Background: Over the last two years of unemployment I've been developing various projects incrementally. My lack of focus has left me destitute since nothing has made it into production, and I'm unsure where to direct my efforts. I've come to realize people don't care that much about supporting causes, they're more for reaping the results ie physical goods. So with that said, it's time for me to get on the ball. The caveat: I'm long out of savings, and at the end of my "line of credit" so my only options at this moment are to go out and literally beg for a low-responsibility retail job (computer engineers aren't exactly in demand here in the North East) so I can finance my own production in my down time, or take the plunge into entrepreneurship by reaching out via Kickstarter etc. Obviously #2 sucks less, so... Which of the following projects would you (financially) support? Console game stuff: Another Famicom/NES flash cart - approx. $100, would need approx. 200 backers. Why when there are two currently in development? Well, I don't like their hardware designs from the teaser pics, and since FC is probably my favorite console I'm confident I can do as good of a job, if not better. With enough support I will make it open-source. Low-level Famicom disk dumper and analysis software - would require 200+ x $10 donations. The project would be more about research and dumping disks than hardware, but I would send hand-built hardware to those most eager to be part of the effort. Highly-accurate open-source Famicom/NES clone *DESIGN* - would need approx. 600 x $10 donations. This project is about transcribing the chip layouts made available by the Visual6502 team and Quietust into transistor schematics, then from transistor schematics into logical components and then into a hardware-agnostic HDL design. All work will be released and part of the funds will be offered to the others involved in the prior work. -OR- Highly-accurate open-source FULL Famiclone *HARDWARE*, w/ RGB output etc - approx. $100 ea, would need 200-250 backers. This would probably be a PCB replacement for the original Famicom, but offer RGB output, and maybe some optional development features allowing you to roll your own debugger or game dumper. FC/NES *PPU* clone board, w/ RGB output - approx. $100 ea, would need approx. 200 backers. Save the Playchoice PPUs! Neo Geo AES/MV-4 schematics, custom chip hints, etc. - would need approx. 300 x $10 donations to finish. NG HDMI upgrade board - approx. $100 ea, would need approx. 200 backers to start. NG MVS *FULL hardware clone* w/ HDMI output - approx. $400 ea, would need 100+ backers to start... I'd be under immense pressure with this one, but it could pay off. I could alternately do an AES board that fits into the original shell, but it'll be even more expensive... Open-source Sega Saturn modchip, based on the older designs - approx. $20 ea, would need approx 300 backers. Yeah, kind of a long shot... Windows software: High-performance 2D game development framework (open-source) - a lightweight API that will let you get down to the nitty-gritty of game development. Think SDL but lighter and highly influenced by console games. Would need approx. 300 x $10 donations to start. ^The above, but further developed into a multi-emulator system a la MAME/MESS. MAME is a MESS, so this project shouldn't be thought of as a MAME rewrite or clone, it'd be an original implementation of a similar idea, with a different mission and design guidelines. My vision is something like an end-all emulator for not only arcade platforms, consoles and computers, but by modularizing things it could be turned into a powerful SYSTEM simulation/development tool at any level of abstraction. Note that I cannot possibly replace all of MAME's drivers! Would need approx. 400 x $10 donations. -OR- ^The above, but further developed into a PC based arcade platform like Taito Type X. The project would be expanded to include a USB input/security dongle device. Would need approx. 500 x $10 donations... Tech hardware: Open-source logic analyzer - approx. $100 ea, would need approx. 200 backers. There are a few open-source logic analyzers out there now, but they suck. Their hardware sucks, their FPGA designs suck (they have low timing sample rates, they don't support state analysis), and their software sucks. I think I can change all that by bringing both a novel FPGA design and useful waveform software. Open-source parallel port <-> USB bridge - approx. $50 ea, would need approx. 200 backers. This would would let you interface with parallel port devices over USB, or you could use it for GPIO in whatever project. There is one similar device out there, but it has a lot of limitations: it uses obscure hardware, it's over-engineered and difficult to build, the learning curve is high and it's slow. I can make the hardware simple, the API simple, and the performance high by getting around the USB latency issue. Ideas? Comments? (Yes I know real jobs pay more, but this stuff is my life.)
