From CPCWiki - THE Amstrad CPC encyclopedia!
Jump to: navigation, search

FPGAmstrad cc2.jpg

This is a VHDL version of Amstrad CPC 6128 running on FPGA starter-kit NEXYS2 500k-gates from Digilent. A starter-kit is a board made for learning FPGA, so it is a standard FPGA development board.

Please refer to [MiST-board CoreDocAmstrad] for the final user version, running on MiST-board platform.

Some games running on it : 1942 1943, 3D grand prix, A view to a kill, Action Fighter, animated strip poker, Antiriad, Arachnaphobia, Arkanoid, Arkanoid Revenge of Doh, Asphalt, Axiens, Barbarian MH, Barbarian PA, Barbarian II, Battleships, Bomb Jack I and II, Boulder Dash, Bruce Lee, Bubble Ghost, Buggy Boy, Bully's Sporting Darts, Cauldron I and II, Chase HQ, Chicken Chase, Classic Axiens, Classic Invaders, Crazy Cars, Crazy Snake, Dan Dare I II and III, Dizzy 7, Donkey Kong, Double Dragon I and II, Druid, E-Motion, Exolon, Express, Fruity Frank, Ghost'n'Goblins, Golden Axe, Gryzor, Heart Land, Hold-up, Hyper bowl, Ikari Warriors, Impossible Mission, Invasion of the Zombie Monsters, Iron Lord, Killapede, Klax64, Laser Squad, Light Force, Locomotion, Lode Runner, Macadam Bumper, Mange-Cailloux, Mario Bros, Maze Mania, P47, POP-UP, Prince of Persia, Prohibition, Realm, Rebel Star, Rick Dangerous I and II, R-Typeee, Rock Raid, Rogue, Rygar, Salamander, Sapiens, Shinobi, Sim City, Sorcery, Spectra, Spherica, Spin Dizzy, Star Ranger, Star Quake, Star Wars, Strider II, Super Ski, Superkid In Space, Tank Command, Tempest, Tetris 95, The Duct, The Empire Strikes Back, Trail Blazer, Trakers, Turbo Tortoise, Turrican I and II, Venoms, Victory Road, Vixen, WEC Le Mans, West Bank, Wizard's Lair, Word Cup Challenge, Word Series Baseball, Xevious, Xor, Zolyx.



How to assemble it

NEXYS2 Xilinx version is obsolete, it is still describe here for history reason (showing the prototyping part). Please refer to [MiST-board CoreDocAmstrad] for the final user version, running on MiST-board platform.

You need:

  • a "NEXYS2 500kgates" starter kit from Diligent [[1]] (1200kgates should be better for future version => in fact I choosen NEXYS4, it is same RAM inside) also in France : [[2]] or in Germany : [[3]]
  • a "PMODSD" module for reading sdcard [[4]]
  • an alimentation (cause they don't give it with starter kit) [[5]]
  • optionally a DIGILENT USB JTAG (normally starter kit can be programmed directly by usb, but I don't have tested this way) [[6]]
  • a 4GB SDCARD (no more), I have exactly a "SDHC 4GB class4 Verbatim"
  • the binary of this project (candidate 002) : source+500k-gates binary [[7]] ; 1200k-gates binary [[8]]
  • several ROM files: OS6128.ROM BASIC1-1.ROM AMSDOS.ROM (from JavaCPC[[9]]); MAXAM.ROM[[10]]

A package that contains minimum set of ROM and DSK for filling simply sdcard [[28]] (OS6128, BASIC1-1, AMSDOS, MAXAM)

You have to:

  • program FPGA with the binary file "amstrad_switch_z80_vga_sd.bit" of this project, for it I use Digilent Adept software and my USB JTAG cable
  • format your 4GB SDCARD in FAT32 4096 byte allocation size
  • copy ROM and DSK on SDCARD
  • plug PMODSD on slot JC1 of starter kit, and set all 8 switches to 0
  • plug VGA, and turn on starter kit

FPGAmstrad minimal plug.jpg

You can:

  • plug a PS/2 keyboard, and type "cat"
  • increment switch to select another disk at boot, if screen became RED, it's that binary value done by switches is too big, leds are doing a small animation when a disk is correctly loaded
  • plug principal joystick on slot JB1 (Vcc 3.3v is common)
  • plug another joystick on slot JA1 (Vcc 3.3v is common)
  • plug a jack on slot JD1, one at upper GND plug, and second wire at next plug, just at left of it (if it was 3.3v Vcc, choose the right one instead)

FPGAmstrad sound plug.jpg

Wires are plugs at upper part of pmod D.

From bottom to upper: [nothing: Vcc 3.3v] [blue: GND] [red: sound]

Sure you can connect a jack...

And two joysticks. On pmod B: bottom is (both) 3.3v, so joystick's common, next is (both) GND (not used), and others are joystick connections, to test manually :)




Last news about this project

In MaY 2020, I add Sorgelig formula for WAIT_n=quick, no more table of instruction time in code, just a simple WAIT_n formula.

In MaY 2019, I add cassette feature.

In August 2018, totally desperated -around time and level of efforts- about reaching next step focus around Z80 range, here came Sorgelig, he is working around port of FPGAmstrad into the MiSTer FPGA platform, and make during his step an intermediate jump step on MiST-board called "Amstrad_MiST" full of verilog as he seems to love. And, as a specialist of Z80 core, I just send him Z80 testbenches I collected since, he then corrected the Z80 fully this way, I merged, resulting this next current checkpoint.

In May 2018, I programmed my first CPC game http://www.pouet.net/prod.php?which=75855 following JDVA youtube tutorial since january, they are based on CPCMania 2005's website knowledge about programing in CPC using SDCC. I think that if I do progress this way enough, I'll implement my own CPC testbenches, needed for reaching next realise of FPGAmstrad (I did it : Moustache testbench)

Image converted to CPC by SuTeKH/Epyteo

In January 2018, Jepalza has ported FPGAmstrad from this wiki (Xilinx version, principe of concept 2011) on spanish ZX-Uno low-cost FPGA final platform (three times cheaper than MiST-board/same Xilinx chip poc 2011/chip used at 100%). So I bought a ZX-Uno to help around this fork, merging components. Normaly I can go a little further later (CRTC0, joystick), and then go back to MiST-board :)


In May 2017, FPGAmstrad TV mode is validated using a TV from Tetalab group.

In February 2017, CRTC1 is also implemented following JavaCPC's source code, now you can choose between CRTC0 and CRTC1 in the OSD menu.

In January 2017, scanlines mode is implemented, you can select it from the OSD menu.

In December 2017, implementing green screen, using "Les Sucres en Morceaux" tutorial.

In September 2016, FPGAmstrad does use external RAM as RAM+VRAM, no more "LowerVRAM/UpperVRAM" switches to select in the OSD menu.

Using 64K of VRAM...

In August 2016, FPGAmstrad does pass some arnoldemu's testbench : PPI PSG CPCTEST.

In February 2016, FPGAmstrad has a TV mode (original signal)

In September 2016, FPGAmstrad has write access on both disk drives, and can change dsk.

In March 2015, FPGAmstrad is stable on MiST-board, please refer to MiST-board CoreDocAmstrad and mist-board amstrad bin core

In December 2014, FPGAmstrad is stable on NEXYS4, WIP, BOOTLOADER improved, VRAM improved (with BORDER colors !), RAM relaxed, no more timeout in keyboard key-press, Ghouls 'n' Ghosts, Macadam, a lot more to do. 14% of platform is used.

In November 2014, I bought MiST-board, with two USB pro joysticks.

In September 2014, I bought NEXYS4, more powerfull than NEXYS2, with same external RAM, internal mini-sd, no PS/2 (it is a pmod option)... I have some patchs to make (MSB FAT32 offset). I would like to make a USB sniffer also with it (usb to ethernet (wireshark))


I am Freemac, my IRL name is Renaud Hélias.

You can contact me by mail to renaudhelias @t gmail d0t com, if you have questions about assembling this project or running it in MiST-board final platform.

Follow me on my google plus account!

Tests done

Great games that are running properly are listed at top of page.

On MiST-board CoreAmstrad version

Games that doesn't run are :

  • 007 The Living Daylight.dsk: problem with VSYNC. Certainly two VSYNC per frame, GA ignoring the second one.
  • ACPC_logon_system.dsk: text scrolling lag. This demo will be used for horizontal ink calibration (when I’ll buy a luxurious FPGA platform... I need in fact 224KB of internal RAM to do it), and CRTC overcounts.
  • split ink demo.dsk: (from cpcrulez) : may help about ink raster calibration.
  • Sultan's Maze.dsk: does need the right part of keyboard (F0-F9 are used for directions in this game)
  • Orion Primes.dsk: does display "secteurs entrelacés" - "vérifiez votre copie", a FDC problem, perhaps "sectorId++" is not the good way to reach next sector, or else two tracks in one track. Does pass on Sorgelig fork.
  • Batman_Forever.dsk: some problem during flying chip demo part (one garbage line), and several rupture showing ghost lines around Vcc=0. Rupture solved in r005.8.16c4, but flying chip now show a half garbage of pixels. Batman seems CRTC1. Problems with FDC in r005.8.16 (does slow animations, like if I missed some "not ready" signal ?)
  • 30YMD.dsk: in Benediction demo, at bottom some time you see some ghosts of central animation (too many HSync per screen ?), solved in r005.8.16 (CRTC0 seems perfectly implemented). 30YMD seems CRTC0, running fine except that changing disk feature does still fail (inserted/not inserted/inserted signal ?)
  • imperial_mahjong.dsk: modern EXA/EXA2 resolution not passing my color pallet heuristic :p - does pass on Sorgelig fork.
  • rtypeee.dsk: at begin of presentation, a draw of "jack plug" is done in a strange video mode, more than 200 pixels of height !, see flipping lace
  • S&Koh.dsk: LOGON SYSTEM, black screen in r005.8.4... damn
  • Fres Fighter II Turbo.dsk: FDC problem, cannot be launched.
  • Seascape.dsk: Devilmarkus, using scandb50Hz and MEM_wr=slow, does display, but a flower petal at bottom is drawn in blue. A good raster test. Test on real 6128 forum: Seascape CPC by Impact - on original CPC 6128 ? - by Emashzed : type 1 is perfect (no blue on bottom right flowers, no blue on middle triangle rock), type 2 has bright cyan squares (one on bottom right flowers, and one on middle triangle rock. Calibrated OK in r005.8.16.5 using OSD VGA:scandb50Hz. Does pass on Sorgelig fork.
Seascape (type 1) - r005.8.16.5
  • Megablasters[original].dsk: has a 2 pixels glich border on left side. Certainly final HSYNC offset problem, as on Super Cauldron normaly the right (not left) border has to be selected to get a nice bottom bar in game (actually too centered)
  • Edge_Grinder.dsk: screen not stable horizontally, music change speed during game ? Does pass on Sorgelig fork.
  • Welcome To Amstrad CPC 6128.dsk: does display "Incompatible BASIC installed" message.
  • phX.dsk: does begin to pass on r005.8.16. Does pass completly on Sorgelig fork (scandoubler). Does finish on amstrad_180804_r005.8.16.5. Does show vertical bars with CRTC0, no display during end scroll part. Does freeze at middle on amstrad_180804_r005.8.16.6 (doesn't launch the read of disk), does pass on Sorgelig fork but song is 2 times slown down before reaching this part. Did pass one time on amstrad_200527_r005. (CRTC0 WAIT_n:quick (Sorgelig GA simple formula)), but I was lucky. Did pass one time on amstrad_200527_r005. (CRTC1 WAIT_n:quick) except vertical bars of begin. Seems that launching Pinball Dreams CRTC1 (until menu of boards) before (soft reset (page up key) and) PhX CRTC0 does unlock PhX.
  • Ghouls'n'Ghost.dsk: does fail on r005.8.16.2 : time going to zero in 3 seconds, is fine in r005.8.15.2, fine also on Sorgelig fork... also unlocked by Sorgelig in r005.8.16.3 !
  • Ultimate Megademo (Face Hugger).dsk: First part is better using CRTC0. Ending part (one just after Lemmings part), a double VSYNC problem (a small scrolling text instead of... a lot of things), music OK during this ending.

Arkanoid.dsk stars use rupture address (changing address several time during display of one image), it is now supported on "candidate 001" version of FPGAmstrad. Run better in r005.9.11e than in r005.9.11.

Gryzor and Prince of Percia use rupture ink/mode (changing ink and mode during display of one image), it is now supported on "candidate 002" version of FPGAmstrad

Gryzor is really sensitive to Amstrad general stability : do press esc at start menu does activate music during game, if no music, FPGAmstrad is in an instable version (last known stable version r003.8)

Crazycars2.dsk first car image use 32KB of VRAM, it is now supported on "candidate 001" version of FPGAmstrad.

Ghouls'n'Ghost.dsk / Ecole.dsk does need RAM write when writing in ROM (RAM is beside ROM, hard to emulate with asynchronous SDRAM controler, MiST does use a hacked synchronous RAM done for that)

moktar.dsk / super_cauldron.dsk does run fine since r004.8.1.1. Morkar run fine in r005.8.16c4 using CRTC0 and MEM_wr=slow. Super Cauldron bottom bar is fine in r005.8.16.4 when we select "right border" (instead of default left one) during "screen synchro" menu welcome.

CPC Aventure does run fine since r005.2 (message about turning disk now displayed)

prehistorik.dsk does run fine since r005.2 (key can now be pressed in intro demo)

Sim City and Hero Quest run fine since r005.2 (it was about "key always pressed"), but does need a hard boot (not quick reset keyboard key), certainly a small problem of component init state. Sim City hack demo intro does run better using A-Z80 instead of T80 (2/3 of luck).

A lot of demos don't pass in NEXYS4 FPGAmstrad's version (I need to implement back the SDRAM hacked in MiST board FPGAmstrad version). CPCRULES demos is a cool ressource as it contains simple dsk formats.

crazycar.dsk does use lowerVRAM=00 (&0000-3FFF and &C000-FFFF area for VRAM)

crazycar2.dsk does use lowerVRAM=10 (&8000-BFFF and &C000-FFFF area for VRAM) while overscan presentation, and lowerVRAM=00 (&0000-3FFF and &C000-FFFF area for VRAM) during play. Run fine since r005.7

buggyboy.dsk: I found one dsk running correctly and another not : problem on this last one is about turning left/right effect : the car continues straight (it seems a prototype bad version)

superski.dsk : don't forget to disable autofire, in jump, do press fire+up or down.

protext.rom: All right since r004 (before, in edit mode, if I pressed continuously one key, it did write ten letter and freeze/crash. Do note that a number is incremented in live at top right, so this is a complex software, good for keyboard vector stability tests)

atomdriv.dsk: unlocked since r004.5 (the game and music did freeze (ROM unplugged shall return xFF))

rtypeee.dsk: unlocked since r004.7. Does use special read FDC cmd, with BOT different of EOT (begin of track/end of track), so reading several 512 bytes blocks in one FDC command only.

