News

may-24-2011: new ZX-One core emulates PAL ZX Spectrum 48k
Alessandro Dorigatti has made a new core for the C-One with FPGA extender: The ZX-One core will emulate a PAL Spectrum with 48k memory and the ULA+ supporting 64-colour mode. In addition to that, the Kempston joystick interface is emulated. Software can be loaded using TAP files. TZX files are currently not supported. Please note that this core requires a monitor that supports 50Hz - if you already had a PAL Minimig running, you're good to go.
Operating the core may be a bit difficult if you have never owned a ZX Spectrum before. You will find an essential picture in the download archive: The keyboard layout. The ZX Spectrum had it's very own philosophy of reducing the number of keys, and you'll need to get used to pressing shift-0 for "backspace". You may want to take a peek into the one of the many books that worldofspectrum.org has put online before you take yourself back in time with the new core.

nov-30-2009: New Minimig core V2.01 fixes minor bug in OSD menu
Tobias released a new Minimig core today that fixes a minor bug that occured every now and then in the OSD menu. Random/phantom keypresses should not occur any more with this new version. As always, it's available from the download section of this website.

nov-26-2009: New Minimig core V2.0 includes ARM replacement
Tobias Gubener has done it again: A new port of Minimig is available in the download section of this website. V2.0 is based on Jakub Bednarski's sources of september 2009. Features include:
  • emulation of four floppy drives (DF0: to DF3:)
  • emulation of two harddrives using two image files of up to 4GB each
  • FAT16 or FAT32 supported for the CF card
  • long filenames for ADF images allowed
  • max. memory 11.5MB: 2MB chipmem, 1.5MB slowmem and up to 8MB fast at the same time!
  • horizontal and vertical image interpolation in VGA 31kHz mode
  • scanline effect emulation like on Indivision flickerfixers
While some of these features cannot be used on the original Minimig hardware at all, others are possible using Jakub's ARM board as a replacement for the PIC chip. However, Tobias' implementation uses a different approach: A second instance of the 68000 processor replaces the ARM processor. This makes the C-One with FPGA extender the best-equipped Minimig on the market!

july-18-2009: CPC core at insane speed with dsk support
Tobias Gubener has finished his new CPC core, and it's an impressive demonstration of the C-One's reserves: Not one, but two high-speed CPC cores are running at the same time, serving as a perfect emulation or as a high-speed clone of the original Amstrad/Schneider CPC computer. While one of the two virtual computers takes care of dual floppy drive emulation using DSK files and a menu, the second computer is the one for the user. Three speeds can be selected: Clone-Speed at 4MHz, Fast speed at 24MHz and "insane speed" that reaches up to 80MHz, depending on cache-hits of the code.
The new CPC core archive also contains SymbOS, the operating system by the German programmer Prodatron. SymbOS is used for the menu that lets you select the type of hardware and ROMs, which gives the comfort of using mouse and/or keyboard for controlling the menu. SymbOS gets even more impressive if run on the main CPC machine at "insane speed" - you'd never guess that you're using an 8-bit machine. Find the core and the source code, which has been put under GPL by Tobias Gubener, on the download page of this website!

june-2-2009: new documents online
The updated schematics of the Instant-On board have already been uploaded to this page yesterday, and the new C-One core development manual by Peter Wendrich followed today. Both are located in the tech section of this website.

may-31-2009: 6526 CIA and C-One reverse engineering
Yes, you've read it right: The makers of the C-One have to reverse-engineer the C-One. We're specifically talking about the configuration CPLD that launches the 1k30 FPGA after a cold-start. Jeri lost the source code and only the binary file was left. We programmed the binary into a EPM3032 chip, wired it to a C64 and displayed all interesting lines with LEDs. Further, the instant-on board schematics have been corrected and will go online tomorrow.
An 8BitBaby board was used to connect the config CPLD and a 6526 CIA to the 44-pin expansion slot of the C-One. The board lets us control the CIA and the config CPLD from software that is running on the C-One. A Special core can now measure and record the behaviour of the chips. We will hopefully find the last secrets of the CIA chips this way, just like we found the last secrets of the 3032 CPLD. Our findings already benefit the core development of the Chameleon core, and Peter will be publishing source codes for having full control over all three FPGAs of the C-One. Two new pictures have been added to the pictures section of this website.

february-6-2009: Minimig core V1.29 makes IDE mod obsolete!
The new Minimog core implements soft-IRQs, so the IDE ports can now be used without having to solder on the C-One! It is still recommended to do the first part of the mod (cut off the CSEL pin from the IDE ports for safety reasons), but that can be done without any soldering skills. Find the core on the download page of this website!

february-4-2009: Core and document updates
Tobias has changed the sync behaviour of the bridge between the C-One board and the extender card. This fixes another possible cause for Minimig startup problems. The Chameleon core also got an update - the CPU now runs at the right speed of 0.985MHz for PAL, and reading from an installed SID chip also works fine now. Find both cores in the download section of this website.
Peter also continued his work on the C-One core developmoent document, which can be found in the tech section of this website.