WOw, it all sounds awesome. almost cannot choose... only I dont like the Tech hardware idea's, another logicanalyzer? another parallel<usb> ? http://xkcd.com/927/ but then in choosing the best opensource hardware... What I do like is your idea on Mame, its bulky, ive been trying to fix some stuff and while you compile stuff its take long... and 50 mb output file? why do I need all the drivers? Also your idea on an 2d framework kinda sounds awesome, another competitor to choose from, but if you make it work on a large scale of hardware and dont start with SDL like scripts but beter code... well, it depends. Full NES clone? Awesome. only part of it? a new pcb ? all sounds awesome and I like it. Only money I dont have much anymore, need to save some for school. But if you and others choose a nice project (or a few) ill see how much I can spare. I hope to see a good Emu/module software or a new NES thing Hope it make some sense, I love your ideas already.
I like all of the projects and would probably need some time to choose a preference order... I would gladly back all of them; however my budget is very limited. I'll most definitely donate what I can once you're ready to start with these. I hope you get as many backers as you need. Good luck!
How about a Neo Geo MV2 shell? I don't need a clone of the board. Just something better to place it into. Involves zero chips, PCBs, FPGAs, wires, or any other electrical/programming/etc matters related to MVS hardware.
These ones for me. But I think you would probably get more people doing the famicom replacement full clone board with RGB/HDMI/Component/what ever
"FC/NES *PPU* clone board, w/ RGB output - approx. $100 ea, would need approx. 200 backers. Save the Playchoice PPUs!" this. instant 100$ from a lot of people.
That would be my choice too, however, there is someone else who started working on this. Maybe the competition would be a good thing, maybe not. Idk http://universalppu.com/
FC/NES *PPU* clone board, w/ RGB output - approx. $100 ea, would need approx. 200 backers. Save the Playchoice PPUs! Definetely, that would be awesome!! w/ VGA or HDMI?
+1 vote for a Famicom in an FPGA. I'd be interested in helping with some retro machines on an FPGA (Verilog only though). I tried to get the minimig and Genesis to work before, but alas, my cheap FPGA lacks the SRAM which most of the existing projects need. They're getting close to a full system with some of these now though... http://danstrother.com/fpga-nes/#new http://cegt201.bradley.edu/projgrad/proj2006/fpganes/ https://rm-rfroot.net/nes_fpga/ http://hackaday.com/2009/10/17/nes-processor-cloned-on-a-fpga/ A Verilog / VHDL core for the Visual6502 would be great too... http://visual6502.org/wiki/index.php?title=6502_-_simulating_in_real_time_on_an_FPGA (this one looks fairly complete)... https://github.com/pmonta/FPGA-netlist-tools/blob/master/verilog/chip_6502.v NeoGeo, crikey! I looked into the schematics for this a while back, it would be a mammoth project to say the least. The secret of the NeoGeo seems to be the HUGE cartridge ROMs though, much of the hardware is then used for moving / scaling / rotating the huge ROM sprites. As for a logic analyser, the SUMP FPGA one looks OK, and there are a few others out there... http://www.sump.org/projects/analyzer/ http://dangerousprototypes.com/2010/02/25/prototype-open-logic-sniffer-logic-analyzer-2/ http://gadgetforge.gadgetfactory.net/gf/project/lax/ http://www.bastli.ethz.ch/index.php?page=BitHoundEn http://www.opalkelly.com/experts/logic/ http://www.christoph-lauer.de/Homepage/Blog/Eintrage/2010/10/29_FPGA_Based_Logic_Analyzer.html The USB-to-parallel bridge could just be a FT2232 module. They are cheap, and could be quite simple to hook up to 5V stuff with a couple of QuickSwitches... http://www.idt.com/products/memory-logic/bus-switch/50v-quickswitch Just some thoughts. :encouragement: OzOnE.
competition is always a good thing. i will buy a famicom with top-loader+this ppu+NES Everdrive from Krikkz. perfect combination and goodbye to emulators.