Arkanoid Revenge of Doh and Asphalt are two games that doesn't support ROM extensions : Doh starts to show small bad layers, and Asphalt does crash and reset before main menu... if you add another gadget ROM.

antiriad.dsk: no keyboard/joystick between r004.5 and r004.7. By adding pull-up at r004.5 I lost this game, by implementing a better PPI in r004.8 this game run fine : Antiriad is back !

-circles.dsk: this demo freeze does since r004.8 (PPI border effect ?) and is back since r005.5, it was nice to calibrate SOUND clock : I did generate 8 candidates of synchronizing this clock (1MHz from 4MHz : 1100 0110 0011 1001, and 0.5 deltas : 1100i, 0110i, 0011i, 1001i), only one does not freeze -circles... so I release r005.5 candidate. This demo is a great one around calibrating Yamaha clock.

Nigel Mansell's Grand Prix.dsk: Only one race track seems ok : Monaco (Brazil track does not start). Unclassified : this disk bug also with other emulators, certainly a bad dsk dump here, TOSEC version of Nigel Mansell does run fine (but some legendary traces of "SK bit purpose" needed by here (in FDC, setting SK does jump deleted disk tracks), perhaps to investigate) - update : some tracks unlocked in r005.8.15c61.

saboteur2.dsk: run fine since r005.5 (nice music and then freeze problem), it was about Yamaha clock generator (generated by Gatearray, versus WAIT_n added in short Z80 instruction to let them during all 4 clocks (Z80 in Amstrad does use 4T or 8T instructions (WAIT_n does insert missing T)). Does freeze at welcome since r005.8.7. Back since r005.8.10.

tetris95.dsk : bad in r005.8.9.2 (4 beep while breaking 3 lines (instead of 3 beep while breaking 3 lines), was correct in r005.8.4. Back since r005.8.10.

Battro.dsk seems also CRTC1 and does fail completly. Does pass in r005.8.16.

arkanoid2.dsk: don't run in r005.8.13, but fine in r005.8.13e (experimental fork), ok in r005.8.14 (using default OSD value : MEM_WR=quick)

trailblazer.dsk: no more "raster" problem since r005.5, it's now perfect ! Palette heuristic offset (done for unlocking Batman Forever Demo) has a small effect in left (squares are not separated by a black line in first column) - same small defect in TV mode using r005.8.14.2... Thinking about a HSYNC offset of 2 (instead of 1 currently) then also delaying DATA+HDISP of 1 (char) . Has defect on bottom scroll text bar r005.8.16.2, is fine in r005.8.15.2, fine also on Sorgelig fork... also unlocked by Sorgelig in r005.8.16.3 !

commando.dsk: pixels that should be deleted are not deleted (only VRAM &C000-FFFF seems used), on level 1, the moto is not displayed correctly inside the bridge... but after the bridge :/ - unlocked by Sorgelig in r005.8.16.3 !

Pinball_Dreams__PREVIEW.DSK: Does run in version r005.8.16.6 using CRTC1 (and WAIT_n=slow). Sorgelig fork does implement interlace (used an welcome screen - eagle)

Game Welcome Title Game
CrazyCar2 1011 0011
One 0111 0011
Scarabus 0111 0011
Ace 0010
Tintin 0111 1011 0111
Devil's Crown 0001 0011
Mach 3 0111 0111 0110
Paranoia (1994) 1011 0001
World Class Rugby 1011 0001 0001
Miam Cobra 1001
Galivan 0001

XOR.dsk does use lowerVRAM=01 (&4000-7FFF and &C000-FFFF area for VRAM)


On MiST-board CoreAmstrad version - TAPES

https://cpcrulez.fr/GamesTest/legend_of_steel.htm :

a nice game, unfortunately graphic error at 50 and 72 Hz.
My monitor can be 50Hz. 48Hz not.
Why does the game show 48Hz on my monitor?
Best regards, MiC

List of known running TAPE games (on r005. A Magician s Apprentice (txt), A Message from Deep Space (txt), A View To A Kill, Aaargh!, Academy (missile commander 360°), Advanced Destroyer Simulator, Advanced Lawnmoving Simulator (by DevilMarkus), After Shock (txt+img), Aftermath (missile commander), Air Attack, Air Traffic Control - Heathrow, Air Traffic Control - Schiphol, Airborne Ranger (aircraft or walk), Airwolf, Aladdin s Cave, Alex Higgins World Snooker (pool table), Algebre, Ali Candil Y El Teroso dde Sierra Morena, Alien TurboAmstrad, Alien Syndrome, Aliens The Computer Game, Alkahera (spacecraft simulator), Alpine games, Alsim, Alternative World Games, Amaurote, Amsgolf, AmstradMagazine Le Survivant, Amstrad Shuffle (card games), Amstrad Tecla a Tecla, Amstroid (txt), Amstroids, Angelique A Grief Encounter (txt), Angleball (pool table), Animal Vegetable Mineral, Annals of Rome, Antalex (txt), All Points Bulletin (little car), Apocalipsis New (txt), Arcade Fruit Machine, Arcos, Ariane, Arnhem, Asphalt, Assault Course, Asterix and the Magic Cauldron, Astro Plumber, Atahualpa (angel), Athlete, Atlantida 3000 (platform), Atlantis Anirog, Atlantis (txt), Atomic Battle (asteroids), Atomic Driver (little car), Auftraq In Der Bronx (txt), Aufwarts Zur Rettung (donkey kong), Avenger Way of the Tiger II, Aventure Au Chateau (adventure), Azar Menu, Abu Simbel Profanation, Ace of Aces, Advanced Pinball Simulator (CPC-Power), After The War, Afteroids (CPC-Power), Agent X 2, Alien Highway, American Football, Amo Del Mundo, Amsgolf, Amsoccer (del key to launch party), Amstrad Unlocked, Animated Strip Poker, Aquad, Arkanoid, Arkanoid II, Arkos, Army Moves, Artura, Ashkeron (txt), Atlantis Anirog, Atom Ant, Auf Wiedersehen Monty, Autocrash, Auxilio Aereo.

Games that doesn't run are :

  • Action Force.cdt: freeze during loading, black screen with green border, nothing more. glitch in r005.
  • Adidas Championship Football.cdt: freeze after loading.
  • Afterburner.cdt: nothing after load (still black screen with colorfull border) reboot
  • Alien Attack.cdt: simple tape, but freeze (idFF : end of tape) saying "is loading". Same using WinAPE.
  • Alien Legion.cdt: not enough in line 19 : memory &9FFF. Run in r005. but no key ?
  • Alta Tension.cdt: 007 Living in fact, same vsync problem than disk version.
  • AMC Astro Marine Corps.cdt: welcome picture not drawn completely, do load data for nothing and then freeze.
  • Anatomie.cdt: end of tape idFF.
  • Arnold Goes to Somewhere Else.cdt (txt): Rewind tape.
  • Atlantida 3000: welcome picture seems not fully loaded, but game starting. Same in WinAPE.
  • ATV Simulator.cdt: nothing after load (still black screen with colorfull border), reboot
  • Automec.cdt: reboot during game (second tape of three)

Back-Tron, Bacterik Dream (puzzle), Bactro (tron), Bactron, Balade Outre-Rhin (txt), Ball Breaker II (breakout), Ballon Buster (press space to unlock black screen, and another time after welcome screen displayed), Barry McGuigan Word Championship Boxing, Batalla de Inglaterra, Batalla Naval, Battle ships, Battle Tank Simulator, Batty (breakout), Beat the Clock (scrabble), Behind The Lines (txt), Berks 3, Bestial Warrior (platform), Beta-2515, BeTiled (Bejeweled), Better Spelling, Big Trouble in Little China, Blade Runner, Blagger, Blindado, Blitz AmstradAction, Blockbusters, Blue Tomb (bomberman), BMX Ninja, Bob Morane Espacio, Bomb Jack, Border Harrier (adventure), Bosconian 87 (Asteroids), Boulder Dash 3, Brian Jacks Superstar Challenge (run), Bride of Frankenstein, Bridge (card), Bronx, Bubble Bobble 4, Bubble Dizzy, Buffalo Bill s Rodeo Games, Buggy II, Buggy Boy, Bugs Buster, Bullseye, Bump Set Spike, Bundesliga Manager 3 (tool), Buscate la vida, Business Plus, Buster Block, Ball Crazy, Barbarian, Bataille Pour Midway, Battle for Midway, Beach Buggy Simulator, Bedlam, Beyond the Ice Palace, Bivouac, Black Beard, Blasteroids, Brawn Free (txt), Bruce Lee, Bubble Bobble (2014-06-20), Buitre-Emilio Butragueno Futbol

  • Back to Reality.cdt: Rewind tape.
  • Barbarian II.cdt: black screen blue border. Welcome screen with glitchs
  • Beach Head 2.cdt: freeze load. Reboot id13
  • Blood Valley.cdt: welcome screen, "Read error b"
  • Bomb Jack 2.cdt: welcome picture not load completely, freeze.
  • Boom.cdt: load until block 16, then freeze. Reboot at block 21.
  • Booty.cdt: second welcome screen then freeze. id15 (ear)
  • Boulder Dash 4.cdt: not a game but a tool kit,
  • Brainache.cdt: no pause at begin of tape, welcome screen turn into black screen at end of tape (idFF)
  • Bugs.cdt: "Rewind tape" message

Cap Horn, Cap sur Dakar, Cavebola, Cedric y los Juguetes Perdidos, Chessmaster 2000, Civil War, Cluedo, Cobra (snake), Coloric, Concurso, Conflit en l an 2000, Crystann Le Donjon de Diamant

  • CORE Anatomie: end of tape idFF.

Dan Dare 1, Dances With Bunny Rabbits (txt), Danger Adventurer at work (txt), Danger Mouse In Makin, Dark Side, Dark Star, Darts, De la Terre a la Lune, Deactivators, Deadenders (txt), Deadly Evil (walk platform), Deathkick (right part of keyboard), Deathscape, Dedalos, Defcom1, Demon Attack, Dempsey and Makepeace, Dernier Metro, Desert Rats, Devil s Castle, Diamond Mine, Diamond Mine 2 (same ?), Die Heilige Rustung Des Antiriad, Die Tiefe (txt), Die You Vicious Fish (txt), Digger Barnes, Dimension Omega, Dirk, Dizzy 2 Treasure Island, Dizzy 4 Magicland, Dizzy 5 Spellbound, Dizzy 6 Prince f the Yolkfolk, Dizzy Down the Rapids, Dominoes (keyboard F1-F7 + space), Doodle Bug (pacman+), Doomsdark's Revenge (txt), Dr Jackle and Mr Wide, Dracula (txt), Dragon's Lair, Dragons of Flame, Driller, Druids Moon (txt), Duel A Abilene, Dun Darach, Dungeons Amethyst (txt), Dynamite Dan, Dynamix Daley Thompson Super Test 1, Dark Fusion, Death Stalker, Death Wish 3, Desperado, Die Alien Slime, Dizzy Amstrad Action Special Edition, Dizzy Cristal Kingdom, Double Dragon, Dragon Spirit, Dragontorc, Dustin

  • Daleks.cdt: use of right part of keyboard as arrow (not mapped)
  • Daley Thompson s Olympic Challenge.cdt: reboot
  • Damas.cdt: Rewind Tape
  • Dark Sceptre.cdt: Rewind to 04
  • Dark Man.cdt: black screen pink border
  • * on r005.
  • Death Pit.cdt: welcome screen not loaded completely idFF
  • Deliverance.cdt: black screen pink border, reboot id13
  • Demon s Revenge.cdt: "Searching 00" freeze
  • Des Chiffres et des Lettres.cdt: reboot
  • Dick Tracy: black screen id15 (ear), idFF (end of tape)
  • Dizzy 3 Fantasy World: black screen
  • Dominator: black screen pink border
  • Domino: Rewind tape
  • Don't Panic: no pause, freeze message "SEARCHING", id15 (ear) idFF (end of tape)
  • Dragon ninja: welcome screen ok, message "128K MACHINE DETECTED" - "PLEASE WAIT" too long, slower than WinAPE id11 read (datalen=0 does ignore data read ?)
  • Dynamic Duo: "Searching 00" freeze

E.X.I.T, Eagle AmstradVideoPlay, Eagle, Edd the duck, Eden Blues, Egg Blitz, El Comecocos, El Cuerpo Humano, El equipo A (A Team), El Formamento, El Foso, El Misterio del Milo, El Prisionero (txt), El Secreto de la Tumba, Electric Fencing, Electro Freddy, Elevator Action, Elven Warrior, Emerald Isle (txt), Emilio Butragueno Futbol, Emlyn Hughes International Soccer, En Busca Del Arca De ma Alianza (txt), Encyclopaedia of War-Ancient Battles, Endurance, Enduro Racer (moto), Enigme a Oxford, Enterprise, Equinox, Escape-Edisoft, Escape from Khoshima (txt), Espacial, Espana-Comunidades Autonomas 1, Espionage, Espionage Island (txt), Euro Boss, Europa Teatro de Operaciones, European 2, European Champions, European Soccer Challenge, Evening Star, Everyone s a Wally, Exolon, Exploring Adventure on the Amstrad (txt), Exterminator, Eye, E-motion, Echelon, El CID, El juego de la Oca, El Ladron del Sol Purpura, El Poder Oscura, Elektra Glide, Elidon, Emilio Butragueno 2, Emilio Sanchez Vicario Grand Slam, Empire, Enchanted, Endzone, European 5, Every Second Counts, Eye Spy