february-1-2009: New documents and two new cores for the C-One!
By popular request, we have simplified the necessary modifications to the C-One board that are required to activate the primary IDE port for use by the Minimig core. Even these simplified modifications are not easy to do for beginners, so if you don't feel comfortable soldering on your board, please be honest to yourself and get help from an experienced friend or a professional.
The new Minimig core V1.25 requires this modification, otherwise it will not start properly. Sorry to everyone who is not up for soldering today, as this core also contains a major bugfix in the SD-Ram controller. All previous cores have over-clocked the SD-Ram on the extender board, causing startup errors and crashes where you would not expect them. Since this over-clocking was extremely moderate (timing was only off by 400 picoseconds, that's 0.4 nanoseconds!), this bug did not show at all temperatures and not on all boards.
Another new core is the preview to the forthcoming C64 product Chameleon. While the Chameleon hardware does not exist in it's final form yet, this preview core already shows some of the capabilities. The C64 works with the IEC bus, a 2MByte cycle-exact REU and a framebuffer-based, flicker-free VGA output. Since this is a preview, many functions are not available yet, such as the cartridge port, turbo function, drive emulation or the extended register set for VGA control. Peter will of course continue working on this core, and as more features become available, they will be made available to C-One owners for free.
While working on the Chameleon core, Peter also started writing down development information for the C-One, which is scattered all over this website, some mailing lists and the brains of all involved developers so far. His work aims at concentrating all this info in one document that answers all questions someone might have who is writing a core for the C-One.
All documents can be found in the tech section of this website, and the new cores can be downloaded from the download section of this website.

january-7-2009: Minimig gets a harddrive!
An internal core (not for release) now works with the primary IDE port of the C-One. Minimig V1.23 is a proof of concept that will only run will small board modifications, so it's not meant for general release. A separate IDE device is required (for example a CF card with an adapter, a real harddrive or an IDE flash module), which needs to be Amiga-formatted. Our goal is to use hardfiles directly from the CF card that you already have. Take a look at the pictures section, where we have added three new pictures.

december-27-2008: Minimig bugfix
V1.21 that was published for xmas was accidently published with an old version of the 1k30 core in the archive. This caused audio problems in some cases - we have updated the archive and called it V1.21b - download from the download section of this website!

december-23-2008: Minimig gets more memory!
The best way of saying "merry christmas" is with a little gift: The Minimig core V1.21 which is available from the download section of this website, makes use of the large memory chip on the FPGA extender: In addition to 2M Chipmem and 1M fake-fast memory, there's now another 8M fastmem usable from any program, resulting in a total of 11 megabytes of usable memory. The update also fixes a few memory-timing problems, so any startup-problems should finally be a thing of the past. Merry Xmas from Tobias and Jens!

december-8-2008: New version of Minimig core online
The new version V1.15 fixes a number of smaller bugs that might have kept the Minimig core from starting properly: The SD-Ram is now left in a known state by the bypass-core, SD-Ram timing has been adapted, timing of the IO-bridge has been tweaked and a possible problem in the directory recognition of the Minimig has been fixed. Please report back how the new core works for you! As always, the new core is availablefrom the download section of this website.

november-29-2008: Known bugs and fixes
The first boards and extender cards are reaching customers now, and a few smaller bugs have been found. If you are using an extremely small CF card (16MByte or smaller), it may be FAT12 formatted, causing the Minimig core not to work. All other cores start fine from FAT12-formatted cards. Tobias is working on a fix, but for the meantime, it's enough for you to use a FAT16 or FAT32 formatted CF card. If you are unsure, choose FAT32 in the format-dialogue of Windows.
There's also a small bugfix for Newboot: Some C64 programs caused problems if loaded from CF card, and many of these problems are fixed with the new boot.bin file. Since there are many boot.bin files in a Newboot installation, we have extended the full Newboot archive with the right files in the right places. The new archive also contains the Minimig core now, so new users only need to download and unpack a single archive from the download section of this website.

november-22-2008: Minimig core released
Tobias Gubener has ported Dennis van Weeren's Minimig core to the C-One with FPGA extender card. The version that's available from the download page of this website is the one with the latest patches by Jakub Bednarski. The big change to the open-source hardware that Dennis has published is that this core does not require a physical 68000 CPU. Instead, this core uses Tobias Gubener's TG68 CPU core, which is available under LGPL. Also, the PIC controller that is physically mounted on the Minimig board has moved into the FPGA in this version of Minimig. Instead of using a PIC core, Tobias has used another TG68 core to do the PIC controller's job, so this Minimig actually uses two 32-bit processors, and the huge Cyclone3 FPGA is only about 65% full! Yes, you can expect updates to Minimig...

november-21-2008: FPGA extender production finished, delivering NOW!
After the C-One's startup system got a new polish, the hardware will now get it's share: The first production run of the FPGA extender card has just passed quality control and is ready to ship. The FPGA extender card brings the latest in FPGA technology to the C-One: A Cyclone3 FPGA with over 24000 logic cells using 65-nanometer low-power semiconductor technology makes the heart of the extender card. Right by it's side is a 16MByte SD-Ram that can run at up to 166MHz, giving 330MByte per second peak performance. The FPGA extender has direct connection to the Video DAC and the VGA sync signals for direct and crystal-clear VGA control. It has a high-bandwidth connection to the 1k30 FPGA, which is the gateway to the C-One itself. Two new clock sources allow generation of exact multiples of PAL and NTSC frequencies for even more precise emulation of the original systems. The clocks are available to all parts of the C-One. In addition to that, a stereo delta-sigma audio converter extends the C-One's audio capabilities. Installing the FPGA extender is extremely easy: It is plugged to pin headers on the C-One board, as you can see in the pictures section of this website.
Everyone who has ordered in the past few days will now get an eMail with payment details. We apologize to everyone who had to wait a few days, but it's company policy not to take any money in advance for things that we cannot deliver right away. It has come to our attention that people have pre-paid for the C-One way in advance, and probably before Individual Computers entered the C-One project late 2002. Should you be among these persons, we'll help. Please contact Individual Computers.