A full famiclone from scratch that supports more games than a NOAC, plsu offers higher quality audio and video output than a typical NOAC, while support original carts and accessories, and without some retarded as hell looking casing? I'd be in for sure, especially if you could integrate video upscale in the fpga for HD level output (wishful)
I'm afraid "a lot" of people is more like 100-200 total in the world. The $12k earnings off 200 sold isn't much to develop such a product over. I think this project might be for someone else in their spare time rather than 8 hours a day. HDMI would be possible with more time, but I wouldn't be thrilled about adding both analog and digital outputs. HDMI will take more hardware, a more expensive FPGA and possibly frame memory (which would be a dealbreaker since the thing would cost too much and take too long to finish). While all these projects work to varying degrees, they are not faithful implementations of the real hardware and AFAIK none have the accurate timing or sequencing to interface with real cartridges. Some don't even attempt to implement the NES hardware as proposed in prior documentation, but instead go for "high-level" approaches. Now that everything has been opened up and the simulations work nearly perfectly, it's just a matter of doing the work. The netlist 6502 core is very interesting, it looks like they took the Visual 6502's transistor netlist and turned it into a synthesizable design. I'm not sure if it works, but either way I don't think I'd be using it since it's not readable. I've already turned the netlists into unreadable schematics myself, and auto-captured some of the logic, but there's no (easy) way around rearranging design manually unfortunately. It's true Neo Geo is a big undertaking, but most of the difficulty lies almost entirely in maintaining the huge circuit's integrity rather than the logic which isn't much more complicated than the NES. I have a lot of the system already designed, but no way to test or work on it without hardware built, or sinking even more time into software simulation. AFAIK they're all based on Sump. The Sump client is usable but it's so basic, it lacks a lot of convenience for analysis, and well it's in Java... The BitHound has a lot of memory, but I think it's the only one that does however, and the hardware itself doesn't look very promising. The rest are just typical blah designs because of 1) simple input buffers and 2) their logic isn't the least bit creative. My goal would be 1+ GS/s on all channels, sufficient analog bandwidth to capture 200 MHz signals (basically any TTL/CMOS signal), glitch detection, RLE compression, better triggering, 128 MB divided among the channels, high-speed USB for rapid acquisition. I've thought of this, with an external O-C driver for the control lines, but you'd still be limited to 1k read-modify-writes/second, which sucks for some LPT devices. With a MCU (or CPLD state machine) you could hand off part of the protocol to it to achieve dramatically faster rates. It would be nice to use FTDI's USB drivers, but I think PIC unfortunately is the way to go with this project. The high project estimate is mostly around working on Microchip's USB stack which is frightening... Hopefully as accurate as a NOAC, but with proper cartridge wiring and sound. HDMI is possible but not factored into the cost... That's something that would have to be decided much closer to the end, as would cases. I have no experiences with cases. Maybe it can be 3D printed?
The accurate NES clone with RGB (or even without) and the NeoGeo MVS clone with HDMI are my favorites. Probably the NES over the NeoGeo but I'd like to see both. For NES I would imagine that the RGB output would offer for those interested in Component or HDMI much better results than using the original Composite. Plus with RGB output for those that have S-Video as their top option that would be possible. Also it would be nice to like many clones have both 72pin NES and 60pin Famicom cartridge ports available. I do think the NES Clone with RGB (since its a great gateway to all other video outputs) might be the most popular. Particularly since current clones suck. Did you think at all about including in such a clone a built in flash cart? Even if it were not some fully capable device, but just something that could play the much more simple cartridge games (non ASICs) would make it a stand out clone. Ex: NROM,CNROM,AxROM,UxROM,BNROM. It would allow for those that get the new cloned system a way in which to play or develop new software without having to also invest in a more expensive flash device.
Tell me about it. I still have a mini PIC18F2550 module that's been sat here on the desk for about a 18 months. I planned to use it for a "simple" projection screen controller, but I gave up trying to figure out their ridiculous USB library. (In the end, I just stuck a bog standard 16F88 in it with IR remote control.) I'd still like to see an affordable breakout board for HDMI. There are some available from FPGA board sellers, but they are WAY overpriced. A good HDMI chip costs around $5 these days, but I've never seen a breakout for one. If there's a decent PPU replacement out there for the NES (in Verilog), then I don't think it would take much to add HDMI to it assuming you can get the PCB made. I really must try harder to learn how to use Altium Designer. I still like Eagle, but it has it limitations. EDIT: @MottZilla - It wouldn't be difficult at all to emulate a NES cart if you stuck an FPGA and some SRAM on the board. Actually, I still have the code for emulating an N64 cart (from when I was helping Marshall with the 64drive). This is yet another thing I really should test - my FPGA board is already hooked up to an N64 cart board for the 64DD! OzOnE.