  • Eddie Edwards Super Ski: rewind tape
  • El Caldero Magic: return to basic during game, bad charset
  • El Capitan Trueno: slow block load, reboot
    • on r005.
  • El Gerente: rewind tape
  • El Jabato: blue screen, reboot
  • El Laberinto del Sultan: rewind tape
  • El Tute: after a certain time doing nothing, "Read error b"
  • Election: black screen with blue border, message "LOADING: Please Wait"
  • Eliminator: black screen with blue border
  • Elite: "Searching 00", id15 (ear), with somes id11 before and after it.
  • Energy Warrior: game launched but bad horizontal alignment
  • Er-bert: "Press PLAY then any key"; end of tape idFF
  • Erik the Viking (txt): end of tape idFF, no key
  • Escape from The Planet Of The Robot Monsters: welcome image OK, then black screen border green
  • Eswat-Cyber Police: Cannot insert face B
  • Execution: gray screen
  • Exploding Wall: black screen red border
  • Explorer: reset to BASIC
  • Express Raider: yellow screen blue border, message "LOADING: Please Wait"
  • Extreme: black screen blue border

Fantasia Diamond (txt), Federation (txt), Feliz Navidad, Fernandez Must Die, Ferry Captain, Fifth Quadrant, Feud (OK on r005., Fighter Pilot, Finders Keepers, Fire Ant, Firescape (txt), First Steps With The Mr Men, Flash, Flight Path 737, Flight Simulation, Fluglehrer, Flunky, Fly, Fly Spy, Football manager, Football manager III, Forbidden Planet, Force 4, Forces, Formula 1 Simulator, Formula, Fourth Protocol, Frank Bruno s Boxing, Frank N Stein, Frankenstein (txt), Freedom Fighter, Friss Man, Frontline, Frost Byte, Fruity Frank, Fu-Kung in Las Vegas, Fusion 2 F-1, Fernando Martin Basket Master, Fiendish Freddy s Big Top O Fun, Firelord, First Past The Post, Five a Side Football, Football Champion, Footballer of the year I, Forgotten Worlds, Formula One, Frankenstein Jnr/Junior, Froggy, Fuego Curzado, Future Knight

  • F15 Strike Eagle: black screen blue border, xFF (end of tape)
  • F16 Combat Pilot: black screen blue border, xFF (end of tape)
  • F16 Fighting Falcon: light gun game
  • F1 Tornado Simulator: some vsync problem ?
  • FA Cup Football: Rewind tape slow face of tape OK
  • Fairlight 1 A Prelude: Rewind tape
  • Fast Food Dizzy: black screen, a dot
  • * on r005.
  • Fighter Bomber: black screen pink border, reboot
  • Fighting Soccer: black screen blue border, reboot
  • Fire and Forget 1: Rewind tape, idFF (end of tape)
  • Firezone: use right part of keyboard
  • Flash Gordon: Read error b
  • Flying Shark: Searching 00, strange id60 (end of tape). Rewind to 12
  • Football Director: welcome picture not load completely, freeze, xFF (end of tape)
  • Football Frenzy: Rewind tape
  • Football manager II: welcome image and music, then Rewind tape
  • Football Manager World Cup Edition: welcome screen, no more idFF (end of tape)
  • Footballer of the year II: glitch
  • Freddy Hardest 1: black screen blue border
  • Freddy Hardest 2: reboot
  • Freestyle BMX Simulator: black screen
  • Friday The 13th: rewind tape at idFF (end of tape)
  • Fruit Machine Simulator: black screen blue border
  • Fruit Machine Simulator II: black screen, reboot

G-LOC R360, Galachip (space invaders), Galletron, Garfield 1 & 2, Gauntlet-Micropower, Gazza s Super Soccer, GBA Championship Basketball 2 On 2, Geasa-Parabola, Gem, Geoff Capes Strongman, Ghost Hunters, Ghostbusters 1, Ghostbusters 2, Ghouls, Gilligan s Gold, Glass, Glen Hoddle Soccer, Glider Rider, Golf Trophee, Gorbaf El Vikingo, Graham Gooch s Test Cricket, Grand Prix 500cc, Grand Prix Driver, Great Gurianos, Grell and Falla, Greyfell, Grid Trap, Ground Zero, Guadalcanal, Guardian 2-Revenge of the mutants, Guerre des galaxies, Guerrero Espacial, Gunfighter Galaxia, Game Over I - 1st., Gary Linekers s Superstar Soccer, Gauntlet II, Gems Of Stradus, Get Dexter 1 & 0, Go for Gold, Grand Prix, Grand Prix Simulator 2, Grand Prix Tennis, Grange Hill, Green Beret, Gregory Loses His Clock, Guardian Angel

  • Gabrielle: Rewind tape
  • Galactic Conqueror: Rewind tape, idFF (end of tape)
  • Galactic Games: "LOADING: Please Wait"
  • Galaxia SPANISH: crash during play (when crashing on a wall : reboot)
  • Galivan: black screen
  • Game Over II: black screen orange border
  • Gauntlet I-The Deeper Dungeon: nothing displayed
  • Gauntlet III: welcome image, no proposition to change tape side. At end of first tape I insert the second one... Not loading the 3rd tape.
  • Gauntlet: reboot
  • Gazza II: "Loading error - Please retry"
  • Gee Bee Air Rally: return to BASIC
  • Gemini Wings: black screen pink border, id13 reboot
  • Ghost n Goblins: Rewind tape
  • Gi Hero: Searching 00 Loading 00...07 Rewind to 0C (to retry ?)
  • Gilbert-Escape From Drill: glitch, and then black screen
  • Gladiator: Rewind tape
  • Golden Axe: reboot id14
  • Golden Basket: welcome image then reboot.
  • Goliath-Le Defi: problem during load of welcome image
  • Golpe en la Pequena China: no key
  • Grand Prix Circuit: tape stopped with strange id value : 00
  • Grand Prix Master: blue screen
  • Grand Prix Simulator 1: idFF (end of tape)
  • Great Courts: blocked by a password
  • Gremlins-The Adventure: message "Resume a saved game ?" - no keyboard (same in WinAPE), version ripped in 2018 in CPC-Power does run fine.
  • Gremlins 2: reboot (Turbo Data + Standard Data) - running ok in WinAPE
  • Gryzor: reboot (Turbo Data + Pure Tone + Sequence of Pulses)
  • Guillzemo Tell: no keyboard, 4 Turbo Data, last one is big - running ok in WinAPE

Hacker 2, Halls of Gold, Happy Letters, Happy Numbers, Happy Writing, Hard Drivin, Hard Hat Mack, Harricana-Raid International Motoneige, Harrier Attack, Haunted Hedges (pacman), Haunted House (txt), Havoc, Hawk Storm version 64K, Heartland, Heathrow International Air Traffic Control, Heavy on the Magick (txt), Helicoptero 2000, Heliot, Helvera-Mistress of the Park (txt), Herberts Dummy Run, Here and There With the Mr Men, Hero, Hero of the Golden Talisman, Herobotix, Hexenkueche 1 & 2 (Cauldron), Hi-Q Quiz, Hi Rise, High Moon, Highier, Hijack, HMS Cobra, Hobgoblin (GNG), Hold-Up, Holdfast, Hollywood Or Bust, Hollywood Palace, Home Runner, Hong Kong Phooey, Hot Shot, House of Usher, How To Be A Hero, Hunchback, Hundra, Hunter Killer, Hustler, Hydrofool, Hyperbowl, Hyperspace 4 (txt) Hacker 1, Hammer-Head, Hammer boy, Hercule-Slayer Of The Damne, Heroes of the Lance,, Hideous, Humphrey, Hunchback 1, Hunchback 2, Huxley Pig, Hypsys

  • Hammerfist: white screen blue border, reboot while starting drawing welcome screen id13
  • Handicap Golf: Rewind tape
  • Hanse: Rewind tape while loading welcome screen
  • Hardball: Rewind tape
  • Harvey Headbanger: id11 id11 id12 id13 - Sequence of pulses of different length - id9C (crash) + Searching 00. Rewind to 0D
  • HATE-Hostile All Terrain Encounter: reboot
  • Having Fits Of Madness: BASIC 1.1 - to retest in 6128 model version. black screen green border.
  • Head Over Heels: black screen with message "LOADING: Please Wait" id11 id11 id12 (Pure Tone) id13 (crash)
  • Heavy Metal: reset while loading welcome screen
  • Helichopper: id19 - Generalized Data Block. Rewind to 03
  • Hero Quest: black screen, message "level 12"
  • Heroes of Karn: "LOADING ERROR" back to BASIC.
  • High Steel: id11 id11 id12 id13 id70 (crash). Rewind to 0A
  • Highlander: message "LOADING: Please Wait" id11 id11 id12 id00 (crash)
  • Highway Encounter: message "LOADING: Please Wait" id11 id11 id12 id00
  • Histo-Quizz: Rewind tape (after loading a party) idFF (end of tape)
  • Hive: id13 id2E (crash). Rewind to 02
  • HKM-Human Killing Machine: reboot after a count down
  • Hockey: idFF (end of tape), no keyboard
  • Hopper Copper: Rewind tape (to retry ?)
  • Hoppin Mad: Rewind tape (to retry ?)
  • Hostages: welcome screen and freeze (tape stopped), no keyboard
  • Hot-Rod: black screen pink border, reboot id14
  • How To Be A Complete Bastard: white screen, blue border
  • Hudson Hawk: grey screen, pink border
  • Hunter or Hunted: Rewind tape
  • Hybrid: id13 id7F (crash) "Searching 00"
  • Hydra: black screen border blue
  • Hyper Sports: welcome screen, and then nothing

i alien, Ikari Warrior, Impossaball, Inca Curse - adventure B (txt), Indigestion, Indoor Soccer, Inertie, Infernal Runner, Inquisitor-Shade of Swords, Interchange, International Football, International Karate Plus, International Manager, International Rugby, International Tennis, Into Oblivision, Into the Eagle s Nest, Invasion of the Zombie Monsters, Invasores, Invierte y gana (manager), Invitation (txt), ISS-Incredible Shrinking Sphere, Iznogood Ilogic All, Impossible Mission 1, Impossible Mission 2, Indoor Race, Inside Outing, Invasion, Isoleur, Italian Supercar, Ivan Ironman Stewart s Super Off Road

  • i Ball 2 - quest for the Past: Rewind to 8
  • i Ball: Rewind to 7
  • Ice Breaker: reboot
  • Imagination: Rewind to 7
  • Impossamole: "LOAD ERROR" level 1. A lot of id14
  • Indiana Jones 3: black screen during second tape load
  • Indiana Jones 4: Please rewind
  • Indiana Jones: counter down : tape stopped in middle of id11 transmission, noise, reboot
  • Infiltrator: message "LOADING: Please Wait", glitch idFF (end of tape) idem second tape.
  • Infodroid: reboot
  • Interdictor Pilot: game not running - no key
  • Interieur: "Rewind tape", end of tape id=FF
  • International 3D Tennis: 64K version : reboot at begin of welcome image.
  • International Karate: reset to BASIC
  • International Speedway: rewind to 15 - great for comparative speed test :)
  • It s a Knockout: message "LOADING: Please Wait" id14
  • Italia 90 World Cup Soccer: "WARNING" message, reboot
  • Italy 1990: game not launched despite of inserting face B

Jack and the beanstalk, Jack the Nipper II, Jackson City, Jail Break, Jammin, Jet-Boot Jack, Jet, Jet Set Willy-the final frontier (password needed), JetPac, Jeux de Kim, Jigsaw Rescue, Jimmy Business, Jimmy s Soccer Manager, Jinks, Jocky Wilson s Compendium of Darts, Jocky Wilson s Darts Challenge, Joe Blade I, Joe Blade II, Joe Blade III, Johnny Reb II, Jolly Poppa Down (txt), Jolly Roger s Dungeon Escape, Jonah Barringtons Squash, Jonny Quest, Juggernaut, Jump, Jump Jet, Jumper, Jumpman, Jungle Warrior, Justin Jack Nicklaus Golf, Jai Alai,

  • Jack the Nipper I: game not launched. xFF, still at welcome screen.
  • Jackal: grey screen, then reset
  • Jahangir Khan World Championship Squash: "Tape Loading error..."
  • James Clavell s Shogun: id11 6 times then id10
  • James Debug dans Le mystère de l'Ile Perdue: joystick misunderstood
  • Jaws: id11 4 times then id10 - "Rewind to 04"
  • Jet Bike Simulator: id13 pure data, pure tone, loop
  • Jet Set Willy II: Rewind tape (all id11)
  • Johnny Proot II: Rewind tape (end of tape idFF, all id11)
  • Jungle Jane: blue screen, idFF (end of tape)
  • Jungle Warfare: game auto launched, no joystick/keyboard

K Y A, Kane, Karl s Treasure Hunt, Kart 3000, Kat Trap, Ke Rulen Los Petas (txt), Kenny Dalglish Soccer Manager, Kenny Dalglish Soccer Match, Kentilla (txt), Kentucky Racing, Kick off II, Killapede, Killer Cobra, Killer Gorilla, Killer Ring, King Leonard, Klax, Knight Ghost, Knight Lore, Knight Orc, Knight Tyme, Knightmare, Knights and Demons, Kobayashi Maru, Kokotoni Wilf, Kong s Revenge, Krakout, Kung-Fu Master, Kwik Snax, KYA Karnov, Kung-fu

  • Kaiser: Rewind tape
  • Kettle: reset id11, id13 not reached
  • Kick off I: Rewind tape
  • Killed Until Dead Murder At Midnight: id11, id11, (id12 id13) several times
  • Kinetik: "Rewind to 03"
  • Knight Force: Rewind tape
  • Knight Games: id13
  • Knight Rider: id13
  • Konami s Golf: id13
  • Kong Strikes Back: id13
  • Kotoran s Shadow: invalid CDT format
  • Kristal: Rewind tape

