This is a special opportunity to discuss what transpired during an integral part of the development of the RETRO Video Game System (now known as the Coleco Chameleon) by RETRO Video Game Systems Inc. It is clear that Mike Kennedy is the primary owner, as far as we know, of RETRO Video Game Systems Inc, he is also listed on their website as the president of the company. Mr. Kennedy has also been the voice of the project for quite a while as precious few of the other members are giving interviews to the press. Mr. John Carlsen is the former hardware engineer that was working on the RETRO Video Game System up to late 2015. After a failed IndieGoGo campaign, Mr. Carlsen disappeared from the gaming world. Having gone silent after a professional departure note was posted on the RETRO Video Game Systems Inc Facebook page, it was unclear as to what role Mr. Carlsen really played in the price of the console. Over the following months, Mr. Kennedy has done many interviews where he has placed the high cost of his console solely at the feet of Mr. Carlsen. In the pursuit of the truth and staying within my boundaries of a journalist, I contacted Mr. Carlsen and asked if he would like to discuss his position, activity and thoughts involving the RETRO Video Game System. What follows is that interview, edited as little as possible. The only things removed are formalities, personal discussion (life, traffic, website hosting, etc). This is as complete an interview as is available with Mr. Carlsen at this time.
John Carlsen: I have a vested interest in seeing Mike succeed, as that’s the only way I’ll get paid for my time and expenses on the project. But, I would also like to refute some of what he’s said as untrue. There’s a lot to the story…
Carl Williams: I agree, while not invested in this thing, other than time, if it fails, I have one less polarizing thing to write about. Let’s start with your career, you have a storied career- how did it start? What was your motivation?
JC: In short, I grew up in Silicon Valley, about half way between Atari and Apple, and was still in grade school when those two companies caught my attention and started me thinking about building my own computer and a company. I got my start trying to keep running on my Atari 800 a disk drive that someone had made out of salvaged parts. It needed a fair bit of TLC, and I learned to keep it running so that I could write code and play games. So, my motivation was mostly necessity from being a broke kid wanting to play with a new toy.
I started running a little fix-it shop from my parents’ house, and when I was in junior high school, I managed to buy the Atari field service kit and inventory from the national retail chain Video Concepts as it closed.
Between my last years of high school, I managed to get a summer job providing technical support at Atari (often as the only one who did). There, I got to know the Tramiels.
After high school and through my first semester as an engineering student at San Jose State University, I worked for IBM at its Almaden Research Center (one of IBM’s two large research campuses in the USA, which had opened only the year before).
Then one of my friends from a local BBS mentioned that his brother had attended a lecture at Stanford by Nolan Bushnell, who was starting a new robotics business. I looked him up, got hired, and worked for him for about a year. By the way, my parents made me get rid of a coin-operated Pong game that I had been restoring with a friend; after hearing that Nolan had given his last Pong game to the Smithsonian museum, for Christmas I gave Nolan my old Pong game, and still see it occasionally in his press photos.)
CW: It is safe to say, hardware is in your blood so to speak.
In my previous interview when you were with the RETRO Video Game System team, I believe you mentioned having worked on a reworking of the PlayStation and at Iguana. Could you elaborate on these and any other areas of interest for the readers?
JC: Yes. My parents met while working for IBM, so in a way I was born into–or at least from–the computer business.
Sony released the PlayStation in late 1994, while I was at Iguana Entertainment and already more than a year after we had moved the company to Austin (that was about a month before Iguana was sold to Acclaim Entertainment). There, my primary responsibility was to create the tools we needed in order to produce games. Back then, that involved reverse-engineering game consoles enough to build the interfaces.
Immediately after the PlayStation was released in Japan, I was sent one to take apart. It took me roughly six months to map out the system hardware, disassemble its ROM, and build interfaces that would allow us to develop our games cheaply.
At Iguana, I also made similar interfaces for the SNES, Saturn, and Jaguar. It’s funny that I was building the same kinds of interfaces that Activision had used to create its great games, and that I had seen those old units when I worked there just a few years earlier (when it was called Mediagenic), and that the first SNES development systems I ever saw were the two in Steve Woita’s lab there.
