Coleco Chameleon Is A Kickstarter Scam

Retro gaming consoles exploded with the introduction of the Raspberry Pi and other similar single-board Linux computers. They all work the same way in that they emulate the original game console hardware with software. The game ROM is then dumped to a file and will play like the original. While this works just fine for the vast majority of us who want to get a dose of nostalgia as we chase the magic 1-up mushroom, gaming purists are not satisfied. They can tell the subtle differences between emulation and real hardware. And this is where our story begins.

Meet the Coleco Chameleon. What appears to be just another run-of-the-mill retro gaming console is not what you think. It has an FPGA core that replicates the actual hardware, to the delight of hardcore retro game scam_04enthusiasts around the world. To get it to the masses, they started an ambitious 2 million US dollar Indiegogo campaign, which has unfortunately come to a screeching halt.

Take a close look at the header image. That blue circuit board in there is nothing but an old PCI TV tuning card. To make matters worse, it also appears that their prototype system which was displayed at the Toy Fair in New York was just the guts of an SNES Jr stuffed into their shell.

This scam is clearly busted. However, the idea of reconstructing old gaming console hardware in an FPGA is a viable proposition, and there is demand for such a device from gaming enthusiasts. We can only hope that the owners of the Coleco Chameleon Kickstarter campaign meant well and slipped up trying to meet demand. If they can make a real piece of hardware, it would be welcomed.

67 thoughts on “Coleco Chameleon Is A Kickstarter Scam

        1. Because ever body at each level washes their hand of liability to save their own necks?
          Like the UK political system, it is rare to have a direct path to responsibility anymore. :/

      1. Wonder if calling it the Coleco Chameleon was an inside joke at the backers expense as Chameleon’s change their color to pretend to be a branch, leaf part of a tree etc.

      2. From what I understand, it started on Indigogo but didn’t reach the desired amount and the guy was set to try again on Kickstarted near end of past February when it was pulled unexpectedly after the eagle eyed members on Atariage forum noticed the mystery “FPGA” board inside the shell were in fact a capture card.

    1. Most of the modern emulators still use HLE, meaning that they cheat on certain routines to achieve speed requirements. Replicating the hardware more accurately in an FPGA seems attractive since you’d be closer to playing on authentic hardware. It also becomes attractive because some systems or components will get more scarce over time. This is the case with the PPU in the NES- people figured out that you could get RGB out of the NES by using a PPU from a Playchoice-10 board, which means that people started to cannibalize the PC-10 hardware. Eventually we could run out of PC-10s, as there are far fewer of those than there are NES consoles.

      All that said, I’ll still go for emulators, since 99.9% accuracy is good enough for me. =)

      1. That’s a common misbelief but emulation on FPGA is not necessary more LLE or more accurate than software emulation, it entirely depends on the implementation and knowledge of the inner working of the console chips at cycle level. You still have to take shortcuts or approximate the original hardware design and functionalities, even with a FPGA, it’s just that FPGA design makes it easier :

        1) to run and sync multiple emulated chips (parralelism)
        2) to replicate chips I/O signals in order to interface with real hardware (cartridge, controllers, etc)

        Actually, most current FPGA implementation are less accurate than the current best emulators, because
        1) their implementation is still heavily based on emulator code rather than deep hardware reverse-engineering
        2) it’s difficult to figure exactly how the chips are internally designed and working, most of the time you have to guess based on functional behavior through testing
        3) accurate emulation require a lot of gates, sometime it’s more cost effective to emulate on CPU to get similar result

        1. Exception of course is the FPGA NES. That system is documented to death since last few years. Check out the things that Kevtris has done with SID’s, too. :)

    2. Maybe [Xerxes3rd] is correct, on the other hand, I figured it was because a modern 32/64bit MCU running at 80MHz or so would execute / re-interpret the old firmware/software and still need to have a number of WAIT states thrown in to accurately imitate a 1-8MHz 8-16 bit CPU of yore.

      1. Emulating the CPU is usually not where the problem is for really old systems. It’s accurately simulating the rest of the hardware.

        I’ve done a gameboy emulator for example. The CPU, easy. The real fun comes with the display, that hardware is quite clever and has quite a few things you need to emulate. And then you discover that quite a few games modify the display state WHILE the display is drawing to get certain effects. So your timing of that piece of hardware needs to match the CPU timing.
        Sound is one level harder even, if you mess up the timing there, it sounds like a mess. And you need to emulate filters that where implemented in hardware as well.

