Real Commander 2.2
ZXNet echo conference «real.speccy»
Meet the new version of Real Commander 2.2, now the commander
is distributed in full (together with the Doctor and RCGEN modules).
In addition to general bugfixes, RC has several new features, in particular
it now supports CMOS of the Scorpion computer (thanks to the authors of the emulator
Real Spectrum is an excellent emulator, but the layout lets us down keyboard),
also includes a new module from Imperio/PHG - "Multi Rename".
* Crossposted in REAL.SPECCY
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
Monday 04 February 2002 21:57:17 Pawel Kislyak -> All:
PK> Meet the new version of Real Commander 2.2, now commander
Received and archived.
PK> distributed in full set (together with modules
PK> Doctor and RCGEN).
PK> In addition to the general bugfixes, RC also has several new
PK> features, in particular
PK> it now supports CMOS of the Scorpion computer (thanks
Glitches there. As far as I understand, RC makes some changes
to the CMOS memory. The shadow guy stores some of the variables he needs
in CMOS and calculates the checksum. So - after RC
the sum does not match and the shadow resets the memory upon reboot,
entering data there (date & time) by default, which is somewhat
frustrating ;)
So let it lie in the archive for now.
[REAL ZX]
NA> There are glitches there.
Everything is in order with CMOS, tested on the Real Spectrum emulator.
NA> As far as I understand, RC makes some changes to the CMOS memory.
Well, yes, it saves the panel configuration in two CMOS bytes,
this is all documented in Help.
NA> Shadowvik stores some necessary ones in CMOS it
NA> variables and calculates the checksum.
As far as I know, it doesn’t store anything there.
NA> So - after RC the sum does not match and the shadow resets the memory to zero when
NA> is rebooted, adding data (date & time) by default, which
NA> is somewhat frustrating ;)
You probably have an old version of the ROM, because in the emulator it is in
emulation mode Scorpion had no such problems.
PK> As far as I know, it doesn’t store anything there.
PK> You probably have an old version of the ROM, because in the emulator in the
PK> mode of Scorpion emulation there were no such problems.
IN THE EMULATOR...
PK> Everything is fine with CMOS, tested on the Real emulator
PK> Spectrum.
Tested on real life - there are glitches!!!
Shadow writes “CMOS Checksum error” when restarting and writes default values into
it.
[ cut dick ]
PK> You probably have an old version of the ROM, because in the emulator in
PK> mode
PK> Scorpion emulation there were no such problems.
The latest version is 5.00G
[REAL ZX]
February 17, 2002 (and it was 21:46 then)
Nikolaj Amosov in his letter to Pawel Kislyak wrote:
NA> The latest version is 5.00G
What kind of ROM is this? For GMX?
Vega / ex-Style Group. February 18, 2002
18 Feb 02 04:18, Vlad Sotnikov Nikolay Amosov wrote that:
Monday February 18, 2002 04:18:00 Vlad Sotnikov -> Nikolaj Amosov:
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
PK> You comrades, let's get to the point - in which cell is the check
PK> amount stored and by what method is it calculated.
In general, I'm on ZX.NET've already been enlightened about this issue,
but the problem is that there is only one sign for CRC calculation
occupies 176 bytes (procedure from ZX-News 54), resulting in an unreasonably
large memory consumption. Who will have what options? You can write a
module that will set the correct CRC.
19 Feb 02 22:39, Pawel Kislyak wrote to Nikolaj Amosov:
PK> You comrades, let's get to the point - in which cell is the check
PK> sum stored and by what method is it calculated.
There is no need to write down just anything anywhere in the CMS.
Nowhere is it written that anything can be written there at all, it was stuck there
not for
your commander, but for the shadow service. For you, there is RST 8 through which
you can count
the time and that’s it.
PK> So you need to downgrade ;), I had 2.40 in the emulator :)
You don’t even know that the actually working non-“professional” version is
2.95.
I have never seen version 2.40 at all, and 2.50 is old and glucodrome.
There is no support for SMUC in non-professional versions. And the “professional”
versions are
from 3.00 to 5.xx.
PK>> :)
this is difficult, since the firmware, in addition to the monitor itself and regular
cans, contains firmware of different circuits, and replacing it with the old one by substituting a floppy disk with the new firmware version will not work
- it is protected by CRC and simply will not be flashed into the flashrom.
[REAL ZX]
21 Feb 02 23:25, Pawel Kislyak wrote to All:
PK> In general, I like ZX.NET've already been enlightened about this issue,
PK> but the problem is that there is only one the plate for calculating CRC,
PK> takes 176 bytes (procedure from ZX-News 54),
You, of course, read these newspapers, but don’t take them too seriously as an absolute truth.
By using a certain amount of brains, the plate can be thrown out (it is
calculated).
In addition, MOA was bugged there - CRC was intended, but it turned out unclear what happened.
PK> this results in an unreasonably high memory consumption.
RTFM some kind of trick for calculating the 16-bit CRC for an X-modem - the same
(in the sense of a sign).