In making those interfaces, I had learned a lot about the PlayStation, perhaps even more than anyone outside of Sony. So, when in 2010 Sony wanted to build a low-cost PlayStation to develop foreign markets, I got a call, and a really great project. I created a PlayStation that could be built into its own controller with a bunch of games, be plugged into a TV, and run on batteries (to keep the cost down). I made sure that the hardware supported 100% of the games for the original PlayStation, including those that could use multi taps and the link cable. I built the prototypes out of gray PlayStation analog controllers I bought on eBay (possibly from some of your readers), and they worked great.
The project got so popular within Sony that the different offices wanted to sell it in all of the BRIC nations (Brazil, Russia, India, and China). Unfortunately, it got more popular than a low-margin device should. Sony’s top executives determined that the risk was too great that it might be imported to compete with other products in its primary markets, so Sony killed the project.
CW: Wow, so your career is the continuation of the family legacy. That is awesome. By trade my family are more blue collar.
JC: Most of my family are more blue collar, too. My mother is from the Netherlands, where she and my father met while working for IBM; she and most of her family worked in the national steel mill, including her father, who worked his career there as a security guard. My father’s family were mostly fishermen (including his father), with a few school teachers (including his mother).
CW: Let’s get to the elephant in the room here. The Coleco Chameleon, formerly the RETRO Video Game System. Can you give us some insight as to how you became a member of the team on this console?
JC: Probably closer to a year ago, when I was still living in Palm Springs after working for another client in that area, I recall finding Mike online through either some news about the project or possibly even some early advertising. I was intrigued by the idea of reusing the Jaguar design, which I recalled from my days at Iguana.
I was also surprised to reconnect with Steve. We hadn’t really gotten to know each other when we both worked at Mediagenic.
After resigning from the project as our funding campaign floundered, I hadn’t really kept up with the project until I heard that a new prototype was going to be displayed at the Toy Fair. The Coleco Chameleon seems to have little in common with what I designed, other than the use of some 3-D renderings of one of my early designs for the rear panel.
CW: I understand you can’t comment on the current status of the project but would you like to address any criticism from when you were a member of the team? Such as comments made by project manager/owner, Mike Kennedy in an ArsTechnica interview he did?
JC: Obviously I haven’t been privy to what Mike has been doing since I left the project, but I think my work on the project gives me pretty good insight into what he can offer.
CW: Yeah, it seems that after you left the team, things changed quite a bit for the “prototypes” they were showing the public.
JC: When I first approached the Retro VGS project, I gathered requirements, as I do when starting many new projects. Mike wanted a console that would fit in the shells that could be made from the tooling used for the Atari Jaguar (after modifications for use in dental imaging) and play retro-style games, many of which were already being developed for existing old consoles. Steve added two more requirements: the console had to include an FPGA that could be used to emulate (although he refused to use that word) the older systems in hardware, and the console and games had to be able to work right away after being stored in a closet for 50 years. Mike and Steve had seniority, and they made it clear that these requirements were non-negotiable.
After I had researched the best (and lowest-cost) solutions, put together initial designs, and estimated costs and a target price for what he and Steve required, Mike wouldn’t believe me when I told him how much it would cost. He still wouldn’t believe me when I compared the prices and apparent manufacturing costs of products that would do similar things, including the FPGA Arcade Replay.
Using an FPGA for hardware-based emulation and the 50-year data retention requirement were the major cost adders, and the cost of building modern hardware into the Jaguar shell was also significant. In the Ars Technica interview, Mike had overstated the PCB cost I had given him by about three to five times.
It turned out that we might have been able to release a system for $150, but we wouldn’t make any money and it wouldn’t include an FPGA.
It certainly took him a long time to believe my opinion of what it would cost to build what he and Steve demanded. As [Mike] did, we slowly became able to scale back those demands to offer a lower price, but keeping it in the Jaguar shell kept the minimum price up. Plus, making something so big made it expensive to ship them individually.
Mike caused me to distrust him. When I first joined the project, we had agreed on a timeline, which of course slipped; I had promised six months and put in seven. At the end, Mike, Steve and I had agreed to quietly let the crowdfunding campaign fade as I left the project, but Mike announced only a day or two later that they were making changes to the hardware team; Mike’s announcement caused me to make mine, clarifying that I had left on my own accord. Along the way, Mike had let out that I had been researching putting microdrives into the cartridges–only days after I explicitly got him to promise not to–and that I was developing low-cost adapters to play cartridges from other systems.