november-16-2008: New startup system V2.0: Newboot
The C-One comes in completely new clothes - Newboot is here! It's faster, more colourful, easier to use and makes more use of the C-One's revolutionary hardware concept. The all-graphical menu system will also let you load programs directly with the cores, and if you don't want to see a menu at all, launch the C64 core in about six seconds from power-on. Read all about newboot - upgrade today!

november-13-2008: C64 PAL/NTSC core V0.28 with stereo-SID support!
The biggest change on this new version is probably the possibility to use two SIDs from the C64 side. The second SID is located at "the original" location where the famous German 64er magazine has located a second SID as a DIY-project in issue 11/86: Right after the first SID at addresses $d420-$d43f. The tool for changing existing SID music to Stereo that has been published in issue 6/87 of that same magazine will produce amazing surround-like sound with the second SID!
Peter also did more work on the cartridge port; now the IRQ and NMI lines are bi-directional, just like on a real C64. However, there are still issues with the MMC Replay cartridge, but we're working on that. The VIC-II implementation has now been taken from the "Chameleon" project, which is one of the next products to come out from Individual Computers. This new VIC-II has better graphics compatibility, but since Chameleon is a commercial product, the source code of the core is no longer available. However, with development of Chameleon, we will also publish new documentation of the VIC-II chip that supersedes all other documentation out there. This documentation will be public and also available to people who don't have a C-One or a Chameleon. We assume that this is even more valuable to the C64 community than a VHDL source that hardly any programmer understands!
The new core is available on the download page of this website.

august-20-2008: C64 PAL/NTSC core V0.26 now with speedloader support!
Yes, this site update is late. Almost one month. Peter has improved the PAL/NTSC core one more time with improvements in many places. First of all, the final CPU issue is solved, and that's the decimal mode of the ARR instruction (an illegal opcode). Next, reading from the SID chips is now supported, so games like Paradroid do not require the emulated registers any more. You are encouraged to install a SID chip now, as the emulated SID registers are now totally gone!
Pressing F12 now switches not only between PAL and NTSC, but also adapts to the proper CPU and VIC frequencies. This is a major improvement over any other core that was released before, as the correct frequencies now allow using fastloaders with 1541 disk drives. The correct frequencies have one drawback, though: Some VGA monitors don't like 50Hz. However, we believe that this core is a major improvement, therefore you are encouraged to download it from the download page of this website.

january-31-2008: Major update to the C64 PAL/NTSC core
FPGA64 for the C-One is now available in V0.23 - Peter has re-written large parts of the VIC-II code with the result that DMA-delay now works perfectly! Also, VIC-tricks like displaying 9 sprites in the same row are working correctly, and Mark McDougall has contributed a joystick emulation to the keyboard interface, so the numeric keypad can be used as joystick. As always, the new version of the core can be found on the download page of this website. Thanks, Peter and Mark!
Final core bounty: We really want the board to leave beta state now. The final thing to test is the 65816 processor, therefore we have a bounty of 500,- EUR for the first person who is making a VIC-20 or C64 core using the 65816 processor on the CPU/RAM card. The bounty is good until march 20th of this year! All necessary technical information has been published in may 2007 in the cone_cores Yahoo group. Read message number 89 for all the details. That Yahoo group is a very good place for finding technical answers. If you want to claim that bounty, joining this group will increase your chances!

july-12-2007: another New C64 PAL/NTSC core
Peter Wendrich and Jens Schoenfeld have done more in-depth research on the C64. This time we took a close look at the behaviour of the CPU when an NMI interrupts an IRQ or a BRK command. Further, Peter has implemented five new illegal opcodes and corrected some CIA timing. The new version of the core has version number 0.21 and can be found on the download page of this website. Thanks, Peter!

may-12-2007: New C64 PAL/NTSC core
Finally, this website gets the first update for this year. Several new versions of the C64 core were released on Peter's site in the meantime, we're sorry for the slow updates here. Find the new version on the download page of this website. Changes include improved CPU timing (especially on IRQs and NMIs), a new undocumented opcode (LXA), sprite corrections (can now be displayed at the very bottom of the screen) and in-depth research of proper CIA timing behaviour. Thanks to Peter Wendrich and betatesters!

october-28-2006: FPGA extender card prototype arrived!
The FPGA extender cards have arrived and the first tests have already been made with it. See the first picture of the cards in the pictures section of this website, and join the C-One cores mailinglist for detailed information and free discussion of the board. The two most active C-One developers Peter Wendrich and Tobias Gubener will have their FPGA extender card shortly. Mass-production of the card is only a few weeks away!

october-7-2006: FPGA extender card prototype ordered
The often-discussed FPGA extender for the C-One is entering alpha state now. With a little luck, it's going to be shown at the Amiwest show in Sacramento on the 21st of this month. Here's a first picture of the top layer of the board. With the new card, the C-One can be expanded beyond insanity - read all technical details on the new C-One cores mailinglist. Subscribe now :-)

september-25-2006: C64 PAL core leaves alpha state, first 1541 emulation
Peter Wendrich has published the first C64 PAL core that is not considered to be alpha any more. The CPU registers $0000 and $0001 are fixed now, CIA timers are fixed (including TOD registers), the IRQ/NMI timing was too fast under certain circumstances (fixed), and the VIC<->CPU timing has been improved so even the sideborders can be opened now. Peter is now accepting bugreports under the eMail address mentioned in the readme file.
Tobias Gubener has released a new system flash that can now emulate a 1541 drive. The 1k30 FPGA can be fed with a new rom file containing the drive emulation code. Every core that uses the IEC port can make use of the emulation, including Jeri's NTSC core, Peter's PAL/NTSC core and Tobias' VIC-20 core. Just hit F8 to get to choose the D64/PRG file, change the device address (default: 9), or deactivate the emulation. Many thanks must to Tobias, the recent updates are giving another demonstration of the true power of re-configurable computing.
The new core and system flash can be downloaded from the download page of this website. The sourcecode to the C64 core as a full Quartus project is included.