it all depends. Krikzz is making his living out of the everdrives he is selling afaik. of course i do not expect you to pay your bills just by the RGB NES, but combined with other projects it may be worth. Sooner or later a RGB supporting NES clone with a price tag about 100 bucks will hit the shelves. and i am sure it will sell well.
I don't think I'd implement component or S-video since they're both obsolete, those would have to be provided by external encoders. It won't be long before component is dropped from TVs, but composite will still linger around. So it'd be composite for "authenticity", RGB for RGB fanatics (15/31 kHz), and HDMI (strictly 720p) for modern TV gaming.. I can't promise anything about PAL support either, maybe I can hack on a Dendy mode though. I'm not sure about 72 pin availability but I'd consider it. I'd be worried about misuse... I have, but that's not really the goal. I want to keep these projects as clean and uncontroversial as possible. The FPGA will have enough block RAM to store WRAM, VRAM, a line buffer for upscanning, the FIR filter, RGB lookup tables etc. Not enough for ROMs, except maybe a boot ROM used for debugging. Signal integrity would be an issue for a breakout board, plus you'd need a FPGA with SERDES and differential outputs, yet without a TMDS mode. All FPGA now have TMDS-friendly in/outputs I think, but few recent FPGA of any use are affordable. Also unfortunately a breakout board would be 3/4 the way to a general dev board in difficulty. I would like to design my own dev board built around the Artix 7, but they aren't available from suppliers yet and well, it's really damn hard to do.
I've thought of this a lot and Krikzz and others "living the dream" are an inspiration, but there are some key differences: flash carts are relatively easy to design; most of the necessary information about various components/decoding/mappers has been previously reverse engineered and documented freely online, at the very least in open emulator source. Second, it's innately harder for me to compete since I live in the US with a substantially higher cost of living, so I can't afford the time to produce a lot of projects much less develop them. If someone's going to buy my product, which will likely be more expensive, it has to be held to a higher standard, which is difficult to show customers. I live quite frugally so I know it's possible in theory to support myself on game stuff/sponsored software, but I need a few big projects first to pay off debts and lower my expenses even further. Eventually I'd like to complete a number of projects I listed (and more, I listed only about 1/4 of the projects I've started tinkering with), but that's a long term plan. I'm involved with some other collaborative things too but so far I haven't received any royalties for my work, and I'm not sure when or how much I'll see since I haven't been involved in production. Maybe if they or one of my projects pays off and I get a little passive income I can take some chances on the less desirable devices. A case for the MVS board? How many people have MV-2s? I think I'd stand to lose a lot of money with that one. Maybe you should look into 3D printing, though with an object that size it won't be cheap.
When I mentioned S-Video or Component I actually meant exactly that, with RGB anyone can get an encoder to go from RGB to S-Video or Component Video, or whatever else they could need. What do you mean about misuse? I would assume plugging in two carts at once? You'd have to design it so that wouldn't happen. Didn't the Twin Famicom deal with that problem?
Exactly, usually it's a mechanism in the case that protects against both carts being inserted. If I made two carts available I probably wouldn't protect against foolishness because it would be ugly engineering and it would also prevent interesting things, like a dual-cart system for dumping or music making. It looks like a NES clone is what people are most interested in. I think it might be a tough sell on Kickstarter though which is tightening down on what's allowed. I also don't want to step on anyone's toes since numerous people have worked on similar projects, but I think that might be inevitable. I'd like to see this project get completed like everyone else, and I think the fastest way is for money to be involved. It's been 7 years since Kevtris showed his near-finished prototype, and even then it's quite a different concept. If anyone has any other game devices they'd like to see made I'm all ears.
Yeah that's right, it doesn't matter or not whether I actually have enough money to foot ideas of that nature from scratch. No one would believe me even if I claimed to. I lack history of expensive purchases around relevant communities to convince people.
