Sprinter hard & soft

;/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Clocking in ZX-Spectrum mode.

Pentagon mode:
1. 3.5 MHz
2. No WAIT
3. 320 lines per frame.
4. INT signal on the 80th line after the bottom edge of the screen.
Starts at the beginning of the screen blanking pulse. Ends
in 12 μs or on a signal (/IORQ or /M1) - response to INT.
INT is configured according to the RAGE demo program with Enlight-97 for border
effects.

Scorpion mode:
1. 3.5 MHz
2. There are no WAIT
3. 312 lines per frame.
4. INT signal on the 64th line after the bottom edge of the screen.
Starts at the beginning of the screen blanking pulse. Ends
in 12 μs or by a signal (/IORQ or /M1) - response to INT.

Port configuration in ZX-Spectrum mode.

Pentagon mode:
1. #FE - keyboard/BORDER
2. #FF - screen attributes (port FF)
3. #1F/#0F - Kempston port
4. #7FFD - port of the 128th mode decoding via A15,A7,A6,A5,
A2,A1,A0 commands work OUT (#FD),A

Scorpion mode:
1. #FE - keyboard/BORDER
2. #FF - screen attributes (port FF)
3. #1F/#0F - Kempston port
4. #7FFD - 128th mode port decoding by A15,A14,A7,
A6, A5,A2,A1,A0
5. #1FFD - 256th mode port decoding by A15,A14,A13,
A7, A6, A5, A2, A1, A0

Bits, naturally, the same as in Pentagon and Scorpion.

Ports occupied in Z84C15:
(!!! This is a given and cannot change!!!)
#10..#1F - timers, parallel and serial ports
#EE,#EF,#F4 - ports control the operation of the processor
Port #1F is redirected to address #0F. In all
Spectrum mode programs, the
IN A,(#1F) and OUT (#1F),A commands are allowed. In Sprinter mode, the program
must read Kempston data from port #0F.

Port configuration in Sprinter mode:

1. All Scorpion configuration ports.
2. Processor window ports:
#82 - PAGE0 - changes the page connected to The 0th bank
when it is connected via bit 0 of port #1FFD
#A2 - PAGE1 - changes the page connected to the 1st bank
#C2 - PAGE2 - changes the page connected to the 2nd bank
#E2 - PAGE3 - changes the page connected to the 3rd bank
Changes the number of the bank bank that is currently connected
via #1FFD/#7FFD
Changing the page number in the 3rd bank does not change
the page numbers in the others. That is If you did
OUT (PAGE3),A when the 5th page
of the 128th mode was in the 3rd bank, then what was connected in the 1st bank
of the processor will not change. Same with the second and zero.
3. Reading from ports is available#1FFD, #7FFD and from PAGEi
4. There are many other ports hidden in TR-DOS and
managing configurations. All new ports can be
disabled completely by software.

Memory allocation:
The Sprinter ROM has a set of functions for
memory allocation. The programmer does not need to think about which
pages are occupied and which are not. Upon request, the program in ROM
will carry out all the necessary calculations, search for free
pages, tell how many are still free, provide the required
number of pages and provide a list of numbers
of the pages provided. Using this program does not
entail a decrease in the performance of the program, since
all actions to connect pages can be performed
using OUT (PAGEi) commands, where in A the
provided numbers are substituted.

Graphic mode screen configuration 320x256x256
1. The ROM has a function to open the full screen in this
resolution. Called via point #3D13. There are two
such screens.
2. The ROM has functions for drawing simple horizontal
lines and dots.
3. The ROM has a function for setting the palette. (selection of 256
colors out of 16 million)
4. Structure of screen pages.
Graphic RAM pages - #50..#5F
On each page there are 16 lines 1024 bytes long.
To the graphic screens on theseThe lines include segments
#040..#17F and #180..#2BF
The first segment lies on the first screen, the second - on
the second. To draw horizontal lines from right to left
you can use the LDIR command
To draw vertical lines from top to bottom you need to
add the value 1024 (#400) to the point address, and when
you cross the block boundary, switch it to the next one
5. After closing the graphic screen, switch to
Spectrum-SCR, print to Spectrum-SCR and return to
graphic screen, the picture of the graphic
screen does not change. The palettes of the graphic and
Spectrum modes are separate