september-13-2006: new C64 PAL core version alpha 14
Peter Wendrich has released yet another new version of the PAL+NTSC C64 core. Updates include improved video sync behaviour, SID registers for games that use the SID as random number generator, more CIA registers, improved sprite timing for sprite multiplexers, correct VIC-II behaviour in idle cycles, cycle-exact IRQ behaviour in CPU and VIC-II communication, and last not least more illegal opcodes for the CPU. A real big thank-you goes out to Peter for his incredible work; he is giving an impressive demonstration of the real power of re-configurable computing. The new core can be downloaded from the download page of this website. Sourcecode as a full Quartus project included!

august-23-2006: new C64 PAL core version alpha 9
Another new version of the C64 PAL core by Peter Wendrich! This time, only the CPU got a few bugfixes, where the biggest of them is that NMIs are now enabled. Betatesters: Please do some in-depth testing, as there's still a bug that only occurs when certain opcode combinations are used. All single opcodes now pass all tests, but some games still don't work properly. The new core can be downloaded from the download page of this website. Now that the CPU has been completed, the source is also available with this release.

august-19-2006: new C64 PAL core version alpha 8
Peter Wendrich has published version Alpha8 of his PAL core with lots of bugfixes on the CPU. Further, the VIC II has been optimised to use less FPGA space, so the design fits more comfortably in the 1k100 FPGA of the C-One. More details about the changes can be found in the readme file of the archive. They can be obtained from the download page of this website. The C64 core is currently binary-only, therefore the last archive that includes the source is also available there (version Alpha6).

august-18-2006: new C64 PAL core, new VIC-20 core
Peter Wendrich has published version Alpha7 of his PAL core with huge changes on the CPU. It still has known bugs, but the new CPU core uses much less FPGA space, so things that have been missing on the VIC are now included. Tobias Gubener also made a new VIC-20 core: Improved display, working joystick and IEC port, 32K ram expansion and the possibility to use rom images of games and tools. Thanks must also go to Dirk and Oliver for their help. Find the complete archives on the download page of this website. The C64 core is currently binary-only (source codes will be published when known bugs are removed), but the new VIC-20 comes with complete sources.

july-28-2006: C64 PAL core, version alpha 6
After a long time, a new update of the C64 PAL core. Finally, games can be played! Sprites work, collisions between sprites and sprite/characters are properly recognized, and it should be cycle-exact now. The flickerfixer has been improved, and should work with all monitors now that support 50Hz vertical rate. This version of the core does not have the proposed MMU of the CPU, it uses the same memory layout as Jeri's NTSC core. Press the F12 key to switch between PAL (50Hz) and NTSC (60Hz) mode. Many thanks to Peter Wendrich for his continued work on the core, but also to Tobias Gubener who helped finding a bug in the IEC port. Find the complete archive including the source codes on the download page of this website.

April-3-2006: C-One, the truly re-configurable computer!
Tobias Gubener has achieved a major breakthrough in C-One core technology: Now both FPGAs can erase and re-configure each other while the machine is running. The usual configuration scheme before was to load the small FPGA from the flash and leave the contents unchanged for the complete power cycle. Now a new (internal) test-core for the 1k100 FPGA (the big one) has successfully erased and re-programmed the 1k30 FPGA (the small one, also referred to as the slave FPGA). For future cores, this means that the authors can completely control the behaviour of the machine. They do not depend on the GBridge if they don't want to. In numbers: 30% more FPGA space for any future core and as you would expect from a truly re-configurable computer: No soldering required. Just download a file from the internet and feed it to your C-One.
To understand the true meaning of these news, please imagine a mainstream PC as you know it today. Imagine you're starting a software that prompts you to exchange the video card for something that has a feature your current gfx card doesn't have. You open the case, pull out the AGP card, insert a new one, and continue to work as if that's the most natural thing. This is of course impossible with today's PCs and you should not try this at home. However, the C-One has made a step towards such a scenario, because there's hardly a single bit of static hardware behaviour left in the machine.
Re-configuring the 1k30 FPGA takes a fraction of a second, and during this time the monitor is out of sync (flickering). After that, the FPGA can immediately start working with the new contents. Interested developers will get the source code for free - we'll publish the sources on the technical page shortly.

february-17-2006: New C64 core, now PAL/NTSC switchable
Peter Wendrich has finished the next Alpha version of his C64 core called FPGA64. The new core can now be switched between PAL and NTSC by pressing F12 on the PS2 keyboard. Also, there has been a change in the VGA display that should get rid of the "rolling picture" phenomenon. The MMU of his CPU has also been debugged, it now works as described in the docs. The core still has no correct sprite display, but Peter is getting close. As always, Peter is giving out his complete source. Follow this link to the download page, where we have prepared two archives for you; One with only the binary, and the other with binary and source files for the free Altera Quartus FPGA development software.

january-10-2006: C-One available in Europe again, new cores for the board!
As many of you have already read on the individual Computers homepage, individual Computers and E3B have founded a new label in order to split the high cost that the new German law causes. Read more on www.conbridge.de (German language only). That means: The C-One is available to European customers again.
More good news from the cores side: Tobias Gubener and the crew around SymbOS have been working really hard to present you a new turbo CPC with full CF card support, lots of memory (SD-Ram required!), and a fully graphical OS for the Amstrad CPC. Every owner of the C-One should download the archive, as it also contains the new system flash version 0.61. Sorry for the inconvenience, separated archives will be published at a later stage. Tobias will be in Aachen next month, and we'll dedicate lots of time to bugfixes, documentation, and improvements of the GBridge for Jeri's C64 PAL core.
Follow this link to the download page!