#define CRCPOLY 0x1021
/* #define CRCPOLY 2110
*/
unsigned short crc_tab[0x100];
int mkcrctab()
{
unsigned char cbyte, b, c;
unsigned short crc, ptr;
ptr=0x100;
cbyte=0;
do {
crc=cbyte<<8;
crc-=0x100;
for(b=8; b>0; b--) {
c=((crc&0x8000)!=0);
crc<<=1;
if (!c) crc^=CRCPOLY;
}
crc_tab[--ptr]=crc;
} while (--cbyte);
for(ptr=0; ptr<0x100; ) {
for (b=0; b<8; b++) printf("%x ", crc_tab[ptr++]);
printf("\n");
}
return 0;
}
Here, in this echo, the question already popped up and the algorithm was thrown, but IMHO
it’s not working:
uchar crcb (uchar byte)
{
uchar d, a;
d=(byte>>4)|(byte<<5);
a=d & 0x0e ^ byte & 0x1f;
if (a&1) {
a=a>>1;
return (a<<4) ^ d & 0xf0 ^ d;
}
else {
d=a>>1;
a=(d>>2)|(d<<6) & 0xc1 ^ byte;
a=(a>>1)|(a<<7) & 0xf0 | d;
return a;
}
}
PK> You can write a module that will set the correct CRC.
For starters, there’s no need to go there. There's RST #08 for you. If you desperately need it, then
you can
provide procedures for reading/writing time directly. In the remaining parts of the memory
of the RTC
chip, your commander has nothing to do (unless it is a CMOS memory editor) -
the MOA configuration
is stored there. You'll screw up the config, but the user's screw may die fromэтого.
PK> checksum and what method is it calculated.
PK>
PK> In general, I'm on ZX.NET've already been enlightened about this issue,
PK> but the problem is that there is only one sign for Calculating CRC,
PK> takes 176 bytes (procedure from ZX-News 54), resulting in an unreasonably high
PK> memory consumption. Who will have what options? You can write a
PK> module that will set the correct CRC.
bla bla bla:( Back in the summer I threw out a new calculation procedure:
;CRC CALCULATOR ;-)
;easy generated by J7n/4D/PeRM/RuSSIA/30.07.2001
.LOCAL
;INP: HL - BUF, BC - LEN
;OUTP: DE - CRC
CRC_CALC
PUSH IX
LD IX,#FFFF
...0
LD A,(HL)
XOR LX
LD E,A
CALL ...1
XOR HX
LD LX,A
INC E
LD A,#88
CALL NZ,...1
LD HX,A
CPI
JP PE,...0
EX (SP),IX
POP DE
RET
...1
LD A,E
RRA
RRA
RRA
RRA
LD D,A
AND #0E
XOR E
AND #1F
RRA
JR C,...2
LD D,A
RRCA
RRCA
AND #C1
XOR E ;BIT 0,E=0!!!
RRCA
AND #F0
OR D
RET
...2
ADD A,A
ADD A,A
ADD A,A
ADD A,A
XOR D
AND #F0
XOR D
RET
DISPLAY $-CRC_CALC
J7N/4D
PK>> checksum and by what method it is calculated.
KF> There is no need to write down just anything anywhere in the CMS.
KF> It is not written anywhere that you can write anything there at all, it was stuck there
KF> it was stuck there not for your commander, but for a shadow service. For you there is
KF> RST 8 through which you can count the time and that’s it.
Actually, recording something in CMOS was not my idea initially,
there was a request to record the number of the current drive in the #10
CMOS cell, I did it - it turned out to be very convenient (when using the
rom version), which disk we worked with and ended up with it after the reset.
Already on my own initiative in RC 2.x, I decided to use
another cell (#15) to store the panel configuration, its number
I previously agreed with the people in the echo, so that no conflicts would arise. It is on the PC that hardware settings are stored in CMOS, while on Spectrum CMOS is used mainly as a clock, perhaps only on Scorpion it stores any shadow monitor settings or hard drive parameters. I don’t see anything wrong with
using one byte to store the RC configuration (storing
the current drive in #10 was not invented by me), everything is done solely for the sake of convenience.
I can completely remove the recording procedure in SMUC CMOS from the commander,
but this will significantly affect the convenienceuse.
23 Feb 02 18:06, Vassili Klimov wrote to Pawel Kislyak:
VK> bla bla bla:( Back in the summer I tried a new calculation procedure:
VK> ;CRC CALCULATOR ;-)
There is no CRC. Moreover, the results do not coincide with the original
algorithm. Or is there something wrong with my hands?
KF>
VK> ;CRC CALCULATOR ;-)
KF>
KF> There is no CRC. Moreover, the results do not coincide with the original
KF> algorithm. Or is there something wrong with my hands?
well, crc is something else. Out of habit, I call this a small chain of bytes resulting from processing some data array. And
it works absolutely the same, I checked.
J7N/4D
no, really, why do you absolutely need this setting to be saved in
CMOS?
why not where all the settings are, other RC settings?
All the best, siril.4d.perm.ru.
PK>> You comrades, let's get to the point - in which cell is the
PK>> checksum stored and by what method is it calculated.
PK>>
PK>> In general, I'm on ZX.NET've already been enlightened about this issue,
PK>> but the problem is that one only the plate for calculating CRC,
PK>> takes 176 bytes (procedure from ZX-News 54), it turns out
KF>> they stuck it there not for your commander, but for a shadow service.
PK> Actually, recording something in CMOS was not my idea initially,
PK> I received a request to record the number of the current drive in the #10
PK> CMOS cell, I did it - it turned out to be very convenient (when using the
PK> rom version), which disk we worked with and got it after reset.
There are machines where you can do this, and others where you can’t. Make a setup
that will generate a customized version of the commander for use on a
specific machine. A customized commander on another machine should either
not work at all, or disable all incompatible features. In the setup
, provide for setting the location for saving the configuration.
There are only a few standards for the RTC chip:
Glitch SMUC ZXNEWS PROFI(?) NEXT(?) ATM-TURBO2+(?)
There is no information about the last 3.
PK> Already on my own initiative in RC 2.x I decided to use another
PK> one cell (#15) to store the panel configuration, I previously agreed on her number
PK> with the people in the echo,
The opinion of the people in the echo is far from the truth.