CW: So it wasn’t you that was wanting the Super FPGA then?
JC: No, it wasn’t me. The FPGA was one of the two requirements that Steve added. (The other was 50-year data retention.)
From the start I wanted to get rid of the FPGA, or at least get Steve to give us some rational marketable justification for the amount it increased our cost and price.
At first, Mike even wanted to be able to use the FPGA to emulate everything up to and including a Neo Geo, and emulating a Neo Geo in hardware got expensive because I had to add more RAM and RAM buses to the FPGA. By the way, that’s a great example of Mike causing the cost to increase, not me, as he claimed. The system hardware would have been very capable, but expensive and wasted if we had only used it for emulation.
CW: There has been a lot of speculation about the prototype video that you made available around the time of the IndieGoGo campaign. Can you elaborate on that video a bit? What was under the clear Jaguar shell top for instance? Was that video your idea or was it a team effort?
JC: Mike had insisted that we needed to demonstrate something, so I suggested that I could put together a quick video. Steve insisted beforehand that I refer to my kitchen table as our “lab”, which I did with some sarcasm, and many people caught on. Other than that, I put it together myself using what I had been working with. Unfortunately, with Mike pushing me to get the price lower, I had just made a major design change about a month earlier, and that’s all I had to show.
Our interview with you came at a particularly bad time after that design change, as it affected much of the software development path and I wasn’t prepared to talk about the issues that I was still resolving. Mike insisted that we do the interview, but we were totally disorganized at that point.
After putting together the video, I shared it with Mike and Steve, and got their approval prior to publishing it.
That was the only thing I had ever published on YouTube, and soon I noticed that Google was advertising on my channel some of the other YouTube videos I had been watching. By that point, I had already left the project and just found it easier to kill my YouTube account.
What was significant about that video was that I was using the actual microprocessor that I had budgeted for and was designing in, and outputting HDMI as well as analog A/V. (I had to enlist help from a friend in China even just to open a dialog with the maker of that microprocessor!) At that point in development, there were lots of parts on evaluation boards that were loosely wired together. Still, it worked, and gave us lots of computing power at the right price.
CW: I could tell that the interview I did with you, Mr. Woita and Mr. Kennedy was not a popular thing for all involved. I actually just put that interview request up as a joke, people never thought the current team would do it based on my critical articles on the project to that point.
JC: Funny that. Apparently, we had drawn criticism for Mike’s ties to a previous interviewer, which I hadn’t known about until after that interview. Mike had said that our interview with you was important because you would bring a neutral perspective. Mike didn’t tell us much about who you were, nor anything about what we would discuss or even how long the interview would run.
CW: What was the platform shaping up to be at this point in time? Were there evaluation boards that were off the shelf being used?
JC: Major semiconductors these days are usually available on evaluation boards sold by the semiconductor companies and/or third parties. Evaluation boards are important development tools, especially given the high cost of designing and printing circuit boards.
To get costs down, I had proposed a base design with a fast and very inexpensive four-core microprocessor and only minimal programmable logic–not enough of an FPGA to emulate game consoles. The plan was to use stretch goals to scale up the size of the FPGA, rather than offer a cheaper version without the FPGA and a more expensive one with it, which was an idea we had previously considered and thrown out because it would have caused problems for software developers and consumers.
CW: I was not aware of all of the interviews that Mr. Kennedy had done prior (he had done many audio interviews already). His lack of sharing information seems to run deep with his projects, compartmentalized as they say.
Are you still involved with the project in any way?
JC: No, I left the project about a week into our crowdfunding campaign.
Mike seems to share information freely (even too freely) when it appears to be in his interest. As I had mentioned, I learned that I couldn’t trust Mike to keep anything secret.
CW: Are you aware of the latest events concerning what is now the Coleco Chameleon?
JC: I hadn’t kept up until recently hearing that the Coleco Chameleon was to be shown at the Toy Fair, the Ars Technica article in which Mike blamed me for his woes, the pictures and videos from the show (of what appeared to be a SNES misrepresented as a prototype for a $150 system with an FPGA), and that the Kickstarter campaign got pulled the day it was to go live. Is there more?
CW: [I share this link with Mr. Carlsen] [RGM is down at this point so I shared the main pictures- below- with Mr. Carlsen]
Mr. Kennedy has apparently decided to try and pass a DVR capture card as the latest prototype for the Chameleon.