november-24-2005: C-One currently not available in Europe
Due to changed German laws, individual Computers is currently not allowed to sell the C-One to countries of the European union. We're working on resolving all the legal problems, and hope to have the C-One available again soon. Read all about the German implementation of the EU directive for electronic products recycling on the individual Computers homepage.

august-15-2005: PAL C64 core version Alpha 3
Peter Wendrich has published a new Alpha version of his C64 core. This time the CPU has experienced a few updates (decimal mode and a few illegal opcodes), and an experimental MMU is included now that will be useful for future expansions like using the SD-Ram, which is much bigger than any 6502 processor can handle without additions to the hardware. Still no sprites in the VIC. The new files and the full source including a demo source for using the MMU can be found in the download area of this website.
Please note that this is still an alpha version that has lots of known issues. Please refrain from reporting bugs until it has reached beta status.

july-29-2005: Welcome Peter Wendrich to the C-One project!
Peter Wendrich from the Netherlands has already published a PAL C64 core for Xilinx FPGAs, and has now ported his core to the C-One using the free Altera development tool Quartus. The core, as well as the full source code, have been added to the download area of this website. Please note that this is an early alpha version that has lots of known issues. Please refrain from reporting bugs until it has reached beta status. Peter is working hard to finish his core, and will publish the current state of development on an irregular basis.
Credits must also go to Tobias Gubener, who has done a great share of porting the core in a fairly short time.

july-22-2005: New technical document: GBridge
The communication between the two FPGAs on the C-One board involves a fairly complicated protocol that takes care of all the additional IO pins handled by the external CPLDs, bi-directional communication for ROM transfer and graphics multiplexer. As a service to core developers, the GBridge handles all the multiplexing and de-multiplexing between all the C-One's chips, so you can focus on your FPGA designs. GBridge replaces the old GBmux, and is available with it's full source.
The GBridge source and a documentation of all it's signals have been added to the tech section of this website.

july-10-2005: New versions of system flash and example cores: Graphics viewer
The new version of the system flash now allows transferring even bigger rom files. This is demonstrated by two graphics viewer cores that shall serve as examples on how to use the new graphics modes that the gbmuxer offers. All cores are available with the full sources. Developers should update the Altera development software "Quartus" to the new version V5.0.
All files have been added to the the download area.

june-16-2005: New versions of system flash and Turbo CPC
The Turbo CPC by Tobias Gubener has experienced another update, this time with a new graphics mode that is provided by a new version of the system flash V0.47. The new mode increases the pixel rate to 33Mhz (before: 25Mhz), giving the possibbility to display a higher resolution. The old 25Mhz mode is also still available, because the new mode reduces the number of colours to 4096 (12 bits). All files have been added to the the download area.
Meanwhile, Jeri Ellsworth has confirmed to betatester David Aldrich, that, although she has been paid in full for the C64 cores of the C-One in May 2003, she has agreements not to release them yet. Thanks to Jeri Ellsworth for finally being honest, and giving an outlook to the release of the PAL C64 core, which is hopefully by the time the PAL DTV is released to the public.

may-19-2005: New cores added to download section
Tobias Gubener has continued his work on the Turbo CPC core, and has ported the VIC-20 core by MikeJ to the C-One. Along with these cores comes a new system flash V0.46. All cores have been added to the the download area.

jan-15-2005: New system flash V0.44b
This minor update fixes a bug in a rarely used stack command of the Z80 emulation. SymbOS can now run properly, and start the same program several times without messing up the stack. The new system flash has been added to the betatest archive and the flashsys.zip archive. Both are available in the download area.

jan-8-2005: Turbo CPC design study
Tobias Gubener has taken the OpenCores Z80 processor core by Daniel Wallner and designed a turbo CPC around it. It's called a 'design study', because mass storage devices are not supported - only basic programs can be entered with the keyboard. The complete source is included with the package, so you can learn from it how to...
  • access the SRAM on the CPU/RAM card
  • mix CPU and video cycles
  • use a PS2 keyboard
  • make a line doubler (AKA scandoubler)
  • use the gbmuxer
  • use the Z80 core
    The Z80 in this core runs at 16Mhz, so it's four times faster than a native CPC. If there's enough interest in such a turbo-CPC core, work on it might be continued, so disk images on the CF card will be supported. The new core can be found in the download area.

    jan-2-2005: Happy new year!
    The new year starts with good news for the C-One: The new system flash improves compatibility with flash cards, fixes a bug with flashcards that contain a lot of files (including deleted files), and the IDE/flashcard routines have partly been translated to RISC assembler, speeding up boot times by about 25%. Further, the future compatibility of the clockport with the Silversurfer and future devices has been ensured: Specialized devices can now carry the Silversurfer on a pass-through clockport, so two devices can be connected at the same time. The CPC core is also updated with this new system flash: All CPC cores now have the Symbiface IDE interface integrated. A real CPC can transfer up to 160K per second with this interface, the performance on the C-One is not yet known. Please report back! Get more information about Symbiface on this website. Get the new system flash in the download area.
    A new tutorial has been put online. C-One betatester Frank Zoll has written a very detailled tutorial about video/VGA picture generation on the C-One, using the gbmuxer and the free Quartus development software. Read more on Frank's website

    dec-12-2004: huge update!
    The 'long C-One work-weekend' is over, and the results are all avaiable on this website now:

    Flashing made easy
    Updating the system flash always required the user to load a CPC core and type a command before the update process could begin. The new and easier method is a core that does the work for you. Just choose 'update system flash' from the cores menu on startup, and everything runs automatically.