    3. I don’t consider FPGAs as “software defined hardware.” You’re configuring the hardware not writing software.

      If the hardware core was implemented correctly and very well, you end up with a more authentic gameplay. However, as anyone that does FPGA development knows, it takes an incredible effort to reproduce these machines. While the CPU core might have been a 6502 variant, they all had custom instructions along with custom display hardware and custom sound.

      What I never understood is how they were going to handle sound generation, which was done with analog ICs.

      When this thing was first announced, the idea of being FPGA-based was a selling point for developers. If someone were developing a homebrew game for a retro-game system, then this unit could potentially be a target for that homebrew game. “You can sell to retro and modern gamers!”

      Retron 5 is an Android-based emulation machine. It turns out they were using Open Source emulators without giving proper credit. So I suspect the RetroVGS team thought the FPGA route was one method around that issue.

      1. It’s actually possible to generate analog signals using a FPGA. At the very least, you just attach a DAC like you would with a PC emulating a NES through software. Sound cards obviously have DACs. Also, a lot of the remakes are for digital output like HDMI, and thus can use DSP code to create the filtered stream for the digital output.

      1. Was just about to post this. Byuu is a crazy person, but he’s the kind of crazy we need more of to truly preserve old console games. Sooner or later–hopefully *much* later–there won’t be any genuine, working hardware left.

    1. The reason they were using them is because they got ahold of the manufacturing molds for the jaguar for cheap. I don’t remember the whole story as I read is back when they were retro vgs.

      1. Such a thing does not exist!

        Well, actually, there were a few third-party keyboards available in the UK at least. Often as a replacement case, you took the board out of your Spectrum and put it inside, just connect a couple of membrane cables into slots. Better than the original, though still not brilliant, at least the one I’ve used.

        Expandability though, you’re covered! There’s SD card and Ethernet interfaces from some Scandinavian geniuses who’ve taken Sinclair’s heritage into their custody. Searching for ZX Spectrum, the UK / Euro name, would be more succesful than Timex. The later Spectrums, +2 and above, had a proper keyboard and RGB monitor output so you wouldn’t have to deal with PAL.

        And some clever chap has implemented the Spectrum on a Papilio FPGA board, he calls it the Pipistrello. That has HDMI.

    2. I always felt using the Jaguar molds to house a Coleco emulator, of all things, was an ass decision anyways. Like sticking a Honda engine inside a damn sexy Ferrari.

      I would love an FPGA board wrapped with game supporting design choices. Get some sexy hardware and let a third party come up with the case.

      1. Prior to that, the Jag moulds were used by a company to make some sort of dental camera system, believe it or not. It’s up on Atariage, or somewhere. Here’s video, bizarrely…

        I’m informed that the dental people since sold the moulds to these Coleco-scam idiots. Bit silly all over really, to buy the moulds before you’ve even got a circuit working. Wikipedia has a lot on this case,

        Although, as ever with Wikipedia, it seems to be an article written by one or two guys who’re getting their info from some gaming website somewhere. Atariage is where this story of the fake originally came out, and where a lot of the drama occurred. There’s a post from 2010 on Atariage about the dental people wanting to sell the Jag moulds. Perhaps that’s where this silly idea originated.

        There was the Neo Geo X a few years ago, a handheld Neo Geo with a box to plug into the TV. It’s another of the anonymous ARM handhelds with a bit of SNK branding and some official ROMs. That’s the past, present, and future of commercial emulator units: running in software on an ARM, and not selling a whole lot of units. Particularly cos just buying unbranded hardware and downloading the ROMs is a lot cheaper and more sensible.

        In a market competing with the Chinese companies who invented cheap emulator platforms, putting a brand name on it and charging 3x as much isn’t a winning strategy. Using an FPGA is going to make just about no difference, especially for such an old console, which can be emulated extremely well in software. An FPGA isn’t the same hardware as the original, doesn’t have the original passive components and PCB layout and everything else. It’s not the same. It’s a minute amount of difference, but whatever that’s worth, an FGPA Colecovision is on the same side of the gap as software emulation is. May as well just use an emulator.