Configuration of the text screen 80x32: The internal structure of the text screen
is quite complex for
a simple description, so all work with it is done at the
BIOS level. BIOS provides:
- mode opening functions
- printing characters with attributes, without attributes and printing
only attributes
- printing identical characters with attributes, without attributes
and attributes alone
- printing lines of a certain length and up to the control symbol
- positioning function
- functions for scrolling the screen up/down indicating
the lines to be shifted
- function for clearing the entire screen and rectangular areas

;/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

I hope this informationwas interesting. I will accept all questions
and consider all proposals.

░▒▓██ Best regards, Ivan. ██▓▒░

TR-DOS 5.04Ei

Somehow Thu Feb 12 1998 05:23, Ivan Mak wrote a letter to All:

---cut---

IM> Scorpion mode:
IM> 1. 3.5 MHz
IM> 2. No WAIT
IM> 3. 312 lines per frame.
IM> 4. INT signal on the 64th line after the bottom edge of the screen.
IM> Starts at the beginning of the screen blanking pulse. Ends
IM> after 12 microseconds or by signal (/IORQ or /M1) - response to INT.

---cut---
And where is Scorpion Turbo and Turbo+ ???
(True, there were 5 modifications)

Bye... Bye bye, Sersh!

Dirol with xylitol and urea

SZ> And where are Scorpion Turbo and Turbo+ ???
SZ> (True, there were 5 modifications)
Scorpion mode - only by memory allocation and the presence of Expansion ROM. (The very
first Scorpion)
Turbo mode in Sprinter-e is at 21MHz. Expansion ROM is naturally ours, and not from Scorpion. We are not going to violate copyrights.
Actually, I do not have complete information on Turbo and Turbo+, so these
modes will not exist. (Unless you wish :))

IM> Memory distribution:
IM> The Sprinter ROM has a set of functions for
IM> memory distribution. The programmer does not need to think about which
IM> pages are occupied and which are not. Upon request, the program in ROM
IM> will make all the necessary calculations, search for free
IM> pages, tell how many are still free, provide the required
IM> number of pages and provide a list of
IM> numbers of provided pages. The use of this program does not
IM> entail a decrease in the speed of the program, since
IM> all actions to connect pages can be performed
IM> using OUT (PAGEi) commands, A where the
IM> provided numbers are substituted in A.

The procedure for setting a page by number would not hurt, i.e. I know that I have N
pages and the ROM knows which ones are free; I ask it to set me page
1,2,3, and it itself rummages through its tables and sets the first _free_
page, and not the physical one. For greater convenience, you can combine the procedures
of installation and finding out the number of pages - so that the output
accumulator contains the number of pages available to me.

Huh?
Yours truly, Hexman/BYTEX.
[Team ZX ASM] [Team Metallica] [Team Art Works]

Thu 12 Feb 1998 05:23, Ivan Mak wrote to All:
[skip]

IM> Port configuration in Sprinter mode:

IM> 1. All Scorpion configuration ports.
IM> 2. Processor window ports:
IM> #82 - PAGE0 - changes the page connected to the 0th bank
IM> when it is connected via bit 0 of port #1FFD
IM> #A2 - PAGE1 - changes the page connected to the 1st bank
IM> #C2 - PAGE2 - changes the page connected to the 2nd bank
IM> #E2 - PAGE3 - changes the page connected to the 3rd bank
[skip]
IM> 3. Reading is available from ports #1FFD, #7FFD and from PAGEi
IM> 4. There are many other ports hidden in TR-DOS and
IM> that control configurations. All new ports can be
IM> programmatically disabled completely.