PK> so that no conflicts arise. It is on the PC that the
PK> hardware settings are stored in CMOS, but on the Spectrum CMOS is used mainly as a
PK> clock,
Basically as... you don’t know what’s there and how. Where is it written that there
only a clock, and Vasya Pupkin cannot store the hard disk partition table there?
PK> I can completely remove the recording procedure in SMUC CMOS from the commander,
PK> but this will significantly affect the ease of use.
Make it disabled. Make a kmos editor from a disk editor (1 sector).
It will be more useful. Some chips have not 123 bytes of memory, but 241.
Wed 24 Feb 2002, at 21:20:45 Pawel Kislyak told Vassili Klimov about Real
Commander 2.2
PK> I made a CRC recording in RC in SMUC CMOS, everything works as it should,
PK> tested on an emulator with ROM version 4.01, so soon
PK> wait a while for version RC 2.3.
Where does this RC fly anyway? I still have 1.95
Bye, Pawel.
WBR, Vivid^Brainwave.
O.K. Just adjust the volume of the buffer in the 256 kilo mode on Scorp.
Again, the eighth bank is not excluded :( Or make a point in the
setup - disabling the 8th bank so as not to cause hemorrhoids.
[REAL ZX]
NA> Ok. Just adjust the volume of the buffer in the 256 kilo mode on Scorp.
NA> Again, the eighth bank is not excluded :( Or make a point in the
NA> setup - disabling the 8th bank so as not to cause hemorrhoids.
Why waste memory on an absolutely unnecessary item?
The 8th bank on Scorp is excluded by default. If
it were not excluded, then as far as I understand, shadow
after it was damaged, it would no longer be possible to enter - I checked
everything is ok on the emulator.
Another thing is if the memory on the computer is more than 256k and you enable the
limitation of used memory (256k), the volume of the buffer
will be the same as if the 8th bank was not excluded.
Since the limiter works with logical numbers
of banks, then with the 8th bank removed he takes a jar for 256k
another jar over 256k ;). I will most likely redo this mechanism, the 8th bank will not be deleted from
the table of banks, but will simply be reserved.
As for GMX, as far as I understand the essence of the problem
is that when defining the configuration as Profi,
after working with the screw, the required
bank is not installed, since ProfPZU does not know about #dffd...
But then the question is what is this character in general
GeMеX needed? After all, not a single program that uses
Profi memory and works with a disk will work?
RC suggested swapping the Profi
and Scorpion definitions, but then someone else said that this
cannot be done :). INIn general, with the glitches of ProfPZU and GMX
deal with it yourself, RC has nothing to do with it.
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
On 02 Mar 02, Pawel Kislyak wrote to Nikolaj Amosov:
[Killed.]
PK> As for GMX, as far as I understand the essence of the problem
PK> is that when defining the configuration as Profi,
to the point. :)
PK> after working with the screw, the required
PK> bank is not installed, since ProfPZU does not know about #dffd...
yep
PK> But then the question arises: why is this
PK> GeMeX needed? PK> Profi memory and working with a disk will not work?
It will if it works with a floppy disk, and not with a screw. By the way, Profi
does not know how to work with a screw under tr-dos, even more, NO MACHINE EXCEPT SCORP
CAN WORK WITH A SCREW VIA TR-DOS. So don’t hide behind the pros, GMX is
first and foremost a Scorpio.
PK> RC suggested swapping the definitions of Profi
PK> and Scorpion, but then someone else said that this
PK> can’t be done :).
And why? His pro is defined as Scorpio? Nonsense!
PK> In general, deal with the glitches of ProfPZU and GMX
PK> yourself,
And you figure out why I can’t work with disk drives when I boot from a screw
. And not on GMX, but on the green Scorp. Don't refer to emuls, I don't use them on principle. I need a subject in real life, but on PC there is Far.
PK> RC has nothing to do with itthan.
Remember what I called the people who wrote bootloaders that were turbo and
worked only from drive A:? What are the names of the people who write special programs that
do not work on machines with full addressing? Do you want to be called the same?
Don’t bend your fingers, that the subject has nothing to do with it, WHAT DOES IT HAVE TO DO WITH YOU, because. accumulated
subject. Fix it, don't push it.
PK> In ZX.SPECTRUM, messages about the release of new versions are also posted here.
PK> If _/MODERATOR_/ doesn’t mind, I can post RC 2.3 here too.
I think it won’t. You ask with soap. And what about uploading, did you fix all the bugs?
Or “we’ll fix the bugs and then remove the features”? (c) someone on irc.forestnet.org
#z80.rus ;)
ZXNet: 500:8462/1 ~/ <~ http://zs80.chat.ru/
Fido: 2:5057/21.93 /_ _> SkyNet: 71:170/21
PK> cannot be done :). In general, deal with the glitches of ProfPZU and GMX
PK> yourself, RC has nothing to do with it.
Well, as you know...
[REAL ZX]
Greetings, Sergey!
SZ> NO MACHINE EXCEPT SCORP CAN WORK WITH A SCREW VIA TR-DOS. Well, this is obvious a bend, and I couldn’t stand it! The printer works via TR-DOS with a screw. And it can - Divorce the scheme, big and small... [Sprinter] Sprinter-II [Forth-CPU] ... ivan...@mail.ru * http://st-rektal.chat.ru * http://www.petersplus.ru
save/read floppy disk images using TR-DOS_ commands, and not by pressing a button on ....
The Protoss escaped. The zerg dug in. It's time for me to call it a day. Ivan.
