pc keyboard to zx

Press RESET immediately, All! This can be upgraded a little and obtained on one controller keyboard and mouse. There may also be a serial port in addition to mouse "input-only" port. We need a controller with a larger amount of RAM and ROM and preferably with an EEPROM to save settings and a UART module. The classic i8051 is not suitable due to the need for a large number of additional elements of fine logic. Need something with a ROM capacity of 1-2 kb, RAM 128-256 bytes, performance ~4MIPS and more the more the better - the shorter the WAIT cycle when accessing a port. - Emulator people hang out (2:5030/827.2) --------------------- ZX.SPECTRUM - Msg: 1963 of 2000 From : Wlad Sorokin 2:5020/6664.14 22 Feb 00 To : All 24 Feb 00 Subj: Mouse on ZX Big Hello All , &#%^& &&^# !!!!! Somehow the Kempston mouse interface diagram from ZX format flew by here, and remembered how once, two years ago, I connected an AMIGA mouse to my Pentagon and thought that I should build something similar again, especially since the software is under the mouse is much bigger now, and the spec with the mouse would be more impressive looked... :) Immediately a problem arose - where to get an Amiga mouse from (from "search" Now it’s a problem to find, and it doesn’t look very good...) as usual, there was no mouse nearby, except for the active one, but I didn’t have the chance to break it :)Well, I decided to connect it... And this is what happened - only 3-4 microcircuits, normal movement speed (unlike many passive mice), fully compatible with Kempston, (except for the missing middle button).... Unfortunately, the circuit uses the processor's WAIT signal, although it does not last long. and only while reading from the port... Below is the circuit and firmware for the PIC16c620 microcontroller, (osc=hs wdt=off) although it will get along well in 16c84,16F83,16c622 etc, the only thing is that all of them should be at an operating frequency greater than 10 MHz... By the way, at the moment I don’t have 561la9 installed... I got 561lp5 in the Pentagon itself, on the video output (it’s like an inverter there) and to check I typed- 10 PLOT IN 64479,IN 65503:GO TO 10 cool! (if you don’t go beyond the top or bottom of the screen) if instead of certain lines the trail goes in an unclear direction - the problem is the clock frequency (a separate quartz will help) and also, on the Spectrum data bus there are resistances of 10k to 5v, so inside they cost a controller too! hence the conclusion...(you can tear it out) and also, 14 MHz from the board - you need to skip through a SEPARATE inverter or hang additional quartz on 15-16 legs of PIC-a (from these legs you will also have to install two 10-30 pF capacitors to ground) By the way, with this solution you can remove the “native” quartz from the board... ps. The firmware is written for 14 MHz, but it works fine at 14.31818 MHz(fortunately there are a lot of such quartz on dead 286-386) The Wait signal sometimes needs to be passed through a diode (plus on 24n.Z80) P.P.S. This is a beta release! System bus || || ||-A10-------------------------+ ||-A8------------------------+ | ||-RESET---------+ x+5v | | x+5v || 4| |14 | | | || +---++-++---+ | | | ||-D0-------6|Rb0|CPU|Ra0|18--+ | | ||-D1-------7|Rb1|IC1|Ra1|19----+ | ||-D2-------8|Rb2| |Ra2|1----------)-----+ ||-D3-------9|Rb3| |Ra3|2------+ | | From mouse ||-D4------10|Rb4|pic|Ra4|3----+ | +----+|//_ ||-D5------11|Rb5|16c|---| | | (o o o|o o) ||-D6------12|Rb6|620|xt.|15 | | o o|o o / ||-D7------13|Rb7| |xt.|16-+ | | | || +---+-+-+---+ | | | GND || 5| | | | || __ GND | | | || |1 | | | | ||-14mHz■--+ o--------------+ | | || |__| | | || | | || +---+ | | ||-Wait-11-+=1 +12-------+-)-------------+ || |IC4+13----------------------------+ | || +---+ | || || +---+ | ||-A5----1-+ 1 |9 +---+ +---+ | ||-IORQ--2-+ o-------3-+ & |6 +---11-+ 1 |10 | ||-RD----8-+IC2| +---4-+ o--+---12-+ o---+|| +---+ | +-5-+IC3| +-13-+IC2| || | | +---+ | +---+ ||-A15-------------+ | +---------+ ||-A12---------------+ | || +---+ | ||-A0■■■■1■| & | | IC1-PIC16c620/20p ||-A1■■■■2■| o--------+ IC2-561ле10 ||-A2■■■-8■|IC3| IC3-561ла9 || +---+ IC4-555лп5 || || Copyright '2000 Wlad Sorokin || Spacesoft Hackers Group || Last edition 22.02.2000 22:21 pm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rigin: Угнали ??? Hе надо было ставить CLIFFORD !!! (2:5020/6664.14)