L'Affaire Sydney (txt), L'Affaire Vera Cruz (txt), L'Anneau de Zengara, L Hepiss, La Armadura Sacradda De Antiriad, La Aventura Original (txt), La Caza del Octubre Rojo, La Chose De Grotemburg (txt), La course a la boussole, La foret Infernale, La France, La Geste D Artillac (txt), La Guerra de las Vajillas (txt), La Trilogie du temple d Apshai, La Ville Infernale, Labyrinth Hall (txt), Langolo del Diavolo, Las Joyas Del Nio, Las Vegas Video Poker, Laser, Laser Blast, Laser Tiempo, Le 5eme Axe, Le Bagne De Nepharia (txt), Le Diamant de l Ile Maudite (txt), Le jeu du Roy, Le Millionnaire, Le spectre d'Anubis (txt), Le Survivant, Le Talisman d Osiris, Le Tour du Monde en 80 jours (txt), Le Tournoi du Siecle, Le Tresor de l Amazone, League Challenge, Legend, Lenguaje, Les aventures du KA-Menace sur l Arctique, Les Diamants de la Peur, Lettura Rapida, Life Expectancy Zero, Lifeterm (txt), Light Force, Linkword-French, Logiformes, Lop Ears, Lords, Lords of time (txt), Lorna, Los Angeles Swat, Los Pajaros De Bangkok (txt), Los Planetas-El Sistema Solar, Los Templos Sagrados (txt), Lost Caves and the Tomb of Doom, Loto, Lotus Turbo Esprit, Lucky Fruits L Heritage-Panique a Las Vegas, Lee Enfield Space Ace, Little Puff in Dragonland, Lords of Chaos

  • L oeil de Set: rewind tape
  • La Abadia del crimen: blue screen and grey border
  • La Malediction de Thaar: rewind tape idFF
  • La Tour Fantastique: rewind tape
  • Lala Prologue: black screen
  • Las Vegas Casino: Read error B (too quick)
  • Last Duel: count down, reset
  • Last Ninja 2: glitch, id14 reboot
  • Le Necromancien: red screen, blue border
  • Leader Board: id12 freeze
  • LED Storm: black screen, rewind at start of block.
  • Lemmings: no welcome image
  • Les aventures de Jack Buron-big troube in little china: no keyboard
  • Les Joyaux de Babylone: "ERREUR CASSETTE"
  • Leviathan: reset at id14
  • Line of Fire: inserting second side does nothing
  • Live and Let Die: black screen idFF (end of tape)
  • Living Daylights: same problem than in dsk version
  • Lords of Midnight: welcome image, then freeze

Mach 4, Mach 3, Macrocosmica, Mad Mix Game, Mad Mix Game 2, Magic Clock, Magical Drop CPC, Manager, Mansion Kali 1 (txt), Mansion Kali 2 (txt), Mantis 1 (txt), Mantis 2 (txt), Mapa del Cielo, Master Chess, Match Point, Mathasard, Mathe Stunde 1, Max, Meltdown, Mercenaire, Mexico 86, Minas, Misil Atack, Missile Ground Zero, Missiles, Molecule Man, Money Molch, Monster of Murdac, Monty Mutant, Monument, Mordon s Quest (txt) Mariano The Dragon In Capers In Cityland, Marius Tresor Foot, Masters of the Universe, Mathex n1, Maziacs, Megacorp, Meganova, Message from Andromada (txt), Miami Cobra GT, Microprose Soccer

  • M enfin: sound freeze during welcome picture
  • Mahjong: Rewind tape
  • Manic Miner: crash after loosing one time.
  • Mansion: Rewind tape during first mission
  • Marble Madness-Deluxe Edition: blue screen
  • Marsport: password
  • Master of the lamps: Rewind tape
  • Matchday 2: black screen blue border
  • Mercenary Escape from Targ: welcome screen, no keyboard
  • Mercs: reboot
  • Metropol: back to basic at first round of play
  • Miami Vice:
  • Michel Futbol Master Super Skills: reboot
  • Mickey Mouse: reboot
  • Micro Sapien: Rewind tape
  • Mike Head S Computer Pop Quiz: menu without keys
  • Mineur: Read error a
  • Mobile Man: reboot
  • Monopoly: grey screen
  • Montsegur: reboot
  • Moonblaster: game not launched

  • * Breaking Baud.cdt: no pause, last part "pink floral" heart is missing. As explain at begin of demo, instructions are executed during they are written in memory, speed of tape against speed of Z80...

On ZX-Uno FPGAmstrad version

Games that doesn't run are :

  • adios_a_la_casta.dsk: bad sound. Does pass hacking memory banks this way :
else b"1100" when RAMbank="100" and (A(15)='0' and A(14)='1')
else b"1101" when RAMbank="101" and (A(15)='0' and A(14)='1')
else b"1110" when RAMbank="110" and (A(15)='0' and A(14)='1')
else b"1111" when RAMbank="111" and (A(15)='0' and A(14)='1')
else b"10" &  A(15 downto 14); -- default value
  • Ghouls'n'Ghost.dsk: black screen during game. RAM is not relaxed enought to permit changing address just after knowing if Z80 does a read or a write (ROM written=> does write on RAM hidden behind)
Antonio Villena mail "ZXDOS conversion of your CPC core" 2019/01/29
Hi Renaud
It's for asking if you plan a conversion of your CPC core for this platform. It's like ZX-Uno, but with:
-A bigger FPGA, LX16 instead LX9.
-18 bit DAC
-32Mb of SDRAM, also 512K of SRAM
-dual joystick port
-Separate PS/2 joystick and mouse
If you are interested I can provide the two addon boards, so you only need the LX16 board. Bought from Aliexpress is about 18 EUR.

Effort done


Pixels and VRAM. Palette and rasters. CRTC0


CRTC0 seems the best one, some demo does cry when detecting a poor CRTC1 (CRTC1 seem a low cost version of CRTC0). I have to implement a CRTC0 instead of my current CRTC1...

In fact CRTC1 is the best one. CRTC2 is the low cost version. CRTC0 did appears before CRTC1.

Some demos are running only on CRTC0 and others CRTC1.

Done in r005.8.14. Detected as CRTC0 by WakeUp! - "Enjoy the show" message displayed.

In r005.8.15. WakeUp! (CRTC0/MEM_wr quick) detected as Emu first time, and after a quick reset, does say detected as CRTC0.

In ZX-Uno FPGAmstrad, I implemented CRTC0.

CRTC1 has double sized VSYNC against CRTC0 (moustache test)



VRAM contains 800x300 amstrad pixels (VZoom x2), displayed VGA 800x600@72Hz with fix regular border at 768×576 and fix inside border at 768×544.

In ZX-Uno, VRAM contains 800x300 amstrad pixels (VZoom x2), displayed 640x480@60Hz, with vertical only border.

  • simple_GateArrayInterrupt.vhd (GA to VRAM) parameters : VRAM_Hoffset/VRAM_Voffset
  • aZRaEL_vram2vgaAmstradMiaow.vhd (VRAM to VGA) parameters : H_BEGIN/H_END/V_BEGIN/V_END (theorical fixed values)

To calibrate : VRAM_Hoffset++ does offset one char left. VRAM_Voffset++ does offset one line up. On display H_BEGIN does begin to scan lines of VRAM. But V_BEGIN does not enter in consideration here : vertical=0 does begin to scan columns of VRAM.

In original CPC, top border has 1/2 char more than bottom border. I used Batman Forever default welcome/calibration screen to calibrate VRAM offsets. On ZX-Uno I used Arkanoid to calibrate VRAM offsets.

RAM_palette contains the ink list and the mode for each line of VRAM, sampled at horizontal middle of 800x600 screen, and used at begin of each line.

VGA: TODO : arnoldemu testbench

arnoldemu testbench: crtctest

Adding choice of CRTC 0 or 1 on OSD, and passing this test could be great.

VGA: TODO : winape testbench

winape testbench: plustest

a better border heuristic

Using winape testbench (plustest), test 2 does show somes problems while border does go out of screen, negative border does hide line itself.



(nothing to say here, really ???)


GA: alignment of HSYNC Interrupt

Interrupt are respected since version "candidate 001" of FPGAmstrad, Markus does help me a lot about it.

JavaCPC running norecess.jpg

JavaCPC running norecess's "using-interrupts" code [[29]]

It could be interesting to test this asm code on next version of FPGAmstrad.

GA: Sniffing of a real Amstrad

Cpc plus m1.jpg Code name: Raptor

I listen to some wires of my Amstrad CPC 6128 plus, but I can't access VSYNC/HSYNC output of CRTC, so I have to buy another model in order to do this test. In fact you can listen at clock of Amstrad and transmit it to FPGA DCM component, resulting a accelerated clock sequence, that's it, with FPGA DCM you can overclock output Amstrad clock signal in order to insert more operations, I use this tip for listening signals and save them inside starter kit asynchronous RAM (write, stop write, write, stop write... I’m a perfectionist paranoid...)

You can power Amstrad CPC using extension port, applying 5v. By doing it, power down button of Amstrad doesn’t run. Using this way you reach a common 5v power between starter-kit and Amstrad. I connected wires from extension port directly to FPGA, as they are used just for listening.

GA: WAIT_n generator - currently in r008.5.14

Instruction timing.

I tested instruction timing of T80 compare to instruction timing of JavaCPC emulator. I deduce synchronization of Z80 with CRTC on M1 signal by WAIT_n insertion in order to have a multiple of 4 Tstates per instruction. I deduce also one WAIT_n inserted during MEM_WR operation (yes I log testbench T80, I’m crazy)

I just made a test bench log of T80 (log of instruction's M1, and first M1 coming after knowing that I send a lot of NOP after my instruction), and compare it to a JavaCPC timing array. Some instructions was not tested (interrupt wait, and special timing (instructions with change timing)), but all others passed correctly.

MEM_WR has an OSD menu choice to switch between "quick" and "slow", "slow" mode does insert ONE WAIT_n during detection of MEM_WR. This switch exists because somes games are running in "slow" mode and others in "quick" mode.

An elegant fork by Sorgelig, sum up nicely the "MEM_WR:slow" algorithm + the HACK_Z80 flag of r008.5.14 :

T80pa CPU
.cen_p(ce_4p & (WAIT_n | no_wait)),
.wait_n(1) // (cyc1MHz | (IORQ_n & MREQ_n) | no_wait)
// Current WAIT_n generation is not correct!
// It should use WAIT_n instead (see commented out code above ^^)
reg WAIT_n;
wire acc = (MREQ_n | ~RFSH_n) & IORQ_n;
always @(posedge clk) begin
	reg old_acc;
	if(ce_4p) begin
		old_acc <= acc;
		if(old_acc & ~acc) WAIT_n <= 0;
		if(cyc1MHz) WAIT_n <= 1;

GA: WAIT_n generator - talk about r008.5.14

In fact it exists several instruction making MEM_wr, and adding each one ONE WAIT_n does result in different case of synchronization.

CPC Z80 timing

If it's about managing GA reading pixels, perhaps not only M1 signal are synchronized but also the MEM_RD and MEM_WR accesses at another offset.

Timings instructions Z80 sur CPC

If interruption r52 is regular, even while making a continues MEM_WR, interruption (int<='1') shall be taken into account above WAIT_n insertions ?

In Z80 sequence diagram, an IO_ACK(+M1) is preceded by M1 (single)

cpctest.dsk -Timing Instruction- is different while using mode "MEM_WR=slow" and "MEM_WR=quick". Strangly better using "MEM_WR=quick".

Current version is using "Z80_HACK=true" (parameter set during compilation), that shunt Z80.WAIT_n entry, Z80.clock is slow down during theses WAIT_n. It's the only current way I succeed in slowing down enough Timing Instruction for unlocking Saboteur 2 game.

Key games here are : Saboteur 2 (run fine with "MEM_WR=slow", does freeze with "MEM_WR=quick") and Arkanoid II (run fine with "MEM_WR=quick", too slow using "MEM_WR=slow")

CPC Z80 Commands and how long they take...

GA: WAIT_n generator - plustest-5 - Tests on real CPC (by DanyPPC)

CPCWiki's forum : Need plustest.dsk testbench 5 output on original CPC 6128

So 2A is really during 5 NOPs... perhaps MEM_rd has to be slow down with one WAIT_n like for MEM_wr. Perhaps in this case 5T's instruction has not to be slow down. I have to fork r005.8.16c3 to test that. - update : in schematic GateArray does not has "MEM_wr wire" (but MEM_req and RD, so can deduce WR, but in an evil brain's way) - update 2 : all tests failed except one in testbench using this way, perhaps because RFSH_n does also use MREQ_n during M1 cycle. Perhaps WAIT_n generator can detect the current OP Code fetched (this is conform to Patent US5313621 : Programmable wait states generator for a microprocessor and computer system utilizing it ). - update : principe of concept validated for one instruction (2A), I can slow down instructions one per one, I don't know why I had to insert two WAIT_n instead of one here, but its results a plustest.dsk testbench with 2A instruction validated, that's done on r005.8.16c5f5 (candidate 5 fork 5), but I have to revert it to r005.8.16c3f5 I think before going further.

About testbench border effects, I think that IO_ACKed instructions has to be under same rules (MEM_wr, modulo 4 etc) - update : same result in testbench using this way.

GA: WAIT_n generator - plustest-9 - Tests on real CPC (by GUNHED and Kris)

CPCWiki's forum : Need plustest.dsk testbench 9 output on original CPC 6128

GUNHED results :

System 1: CPC6128, CRTC2:
Test 9 works normal until the &EC codes, there are two errors marked with an "X". 
&ED, &46: 2
&ED, &4E: 2
After &ED, &5D it suddenly stops working!

System 2: 6128 Plus:
&ED, &46: 2
&ED, &48  5
&ED, &49: 5
&ED, &4E: 2
&ED, &50: 5
&ED, &51: 5
&ED, &58: 5
&ED, &59: 5
After &ED, &5D it suddenly stops working! Probably a crash, since a spot appears on screen.

Kris results :

Here are my results (teste performed on CPC 6128 CRTC 1)
Pictures of each screen attached in the .rar file.

In pictures of Kris, after ED5D, it does stop also. Only ED test part has some failings :

  • ED46:C
  • ED4E:C

In WinAPE (by default CPC 6128 CRTC1), ED test does finish its screeen result, with several fails :

  • ED46:2
  • ED4E:2
  • ED66:2
  • ED6E:2

Others screens results after does pass. Relaunching once again in WinAPE, same results.

GA: WAIT_n generator - RET cc and WAIT_n timing analysis

IO_ACK offset into INT (interrupt) should not implicated by WAIT_n generator, and it seems that a WAIT_n during T2 is ignored because of autowait already inserted at this moment... for synchronizing an IO_ACK, I have normaly to insert WAIT_n during T2+2. No way, instruction itself is synchronized, so IO_ACK is synchronized also, you don't have to insert WAIT_n during T2+2.

http://www.cpcwiki.eu/forum/emulators/cpc-z80-timing/ : ~ a WAIT_n does not use RAM access, so does not slow down a "CPC instruction" (hypothesis) - but what about an IO_ACK during NOP in this case ?