SZ> And why? His pro is defined as Scorpio? Nonsense!
PK>> In general, deal with the glitches of ProfPZU and GMX yourself,
SZ> And you figure out why I can’t work with disk drives when I boot from the screw
SZ>. And not on GMX, but on the green Scorp.
Version 2.2? I have no idea what this could be? With TR-DOS 5.03
I figured it out, in RC 2.2 there is now a piece of code that emulates
function #18 TR-DOS.
SZ> Don’t refer to the emuls, I don’t use them on principle. I need
SZ> a subject in real life, and on PC there is Far.
But it was only thanks to the emulation that I made support for SMUC CMOS ;)
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
SZ>> TR-DOS.
IM> Well, this is an obvious excess, and I couldn’t stand it!
IM> The sprinter works through TR-DOS with a screw. And
IM> can save/read floppy disk images using TR-DOS_ commands, and not by pressing a button on
IM> ....
Well, okay. Convinced. :) However, I don’t consider Sprinter a Spec. No offense, but it just
only has a compatible mode, so that it would be easier for users at first.
Isn’t that (or something like that) what I read about Sprinter somewhere?
ZXNet: 500:8462/1 ~/ <~ http://zs80.chat.ru/
Fido: 2:5057/21.93 /_ _> SkyNet: 71:170/21
ZXNet: 500:8462/1 ~/ <~ http://zs80.chat.ru/
Fido: 2:5057/21.93 /_ _> SkyNet: 71:170/21
Greetings, Sergey!
SZ> Well, okay. Convinced. :) However, I don’t consider Sprinter a Spec. No offense, but I’m not offended. I always answer such statements with examples: Amiga-500 is a real Amiga, and everything that comes next is no longer an Amiga, but just A real PC is an i8088 at 4.7MHz - and everything else is not a PC... And the real Spectrum is only 1982, 48kb and tape recorder... Shall we assume so? By the way, by this logic, Scorpion is not a Spec...
SZ> it just has a compatible mode to make it
SZ> easier for users at first. Isn’t that what I read somewhere (or something like this) about
SZ> Sprinter?
“something” that is compatible with an Amiga.
Historically, the Sprinter was built on the basis of the Spectrum, so it is a Spectrum...
If I danced from the MSX, it would be an MSX. If he danced from Orion he would be Orion...
etc.
IM> I'm not offended. I always answer such statements with examples:
IM> The Amiga-500 is a real Amiga, and everything that comes next is no longer an Amiga, but just
IM> only “something” that is compatible with the Amiga.
IM> A real PC is an i8088 at 4.7 MHz - and everything else is not a PC...
maybe an i8086? %)
By the way, Sinclair also produced PC-200 and PC-500 computers based on the i8086. They look
wonderful, not like modern ones... Spekovsky style. :)
IM> And the real Spectrum is only 1982, 48kb and a tape recorder...
Sorry, but there was a ZX128. Also kind of like Spectrum. In Sinclair User magazines the name was ZX Spectrum 128. I don’t mean 2+ and 3+, because... they were produced by Amstrad, and not
Sir Sinclair.
IM> Shall we assume so?
You are thinking correctly. :) Almost like me.
IM> By the way, by this logic, Scorpion is not Speck...
And I’m saying that it’s a clone. In my opinion, the most thoughtful and successful. This is just
imho. :)
SZ>> don’t write it down?
VK> so damn, you document all sorts of kmos, gmhi, etc.
SMUC hours are documented. Described how to work via rst8. This is
enough for me.
VK> And then at first
VK> they say, there is hardware - there is no software. But how to write it if there is no
VK> documentation?
And if there is it, but they don’t read it? :)
VK> For example, how many programs do Gluck’s watches support?
VK> and how many do Smukov’s watches support?
Well, who has more? To be honest, I very rarely see mention of glitchy watches.
VK> although the difference is only in the ports and some kind of CRC.
CRC has nothing to do with the watch. They can be changed without any CRC recalculations - via
rst8.
Konnichi wa, Vassili-san!
Once Friday March 8, 2002 at 00:21:19, Vassili and Pawel were arguing about Real
Commander 2.2
PK>> Okay, I’ll swap the definitions, then blame yourself
PK>> ;).
VK> then be sure to check before writing to #1ffd cell
VK> #0008 in ROM 48 bass,
VK> and if there is a standard value there, in no case
VK> do not use
VK> this port!
forgot about Kay256?
Konnichi wa, Vassili-san!
One day Friday March 8, 2002 at 00:24:04, Vassili and Sergey were arguing about Real
Commander 2.2
VK> PS: by the way, #dffd historically appeared in the pros before
VK> #1ffd and scorpion.
Forget it because it doesn't matter
even in a number of branded bourgeois programs in port #1ffd something was written and tested
:)
Terminator 2
Gryzor 128
Total Recall
and something else
and since the versions were localized
you looked like this
di
ld bc,#1ffd
ld a,nn
nop
a whole bunch of nop s
ld bc,#7ffd
ld a,#10
further the standard definition of #7ffd
experienced hackers assure that the port was used for something and
Zonov only slightly altered it for himself
[Wizard Warz] Yurik aka Jerri / Alien Factory /I \
[No RAR allowed] [SoHm] [Conceptual Minimalism] \ZX/
PK>>> RC has nothing to do with it.
SZ>> Remember what I call the people who wrote bootloaders that were
SZ>> turbo and worked only from drive A:? What are the names of the people who write
SZ>> special programs that do not work on machines with full addressing? Do you want
SZ>> to be called the same?