Hello, Kirill! Once, Fri Aug 30 2002 00:23, Kirill Frolov wrote to All: KF> This can be upgraded a little and obtained on one KF> controller keyboard and mouse. All this is sax, since it will generate wait. KF> Possibly also a serial port KF> in addition to the mouse "input-only" port. Need a controller with KF> larger amount of RAM, ROM and preferably with EEPROM for saving settings KF> and UART module. KF> The classic i8051 is not suitable due to the need for availability KF> a large number of additional elements of fine logic. yes, that's it :( KF> Need something KF> with ROM capacity 1-2 kB, RAM 128-256 bytes, performance ~4MIPS and KF> the more the better - the shorter the WAIT cycle when accessing a port. it must be so that it does not exist ;) options... one is mine ;) with a buffer time division dual-port ram ;) , the other is mcu with a performance of ~20 mips, then you can resolve it without waitstates. AT90S1200/S2313 datasheet gives up to 10 MIPS, some craftsmen we drove them up to 24 ;) legs, however, it’s not enough, we’ll have to steer it with petty logic.costs around 2 bucks. go for it ;) - now you need to act this way. If you count on 7 MHz inside the Spectrum, you need something else. yes things with hundreds of mips, but they cost money. If I had to do it now, I would install CPLD+MCU. it's easier for me this way ;) but for the money this is not suitable for spektrumists :( All the best. Dmitry. [ZX] [Quake] np: The Beatles - Girl

Press RESET immediately, Dmitry! 31 Aug 02 00:58, Dmitry Lomov wrote to Kirill Frolov: DL> one is mine ;) with a buffer time division dual-port ram ;) , Maybe just look for dual-port memory? Or is it a costly deficit? DL> another - mcu with a performance of ~20 mips, then you can sort it out DL> without waitstates. AT90S1200/S2313 datasheet gives up to 10 MIPS, DL> some craftsmen drove them up to 24 ;) legs, though not enough, it will have to DL> steer with fine logic. costs around 2 bucks. go for it ;) - Costs $3-4 imho. 1200 don't have enough brains - it disappears. DL> this is exactly what you need to do now. if you count on 7 MHz DL> inside the Spectrum - you need something else. there are things with a hundred DL> mips, but they cost money. There is no need to do such short I/O cycles as in Scorpio -- AY, VG, maybe screw and who knows what else are buggy. There should be WAIT here necessarily.

Hello, Kirill! Once, Sat Aug 31 2002 03:39, Kirill Frolov wrote to Dmitry Lomov: DL>> one is mine ;) with a buffer time division dual-port ram ;) , KF> Maybe just look for dual-port memory? Or is it a shortage KF> expensive? not scarce, but expensive ;) DL>> another - mcu with a performance of ~20 mips, then you can DL>> resolve without waitstates. AT90S1200/S2313 according to datasheet give up to 10 DL>> MIPS, some craftsmen drove them up to 24 ;) legs, however DL>> is not enough, you will have to use fine-grained logic. costs around 2 DL>> bucks. go for it ;) - KF> Costs $3-4 imho. 1200 don't have enough brains - they disappear. yes, perhaps... DL>> this is exactly what you need to do now. if you count on 7 MHz DL>> inside the Spectrum - you need something else. there are things with a hundred DL>> mips, but they cost money. KF> There is no need to do such short I/O cycles as in Scorpio KF> - AY, VG, maybe a screw and who knows what else are buggy. Here WAIT should KF> must be.ok, I agree. in the turbo - let there be a wait ;) but no one will do it. It’s easier for people to cut 100 buttons and use wires connect ;) All the best. Dmitry. [ZX] [Quake] np: Space - Let Me Know The Wonde

Hello, Kirill! Once, Sun Sep 01 2002 09:41, Kirill Frolov wrote to Dmitry Lomov: Sorry for the pause, I was on a business trip ;) KF>>> Maybe just look for dual-port memory? Or is it a shortage KF>>> expensive? DL>> not a shortage, but expensive ;) KF> How expensive? After all, you don’t need a full-fledged two-port KF> memory, but simply writing from one end and reading from the other. Well type KF> 555ИР16. yes, volume 256*5 ;) DL>> but no one will do it. It’s easier for people to cut 100 buttons and DL>> connect with wires ;) KF> It's really simpler. It's easier to cut and solder 100 buttons than 200 KF> legs of fine logic on the layout. If it were on 2 chips... KF> And also write a program... Do you think it’s necessary in terms of “assembling the keyboard”? IMHO, for anyone smart a person counting on the future, it should turn into “learn something smart, and at the same time assemble a keyboard.” cut 100 keys and a monkey can ;) and here you can also study circuit design, and programming. and microcontrollers and programmable logic. and God knowshow much money does it cost... in three years the topic of Spectrum for a specific individual will disappear, but the ability will remain. you look, and hack work appears, and work. I'm like that myself ;) and my friends There are many of me like that. Best wishes. Dmitry. [ZX] [Quake] np: Metallica - Loverma