In doc, IO_ACK begin after T2, during the two autowait inserted. So no way to detect that an instruction is IO_WAITing before slowing it following "slow down" instruction timing table.


So, it's more correct to not think of it stretching the M cycle, but instead not starting the next one that requires
a memory access until the 4th cycle.

If you think of it like this, it also explains the weird exception that happens for interrupt handling.
Normally, responding to an interrupt adds 1us on the CPC. That's because it actually just adds 2 T states for the interrupt
acknowledge before the next instruction fetch. However, in case where the last M cycle takes 6 T states, the interrupt
acknowledge doesn't delay the instruction prefetch and so the usual 1us delay doesn't occur.

"Simples!" - ralferoo

(4T equals 1us equals 1 NOP => modulo 4 synchronization of M1)

So you slow down instructions following a slowing down instruction table, slowing it the less you can, and then IO_ACK comes or not, and then you synchronize next M1 putting WAIT_n during T2 modulo 4. IO_ACK two autowaits are not prolongated.

ralferro explains also that stretching instruction timing depends of memory used or not by instruction. I know that Amstrad schematics does not use the MEM_WR wire. So it could be hard to deduce if they added 1 or more WAIT_n for certain instructions. But I'm more about 1 WAIT_n inserted at maximum each time (it's more easy to hard implements), and the modulo 4 synchro, let's see results of my current experiment (comparing time instruction of Z80 and plustest.dsk testbench, deducing diff table of "slowing down instructions")

GA: WAIT_n generator - talk about r005.8.16

OSD menu has now "WAIT_n:slow|quick", a WAIT_n generator is implemented adding 0, 1 or 2 WAIT_n per instruction.

I revisited the edges of WAIT_n generator, to insert WAIT_n at 2T (like explain in Z80 doc), so normaly it is retro-compatible with official T80. Doing this way I can set HACK_Z80 flag at false (T80.WAIT_n is now used in r005.8.16)

I also removed the edge detection of IO_ACK on gatearray, replacing it by state detection of IO_ACK, resulting cpctest's testbench back : this test of HSYNC width is now successfull.

plustest.dsk has some "missing tests" but in fact there are the prefixes : CB, DD, ED, FD, used to launch other areas of instructions.

HALT is the only one instruction that will be always OK on plustest.dsk instruction timing testbench. As this instruction cannot be timed.

plustest.dsk testbench 5 does pass, except for two instruction : CPIR and CPDR - btw, in r005.8.16, its instructions are using then same "WAIT_n generator" slower than CPI/CPD : none (no WAIT_n added for theses instructions)

GA: WAIT_n generator - plustest.asm

About 22 pages of source code using 3 columns per page.

(launch tests on several list of instruction, some instructions are tested differently using test functions : normtest, testit (testdjnz), rsttest...)
ld a,#c7
call rsttest
=> C7   RST 0H   4   3
ld a,#cf
call rsttest
=> CF   RST 8H   4   3

rsttest seems a nice candidate to explore, as all its tests are failing here.

(CPC Timing array (first instruction set))

GA: TODO : arnoldemu testbench

arnoldemu testbench: cpctest

forum : amstrad cpc "acid" test => I have uploaded updated tests : http://cpctech.cpc-live.com/test.zip

Tests done here : ppi/psg/cpctest.

Arnoldemu testbench CPCTEST-r005-6.png

In r005.6, I reach successfully some arnoldemu tests to calibrate more efficiently HSYNC interrupt : ppi.bin, psg.bin, cpctest.bin.

Games unlocked by r005.6 : Sigma7, Pac-land, Golden Tail.

In r005.8, Prehistorik is running fine.

In r005.8.4, arnoldemu testbench "cpctest" does fail :/

In r005.8.7, arnoldemu testbench "cpctest" is OK

In r005.8.14 version, using default mode "MEM_wr:quick", is OK. And Prehistorik II is running fine.

In r005.8.16c29, arnoldemu testbench "cpctest" is OK (but it is a wip version :p)

GA: TODO : MODE row buffer

MODE does change at each begin of lines, not at begin of pixel drawn.

Il ne faudrait pas penser que l'on puisse changer de mode plusieurs fois par ligne. En effet. c'est "impossible"! (jusqu'à preuve du contraire, le mode s'enclenche à chaque synchro horizontale (HBL).


GA: Moustache testbench

A homemade Testbench done firstly for helping Sorgelig to calibrate it's port of FPGAmstrad into MiSTer. But as Sorgelig core does run finer than mine (Pinball Dreams did pass ! WAIT_n:slow, CRTC 1, Brand name: Amstrad), I do then take back the good behavior using this testbench, resulting r005.8.16.1

It's a stress testbench around VSYNC/HSYNC/interrupt.

If the cat doesn't catch the line, this testbench does fail (that's a small palette testbench). First array is about VSYNC length comparable between a real CPC and an emulator. Second array is about interrupt length. Stress is done by inserting NOP, NOPNOP, NOPNOPNOP or else NOPNOPNOPNOP instruction before each measure.


jdvpa6_moustache.dsk testbench

GA: Sorgelig formula

GA instruction-timing formula (compteur1MHz is 4MHz mod 4):

--Sorgelig formula : .wait_n((phase == 0) | (IORQ_n & MREQ_n) | no_wait)
if compteur1MHz > 0 and (IO_REQ_R='1' or IO_REQ_W='1' or MEM_RD='1' or MEM_WR='1') then

I put it inside OSD menu WAIT_n:quick in r005. Does pass easily plustest.dsk test 5, but not 9. Mister Amstrad does pass test 9, I misunderstood how its Z80 is hacked.

Another great Sorgelig formula btw :

-- Sorgelig formula : wire acc = (MREQ_n | ~RFSH_n) & IORQ_n;
MREQ<=not(MREQ_n or not(RFSH_n));


Architecture of Z80.

Z80: test of a real Zilog 80

Z80fx2bb.jpg Code name : Z80fx2bb, real Z80@2MHz (instead of 4MHz) on fx2bb extension card.


For it I plug all wires simply from 1 to 40. Some wires are cut, some are Vcc, others GND. Z80 output are directly connected, Z80 input are pull-up with red-red-red resistors (I like red), Z80 is powered 5v (pmod can give 5v using jumper). In fact z80 is so old component that powering it 5v does output 3.3v.

In fact the only difference between T80 of opencore and real Z80 is that T80 runs on rising_edge, and Z80 runs during low state. Test past with little modification of sequencer forcing it to do nothing during low state of z80, resulting a downclock (memory is too overclocked with this sequencer modification), perhaps using buffer on address bus and data bus could solve this detail... but as it runs for me it is not a problem.

Z80: architecture

a) T80.vhdl

17 pages of source codes to read.

Not analyzed yet completly.

Contains the main workflow of Z80: current MCycle and its current TState.

Contains T80_ALU.vhdl and T80_MCode.vhdl components.

b) T80_ALU.vhdl

6 pages of source codes to read.

Not analyzed yet completly. This analyse can certainly be wrong : wip.

Contains flags : C N P X H Y Z S

C : carry - set if result did not fit in the register
N : negative? - last instruction was substract
P : parity or overflow - overflow example : signed, 7F+7F=FE with overflow setted
X : undocumented
H : half carry - set if 4bit first bits of result did not fit in the register
Y : undocumented
Z : zero - set if result is zero
S : sign - it is an input ?


ALU_Op is the basic instructions of Z80 coded here. T80_ALU.vhdl is a slave, a service exposed to T80_MCode.vhdl throw T80.vhdl

[§ Disassembly tables] shall make a cool ALU_Op quick reference card, doesn't it ?

c) T80_MCode.vhdl

First 5 pages, and last 2 pages of source codes to read. Others pages are "always the same" architectually speaking.

Not analyzed yet completly. This analyse can certainly be wrong : wip.

Gives instructions lengh : MCycles, TStates (please remark the 's' at end of theses words...), in [Z80 doc] each instruction is timing described using "M Cycles" and "T States" vocabulary.

It's a "controler" (proof : you have some Set_*_To outputs), does gives orders to T80_ALU.vhdl throw T80.vhdl

Actions of this controler are :

  • ALU_Op : the action !
  • I_DJNZ I_CPL I_CCF I_SCF I_RETN I_BT I_BC I_BTR I_RLD I_RRD I_INRC : actions not for ALU (wiring input/ouput, changing flags...)
  • Save_ALU/PreserveC : an option about register "erased or not" at next instruction

Instructions not coded in T80_MCode.vhdl but in T80.vhdl (strange, barbarian part of code ?) :

  • Jump/E/XY Call RstP LDZ LDW LDSPHL Special_LD ExchangeDH/Dp/AF/RS

Inc_WZ register : take a look at [§ The WZ temporary registers. It's a tmp internal register in fact.

Z80: Some bad instruction timing analysis

Based on [WinAPE>download>Plus test>plustest.dsk] testbench, mapped using [Z80 instruction set - ClrHome], instruction described then in [Z80 doc], against [WinAPE] passing testbench timing.

Solved in r005.8.16.3 (WAIT_n generator)

Z80: Some bad instruction analysis

Based on [Zexall: Z80 instruction set exerciser], running fine in JavaCPC.

Solved in r005.8.16.3 by Sorgelig (T80 fixes) - valided

Z80: ED B9 cpd(r) / ED B1 cpi(r)

Problem here : CPDR and CPIR has same implementation than ED A9 CPD and ED A1 CPI.

Solved by Sorgelig (does not pass plustest.dsk testbench 5 in r005.8.16.2 but does pass it on Sorgelig MiSTer fork version)

In fact it is a more difficult instruction that I was thinking, if you watch at it, it shall take only 4 NOPs maximum but in plustest.dsk it does take 6/4 NOPs. In fact C9 RET seem also used here, as sub instruction. I remark that CPD and CPI does take 4 NOPs, so here CPDR/CPIR does not has same implementation than CPD/CPI (so I was wrong : CPIR/CPIR is implemented in T80)

Done in r005.8.16.4 : just added a TState (TStates <= "110";) on case 3, that way plustest.dsk tb 5 does succeed.

Z80: TODO : cpc-power testbench


Some errors detected in r005.8.4 (test done by Philippe D.)

Some errors detected in r005.8.16.3

2 errors left only in r005.8.16.6 (thanks to Sorgelig hard work in T80)

Z80: winape testbench

WinAPE plustest.zip (including Instruction and Interrupt timing tests)


It's data, insertion of disk.

DSK: Another disk selector

In first version of FPGAmstrad (NEXYS2) I used switches for disk selection. As final FPGA platform doesn't have any switches set, I have to add an BASIC instruction for it, something like "OUT &CAFE,disk_number" could be fine.

Since FPGAmstrad in NEXYS4, disk selection is done from keyboard, using "OUT &CAFE,disk_number" instruction. A reset key was added also. "PRINT INP(&CAFE)" does print current disk selected number.

DSK: FAT32 fragmented files support

Since advanced FDC, dsk files have to be defragmented. Only ROMs are safe with a not defragemented sdcard...

ZX-Uno is using simple FDC, not impacted here.

DSK: TODO : arnoldemu testbench

arnoldemu testbench: fdctest

arnoldemu's testbench to pass : test/fdctest/fdctest/fdctest.dsk

Contains the DFC "SK bit" test.

Have also to fix theses "Bad Command" responses from fdc (it seems that when you don't reach a track, you have to send back the current track instead of this "Bad Command" signal). Test : 30YMD demo, "disk change" message not running correctly, "another disk inserted" is not detected in this demo.

arnoldemu's testbench results :

CoreAmstrad r005.8.15

  • 27FAIL01/29FAIL01 : read_track6/read_track10 - very big sector size counter not implemented (more than 512B)
  • 3DFAIL/45FREEZE : read_data_ov/test_write_ov - using flag simpleDSK.IS_ARNOLDEMU_TESTBENCH=false this test will fail/freeze in final version. It will not be implemented (does slow down some demos : 30YMD/Batman)
  • 41FAIL06 : test_write2 - does corrupt the testbench itself (writing a deleted mark in testdisk.dsk file) using flag SDRAM_FAT32_LOADER.IS_ARNOLDEMU_TESTBENCH=false this test will pass in final version, one time :)
  • 51FAIL02/52FAIL01 : bad5_cylinder/bad6_cylinder - writing data without data is not implemented
  • 59FAIL01 format1 - format command not implemented
  • 5EPASS : check_dtl3 - does pass but well to know that a dtl write less than sector_sector_size will not be taken into account (due to write per block of sdcard)
  • 60FAIL01 : format2 - format command not implemented (this test is slow)

Perhaps dsk does go into sleep after a certain time of no use, and then takes a certain time to wake up when reused : a timeout for turning the motor off. Perhaps overrun of FDC does turn the motor off.


During disk data transfers between the FDC and the processor, via the data bus, the FDC must be serviced by the processor every 27µs in the FM mode, and every 13µs in the MFM mode, or the FDC sets the OR (Over Run) flag in Status Register 1 to a 1 (high), and terminates the Read Data Command. 

arnoldemu's second testbench

http://www.cpctech.org.uk/test.zip arnold test last update. Folder disc/, tests : "seek, recalibrate, sense interrupt status, sense drive status, write protect"


Keyboard detection versus VSYNC signal versus interrupt cycle.

PPI: A better PIO

I'm looking after a great implementation of PIO, in original schematic of Amstrad, keyboard (output, not input) is mapped behind Yahama chip behind PIO. In some emulators, keyboard is mapped directly behind PIO. In original schematic, PIO is the only one component having a low state reset (0), I think that imply a 0 value as state init of internal components variable. Data bus of Z80 seems having a pull-up state (read 1 when nothing is plugged), for example a unplugged ROM does respond xFF in data-bus.

PPI: Yamaha clock

In r005.5 I build the Yamaha clock from GA. Unlocking "Saboteur 2" game.

Yamaha clock is generated by GA.

Yamaha clock (YM2149_linmix_AmstradStereo.vhd) is used only for "sound algorithm", not for setting/getting registers (registers are set using "BDIR BC2 BC1" wires), so I have to overclock the setting/getting register clock to simulate the original behaviour...

PPI: PPI clock

PPI in original schematic does not have clock ! So I have to overclock this one to simulate the original behaviour...

Overclocked at 16MHz.

PPI: arnoldemu testbench

arnoldemu's testbench PPI passed.