SZ> Which, by the way, is also buggy. :) You stopped recording in different registers why not
SZ> then
SZ>? So damn, you document all sorts of kmos, gmhi, etc. And then at first they say,
there is hardware - there is no software. How to write it if there is no documentation? For example,
how many programs does Gluck's clock support? how much are Smukovsky ones? although the difference
is only in the ports and some kind of center.
SZ> Tim Kelly 2:465/240
SZ> if nothing has changed. I always wrote there.
maybe he doesn’t read it since there are no rules ;)?
VK> then before writing to #1ffd, be sure to check cell #0008 in ROM 48
VK> bass, and if there is a standard value there, under no circumstances
VK> do not use this port! them...
Then test the presence of each extension, not forgetting to restore
the default value
after the test (this also applies to caches and general sounds). Disable interrupts,
test program
place in bank 5, preferably from an address that is a multiple of 2^n and don’t forget to
duplicate it in other banks from the 128kb set at the same addresses...
It’s easier to make a manual configurator and forget about pluginplay (because
it won’t be done
competently anyway).
PK>>> Well, I wasn’t just about to upload it, the glitches that
PK>>> I know have been fixed,
PK>>> I’ll wait another week, maybe something else will appear.
07 Mar 02 23:05, Sergey Zotov wrote to Vassili Klimov:
SZ> And they released more Scorpovs. :) We meet more often imho at the moment,
SZ> but we have to keep up with the times. ;)
Keeping up with the times - this is Take Out the Crap for professionals.
The Protoss fled. The zerg dug in. It's time for me to call it a day. Ivan.
- Divorce the scheme, big and small... [Sprinter] Sprinter-II [Forth-CPU]
... ivan...@mail.ru * http://st-rektal.chat.ru * http://www.petersplus.ru
SZ> More precisely about all Scorpios, of which there are more than Profi owners. :)
IMHO, not a fact. It seems to me that it’s the other way around. But I can’t give an argument, just like you ;). By the way, I am a “Profi owner” formally (I only have professional boards), the memory is switched via 7ffd and the temporary Pentagon one.
KF> This is all somehow starting with the letter L... (there is Kai).
uh-huh... but by the way, Kai has the 7th bit #7ffd...
KF> Make a table of the ports to which you will write and a list of default
KF> values for them...
KF> Then test the presence of each extension, not forgetting to restore
KF> the default value
KF> after the test (caches and This also applies to general sounds). Disable interrupts,
KF> place the test program
KF> in the 5th bank, preferably from an address that is a multiple of 2^n and do not forget
KF> to duplicate it in other banks from the 128kb set at the same addresses...
all that remains is to remember the ATM, in which the lower addresses (several pieces)
swap places when switching the video mode.
In general, with you didn’t think of something with default values... We don’t know
there is a port or not, and if there is, then which one - 7ffd, dffd, 1ffd ?
KF> It’s easier to make a manual configurator and forget about pluginplay (because
KF> done correctly
KF> still won’t be).
in scenergy the best one is probably... All that remains is to convince Pasha;). Although
my opinion, you can do everything wisely and without a configurator.
JTN/4D
PK> I don’t know the address, rules don’t post here often.
=== file RULEZ_RS.TXT ===
┌────────────────────────────── ───────────────────────────┐
│ RULES OF E H O CONFERENCE REAL.SPECCY │
│ last edition date 06/16/99 │
└──────────────────────────────── ─────────────────────────┘
This echo conference was created for communication on the topic of the
ZX-Spectrum platform, mainly for people who have access to
Fido using real Spectrum-compatible computers (although, in principle, anyone who will
follow these rules can participate in the correspondence).
1. It is prohibited to mention other platforms (you can mention
them if it is related to the features of the software being discussed and hard); It is also prohibited to compare platforms with each other, as well as to criticize them [*].
1.1 Discussion of emulators is unacceptable, this is an echo for the
_real_ Spectrum [*].
2. Skirmishes between different modifications of Spectrum-compatible computers are extremely undesirable (please respect
the creators and users of all types of our warmly beloved
Specchi) [*].
3. Excessive quoting is punished (more than 70%) [*],
try to leave those phrases to which, in your opinion, you
are giving an answer or comment.
4. A little is allowedadvertising, forward from other information that you think is interesting.
5. Files placed in the echo conference must only be
in UUE encoding and meet the standards of one of the following
categories:
a) standard Spectrum file in Hobeta format (*.$*);
b) text file in 866 GOST encoding (alternative) (*.txt);
c) ROM firmware files in binary code (*.bin);
d) disk images of the Shalaev emulator (*.trd);
e) raster graphics files (*.pcx, *.bmp, *.jpg).
Before posting, the above files _necessarily_
must be packed with Spectrum ZIP (*.$z) or a similar archiver on a PC (*.ZIP), the archive must
have a description of the contents in a text file (it is advisable that this file have the standard name file_id.diz,
descript.ion or read_me).
Please do not send files to the echo conference that have symbols in their names that contradict standards
of the operating systems IS-DOS, MS-DOS (.+,"=, etc.)
Placing files in UUE encoding larger than 20Kb
_only_ with the permission of the Moderator or Commodator! [+]
The size is determined by the source file.
When posting files over 20Kb, at the beginning of the letter it is _mandatory_
it must be indicated with whose permission it is being performed posting.
Request permission only innetmail.
The UUE section should not exceed 100-120 lines!
5.1 The "NeOS" project was declared a "pilot" project of the echo conference, it follows that permission to post new versions and documentation, as well as software designed to work
under NeOS, was issued by Roman Petrov AKA Megus/Brainwave^X-Project
(2:5052/7.37@fidonet). The permission includes unlimited UUE traffic without prior approval by a moderator.