JC: After reading Mike’s comments about me in the Ars Technica article, I expect that he will have a hard time finding another engineer willing to work with him.
Seeing that DVR card made me think back to when Mike was having 3-D renderings made of the clear and translucent cases. He had the artist use an image of the Jaguar PCB assembly without its RF shield. I told Mike not to use those images, and protested that to do so would fraudulently misrepresent what the final machine would look like. Among other things, there was a significant chance that I would need to include a similar low-cost RF shield, which wouldn’t look very good.
CW: I have been following leads on his “current hardware guy” and so far it is ending with this guy being “mysterious” and only known by the name Lee or Li.
JC: I don’t know who that is, but may have heard the name in passing. I find it interesting how Mike lists so many people on his Web site as being on his team, but appears to have no engineers.
CW: Is there anything you would like to say to the community?
JC: As for what I’d like to share with the community:
I’m constantly impressed with the level of attention that this project gets. I think it clearly indicates that there are many people who care about video games, and that many are tired of being robbed of their time and money through Internet-dependent consoles that require software updates before we can play and then try to upsell downloadable content. We shouldn’t need a network connection to enjoy video games.
At this point, I’m also sorry to see that the project keeps getting attention. It no longer has anything to offer. If you want to want to play games from cartridges, there are plenty of great used consoles and games that still work.
CW: Yeah, the staff situation is quite glaring. I know he has outsourced “development” and is simply running what that guy/guys tell and show him.
JC: I can hardly blame the developer. At one point, I had to tell Mike and Steve that I would create a design that met their requirements and quote them a cost, but share no technical details because I didn’t trust Mike not to publicly disclose everything and blow my chance to patent my work.
Have you made mention of the launch day cancellation of the Kickstarter campaign? As you probably recall, the announcement was “We’re delaying the Kickstarter for the Coleco Chameleon to make it even better!”
The Kickstarter rules at https://www.kickstarter.com/rules/prototypes say:
“Projects that involve the development of physical products must feature explicit demos of working prototypes. While you can run a project focused on the creation of a prototype, you can’t offer the product that is under development as a reward.”
[Prototypes & Renderings — Kickstarter
When it comes to projects that manufacture and distribute hardware, gadgets, and other products, it’s important to make sure backers know what stage of development the project is in.]
When I was on the project, someone outside our team had apparently pointed this out to Mike, who had claimed (to me, at least) to be a Kickstarter expert. Clearly Mike hadn’t done his due diligence.
Still, via email, Luke Crane at Kickstarter appeared willing to give us a pass. However, in an email message on September 9, Mike flat-out lied about us having a “working PCB”. What Mike wrote was: “We have our console and cartridge shell injection mold tooling and real samples, real controller prototype (Ouya showed a block of wood) and a working prototype PCB.” I knew that we didn’t have a “working PCB” and I called him on it.
Mike shared an email message from Luke that read:
“Ouya came through during a very different time on Kickstarter. We’re constantly updating our rules and procedures. Under our current rules, you can show illustrations and wireframes and animations, but no photorealistic renderings. Sounds like you’ll have no trouble meeting those requirements.”
I told Mike that Luke’s “pass” wasn’t valid because it was based on a statement from Mike that was untrue.
I expect that Kickstarter received complaints about Mike’s apparent fraud before his campaign was to go live.
I’m glad the campaign got pulled. Based on what I had seen online of Mike’s demonstrations and promises, and of course what I had learned about the project, I likely would have been compelled to report it myself.
CW: That is interesting information on the FPGA situation and the situation with Kickstarter when you were involved. Certainly not what Mike said [in other interviews].
Yes, I covered the Kickstarter redaction. It was the article that got my site pulled for too many resources used on the server. I am working to get a decent host that can handle these upticks in traffic
There were more than a few people that publicly stated they were alerting Kickstarter to the situation leading up to when they were to go live.
I do want to tell you, thank you for taking the time to talk with me about the situation as you know it.
JC: You’re welcome. It’s my pleasure. Thanks also for bringing me up to speed on what has been going on in the months since I left the project.
This interview is cross posted with our friends over at Retro Gaming Roundup. This was done in an effort to reach as many people as possible with the valuable information that Mr. Carlsen shared with us. Mr. Carlsen was aware that the interview would be cross posted.