Can you manipulate pages not equal to #4000, but for example #800?
Some problems are convenient to solve with small pieces of memory.

IM> ░▒▓██ Best regards, Ivan. ██▓▒░

IM> TR-DOS 5.04Ei

IM> -+- GoldED 2.50+
IM> + Origin: The Rectals lived on the planet Rens for 2 million years. (2:5030/529.24)

Sincerely, Ilya Trofimov

...
IT> Is it possible to manipulate pages not equal to #4000, but for example
IT> #800 ? It is convenient to solve some problems with small pieces of memory.
No. All pages are 16k and at the same addresses.
However, for some tasks you can make some firmware, which
will contain the necessary addressing and the required page size. (Imho, this is already perverted)

Ivan

IM> However, for some tasks you can make some firmware, in
IM> which will have the necessary addressing and the required size pages (Imho, this is already
IM> a perversion)

this is not a perversion, but a cruel necessity. (for flexibility and speed)

Good Luck...

... "Beam me aboard, Scotty!"
"Aye, Cap'n! Uh, will a two-by-four do?"

[skip]
IM> No. All pages are 16k and at the same addresses.
IM> However, for some tasks you can make some firmware, in which
IM> there will be the necessary addressing and the required page size. (Imho, this is already a perversion)

Let's say that partitioning by #4000 is just a tribute to habit, but complex tasks,
for example, the OS wants the ability to project a piece of memory of arbitrary
length from an arbitrary address, but the problem of memory defragmentation arises,
unknown with pages of a fixed length.

And how quickly the firmware can be changed, are algorithms available? and
reconfiguration procedures, what are the general limits of the system's changeability?
(sorry for possibly incorrectly posed questions, but information
is sorely lacking)

By the way, mail dated 02/03/98 subj: Has Nica ICS arrived completely or has it died with the HDD?

IM> Ivan

IM> -+- GoldED 2.50+
IM> + Origin: Rectals lived on the planet Rens for 2 million years. (2:5030/529.24)

Sincerely, Ilya Trofimov

MR> this is not a perversion, but a cruel necessity. (for flexibility and speed)

for some reason they think that by adding a little crap (covox, axel with unnecessary commands) they will do something cool. but they
don’t want to do all sorts of useful features.

MR> ! Origin: My computer is 100% !ntel Free (2:5030/659.24@fidonet)

I have it too :)

▌█▐▐║▌█▐▐▌▐ W!7h 7he Be$7 W!$he$ fR0M CREATOR product
▌812/03.00▐ Powered by Cyrix 6x86L 200+/16DIMM/S3TrV2 2mb/SBpro/Q1.6Gb/USR28.8

... Change you pampers.

IM>> which will contain the required address and the required page size. (Imho,
IM>> this is already a perversion)

IT> Let's say that partitioning by #4000 is just a tribute to habit, but complex
IT> tasks, for example, the OS would like to be able to project a piece of memory
IT> of arbitrary length from an arbitrary address, but the problem
IT> of memory defragmentation arises, which is unknown with pages of a fixed length.

IT> A How quickly does it take to change the firmware, are
IT> algorithms and reconfiguration procedures available, what are the general limits
IT> of system changeability? (sorry for the possibly incorrectly posed
IT> questions, but there is a catastrophic lack of information)
Reconfiguration with a change in the circuit occurs in about half a second.
Switching ROM firmware (previously loaded into RAM) is simply output
bytes to the ports. Occurs in units, maximum, tens of microseconds.
Reconfiguration of ports depends on how many ports need to be changed. There are
four loaded configurations that can be switched through the port, and
changing the configuration by overwriting data means sending 4 kilobytes
of information.
The limits of system variability lie within the boundaries of 4Mb of memory. You can
load something else from the HDD, but it most likely won’t be needed.
And judging by the changeability of the scheme, the boundaries are simply not visible. Certainly,You can't add much to it, but you can do a lot.
Advanced graphics occupy most of the circuitry in PLM. It can be
simplified a little and some circuits can be added to make it easier to perform specific
operations needed at the moment.