    New system flash V0.42
    The new system flash has a workaround for the TI bug in the C64 core. Click here to read about the details of how we tracked down the bug. Caution - this description is very technical! Another change is the implementation of the 1k100 'support core' in BigBoot. The new RLE data crunching routines have been used to fit a 1k100 core into the flashrom that is needed to prepare the S-Ram for the C64 core, and to run the CPC core. Since it does not have to be loaded any more, the loading times have been reduced dramatically.
    A change in the IDE timing will hopefully fix the problems with incompatible CF cards. Please report back on this, because we don't have an incompatible CF card here!
    Last not least, the system flash now has a safety feature that checks the flash contents and - if the necessary pattern is not found - immediately launches an 'emergency flash' routine that tries to fix any problems without user interaction.

    New betatest archive
    The new system flash has changed the order and the number of necessary files for launching a core. All cores and rom files have been updated - please don't use the old cores any more.

    New C64 core
    This core fixes the TI/TI$ bug. Since the bug is really fixed in this core, it should also work with older versions of the system flash, but you will have to use the old naming conventions in order to try this. We recommend not to mix software versions, and always keep your C-One up to date. Another rom file has been added that keeps the C64 core from re-starting reset-proof games. A reset will still behave as on a real C64 (that is, re-start games that don't want the user to reset the machine), but on a re-config, the memory is properly initialized.

    Tutorial online
    The technical section of this website has been extended with a tutorial about how to make your own cores for the C-One. Thanks to Jean P. Nicolle for permission to use his Pong game for the tutorial. Jean is running the website http://fpga4fun.com, where he is selling FPGA development boards of all flavours.

    We also added a few pictures of the debugging process to the pictures section of this site.

    dec-6-2004: New System flash, new C64 core
    The system flash V0.40 contains two updates: The most important for C64 SID sound lovers is the fix of the SID clock. Now any game, demo and music program uses SID1 of the C-One properly. SID2 is not utilized by the current C64 core. The second feature is untested at this point: BigBoot now supports RLE-packed cores, which should improve core loading times. RLE packed cores will be published within the next few days.
    Jeri Ellsworth writes about the C64 core: "This new core should improve reset issues, enable NMI's and hopefully fix any ?TI$ problems.". Both files are available in the download area.



    nov-27-2004: System flash V0.39
    The new version of the system flash contains a full re-compile of BigBoot and a heavily debugged version of the memory controller. Now 174 out of 178 modules are tested good. Three of the non-working modules have been tested not to be working in other computers as well, so we assume they're defective. The last module has been identified as a proprietary printer module, so from our point of view, the memory controller is fully working now. Sizes between 1MB and 32MB, and speed grades between 50ns and 80ns, both EDO and FPM have been tested.
    If none of the betatesters reports a non-working memory module within the next seven days, we consider this part of the C-One to be final. The Free SIMM offer continues until all standard memory modules are working on the C-One.



    nov-25-2004: C64 core roms separated
    The C64 core used a single rom file until now, which made replacing single parts fairly difficult. The new archive now contains separate rom files that make exchanging single rom files much easier. If you now want to replace the kernal rom with a custom one, just replace the file "0r05e000.bin" with your own kernal. The new archive can be found in the download area.



    nov-24-2004: SIMM module included with every board now!
    We're still experiencing compatibility problems with certain SIMM modules, so a known good SIMM is included with every order from now on, until the compatibility problems are fully solved. Those customers who already have a board will get a free SIMM on request.
    There has been unclear information and questions about Jeri Ellsworth's payment and her share of the profits of the C-One project. We want to make clear that there are no profits on the C-One project. All losses (presumably 40.000,- EUR after all boards are sold) are covered by individual computers. Jeri Ellsworth has not been, and will not be charged. Her payment remains stable, regardless of the amount that individual computers is putting into the project. Individual computers is not affiliated with her parallel development of the C64 DTV joystick that will go on sale tomorrow.



    nov-19-2004: System flash hotfix
    The latest system flash update fixes a bug in the cluster size of large harddrives, large CF cards and CDs. Booting from that media is now possible. The new system flash is available in the download area.



    nov-9-2004: New system flash available
    The new system flash can now be downloaded from the download page. Many improvements on the D-Ram access have been made in order to reach full compatibility with all known SIMM modules. Now any size of module with any number of chips can be used. Updating your system flash is done from a CPC core. Just copy the "cone-sys.bin" file to the root directory of your CF card, and type the |flashsys command at the ready-prompt. The whole process is now done automatically, and should produce this output. All beta boards that have been shipped on or after november 8th already have this system flash. If you have already downloaded the old betatester archive or the old flashsys.zip file from the download area, we recommend NOT to flash the older version, as it has really poor memory compatibility compared to the new version. You might have serious problems re-flashing the new version!



    nov-5-2004: Technical information published
    All those who are interested in the C-One as an FPGA development platform can now peek into it's inner workings. The technical page has been updated with the schematics of the board, the FPGA pin files, and some subdesigns that help getting the communication betwen the two FPGAs to work.



    nov-1-2004: Ordering page for the C-One beta boards is online
    Betatesters of the C-One board can order their boards here.