Using a simple PWM, data is entered at a certain speed, the PWM clock speed.

If you simulate a constant PWM output signal at middle range of voltage (state just between 0V and 5V : 2.5V), it results an alternance of 0V and 5V, that result in a noise sound. In Arkanoid, this defect make some continues sounds instead of silents...

My idea is generating a sound having a frequency upper than dog ultra sound, while I want to simulate a constant 2.5V.

For this I do use two clocks entries in my PWM : one about data entry, and another about algorithm execution.

This result a high quality sound output (in addition to this nice Yamaha sound chip from fpgaarcade)

SOUND: Stereo sound output

STarKos 1.21 running on FPGAmstrad

Sound chip was modified in order to get channel A+B at left, and channel B+C at right. It was tested OK using STarKos 1.21 sound tracker (track "Carpet")

In r005.8.14.1 STarKos does feel better using parameter "MEM_wr:slow" in OSD menu.

run"stk / esc / enter / enter / => / enter / space (wait) / esc / ctrl+F2 / \/ (bottom arrow) / space

(ctrl+F1 to go back into the disk menu)

STarKos seems running PERFECTLY using A-Z80 instead of T80, please do contact me if you want a personalized fork version of CoreAmstrad using A-Z80 (I have just to switch a parameter : USE_AZ80:boolean:=false; in FPGAmstrad_amstrad_motherboard.vhd)


Why not ?

Agile method

This project results of an experiment applying Agile method. Finally this project has taken 5 months. The result is a standalone platform that can run several games of Amstrad. Normally, I had to dedicate 2 months on this project, but as result was so great, I continue to a standalone and better version.

This project was done for my father birthday, so sorry that I can't deliver it yet :^)

One day perhaps I'll write one book, or write a lot of wiki page by there, presenting step by step this adventure :)

But I want really to validate project before doing it. So it will stand a few I think.

Minimal Amstrad Architecture: Build your own Z80 Amstrad Computer

I explain here my first great experiment, having Amstrad saying hello :)

First schematic: Z80+RAM+ROM

Z80 can address from 0x0000 to 0xFFFF.

RAM is from 0x0000 to 0xFFFF.

You have lower and upper ROM, so starting at address 0x0000 you put OS464.ROM, and at address xC000 you put BASIC1-0.ROM.

  • When Z80 do READ MEMORY, you read ROM
  • When Z80 do WRITE MEMORY, you write RAM
  • When Z80 do WRITE IO, you do nothing
  • When Z80 do READ IO, you response it DATA=0x00

When you run this schematic on FPGA, RAM changes!

Second schematic RAM+VGA

With JavaCPC, when you snapshoot, and hex edit result file, you see RAM content starting at a certain address.

Do "paper 2", "cls" on JavaCPC, the screen became RED, and then save a snapshoot, you can see that last part (from 0xC000 to 0xFFFF) had change from a lot of 0x00 into a lot of 0xFF

So last part of RAM is used for video (it's shown on Quasar [[30]] and other legend websites...)

For making my VGA module, I take a look at UNIX "modeline" command that give us all timing for VGA signals, and it run :)

After having a VGA module displaying a RED screen (yeah!), I made it scanning last part of RAM (from 0xC000 to 0xFFFF), and I solved the puzzle.

Crtc puzzle.jpg

RAM contain lines of 0xFF, each finishing by 0x00, but lines are not in great order

Third schematic Z80+ROM+RAM+VGA

Goal is: RAM empty at startup, VGA displays hello after run.

So you put the two last schematics together and tadam... got a problem.

The problem is that two components are accessing RAM at the same time: the Z80 and the VGA, so you had to make a sequencer. A sequencer is simply a counter fed by a clock: 00, 01, 10, 11. And you manage work task like this:

  • 00 RAM WRITE start from Z80
  • 01 RAM WRITE end from Z80
  • 10 RAM READ start from VGA
  • 11 RAM READ end from VGA

You plug sequencer(1) on z80 clock and not(sequencer(1)) on VGA...but another problem appears: VGA uses 25MHz speed for scanning RAM. So Z80 has to use same speed xD

To solve this problem you can use a special RAM done for this problem, a RAM that you can WRITE at a certain speed, and READ at another speed, this magic component is called ramb16_s16_s16. Note that they have no problem to write simultaneously on two RAM components, so that you can dump video RAM content using starter kit's external RAM, and you can display VGA using FPGA'z internal ramb16_s16_s16 RAM.

Build your own z80 amstrad computer.jpg

First and Second schematics video


Here I do program FPGA with a serial port RAM filler (homemade), and then I do upload ROMs (sooo long, you can show several blue progress bars), and then I do program FPGA first schematic : Z80+RAM+ROM, executing it, and then I do program second schematic : VGA is plugged on FPGA platform and does display "Ready".

Just about the "RAM filler", filling a RAM with data is perhaps the more important thing* about using FPGA platforms, do just remember that a RAM does keep its content as long as voltage is entered on it. So you can program FPGA without erasing RAM content. So you can program three FPGA programs : one for filling RAM, one for running a program using RAM, one for dumping RAM. Here I did use a serial transmit, but in fact, in last versions of FPGAmstrad, I do use a SDCARD with a homemade bootloader (filling RAM from SDCARD content)

I did use also Mock components, in first version of FPGAmstrad, FPGA platforms was too small, so I could have FPGAmstrad only with cool sound and low graphics, or else FPGAmstrad with cool graphics and no sound. Both using 100% of this old FPGA platform. A Mock component is a fake component, an empty one, just telling back "I'm OK, please thrust me I do exist" to others components trying to communicate with me (it's a entity with same in out ports as original but using constant output values)

Third schematic Z80+ROM+RAM+VGA video


The three main FPGAmstrad schematics

FPGAmstrad_bootloader_sd schematic

After having a first running Amstrad, I had to turn it into as standalone version. In fact before this step Amstrad ROM was put into RAM using serial port (#RAM_dump), it was slow, and Amstrad ROM was lost when I unplug electricity.

Bootloader FAT32 SDCARD is the only component playing with sdcard. Its tasks, all launched at boot, are:

- deploying ROM file on physical RAM

- deploying Nth DSK file on physical RAM, N being the binary number selected by 8 switches

State machine

Both components of Bootloader, it is to say SPI_MASTER and SDRAM_FAT32_LOADER components, does use several state-machines, one state-machine per process, each process communicating with another one using "MASTER/SLAVE" : the master state-machine does ask a slave to do something, and slave does notify master when its task is finished.

Using VHDL, I implement state-machine using a simple "switch case" on an integer. and before break I just change (increment...) this integer variable value, changing line of "switch case" this way. This "switch case" is encapsulated on a "if do/done do/done do/done" instruction. "do" being a boolean from MASTER, and "done" being a boolean from SLAVE. Each MASTER against SLAVE component has a "do" (input if SLAVE component, output if MASTER component) and a "done" (input if MASTER component, output if SLAVE component) wire. That's all. Like this you can run several sequential instructions, like reading and interpreting severals FAT32 variables using a SPI slaved component solving "read one byte at this address" instruction writen under a really low-level SDCARD protocol language.

Theses state-machines does use led debug : an integer contains the state of state machine, and this integer is displayed on 8 leds so you know where you are, it's for that I add several crash states in order to understand why and where component does crash. On MiST-board, this is displayed using the five 7-segment I just added in OSD, I added also an personalized "OSD menu entry" in order to select one or another state machine during first phases of MiST-board's version of this project (that's why you can still see a mysterious 7-segment still displayed at bottom of OSD, it's used sometime for debug purpose)

FPGAmstrad_amstrad_video schematic


The main component of this schematic is called aZRaEL_vram2vgaAmstradMiaow, due to my first experimentation about drawing a picture on VGA screen.

VGA display component does use the same parameters than unix modeline command, that's all you need, with that parameters you can display something on VGA at the frequency/resolution you choose.

AZRaEL RAM test ok zoom4 decal64 inv.jpg

AZRaEL test vram2vga.jpg


RAM and VGA does not use the same frequency. I add between them a magical VRAM having two clock entries and solving this problem automatically.

The magic RAM in FPGA, getting two clock entries, is not as magical as I was thinking : in fact it does solve clock equations using the clock manager (DCM) and BUFG components (saying phase is freedom between input and output). If you want a set of clocks synchronized do not add a BUFG in one of its wires. If you don't care about synchronize of two clocks, just add it and then it will help to solve finer and greater the clock manager equations of DCM while compiling.

If you seem interested about strange clocks generated during last step of FPGA compile, do look after "time constraints file" and "timing closure".

FPGAmstrad_amstrad_motherboard schematic

This is the core part of FPGAmstrad, it does represent the motherboard schematic of Amstrad, it was aligned to JavaCPC source code.

I fill this schematic component per component, comparing behaviour to JavaCPC components.

JavaCPC is developed in Java, and Java is so cool (Java is better computer language ever, and VHDL is better FPGA language ever =P)

I tickle JavaCPC in order to compare its components to my ones.

Emulator Architecture
 while (¡stop_emulation)
 Figura 2. Basic Emulator Algorithm.

Extracted from the book [[31]],

Using this way, emulators reach a better running time. They don't need to implement the system-bus architecture[[32]] (CONTROL DATA ADDRESS) crossing Von Neumann architecture[[33]] (CU ALU MEM IO).

Component Architecture

Java is an object language, so having new, set, get, for each of its objects.

A Component Architecture in object language has a special cycle life :

  • Build all components (new new new new new)
  • Plug all components together (set set set set set)
  • Run a main component.

Main component is Z80 on JavaCPC. In fact, JavaCPC's Z80 is already configured in order to run each instruction with a certain timing : a timing already synchronized with CRTC (each instruction takes 4 Tstates or 8 Tstates, Z80@4MHz CRTC@16bit@1MHz so drawing 8 colored pixels on mode 1 takes 4 Tstates)

My main component is #Clock_sequence

In real Amstrad, main component is GateArray.

The fact of choosing Z80 as main component just respects the emulation architecture.

Java debug mode

You can run JavaCPC on debug mode in Eclipse, and insert breakpoints.

It's useful for listening wires, and cut them. You can in live pause debug, cut a function and continue run.

Cut a wire, cut a function

Wires are done for sending message, a message in programming is a function call.

When we cut an input wire, we generally plug it to GND or Vcc.

For cutting a function, you have to insert a cut on it. A cut it's a return. You can insert a (very bad) forcing cut as:

 if (1==1) return 0;

everywhere. So function is ended at this moment and next lines became death code. It exists quality code program for checking death code, because it's generally a bug of development, normally we put code in comment.

It is the way I used in order to induce JavaCPC, comparing it with my project.

Clock sequence

When we have to make several components to communicate in a perfect timing, making a sequencer is a nice approach : "It is now your turn to do something".

Clock sequence : first try (prototype)

Original Gatearray of Amstrad is a sequencer (counter plugged with a clock), it manages synchronization between video card and z80 and memory access.

Historically there is a link between CU of CU/ALU, and... control bus and... how making your own sequencer. But I will say no more in order to not disturb these text part xD

Whatever, I made my own sequencer here in form of a bus of 4 wires called CLK4. CLK4 executes a simple repetitive sequence like 0001 0010 0011... CLK4(3), the last wire is directly connected to Z80 clock entry. Components not using explicit CLK4 as clock entry are generally using a not(CLK4(3)) entry, in order to do operations not as same time than z80.

Real Amstrad uses buffer memory in front of each address and data access, and real z80 is clock low state active. Normally if you follow datasheet of z80 you know how to map memory following CU comportment. Or you do as Amstrad, saying that z80 CU sucks, I create my own sequencer, managing all my memories access, alternating CRTC work and z80 work with little synchronization, inserting by the way more pixels that can support my small CRTC...

How to use a sequence in VHDL :

if rising_edge(master_clk) then

 if seq="00" then

 elsif seq="01" then

 elsif seq="10" then


 end if

end if

What not to do :

if rising_edge(seq[0]) then

end if

Because that can auto-generate bad unwanted sub-clocks...

Clock sequence : under time constraints (quality)

In fact, it's better to create your clock sequencer wiring each CLK and not(CLK) directly from DCM, in this case you enter in time constraints norm, and then rules/checks are done on every _edge instruction. Choosing only one sort of _edge (rising or falling) seems better also. Using that way you just have more "bad compiling error" shown, helping you creating a better code (more stable/quality).

Clock sequence using a counter plugged with a clock was in fact a bad practice (but running fine in my first versions of FPGAmstrad as I'm a good blind developer), because output are not under clock constraint : just think about that a "not" component added just after a clock wire is a Time Constraints bad practice... destroying "time constraint" solver (the one telling you when your clock domains are bad (and why), "time constraint" is last step of FPGA compiling process, it is an important step about quality, it shall be respected (generaly in a very last development effort, I shall say in a deploy effort))

Clock sequence : mirror VRAM (performance)

In order to get a better external RAM performance, and getting more luck about porting my project into others FPGA platform, I do now use a "Mirror VRAM" : external is just used by Z80 read and write (no more clock sequence finally ^^'). And a write in video RAM zone (like "poke &C000,255") does just write also in another parallel RAM, a FPGA internal RAM, that I call VRAM, this VRAM can be written at a certain speed and read at another for VGA purpose (FPGA internal RAM can be used like that)

USB Joystick

Before learning final platform and its embedded controlers (USB joystick with a controler, is just 7 wires : left right up down buttonX buttonY buttonZ), and after having destroyed 12 collector original joysticks during tests... I did some research about simply connecting a modern USB joystick into FPGA. It was a part of my Agile Method run, I worked about two months on it.


Sniffing USB frames

USB uses two wires in order to transmit frames, green and white, each with two logical values: 0v and 5v.

Let's plug a joystick on PC, if you listen at its two wires, you can sniff a USB transmission. Finally you can save it for example on RAM.

These two wires can be traduced into one with four states: 00 01 10 11.

One of this states is sleep state, in fact it depends on USB mode you use.

USB mode: USB1 or USB2; low speed, full speed or high speed

For sampling, I speed up five times the saving speed on RAM. I succeed sampling an USB1 transmission: "Logitech dual action USB joystick", and an USB2: "Sony PS3 USB joystick". PS3 joystick is not stable enough with my FPGA, but Logitech joystick is correct.