  1. The $2M attempt was on indiegogo, they couldn’t do kickstarter because they had no working prototype. At least the AVS (FPGA NES with HDMI) from isn’t a scam!

    1. Because, as with anything, they see nothing but dollar signs in their eyes (and they would rather have it in the hands even if it means being shady and underhanded).

      1. That’s the problem, though, we don’t know the intent. For all we know they intended to produce the console they “advertised.” They did make a series of unfortunate decisions in a “rush to market.” No question there were some scumbag moves made.

        Clearly they were deceptive with the Toy Fair SNES and the video card joke. Perhaps their marketing of the product was a scam. But as an overall product attempt, I think calling it a “Kickstarter scam” is incorrect.

        They never released the product; we have no proof they *never* intended to do so, and after all, it never made it to Kickstarter.

        1. They spent the money on buying the moulds, so I imagine they intended to have something to sell. But they went wrong. That’s forgivable. Lying and faking stuff isn’t.

          Although that said, plenty of big companies have pulled similar demonstration-unit scams, running pre-recorded video and pretending it’s live render, etc. If they eventually succeed, nobody blames them for it. Indeed the fakes might have bought them enough time to actually get real hardware working.

          I think, in black and white terms, they were wrong to fake the hardware. But people are so used to companies being dishonest and pulling this sort of shit all the time, that nobody would have minded, had they ended up succeeding. So they were wrong, but no worse than many others. That’s still wrong, of course.

          Expecting other people to provide their cores was also wrong. Seems they bit off a lot more than they could chew, and started taking money for a project they weren’t capable of doing. That’s not necessarily malicious, but it is very stupid. It’s not taking proper responsibility for other people’s money, and that’s wrong as well.

          Meanwhile of course plenty of technically capable people wouldn’t have the brass neck to pull something like this. Apparently every Woz needs his Jobs for the bullshit and off-ripping. This investment capitalism thing, anyone got any better ideas?

          1. They asked the experts in emulation advice, and then promptly ignored it. They made vast overestimates of amount of time and money needed for a working prototype, when other people have made working systems for about what they paid for the moulds. XD

    1. according to the wiki, the current version C64Chameleon can hardware emulate all of the following:
      Atari800XL, Atari2600, Vic20, ZXOne, MSX, PCEngine, Amiga(ECS), PACE (some arcade hardware), and Pong.
      Oh, and the C=64, obviously.

      1. Should be able to do Colecovision then by the sounds of it. There’s the solution!

        As for shipping games on separate cartridges, that’s just a limitation of hardware at any particular time. All computer game cartridges had a ROM capacity appropriate to the technology of the time. The 8K games of the Colecovision were because 8K was a cost-effective amount of ROM to put a game in, and enough data to produce a game on cost-effective hardware. Selling 8K carts in this day and age is insane.

        Real old cartridges have value (some of them financial value) because they’re old. They’re from our childhood, they make us feel all warm and happy. A modern cart with 8K ROM on a multi-megabyte flash chip is just pointless. Their idea to sell single carts is stupid. Having such stupid ideas to start with might’ve been an indicator these people didn’t know what they were doing.

        Their trying to get other people to write the emulators for them, as reported here, is another massive siren. People who can’t program an FPGA trying to sell an FPGA product. If anything it’d have been a dev board with a video output. Which is better than a TV tuner card I suppose.

    1. I’m a huge fan of the MIST. I’ve run C64, Mac Classic, Amiga, Atari ST, Atari 800, NES and other systems. It works better for timing-critical systems than software emulation (Solar Jetman, for example runs perfectly). Better yet, the community and documentation is very welcoming to folks who want to write their own FPGA cores (but you don’t need to at all, if you just want to run old software). Highly recommended!

  2. I could be way off base on this, but I’d have probably started with ColecoVision, and moved on to Adam, before taking on SNES, or any other platform. I’m going out on a limb here as I know very little about the different CPU’s the platforms use, but I would assume that it would be easier to replicate in FPGA the 8-bit Z80, than it would be to replicate the 16-bit 5A22, regardless of the rest of the hardware that they had to ’emulate’ as well.