    oct-29-2004: C-One boards for sale at the Amiga-meeting in Bad Bramstedt, Germany
    A limited quantity of C-One boards will be available to betatesters who will attend the Amiga-meeting in Bad Bramstedt, Germany this weekend. The boards have now reached a state where all updates can be made without special equipment, and as soon as we're back from the meeting on sunday night (october 31st), the betatester registering page will go online.
    The C-One is shipped with expanded flashrom (512K instead of 128K) and with the CPU/RAM card for 269,- EUR. What you'll need to make a working system is an ATX power supply, a keyboard, a CF card, a 72-pin SIMM module (4meg or more) and a VGA monitor. See a video of the machine booting here (caution: 4.6MB!), and see the thing work - finally!



    oct-18-2004: Flashrom filesystem implemented
    The C-One comes with 512K flashrom, where less than half of it is used for the early startup procedure. The flashrom filesystem by MagerValp is now part of the C1Boot project on sourceforge. It is a read-only filesystem that will let you choose a single core to be your "default core", so you can start the C-One without any accessories - just like so many computers of the 80s.
    Writing to the user area of the flashrom is done with the mkromfs tool. It generates an image file that must be transferred through CD or a CF card to the C-One. The early startup procedure will find the image file, compare it with the current contents of the chip, and ask the user if he wants to update.



    oct-14-2004: C-One runs without life support - READY.
    All demos that were shown on the C-One board lately required loading data from a laptop or other source that is not part of the C-One board. Either the data was sent through a Silversurfer, or the early startup routine was completely bypassed, and data was loaded from a dedicated flash chip (that's what the demo on this year's Louisville expo did). Today, the C-One has booted without any "help" - a cold start brings up the early startup program "BigBoot", which loads the C64 roms and the C64 core from a compact flash card in the C-One's CF socket. The C-One reconfigurable computer then shows a READY prompt with the familiar blinking cursor, and is ready to be used as a C64.
    We're now waiting for the CPU/RAM cards to be finished to start shipping the boards to the first betatesters. A website where anyone can register to buy a board will go online before the end of the month.



    sep-25-2004: Welcome Johannes Max Gubener!
    Tobias Gubener, the developer of the CPC core and the early startup procedure, is a daddy now! This morning at 9:27 CET, Johannes-Max was born. The first son of Cindy and Tobias weighs a healthy 4130 gramms, mother and child are doing well. See a picture of Johannes-Max here. Congratulations from the whole C-One development team go to Magdeburg, Germany!



    sep-23-2004: CPU/memory card included for free with C-One board
    The technical specs of the C-One board state, that the CPU card slot can only be used if the CPU cocket for the 65816 processor is not used. Many people complained in advance that they will have to use tools to get the processor out of it's socket in order to make the CPU card slot usable.
    We're now addressing this problem with a CPU card that contains the 65816 processor and 128KBytes of high speed static ram (12 nanoseconds). The memory and the CPU can be used by FPGA cores independant of each other. The latest C64 compatibility core by Jeri Ellsworth (dated september 22nd) uses the memory, but disables the CPU. Later cores can enable the CPU, and use the high-speed memory as cache or main memory.
    Skilled hobbyists can add a crystal oscillator (so-called canned oscillator) to the back of the board to run CPU clock frequencies that cannot be generated with the on-board PLL. The CPU/RAM card is included at no extra cost with the C-One board, the old and new price is 269,- EUR.
    One picture has been added to the pictures section. It shows a pre-production board of the CPU/RAM card.



    sep-21-2004: Early startup core supports 60Hz monitors
    The new early startup core V0.32 by Tobias Gubener now supports a new video mode with a higher vertical monitor frequency. Many of today's VGA monitors and flat screens do not support 50Hz any more, so this increases usability of the C-One, especially in countries where the NTSC video system is common. When the C-One is cold-started, the picture refresh rate is 60Hz, and it can be switched any time while early startup is running by pressing the F5 (50Hz) and F6 (60Hz) keys. Both screen modes use a line frequency of 31khz.
    Along with this change, some updates have been made to the Amstrad CPC core: The 6845 video chip implementation just got much closer to the original, so that almost any demo or game that does not require a cycle-exact Z80 processor should run now.



    sep-9-2004: Amstrad/Schneider CPC core uses disk images
    Tobias Gubener, developer of the Amstrad/Schneider CPC core, has added disk image support to his core. Just like on the real CPC, he places disk routines in a memory space that is reserved for a floppy controller, and uses his own routines to replace a real disk drive. Images are loaded from Compact Flash or from a harddrive. Currently, only AMS-Dos data format images (MV and extended) are supported. These are the 3-inch one-sided disks with a capacity of about 170K per side. The format is commonly known as *.DSK format, and is widely used by emulators.
    In addition to the known CPC commands for using a floppy drive, there are new commands that let you change a disk(-image), and view the directory of available disk images on the Compact Flash card. Plans for the near future are write-support, and support for the bigger Vortex 5.25-inch disks (double-sided 720K format).
    Probably the best-known source for CPC software is this FTP server: ftp://ftp.nvg.unit.no/pub/cpc/. Remember that the C-One is not an emulator. It's a re-implementation of the actual hardware. We like to call it a hardware clone. However, since using up-to-date storage media is much more convenient, Tobias has decided to go with the standard that emulators have set.