USB is just a state machine (welcome how are you today, show me your state, show me your state, show me your state....), encoding (have to read USB manual), you can use some usb sniffer software to decode them (wireshark unix version does it fine). Sniffer software does not show low level messages (ack ko ok) but does show the high level messages (ones that show that a button is pressed or not)

As it is just encoding, you can capture signals and show that they differ only when you do unpress or press a button.

pull up and pull down

If you respect USB protocol, you have to plug some pull-up and pull-down resistors and some capacitors. But as I am a bad electrician, I just simulate then in VHDL, they are important because they cause USB speed negotiations. You also have an electronic mechanism in order to detect presence of joystick plug, I don't care about it.

For reaching which wire you have to pull-up or pull-down, here the tips :

  • For slave (ideal for sniffing) : just take your USB1 joystick without plug it, just supply it (+5v red, 0v black), and test while-black and green-black with voltmeter, if you have got 5v then put a VHDL pull-up, and if you have got 0v then put a VHDL pull-down.
  • For master (ideal for creating a mini-host) : just take your PC USB1 port, and test white-black and green-black with voltmeter, if you have got 5v then put a VHDL pull-up, and if you have got 0v then put a VHDL pull-down. Normally you result two pull-down.

Synchronize, decode and check USB frames

One time sample is done, it is not readable. In fact USB frames are synchronized (they started with a certain synchronization pattern), encoded (NRZI), and checked (CRC). CRC type depends on frame length. Encoding is done for synchronization optimization.

Then using USB HID manual, you can understand type of frames, and author of them, and remark that the author alternates: USB master (PC) or USB slave (joystick)

You can use some "USB sniffer software" in order to understand more easily some frames contain, but they generally don't give all frame, and full frame.

great crc check example in perl - offered by www.usb.org

Build a minimum USB master frames state-machine

Let's just plug a USB joystick on FPGA, directly, permanently, thinking about minimum coding size : we can't implement full HID USB protocol on FPGA ^^'

Objective here is to build a minimum state-machine graph, having for transaction between state a "frame transmission". It is normal on USB protocol to have error of transmission, so you have also to put "error frame transmission" on the graph.

At stabilization, you finally switch between two states, one sending a certain frame that contains at different offset simply certain values of joystick button.

At start, some frames are employed for "next frame description", they can generally be ignored, as our USB architecture is fixed and minimal (one USB joystick, that's all)

go further with USB sniffer

A better way to snif USB could be generation of TCP/IP packets encapsulating USB packets, and to record them directly on PC from a RJ45 plug, using this way I could save more than 10 seconds of information transmission (RAM size is limited on FPGA platfoms)


http://www.ulule.com/usb-paf (unfunded) => but MiST-board final platform does offer USB pro competition Joystick compatibility <3 <3 <3

A fork of USB Joystick by The EMARD

A fork of this minimalistic USB Joystick controler by The EMARD, going further :



Why NEXYS2 500kgates starter kit

Xilinx schematics

Xilinx webpack software permit drawing schematics as book schematics, My point of view is : "For programming a FPGA, you draw a schematic as old books and just press one button. Each component on this schematic can be edited, in a language called VHDL".

My source code is not Altera compatible because of schematics drawn, but webpack can export vhdl code from schematics if you want.

RAM dump

A starter kit that contains a RAM component, that you can dump separaly : you can change schematics without loosing RAM content ! - and so write a schematic for dump only ;)

While power is on you can:

  • programming FPGA with a program/schematics done for filling RAM
  • press reset button
  • programming FPGA with a program/schematics done for using RAM
  • press reset button
  • programming FPGA with a program/schematics done for reading RAM
  • press reset button

My own made program does it with poor serial port, so for dumping all RAM content it takes about 3 hours, and for dumping Amstrad RAM part it is about 15 minutes.

On Diligent NEXYS2 official page, you can download a "Onboard Memory controller reference design" that contains explanation and VHDL source code about dumping on RAM/ROM of NEXYS2 directly from PC (usb port). I didn't tested this yet, but it is certainly a nicer approach :P

FPGA internal RAM size

It's to know that a FPGA chip contain 45KB internal RAM (360Kb for NEXYS2 500k-gates, and 504Kb for 1200k-gates) so you can't insert a dsk inside. This internal RAM is already used in part by T80 (z80 from opencores), by the soundchip, and for special RAM ramb16_s16_s16 (RAM with two different speed one for writing another for reading, in fact two RAM with a common part) that I use for VGA mode.

VHDL components size

T80 (z80 processor) take 100kgates

Yamaha sound chip (from fpgaarcade) take 50kgates

InterruptGenerator + VGA mode take 50kgates

Bootloader (for standalone) take about 120kgates (FAT32 protocol, SPI protocol, DSK protocol)

Actually the project take about 99.9% of 500kgates. But I think that TV mode will take a lower size. A bigger size shall be great for Amstrad CPC Plus version, if JavaCPC evolve, and then if I evolve ;)

Why MiST-board final platform

Final version of FPGAmstrad

MiST-board : CoreDocAmstrad

Altera schematics

Altera does also permit drawing schematics. I love schematics, my top file -gluing components- is sure a drawn schematic.

USB competition-pro Joystick

my favorite one <3

SDCARD entry robust

SDCARD player is nice built. It is not destroyed after 30 insertions. It is also easy to program FPGA : I just have to put my files into a SDCARD, and it runs, that's all.

Metal case

It's a true final platform.

Why ZX-Uno platform

Jepalza port

Jepalza has ported FPGAmstrad on it, A lot of thanks Jepalza !

Same FPGA as NEXYS2 500kgates starter kit

It's the opportunity to update the original simple prototype schematic.

low-cost FPGA

simple and over-documented

As the original, it is using simple components :

  • simple VGA: it is using a 640x480 centered VGA display at 60Hz
  • simple DSK: a dsk here is simply flatten into RAM parts
  • simple bootloader: the bootloader is read-only, loading data using SPI protocol, and slave of a FAT32 state machine deploying this data into RAM just before turning on Z80.
  • simple disk selection: the first disk is inserted at boot, and the "page-up" bottom does reset+insert the next disk.
  • simple GateArray : CRTC0 only

and is over-documented... here !

Xilinx schematics

Schematics, as on original, are quite small, except the motherboard on that is comparable to original CPC motherboard schematic.

fork and merge

This version of FPGAmstrad is a 2011's fork of NEXYS2's FPGAmstrad, merged with last validated components of MiST-board version. This way no useless options are added, and the source code stay clear !

MiST-board - Core Developer's Notes

Here, you'll find all the Amstrad MiST Core development strategy : deployment of FPGAmstrad project on this lovely MiST-board final FPGA platform.

You'll find also the Amstrad MiST Core source code.

Goto MiST-board : CoreDocAmstrad if you want to test final version of CoreAmstrad running on MiST-board platform (a final-user platform)

From Xilinx to Altera schematics

For me, global schematics are really important, for developing and for deploying. A schematic developed in order to be comparable to original documentation schematic is nice. FPGAmstrad is composed of 3 schematics :

 * amstrad_motherboard : comparable to original Amstrad schematic.
 * amstrad_video : does manage a true VGA output, using an internal VRAM.
 * bootloader_sd : sdcard bootloader, in order to load ROM and dsk at boot, from sdcard.

As Xilinx schematics are not compatible with Altera, I do generate "vhf" files, and rename them :

 * FPGAmstrad_amstrad_motherboard.vhd
 * FPGAmstrad_amstrad_video.vhd
 * FPGAmstrad_bootloader_sd.vhd

And then I make a global schematic in Altera, that contains the previous components, and several MiST-board controlers :

 * FPGAmstrad_amstrad_motherboard.vhd
 * FPGAmstrad_amstrad_video.vhd
 * FPGAmstrad_bootloader_sd.vhd
 * sdcard.v
 * user_io.v
 * data_io.v
 * sdram.v
 * osd.v

I create then a main clock component, generating all clocks I want even the not clocks (a good practice, while using "time constraints"), I also add some adapters, about wire/bus range solving :

 * MIST_SDRAM.vhd : each SDRAM has a different RAM bus size
 * MIST_DQM.vhd : just a small wiring helper
 * MIST_RGB.vhd : each VGA output has different count of colors
 * MIST_STATUS.vhd : mapping status wires
 * CONF_STR.vhd : generating OSD parameter

Xilinx to Altera

While generating vhf files from Xilinx Schematics, a lot of small components have to be adapted :

 * INV component became 'not' instruction
 * AND2 component became 'and' instruction
 * OR2 component became 'or instruction
 * GND component became '0' value
 * VCC component became '1' value

The internal RAM, sync and async (with one or two clocks) are to adapt, I use mem_altera_gen.vhd for this purpose.

Special things done during deploy

sdram.v is personalized in order to solve address **after** the write or read event. It is due to Amstrad that permit writing in RAM hidden inside ROM : if I read I read ROM, if I write I write RAM.

sdram.v is also personalized in order having a clkref lower than 4MHz.

RAM optimization

Some efforts done about internal RAM.

As I could not do enter my 32KB VRAM, I used a 16KB+8KB+4KB VRAM, to display the 640x480 output, scanning my 800x600 VRAM : bottom of my VRAM is useless for a 640x480 display.

At this step, I have 0KB of internal RAM free. Now let's do appear 16KB more in order to deploy fully my FPGAmstrad project !

RAM inferred : [in Altera reg is inferred into RAM-block](http://quartushelp.altera.com/13.0/mergedProjects/hdl/vlog/vlog_file_dir_ram.htm) (more), so a reg written like this :

reg [7:0] dir_entry_reg [31:0]

became RAM-block.

In data_io.v : dir_entry_reg does use /**synthesis noprune**/ in order to be not unwired. If I remove output dir_entry_d, RAM-block is inferred. If I let output dir_entry_d, LOGIC-block is inferred.

So I removed output dir_entry_d and set :

(* ramstyle = "logic" *) reg [7:0] dir_entry_reg [31:0] /* synthesis noprune */;

So I continue winning my 1KB internal RAM-block (here 256 Bytes was needed and turn into inferred LOGIC-block)

In sdcard.v

reg [7:0] buffer [511:0];

is a big reg and really important one (speaking to ARM SPI !), so I let it inferring into RAM-block. But about cid and csd I does :

(* ramstyle = "logic" *) reg [7:0] cid [15:0];
(* ramstyle = "logic" *) reg [7:0] csd [15:0];

Winning 2KB of internal RAM for this small 128Bytes reg :)

In VRAM_Palette, I had 16KB. But in fact a raster line is a 2+16+1 RAM palette line, so each line I store 19KB, so in fact 19*600/2=5700 bytes (800x600 VRAM in fact thruly 800x300). So only a 8KB RAM palette only was needed in FPGAmstrad project finally.

At this step I won 10KB of internal RAM (I need 6KB more to succeed in my full FPGAmstrad deployment)

I can nibble 2KB more at end of RAM palette, that's what I does.

Now I have 12KB of internal RAM free :)

And 4KB in VRAM :

800x600=100*300=30KB full
800x480=100*240=24000 24000-16384=7616<8KB=8192

so VRAM with a start vertical offset can be composed of 16KB+8KB only. So I won my last 4KB here.

I did patch my simple_GateArrayInterrupt component by parametering vertical offset :

GA_interrupt : simple_GateArrayInterrupt

generic map (VRAM_Voffset=>38*8-30*8-4*8+4 +0 +15) -- MiST +15 ?

I also patched my aZRaEL_vram2vgaAmstradMiaow component by parametering vertical offset :

 XLXI_476 : aZRaEL_vram2vgaAmstradMiaow

generic map (VOFFSET_NEGATIF =>0, -- MiST 0 VOFFSET_PALETTE=>0) -- MiST 0

Now I can use my 16KB free RAM in VRAM double buffer. Reaching a full FGPAmstrad project deploy on MiST-board, unlocking others games : it is what is done in realise 002 of Amstrad core. I tested ChaseHQ does now run fine.

ZX-Uno - Core Developer's Notes

Why I destroyed the PPL

NEXYS2's FPGAmstrad version is using a PPL (a DCM : Digital Clock Manager), just for half part of clocks generation.

Then comes the sequencer (the counter used to divise time) that does not respect "Timing Contraints good practice", forcing then adding a "I dislike good pratice" sentence on .ucf file like that :


Having half of clocks generated by a PPL results on a project running fine one time on both compilation : you add some normal lines of code, and then you toss a coin.

Then I tryed, as on MiST-board version to manage all clocks from an unique PPL (good practice !), centering all clocks on one component, removing counter and also all logical "NOT" on clocks wires (good practice !), resulting then... in a electronic circuit that does not enter inside my FPGA chip. Damn.

So I go back to dark side, removing PPL. Recalibrating all clocks (this time "rising_edge against falling_edge" instead of "same edges" per component's process), and thinking "no more Time Constraints, no more problems around". And you know what ? I got that :

WARNING:Route:464 - The router has detected a very dense, congested design. It is extremely unlikely the router will be able to finish the
  design and meet your timing requirements. To prevent excessive run time the router will change strategy. The router will now work to
  completely route this design but not to improve timing. This behavior will allow you to use the Static Timing Report and FPGA Editor to
  isolate the paths with timing problems. The cause of this behavior is either overly difficult constraints, or issues with the
  implementation or synthesis of logic in the critical timing path. If you are willing to accept a long run time, set the option "-xe c" to
  override the present behavior.
 Intermediate status: 929 unrouted;       REAL time: 3 hrs 35 secs

Damn, 3 hrs 35 secs of compiling... Then I used my brain and think that it is trying to stupidly clocking my "reset_key" wired between my keyboard clock and my bootloader clock... so I had this set of instructions inside SDRAM_FAT32_LOADER.vhd :

attribute keep : string;
attribute keep of key_reset : signal is "TRUE";
attribute clock_signal : string;
attribute clock_signal of key_reset : signal is "NO";

And tadam, less than half of hour to compile now ! and on a determinist way.

