everything works fine.
RP> Knowledgeable people, tell me - is this an emulator glitch or a feature of the Spectrum?

This is a feature of the BETA disk.
with when ROM-128 is turned on, the hole-dos is blocked by hardware.
in the “interception address control” node, in addition to the addresses themselves, the ROM is also analyzed...
I don’t remember what this is connected with, but it seems due to some
conflict with Some ports were blocked by the company "Technology"
blocked.

RP> Tested on 2 emulators. I just don’t know, should I change the emulator
RP> to write a BIOS, or should I change the functions of working with the disk?

change the functions ;-))
or place a couple of commands
OUT (C), A
RET
in some “variables” and call them for transition
here and there ;-))))

Hey that's it *MAS* with a sledgehammer

From Kirill Frolov To All 28 August 1998

Once on 08/21/98 at 14:30:29 someone Roman Petrov wrote to All
about Question for knowledgeable people and I decided to intervene.

├┤i Roman & All !

[...]

RP> Calling TR-DOS (via #3D2F) with ROM 128 BASIC enabled doesn't
work
RP> (doesn't switch), but with ROM 48 BASIC turned on everything works fine.

You're buggy. In 128 BASIC you cannot call tr-dos.

W!7h Be$7 W!$he$, FK0.

From Kirill Frolov To All 29 August 1998

├┤i Aleksandr & Roman !

[...]

AM> OUT (C),A
AM> RET
AM> in some thread of “variables” and call them to go
AM> here and there ;-))))

No. Out (ts), and addresses are placed on the stack and called.
But this is not best.

W!7h Be$7 W!$he$, FK0.

From Kirill Frolov To All 29 August 1998

├┤i Roman & Roman !

[...]

RM> You can remove this in the Pentagon (as, indeed, in other computers) by recutting
one
RM> track: release leg 11 D75 (LA2) and apply “1” to it.

No need to talk about others! There are PLM, and there are also sprinters...
And it’s not easy in the scorp. And not everyone will climb out to cut paths.

W!7h Be$7 W!$he$, FK0.

From Roman Petrov To All 31 August 1998

RM>> having cut one track: release leg 11 D75 (LA2) and apply to
RM>> it is “1”.
KF> No need to talk about others! There are PLM, and there are also sprinters...
KF> And it’s not easy in the scorp. And not everyone will climb out to cut paths.
I suggest how: on Scorpa you still need to install an additional ROM. So hang up
immediately: when turning on a new ROM, allow access to TR-DOS at #3d00-#3dff,
since in NeOS there is a font (full of 256 characters) in this area.

From Roman Petrov To All 31 August 1998

KF> No. Out (ts), and addresses are placed on the stack and called.
KF> But this is not best.
Indeed, it will slow down. Can anyone suggest a low-memory and minimally slow option for working with TR-DOS from 128 ROM? I have only one option for now -
throw some of these functions into RAM and turn on 48 ROMs for the duration of their execution.
That is, not for each TR-DOS call, but for a group of calls.

From Roman Milukov To All 1 September 1998

KF> cut.

And what are the difficulties in Scorpa?!
And cutting tracks - if they try to install a second ROM in Scorpa - it’s not at all difficult
and cut the track at the same time!
And you can break and restore this connection with the same toggle switch (or port),
which will switch the ROM.

Of course, if a person decides to keep the second ROM in a box, and just
plug them in when they are stuck, then yes...

All the best.
Roman.

From Denis Sotchenko To All 2 September 1998

RM> What are the difficulties in Scorpa?!

GMX...

How do you imagine cutting a track in PLM? :-)
Moreover, on PLM this is generally done programmatically and without a soldering iron, but for
this you need to have a PC and CAD in which to compose new PLM firmware.
In addition, you also need to BE ABLE to do this. And most importantly, if
you don't have the original PLM firmware, you won't be able to do anything at all,
since it can't be decompiled. Unless you “wire” the circuit
manually.

From Roman Milukov To All 3 September 1998

DS> GMX...

Do you have a Scorp with GMX?

...I wonder how many GMXs have been sold?

DS> How do you imagine cutting a track in PLM?
...

Or maybe D. Zonov will release GMX with new PLM firmware? :)
Then he will start changing them (with an additional payment)... It’s a profitable business! ;)

All the best.
Roman.

From Kirill Frolov To All 3 September 1998

├┤i Roman & Kirill!

[...]

RM> Of course, if a person thinks depzhat drowned PZUhu in kopobochke, and pposto
RM> pepetykat, their when ppippdae bilhstnot, then< then> cases.

W!7h Be$7 W!$he$, FK0.

From Valerij Kozhevnikov To All 10 September 1998

Hello Roman! Quoting your msg of [hamster gnawing] to Denis Sotchenko:

RM> ...I wonder how many GMXs have been sold?

50.

RM> Or maybe D. Zonov will release a GMX with new PLM firmware? :)
RM> Then he will start changing them (with an additional payment)... It’s a profitable business! ;)

To the garden. Pentagon/Scorpion is enough for the eyes. Scorpovaya for work,
Pentagon - watch demos.

WBR, Jason.