    jul-10-2004: Amstrad/Schneider CPC core beta version
    The Schneider CPC464 core by the developer Tobias Gubener is in the works since april 2004, and it has reached full usability by now. The Amstrad CPC series of computers were based on a Z80 processor clocked at 4Mhz and had up to 128KBytes of memory. They only had a single custom chip, and used standard components for graphics and sound that can be cloned fairly easily in the FPGAs of the C-One. For the beta versions, the 6845 display chip core has been truncated to the minimum functions that are used by most CPC programs.
    In addition to graphics and sound that are equal to the capabilities of the original machine, a line-doubler (scandoubler) has been implemented, so standard 31khz VGA monitors can be used.
    The Z80 processor is being emulated by a RISC processor inside the FPGA. This processor is an 8-bit RISC that has been developed especially to adapt the Z80 commands. A 1-stage instruction pipeline reduces command execution to an average of less than about 1.5 cycles. In addition to Z80-code, the RISC processor can also execute 6502 programs, starting with version 0.26. In a shadow-mode, 6502 programs can be executed like subroutines from the cloned Amstrad computer: The early-startup procedure that has already been developed for a different core, gains a lot of usability with the display-hardware of the CPC, because we do not need a nullmodem cable any more to display messages of the early startup routines.
    Now that we have a new startup-CPU, Jeri does not have to continue work on JRISC and can completely focus on completion of the cycle-exact C64 core.
    Five pictures have been added to the pictures section, all of them are screenshots from the most recent Amstrad/Schneider CPC core running on an unmodified C-One board with no debug-hardware or additional cards connected. Many thanks go to Magdeburg, Germany, where Tobias has worked overtime in tha past few months!



    feb-14-2004: Valentines update on the C-One development
    A small setback in development lead to a completely new concept of the early startup processor. Today we introduce JRISC to the public, the new processor that launches the C-One reconfigurable computer.



    sep-15-2003: Pictures of the rework added
    The rework PCBs have been shipped today! Three new pictures have been added to the Pictures section. Another test of erasing and re-flashing a critical part of the chip was successful. The developer boards will start shipping tomorrow!



    sep-8-2003: C-One board will be sold to selected developers
    Finally the reworks on the boards are successfully tested, and development on the early startup procedure can go on at a bigger scale. The hardware is now in a state where we do not expect any changes that cannot be done over the internet. If you think you can make a constructive contribution to the project, make yourself familiar with the source code on Sourceforge. You will only be selected to become a developer if you have a track record of 6502 assembly programming. Please note that the disclaimer has been updated. Also note the changes in the specifications, especially the increase in price and the larger amount of flash memory for the early startup. To apply for becoming a developer, send an email to Jens Schönfeld with references to your past experience (links to finished projects!).



    may-4-2003: C-One boards not for sale at the release party!
    The release party is taking place tonight, but C-One boards will not be for sale at the event. Instead, the prototype that was also shown at various shows will be presented to all the journalists at the party.
    The reason for this is a last-minute change of the final boards: As the disclaimer stated (and still states, no time to change it at the moment), the rom for the early startup procedure might have to be exchanged to update to a new version of early startup. Jeri Ellsworth came up with a solution of how to use the flashing capability of the chip that we chose, so even this part of the computer will be internet-upgradable. We hope you all understand that it's worth holding the boards back another week or two until this is all implemented - it's for the benefit of all developers and early customers who want to experience the stages of development "live" as we make them available on this website.
    Check the website of Radwar Enterprises for directions of how to get to the partyplace!



    apr-13-2003: Frontside bus exceeds 100 Mhz!
    Those of you who have followed the development of the C-One since the very beginning may know that the first prototype was running at a bus frequency of 50Mhz. The January "rev 2" prototype which is the base for the currently running production has one significant change: It uses a PLL to generate the bus frequency. This has two big advantages:

    1) the FSB frequency is user-selectable
    2) the C-One will pass CE without any problems

    The choices for the FSB frequency are:

    33.87 Mhz
    50.80 Mhz
    52.92 Mhz
    67.74 Mhz
    84.67 Mhz
    89.96 Mhz
    101.61 Mhz
    105.84 Mhz
    135.48 Mhz

    The board will be shipped in the 67.74 Mhz configuration to ensure proper functionality with every SIMM module, which is mandatory for the early startup procedure. This SIMM module must be at least 4MB in size, and the speed grade must be 60ns or faster. During our tests here, we cranked up the FSB to 105.84 Mhz, and got the C-One to pass early startup with a high-quality 60ns EDO memory module. 60ns FPM modules work up to 89.96 Mhz.

    These are the absolute maximum ratings. As the memory modules get hot, their speed goes down, and pixels on the screen start flickering. Please also note that the horizontal and vertical frequency of the VGA output also changes with the FSB, so your monitor must support the generated frequencies (which are still being adjusted, we'll publish the actual frequencies later).

    The highest FSB frequency of 135.48 Mhz is not supported with the current early startup procedure, and probably will not be any time soon. Other frequencies between the steps given cannot be generated without adding another crystal, which falls under the topic "aftermarket tuning", and cannot be done without a soldering iron.



    apr-10-2003: Production has started!
    After more than two years development time, the C-One production has started. It'll start selling on may 5th in Germany and the Netherlands. Our other European partners should have the boards by may 9th, when the board will also be launched in North America.

    Many projects have dealt with re-configurable computers so far, but none of them is as consistent and flexible as the C-One. Other projects only kept parts of the hardware re-configurable, but the C-One can change the behaviour of it's chipset even during runtime. Therefore, the C-One is the world's first re-configurable computer. The first boards are developer boards. To get some advice if you're unsure to buy at this early stage, read our disclaimer first.



    jan-17-2003: first pictures of the pre-production board
    the pictures section has been extended by 11 pictures of the pre-production board. Basic testing has already been done, all chips are responding positively to diagnostic access. A 100% test procedure will be developed starting this weekend.



    dec-22-2002: specifications page reworked
    some specifications have been brought up to date, pinouts of user-accessible connectors were inserted. Design work is going on, things might still change. The page is a snapshot of the current status.



    nov-28-2002: Website re-launched
    The new design has been done by no-Hype. Hope you like it!