This formula does run also on bus (dsk_info bus wire coming from SDRAM_FAT32_LOADER to simple_DSK)

Internal FPGA RAM (VRAM) config

The internal dual RAM (written at 4MHz by Z80 and readden at 25MHz by VGA) are configured as "WRITE FIRST", "READ DOESN'T CARE".

aZRaEL_vram2vgaAmstradMiaow.vhd (the VRAM to VGA output part) has several manual counter offset calibrations, called "bug_*", it seems this component does not know counting right when reaching 25MHz (in fact it is, "mod" instruction does suffer a lot by here)

palette_D and aZRaEL's counters derailment

Compiler does detect when somes wires of a bus are not used, and when this bus is scanned by several counter derailing it results some data missing (this pixels normaly come from this offset, but is finaly calibred at this offset, so I plug it here and compiler does not thrust me, saying it's plug to an unused wire so does compile all that to GND... black screen)

Solution : using all wires of palette_D, taking care the compiler does not remove an "useless" wire from bus, and do calibrate manualy the derailing counters (all that "bug_*" constants inside aZRaEL_vram2vgaAmstradMiaow.vhd)

Source code

FPGAmstrad source code (Xilinx)

The project binary downloadable on #How_to_assemble_it section contains in fact source code and the binary file (.bit)

This is a simple zip of project folder.

The project was done using Xilinx webpack

It contains some direct drawn schematics, and VHDL components

MiST-board CoreAmstrad source code (Altera)

MiST-board CoreAmstrad source code

Compiling OK in Quartus II 13.0 (Altera IDE), and a few in ISE Design Suite 14.7 (Xilinx IDE) - I have to report back some modifications from my deploy platform(Altera MiST-board) to my dev platform (Xilinx NEXYS4 from Digilent Inc.)

ZX-Uno FPGAmstrad source code (Xilinx)

ZX-Uno FPGAmstrad source code


I explode the main schematic into a task by component, so the schematic is big.

Starter kit use only one RAM physical component for RAM ROM and DSK alignment, so I had to manage accesses (it is possible in fact because Z80 is a sequential processor)

My clock take 4 wires, in fact it exists a clock sequence #Clock_sequence (during 1 z80 tic, I do several things)

RAM is done for being dump, comparable to JavaCPC snapshoots.


sound chip is ym2149 one, patched, and repatched in order to get stereo sound.

PPI chip is 8255 CPCWiki one, patched.

PWM chip is PWM_DAC fpga4fun one, patched in order to get high sound quality (my PWM has two clocks in entry)

And thanks

Certainly first thanks to Markus Hohmann, for having programmed a Java version of CPC, I love Java and VHDL, so this project comes from this Java Amstrad emulator.

Secondary Steve Ciarca, author of "Build your own Z80 computer" (1981), so nice book.

Then the author of the VHDL version of Yamaha sound chip : fpgaarcade. And opencores for the Z80 (T80)...

And websites that give access to so much old Amstrad resources like :

And more :


MiST-board special features

Bulk of effort done/TODO-list, especially for the MiST-board's CoreAmstrad implementation.


ROM/RAM : extension

In r004, you have more RAM +512KB, and you can add ROMs.

  • LowerROM has .eZZ file extension
  • UpperROM has .e00 ot eFF file extension (hexa)

In r005.4, I add another UpperROM set : .f00 to .fFF file extension (hexa). If you press "space" during a reset_key ("page up" key), upperROM files used range from .f00 to .fFF instead of ranging from .e00 to .eFF. LowerROM .eZZ file extension is still used in both case.

ROM/RAM: TODO : RAM 4MB extension

Why not ?



In order to plug FPGAmstrad on TV, and help debugging. And also to test a simple scan-doubler.

r005c17 : experimental version, original signal TV output is running fine, with OSD menu. Have to add a flag in mist.ini instead of using OSD menu. scan-doubler doesn't run ok in mode 2, and has strange offset with Arkanoid (vertical display games), so it unvalidated : only original TV output will be added to r004 in r005.

r005 : VGA 60H/TV 50Hz.

VIDEO: An OSD option to enable scan-doubler

scan-doubler (simple TV to VGA converter) doesn't run ok in mode 2, but there is some many recent demo effect that doesn't pass using current VGA 72Hz implementation. Have to try to insert both VGA implementations (=> done in r005.8.15.2)

On Sorgelig fork, the scandoubler does run ok in mode 2 (but still not centered correctly on VGA 16/9)

VIDEO: A SCART output with border

Original output signal has no border, I have to implement the original border in TV mode.

Priority: HIGH! (asked by Markus Hohmann)

Done in r005.8.14.2

VIDEO: move SCART parameter into mist.ini

Doing like in other cores : do use the global "scandoubler" option in mist.ini to switch between VGA and TV mode.

VIDEO: mix SCART H and V sync into HV sync (sort of C sync)

Amstrad CPC core · Issue #35 · mist-devel-mist-binaries · GitHub :

SCART TVs expect a composite sync. The VGAs vsync is connected the SCART pin used to detect a RGB signal and is constantly driven high. A TV will not cope with a video signal with separate H and V sync.
Bu tit's usually sufficient to xor hsync and vsync to get a csync acceptable for many TVs.

So something like this

Hsync=old_Vsync xor old_Hsync;

Done in r005.8.14.1

VIDEO: refactor of Parrot PAL signal

I found a running 15kHz TV, with mist-board tutorial lesson11 Parrot PAL running fine, but not with CoreAmstrad r005.8.14.1. It's the same TV I used some years ago at festival with original CPC. I have to refactor Parrot tutorial and adapt it on CoreAmstrad in order to generate a better TV signal quality.

Done in r005.8.14.2

In theory, simple_GateArrayInterrupt.vhd shall have :

if hSyncCount=2+4 then

In practice - in r005.8.14.2 - here we have :

if hSyncCount=1+4 then

This way screen is nicely centered but CPCWiki rule "The HSYNC is modified before being sent to the monitor. It happens 2us after the HSYNC from the CRTC and lasts 4us when HSYNC length is greater or equal to 6. If R2=46, and HSYNC width is 14 then monitor hsync starts at 48 and lasts until 51." is not respected.

Test about centering screen are done using "BORDER 0", this way border is ignored and does interact with HSYNC/VSYNC screen synchronisation.


r004.8 : a better CRTC/Gateway implementation, following better JEmu (JavaCPC) one... but it is a CRTC1 (but a better ONE) Some bugs came from PPI also (keyboard bugs in particular), solved in r004.8

VIDEO: CRTC1 detection

I don't remember exactly, but in r005.8.4, one of "Midline Process"/"From Scratch"/"Pheelone" demo does crash due to a "CRTC1 needed" message : my CRTC1 seems not detected as a true CRTC1... If's "From Scratch" that does display this message in fact.

Done in r005.8.14 : Still Rising (Vanity) demo can be launched, better using "MEM_WR:slow" mode.

Test CRTC - Quasar Net

VIDEO: TODO : Interlaced scanlines

Interlaced scanline is an effect existing in CRTC (register R8) used by Wolfenstrad demo Seen also at begin of R-Typeee.dsk ("stereo soundtrack" message's picture), and seem also used in a lot of recent demos as "flipping lace" effect.

Scanline is also used just at begin of Pinball_Dreams__PREVIEW.DSK (eagle draw) - in fact I've got a doubt here, it seems more about a problem of HSYNC edge choice of alignement here.

Les Sucres en Morceaux - Amstrad CPC - Identifier les CRTC

OUT &BC00,8
OUT &BD00,3

Test CRTC - Quasar Net

L'écran passe en 100Hz, les registres 4 et 7 doivent être doublés pour retomber sur 50Hz
VIDEO: Scanlines

Here effect is about simulating CRT (not CRTC.R8) original screen. There is several way to implement it. Here, truly one line out of two is 1/2 darker. By visual effect this result in "a thin full black horizontal line".

FGPAmstrad cc withoutScanlines.png
FGPAmstrad cc withScanlines.png
VIDEO: Monochrome option

Add an option to turn screen into green monochrome mode (in mode TV and in mode VGA)

done in r005.8.9.2 (Soleil Vert demo)

Soleil vert CoreAmstrad.png
Soleil vert CoreAmstrad scandb50Hz.png

Les Sucres en Morceaux - Couleurs - 1 - Les couleurs du CPC

TODO : alternative color, but cool ones : yellow (green) blue orange pink.

VIDEO: Monochrome OSD

Could be great having the OSD in monochrome when monochrome is selected and scanlined when scanline is selected

Done in r005.8.14.4

VIDEO: TODO : Scanline during monochrome + scandb50Hz modes

soleil vert demo display result is best using scandb50Hz mode (r005.8.16c29) because it does alternate two pictures at 25Hz, seeming then like a fixed image for humans.

But my scandb50Hz option does not enable yet the scanline effect that could improve her agains this demo. To do.

VIDEO: USELESS : welcome VGA signal

While bootloader is not fully started, do display a lighter screen output (not darker pixels as original screen color CPC depth using more resistors), as it VGA should be nicely centered at each boot. And then after come back to original CPC pixel depth.

Some VGA does detect FPGAmstrad resolution just if pixels are ligther, so I turn them lighter during start of engine. Normaly a press into reset button (the one front the sdcard entry) does solve directly this problem (you can also turn on screen before MiST-board with this sort of screens)

I tryed also menu core project with my stupid screen, as it normally I can power on MiST-board before screen for FPGAmstrad (switching core does the stuff here also)

Tryed in r005.8.14.4 : lighter pixels during bootload. Also with a full white screen.

This solution does not fix the problem of "stupid screen", but reveals something interesting about the defect (next chapter)

VIDEO: TODO : SAMSUNG 16/9 tests

Using lighter pixels full white screen during bootload show me that screen doubts between two positions : a perfect centered 4/3 with 6.5 centimeters horizontal border each; and a starting 16/9 at left, crop at 6.5 centimeters left.

Without lighter pixels full white screen, the crop of image does change, moving into first displayed characters : in fact in SAMSUNG menu, the position of screen is not 50 50, if you put 50 50 you come back to "lighter pixels full white screen" defect. So here screen begining at first char displayed on screen is a second defect, but a small one, as you just have to set 50 50 in SAMSUNG menu.

So back to previous bug : screen doubt between two positions "a perfect centered 4/3 with 6.5 centimeters horizontal border each; and a starting 16/9 at left, crop at 6.5 centimeters left". When displaying a game, in fact, in found two different case in "perfect centered 4/3" case , this case is not so perfect, it does also doubts between two positions :

  • one time screen does crop at 6.5 left and right, changing the screen vertical position using menu does translate the image cropping left and right at fix position : 6.5 centimeters fix black border. About extra 1 centimeter pixels : image in middle of image does move, but not the borders at all.
  • a second time image does move perfectly (completely/totally) left and right without crop, and if centered has 6.5 black border left and right. This time image seems complete but crushed.

During ZX-Uno merged, I found two bug on VGA implementations (true ones ?), first being horizontal and vertical counter not reaching VTot/HTot (one clock tic missing), and second the horizontal counter limited to 1024 not reaching HTot that seems more than 1024. Perhaps, if this bugs are valided as it, do go back on original 800x600@72Hz modeline formula.


DSK: A advanced dsk drive

Done on r004, I added also a second Drive in order to copy easily files from one disk to another.

Irregular sector size ok.

You just have to select Drive A or B from OSD before selecting another dsk file.

Write is done directly on sdcard dsk file, so you can save games, and write texts...

You can now change disk without reset. And then play games using several disks.

CPCWiki forum - Amstrad CPC hardware - FDC floppy t80ds detection : talk about FDC in MiST-board CoreAmstrad.

Since r004 "mecashark", the FDC implementation has write access !


Add an option in OSD MENU : "SNAP DSK". Does create a copy of current disk in current drive into "SNAP[number].DSK". Heuristic for number : file count (at boot, incremented at each snap dsk done)

DSK: HOWTO: fix message "This program will not run in this environment. Press any key"


The core does not support the bundled CP/M+ software.
With a valid working CP/M+ Disc1 image mounted, the systems returns with the following message after issuing the |cpm command.
"This program will not run in this environment. Press any key"

Due to using wrong language version of CP/M+ disc (cpmpluf1.dsk is french version of CP/M+)

CP/M+ fr disk inserted (cpmpluf1.dsk)

"Wrong disk for your configuration" message seen in one-disk version of "Batman Forever" demo (two separate disk version runs fine), in forum they say that dsk image is using "bad track numbers", in fact when looking at a Track-Info with side 1 (instead of 0), track and side are correct in Track-Info but side is not ok in Sector-Info, normaly track/side are ignored in Sector-Info (Track-Info is used for that)... but still having the message, something else seems also wrong.

Do fix also message "Bad Command" while running a not existing file on disk.

Certainly linked to Orion Primes.dsk loading problem.

DSK: tapes

Do read .CDT files also.

I think @ralferoo had already written FPGA code for tape reading for his FPGA CPC. Maybe you can borrow some code from him?

Caprice32 has a nice tape.c implementation, in fact blocks are just read bit by bit (byte is shifted), at a certain speed. Perhaps starting with a fix CDT small file, reading blocks in loop, could be a nice approach around that.

Some has tryed reading sound directly (on emulator), switching to '1' when level (from 0.0 to 1.0) does pass over 0.5+0.1 and to '0' when level does pass below 0.5-0.1, that's the way @ralferoo uses, but @ralferoo seems also interested around CDT. ZX-Uno 464 is also using an audio jack input.

amstrad_190518_r005.8.16.8 does now read CDT. I've seen that sorgelig implements also the CDT with "Breaking Baud" demo running completely.

DSK: TODO : snapshoot purpose

Like in emulators, do something to go back in time while running a game.

For info, it seems called the "Multiface 2" purpose.


Could be nice around cross-dev.

Transmit: TODO : Ethernet

Integration of "ethernec.v".

Several multiplayer games using several CPC does already exists : Virtual_Net_96.


X/Y: TODO : A X/Y input

I want to work also on screen-pen entry, is there a manner to detect an analog X/Y as pen or gun ? YES : Markus Hohmann does it, he implements the lightgun on JavaCPC-GX4000 using mouse :)


register 11,12 and 13 ?

X/Y: Kempston mouse support

KLNHOMEALONE did ask AMX mouse - sorry about this, finally I added the Kempston mouse model only :p

Merge of Sorgelig kempston_mouse.v done in r005.8.16.6

Advanced Art Studio - Kempston mouse

Advanced Art Studio > Misc.> Input Devices> Kempston mouse

Advanced Art Studio > Misc.> Input Devices> Fast cursor (if you want)

I do not like the AMX mouse, because it can trick a beginner: in fact, in Advanced Art Studio, even if the mouse is already moving, you have to activate the "AMX mouse" on "Misc" menu or else the mouse stay very very slow, the time you understand that our mouse is slow and that it's abnormal, you are disgusted with Advanced Art Studio.

Others tricks

If you aren't ready yet, here somes experiments (youtube) on real Amstrad :