    There’s little doubt that this was always a scam, regardless of their ‘intent’ to deliver a working device at some point or not. From my understanding, it was repeatedly stated that they had the FPGA ‘SNES core’ working, and that’s why they gave the ‘SNES Demo’ at the Toy Fair. The SNES hardware obviously never existed, at least in a functional state. If the hardware existed, they could have easily shown it off, it’s not like the photo showed a connected and running unit. If their development board wouldn’t fit in the case, they could have just shown the board and stated that the final product would be redesigned to fit the case. If it fit the case, but wasn’t functional, that would still have been preferable to trying to fool people with an old video capture board.

    I apologize for my bad grammar and spelling mistakes, and more importantly, if this doesn’t make sense.. I’m typing on zero sleep tonight. I just thought I’d add my 2 cents on this story, as I thought the project sounded interesting, even though I am not interested in this style of retro gaming, I’ve always been fascinated by the interesting hardware projects, even when my knowledge is too limited to fully understand them.


    1. Way it works, for SNES side, is that you first get a CPU working and basic I/O. This lets you test out stuff like implementation of addressing and able to run test code.

      Once you do this, you implement parts of the PPU until you can at least get text to appear on screen. Adding features like the hardware multiplier and various video modes comes next. Only after you actually get it running, do you do massive bug fixes. Adding accurate sound processing is a heroic effort in it’s own right, BTW.

      If they had been serious, they would have had at least a basic CPU/gamepad/2D demo(showing button presses on screen) soon enough, even if it was an ARM miniPC. I do believe they underestimated the learning curve of FPGA, though. As a first project, this is scary to do for a business. I don’t understand why they didn’t offer to license someone else’s code, at least for a demo system. They could have always replaced that code later, if they started selling enough units!

    1. It looks like they were doing a 1:1 PCB mock-up for case and connector fitment, which is legit though not something I’d share with my crowdfunders as evidence of progress.

  3. Unfortunately hackadays coverage of this is very sorely lacking, missing one of the key ‘selling points’ of the Retro VGS/Chameleon. It was going to be a ‘cartridge’ based console, the makers made a huge song and dance about how people wanted to physically own their games, something they could hand down to their kids etc etc.

    > It has an FPGA core that replicates the actual hardware
    They never had *any* cores, from most accounts. They approached several people during it’s development, including people like kevtris, to provide cores. What seems to have happened is they expected to get these cores for free/very little cost and when they managed to piss off one dev, they’d take their knowledge and move onto the next sucker, err I mean dev.

    > We can only hope that the owners of the Coleco Chameleon Kickstarter campaign meant well and slipped up trying to meet demand.
    Again, this article is very poorly written. There was no ‘demand’ as there was no hardware! Most of the ‘demand’ was just people rubber necking this slow motion car wreck, which thankfully has finally stopped moving

  4. Is this Hack a Day or some Financial Review panel or an Investigating Practices of the small developing business, When are the real hacks comming back this has progressively turned into. How to make an Arduino into a glorified LED surely that’s within the scope of the initial Tutorial of the uses of a Arduino, HACK A DAY has become a very dry place to visit now only user comments warrant any merit for providing factual information and useful hacks.


    1. Old computer games are something that interest a few of the readers here. Developing your own product and selling it, or buying someone else’s developed product, is also something quite a few here do.

      One hack per 24-hour period, you know the rules.

  5. Seems like a waste of an FPGA, many of the systems have been remade as tiny CMOS system-on-a-chip now, you could mass produce those cheaper… — Anyway, for an undocumented system, the errors will be the same for emulation — even the hardware based retro-remake consoles have compatibility issues.

    1. In practice, they eventually turn the FPGAs into those, yeah. There’s some major issues with the NoaC’s (Nintendos on chips) that made people not want to use them. I mean, you can get something like the FC3+ (SNES/Genesis/NES) clone which actually does that, but people wanted the ability to have one chip that ran it all and run firmware updates to add more systems. For just a single NES-complexity system, using a FPGA is kind of stupid, I absolutely agree. XD Mentioning updates, can you imagine if this system didn’t support updating cartridges and you had some modern-day 3D shooter or RPG? Oh, and all “EPROM”s are actually Flash ROM so there’s little reason to not support this. Ouch.

      A lot of the emulation difficulties of systems like the NES are the mappers. If you actually implement a cartridge-slot level emulator, you can run those odd MMC5 and other sensitive games. You’re technically not emulating the mapper since it’s implemented using the original cartridge. For running games off of SD cards, though, this is no-go.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.