Coordination of the project and limitation of daily UUE traffic
remains on the conscience of Roman Petrov.
5.1.1. UUE posting, authorized by the NeOS project coordinator, must be accompanied by a corresponding inscription in the first lines of the posting message.
6. Personal correspondence is prohibited! [*]
7. Communication with the Moderator/Comoderator only in netmail,
send all complaints here. Unauthorized moderation
is punished [+].
8. Of course, I also read and write in this conference, so
letters only in the name of Moderator
or CoModerator are considered official.
9. “Hacker” conversations (about hacking programs, disrespect of copyright, etc.) are strictly prohibited [+]. Advertising
hacked products or posting them in echo entails a lifelong ban [!].
9.1 Distributioninformation about all kinds of cheats -
disrespect for the conference subscribers, so I will
punish you mercilessly! [*].
9.2 Distribution of solutions is also not welcome at all! [*] If they ask for help in completing any game,
answer with netmail, because for others it may be completely
nothing!
10. As in ZX.SPECTRUM - the use of
nicknames in the From/To field is not allowed (let's not start a kindergarten,
the “cool hackers” are already making a dazzle in the eyes). The exception is
Moderator/CoModerator :) .
CoModerator has every right to distribute stars and pluses,
beware :) .
Re-election of the Moderator is a malicious offtopic. As long as I'm alive, you will keep order for me here! :)
Language of communication - Russian, Ukrainian, English. Try not to complicate the lives of others, write more clearly, if possible competently, correct the mistakes of others correctly and, if possible, by netmail.
Otherwise - well, everything is described in the echopolis, what is not
prohibited is allowed.
The echo conference is available for gating to all networks
FTN-standard, gate sysops _must_ ensure the correct
content of the "From:","To:" fields in accordance with the standards
of the _FidoNet_ network.
Standard penalties:
[*] - warning;
[*] [*] [*]= [+] - transfer to ridonly for a month;
[+] [+] [+] = [!] - disconnect from echi for an extended period! :( .
Amnesty is provided only for very large holidays ;).
Moderator of REAL.SPECCY - Tim Kelly (2:465/240)
CoModerator of REAL.SPECCY - Roman Milukov (2:46)kon=fayla=29 ===
All the best.
Romance.
[ZX]
YP> Forget it, because it doesn't matter
YP> in a number of branded bourgeois progs in port #1ffd something was written and
YP> checked
YP> :)
YP> Terminator 2
YP> Gryzor 128
YP> Total Recall
YP> and something else
we all know that. The conversation was about switching additional. memory, firstly. There was no trace of her in Amstrad's crafts. And secondly, show me +2, +2A,
+3 (I think it was in the last one) with trdos, for us they are not relevant, so I don’t count them. But this does not mean that I exclude them completely, it’s just that we were talking about
such a specific problem as determining a port in specific (trdos based)
software.
YP> and since the versions were localized
YP> you looked like this
YP> di
YP> ld bc,#1ffd
YP> ld a,nn
YP> nop
YP> a whole bunch of nop s
YP> ld bc,#7ffd
YP> ld a,#10
YP> further standard definition #7ffd
it’s not clear why nops are needed;) ?
YP> experienced hackers assure that the port was used for something and
YP> Zoned it only slightly for themselves remade
slightly, not slightly. It would be better not to touch him at all. For example, I have ports
configurations #80fb and #00fb and they don’t glitch with anything because access to them
opens in a,(#fa) and closes in a,(#fe).
JTN/4D
07 Mar 02 23:39, Yuri Potapov wrote to Vassili Klimov:
YP> in a number of proprietary bourgeois programs in port #1ffd something was written and
YP> checked :)
YP> experienced hackers assure that the port was used for something and
YP> Zonov only slightly modified it for himself
Amstrad 2 or 3 computer. There was 256kb of RAM, CP/M mode and everything on this
port.
It is very likely that the bit layout for the memory coincides with Scorpio.
SZ>> CRC - via rst8.
VK> the point is that since the structure of the clock is the same, it’s easier to substitute different ports
VK> than to write completely different procedures.
It’s not so simple - ports in tr-dos and recording in the clock through one place
is done.
And logically, the Pentagon + modifications are not the Pentagon :)
C uv.Maxony latyshm(at)mail.ru
SZ> Sorry, but there was a ZX128. Also kind of like Spectrum. In the magazines
SZ> Sinclair User
SZ> was called ZX Spectrum 128. I don’t mean 2+ and 3+, because they were produced by
SZ> Amstrad, and not
SZ> Sir Sinclair.
Let's start with the fact that Sir Sinclair did not produce anything, he invented.
C uv.Maxony latyshm(at)mail.ru
KF> Amstrad 2 or 3 computer. There was 256kb of RAM, CP/M mode and everything on this
KF> port.
KF> It is very likely that the bit layout for the memory coincides with the scorpion.
there was no 256kb there, 1ffd was only in +2A, with it you could switch
and turn off ROM.
JTN/4D
KF> It's not that simple - ports in tr-dos and recording to the clock through one place
KF> is done.
However, it is much simpler than through rst8, no extras are needed. books
read.
Moreover, how to debug if there is no real problem. The likelihood of
errors
is many times greater with 2 procedures than with one and port forwarding.
JTN/4D
VK> it’s not clear why nop’s are needed;) ?
for pentagon with its #fd addressing IMHO false positives were removed
VK> slightly, not slightly. It would be better not to touch him at all. For example, I have
VK> ports
VK> configurations #80fb and #00fb and they don’t glitch with anything because
again, it’s difficult to accidentally access these ports :)
VK> access to them
VK> opens in a,(#fa) and closes in a,(#fe).
hmmm, cool :) and turn on/off the turbo by reading do you know from ports?