IT> By the way, mail dated 02/03/98 subj: Did Nica ICS arrive completely or
IT> was lost with the HDD?
It arrived, unpacked normally and the trd-worker itself did not die, because was copied onto the Sprinter screw. Only the addresses that needed to be answered were lost.

Ivan

Flexibility, yes, but speed.. Ha-ha-ha..
To switch the 16K page you need one OUT to the port, and to connect RAM from
the desired address - at least 3. Since you need to set the address for 4Mb.

Ivan

Sun 22 Feb 1998 02:52, Ivan Mak wrote to Ilya Trofimov:

[skip]

IM> Most of the circuit in PLM is occupied by advanced graphics. It can be
IM> simplified a little
IM> and some circuits can be added to make it easier to perform specific
IM> operations needed at the moment.

Excellent! And what brand of PLM is used, I will look for the docks.

IM> Ivan

IM> -+- GoldED 2.50+
IM> + Origin: Rectals lived on Rense for 2 million years. (2:5030/529.24)

Sincerely, Ilya Trofimov

IM>> specific operations required at the moment.

IT> Excellent! And what brand of PLM is used, I will look for the docks.
EPF10K10QC204-4
Actually, there is no need to look for docks using it. I have everything. And the
MAX+Plus II program, in which circuits are automatically routed. All you need is a diagram that can be pushed into this PLM. Or a file in a high-level language. (Language for circuits)
If you are interested, the PLM contains 768 bytes of fast memory, 576 logical
elements with four inputs and one output with any function, and 576 D-type flip-flops.
By the way, I can provide information on the AHDL circuit language here.
Who needs it? ? (Although perhaps completely offtopic)

Ivan

IM> EPF10K10QC204-4
^
Maybe EPF10K10A..., on their website with such parameters the link is only
EPF10K10A.. ?

IM> Actually, there is no need to search for docks on it. I have everything. And the program
IM> MAX+Plus II, in which circuits are automatically routed. All you need is

Documents are good, especially if they are translated. And the full MAX+Plus II, is it available on CD? It is not advisable to download 14Meg WEB-Relise with a limited service life.
The documents that I got from the site made a shocking impression, I suspected
such technology, but did not imagine its vastness...

[skip]

IM> schemes) If you are interested, the PLM contains 768 bytes of fast memory, 576
IM> logic elements with four inputs and one output with any function and
IM> 576
IM> D-type flip-flops. By the way, I can provide information here on the schema language
IM> AHDL. Who needs it? (Although, perhaps, completely offtopic)

Offtopic most likely no longer, due to its direct relation to
the development of Speccy... Only if they kick you out of HARDWARE.ZX :)

2 All : Jast look at www.altera.com & www.zilog.com !
--------------------------------------------

IT> Maybe EPF10K10A..., on their website with such parameters there is only a link
IT> EPF10K10A.. ?
Exactly without A. With the letter A - these are three-volt versions of PLM.

IT> Docs are good, especially if in translation.
I only have a thick book in English. (Protected from copying by its
thickness)

IT> And the full MAX+Plus II, is it on CD?
It is on CD. Only it comes with an electronic key in addition and is sold for
three to four-digit dollars.

IT> It is not advisable to download 14Meg WEB-Relise with a limited
IT> period of operation. Those documents that I got from the site made a shock
IT> impression, I suspected about such technology, but did not imagine its
IT> vastness...
From the Internet, apparently only the demo version is downloaded. (Imho, maximum for EPM7032 and
no more.)

IM>> function and 576 D-type triggers. By the way, I can give here
IM>> information on the AHDL circuit language. Who needs it? (Although, perhaps
IM>> completely offtopic)
IT> Most likely it is no longer offtopic, due to the direct relation
IT> to the development of Speccy... Only if they kick you out to HARDWARE.ZX :)
As soon as I write the file, I’ll upload it..

Ivan