Konnichi wa, Kirill-san!
One day Saturday March 9, 2002 at 00:58:32, Kirill and Yuri were arguing about Real
Commander 2.2
KF> Amstrad 2 or 3 computer. There was 256kb of RAM, mode
KF> CP/M and that's it
KF> port.
KF> It is very likely that the bit layout for memory
KF> coincides with the scorpion.
no, I don’t know about memory, although something similar still pops up in .tzx
where in #5bxx there are also calls to ports via #1ffd
well, it seems that this is not memory, but the ROM was silky there and maybe it’s true CP/M
Sergey Zotov ==========> Pawel Kislyak:
SZ>>> Don’t refer to emuls, I don’t use them on principle. I
SZ>>> need
SZ>>> a subject in real life, and on PC there is Far.
PK>> But only thanks to the emulator I made support for SMUC CMOS ;)
SZ> Which, by the way, is also buggy. :) You stopped writing things down in different registers
SZ>?
What don’t I write down if I miss it ;). The use of two CMOS cells in RC
does not affect the operation of Scorpion in any way. If there are complaints, I will
disconnect.
PK>>>> RC has nothing to do with it.
SZ>>> Remember what I call the people who wrote bootloaders that
SZ>>> were turbo and worked only from drive A:? What are the names of the people
SZ>>> who write special programs that do not work on machines with full
SZ>>> addressing? Do you want to be called the same?
PK>> But why am I, I don’t see any analogy ;).
SZ> And the fact is that the glitches are known and can be easily bypassed. Eliminate the upper
SZ> memory for Pro and 8th bank. It has not been ruled out for you for already
SZ> 2 (!) versions. Reason to think.
The 8th bank was always excluded (the shadow engine would not have been called if the 8th bank
had been damaged during copying). It’s another matter if Profi was determined
and the wrong bank was excluded...
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
Bye!
Pa...@nm.ru www.realsoft.hotbox.ru [RC 2.0] [Real Software]
YP> hmmm, cool :) can you turn the turbo on/off by reading from the ports?
hehe;) well, there’s no need to invent a bicycle here, I did it like on a scorpion +
button. By the way, it’s convenient to debug multicolors in ASMA, but when you start the program it turns off.
JTN/4D
KF>> the place is being made.
VK> However, it is much easier than through rst8, you don’t need to read any additional books.
VK>. Moreover, how to debug if there is no real thing with a smook.
VK> The probability of errors is many times greater with 2 procedures
VK> than with one and with port forwarding.
Do you know how exactly you need to get into the clock registers on SMUC?
Do you know about the update bit and what it guarantees there? And you know
that you need a good perversion to meet this guaranteed time without
turbo mode, otherwise you can catch a glitch (of course it’s easy to get around it by reading the time again and comparing the two read values, BUT WHO THINK ABOUT THIS
UNTIL EVERYTHING IS FUCKING GLUED???)
KF>> Amstrad 2 or 3 computer. There was 256kb of RAM, CP/M mode and everything
KF>> on this port. It is very likely that the bit layout for the memory
KF>> coincides with the scorpion.
VK> 256kb was not there, 1ffd was only in +2A, with the help of it it was possible
VK> to switch and disable the ROM.
Some of the machines had it.
ZXNet: 500:8462/1 ~/ <~ http://zs80.chat.ru/
Fido: 2:5057/21.93 /_ _> SkyNet: 71:170/21
SZ> slowly... :( the same person. He didn’t solder the parts himself
SZ>, although
That’s what was meant!
C uv.Maxony latyshm(at)mail.ru
KF> default values for them... Then test the presence of each
KF> extension, not forgetting to restore the default value after the test
KF> (caches and general sounds also apply).
The default value is restored.
KF> Disable interrupts,
KF> place the test program in bank 5, preferably from an address that is a multiple of
KF> 2^n and do not forget to duplicate in other banks from the 128kb set at
KF> the same addresses...
So far there have been no problems without this.
KF> Do you know how exactly you need to get into the clock registers on SMUC?
approximately, it will be necessary - I’ll find out for sure.
KF> Do you know about the update bit and what it guarantees there? And you know
I know
KF> that you need a good perversion to meet this guaranteed time without
KF> turbo mode, otherwise you can catch a glitch (of course it’s easy to get around
KF> by repeated
KF> reading the time and comparing the two read values, BUT WHO WILL THINK ABOUT THIS
KF> UNTIL EVERYTHING IS GONE FUCK???)
what's wrong? well, even if sometimes it is considered to be the wrong value, spectra are not supposed to be installed in life support systems or spaceships.
JTN/4D
13 Mar 02 00:13, Vassili Klimov wrote to Kirill Frolov:
KF>> it’s easy to get around by re-reading the time and comparing the two
KF>> read values, BUT WHO THINK ABOUT THIS UNTIL EVERYTHING IS
KF>> THE FUCK IS GLUING???)
VK> what’s going to glitch? Well, even if sometimes it’s considered to be the wrong value, Spectrum
VK> seem to be not installed in life support systems and spaceships
VK>.
Today the clock in the comader will glitch, tomorrow some kind of communication
program
like fidomeyler, and the day after tomorrow they have the problem of the year 2000.
10 Mar 02 00:42, Pawel Kislyak wrote to Sergey Zotov:
PK> What I don’t write down is that I miss ;). The use of two CMOS cells in RC
PK> does not affect the operation of Scorpion in any way. If there are complaints, I will
PK> turn it off.
In Scorpio without a professional ROM there is no kmos (and if there is, then according to the zxnews
connection diagram).
In a Scorpio with a professional ROM you cannot write to kmos.
VK> what's going wrong? Well, even if sometimes it is considered the wrong value, spectra
VK> seem to be not installed in life support systems and spaceships
VK>.
IMHO Kirill is right. Once at the beginning you wave your hand - scoop it up later (others)
will get tired.
All the best.
Roman.
[ZX]
RM> IMHO Kirill is right. Once you wave your hand at the beginning, you will have to scoop it up later
RM> (others)
RM> will get tired.
ideologically - yes, practically - no. A programmer cannot write for buggy hardware without having it. In general, we were talking about the subject;) and there the watch is a purely cosmetic thing. Now, if only someone would write a plugin...
speak: why is AXLR with its board, where did it go? even on the Internet
they are interested...
JTN/4D
RM>> then (others) will get tired.
VK> ideologically - yes, practically - no. A programmer cannot write for
VK> buggy hardware without having it.
However, they write, and many do it well. Complete information about the
device is enough (unless, of course, there is a struggle for every clock cycle/byte).
VK> In general, we were talking about the subject;) and there
VK> the clock is a purely cosmetic thing.
The subject I saw only versions 1.5 or 1.6, I don’t remember, I won’t say anything specifically.
VK> Now, if only someone
VK> would write a plugin...
For what? Now, if you use a non-Scorpio screw, then CMOS will be very necessary.
VK> PS: why is AXLR with its board, where did it go? even on the Internet
VK> they are interested...
He is not missing. Right now he’s intensely assembling a prototype, (there are tons of parts there)
yesterday he came to me to scrape the bottom of the barrel, but I don’t have much
left.
We have to wait.
All the best.
Roman.
[ZX]
RM> He is not missing. Right now he’s intensely assembling a prototype, (there’s a ton of parts in there
RM>) Yesterday he came to me to scrape the bottom of the barrel, but I really
RM> don’t have much left.
RM> We have to wait.
The price is the same - $20 for a bare board?
JTN/4D
RM>> is very necessary.
VK> well, they came up with catalogs in holedos, maybe they’ll come up with a creation date
VK>;)
Why the hell do we need holedos? No, we don’t need holes in the screw. It is necessary that it could work with disk images, something like a Spectrum navigator (or whatever it is called?). :)
RM> Same. We have 2 tanks per square decimeter.
so for 20 tanks then 10 dm2 is obtained, and this is a whole landfill and not a fee :)
C uv.Maxony latyshm(at)mail.ru
RM> Why the hell do we need holes? No, we don’t need holes in the screw. It is necessary that it could work with
RM> disk images, something like a Spectrum navigator (or whatever
RM> it is?). :)
Why the hell do we have holes in disk images?
ML> so for 20 tanks then 10 dm2 is obtained, and this is already a whole range and not a
ML> board :)
The range will be very glitchy.
ML> so for 20 tanks then 10 dm2 is obtained, and this is already a whole landfill and
ML> not a fee :)
And it’s almost like that! :) Well, just count how many things there are - ISA
3 pieces, and DMA sound, and a bunch of improvements... and all on simple logic without any
alterations.
All the best.
Roman.
[ZX]
RM> Same. 2 bucks sq. decimeter we have worth.
so what's there with the board? Buy can be already?
Sincerely, Vadim.
... AY now play: Silence.
VC> so what about the board? Can you buy it already?
and don’t feel sorry for yourself: for $20 you can still get hemorrhoids
JTN/4D
VK> and don’t feel sorry for yourself: for $20 you can still get hemorrhoids
so, I’m interested for now... :))
Sincerely, Vadim.
... [IЪZX]AY: Silent now...
RM> And it’s almost like that! :) Well, just count how much is there - and
RM> ISA 3 pieces, and DMA sound, and a bunch of improvements... and all on simple logic without
RM> any alterations.
Which version of zx is taken as a basis? Otherwise I missed something.
C uv.Maxony latyshm(at)mail.ru
ML>> not a fee :)
RM> And it’s almost like that! :) Well, just count how much is included there - and
RM> ISA 3 pieces, and DMA sound, and a bunch of modifications... and all on simple logic without
RM> any alters.
For some reason it seems to me that it will be terribly glitchy, and it will need to be set up
as long and painful as ATM-Turbo.
RM>> could work with disk images, something like a Spectrum navigator
RM>> (or whatever it is?). :)
KF> Why the hell do we have holes in disk images?
And on my hard drive everything is in images. I don’t have SMUK, but a simple non-Mov
connection - there is nothing except is-dos. The image of the hole-dos-disk is loaded into the ram-disk
and you can work directly there. So, it would be nice to do
the necessary operations directly from the commander directly on the propeller.
(although really - why the hell? :)
Good luck.
Roman.
[ZX]
VC> so what about the board? Can I buy it already?
Not yet, wait...
All the best.
Roman.
[ZX]
ML> Which version of zx is used as a basis? Otherwise, I missed something.
Pentagon 1024 on a Simma with a turbo 130 thousand clock cycles (as I understand it, the memory works
at 7Mhz, unlike the classic Pentagon)
JTN/4D
RM>> divorced - and ISA 3 pieces, and DMA sound, and a bunch of improvements... and all on
...
ML> Which version of zx is taken as a basis? Otherwise I missed something.
I don’t know, maybe it’s pent, but most likely something Leningrad-like.
Good luck.
Roman.
[ZX]
RM> I don’t know, maybe pent, but most likely something Leningrad-like.
Leningrad (yellow scorpion) is a very buggy spec.