Newsdee

At this stage a Raspberry PI + a Retrode hacked together seems more viable and honest than a SNES. But then that would be copying the idea of Dragonbox's RetroPi (http://retropi.de/ - not to be confused with RetropiE)

My thoughts is that it's surreal. You can't look away and want to see what happens next...

To avoid going out of topic, I'd suggest we move FPGA discussion to this thread: http://atariage.com/forums/topic/248340-the-mist-fpga/? (or any other FPGA thread if it exists)

Chaos has just released a new version of the Amiga AGA core (1.2). More info and download link at his webpage: http://somuch.guru/minimig/minimig-mist/

60 games for $120-$160 seems super stiff compared to the Flashback offerings. Let's see what else the machine has to justify that...

Whatever happens it's going to be interesting. I presume KS requires them to give some details on how the money will be spent?

Going back on topic, it looks like the Coleco guys still believe hardware is easy or will sort itself out somehow. If you take that perspective, building a "prototype" based on the shells makes more sense because what they really want to show is their molds are viable and allow people to experience cart gaming with their "custom" look and feel. Obviously you could have done the same with any console but I kind of feel that's the logic underneath choices made. It's completely backwards to what most people understand as prototypes though (you'd expect hardware with an ugly shell). "Coleco Chameleon: carts before horses since 2015"

This is why I think open source hardware is rather a big deal. FPGA chips are vendor-specific and they provide the compatibility layer (i.e. they guarantee same code runs in a newer/bigger FPGA from the same vendor), so assuming the PCB outside the FPGA is identical (possible as some projects open their schematics) adapting is not that difficult compared to logic analyzing/replicating an existing ASIC. From what I have seen in open source HDL code, it is considered good form to keep it modular with e.g. whole real chips reimplemented as their own module. This allows to swap them out if you wanted to try somebpdy else's implementation of it, which will compile as long as it uses pure verilog/vhdl. For example the Arcade Replay and MiST projects both collaborate to improve the T68 core from opencores (68K reimplementation). In theory if you wanted to do a project using a 68K CPU you could take that and wire it up to other components, saving you considerable time. This is kind of getting out of topic, but it truly is a very interesting field and it looks like something new will come out of it. But it won't be the same as emulation... more like augmented hardware (e.g. native HDMI out like Kevtris has added to the NES). The development tools are free, unfortunately it's tough if you don't have EE background... I've tried to tweak HDL code in cores and failed miserably (other than trivial changes to e.g. activate palettes on an Apple II core).

It's a little better than that... what's missing is compatibility but many game already work fine in many cores. It does benefit from emulators documenting a lot of weird quirks from different games; what's mostly missing is interested developers to work on open source. Here's the beta Gameboy core of the MiST for example, it has some sound issues but many early games work already:

It will be interesting to see if they include the extra audio hardware from the Famicom. Even if not... it could be patched to have it. My sentiments exactly! It's no magic technology by any means, but there are good projects out there and it's more advanced than people think. (e.g. check foft's thread on AA about his Atari 800 cores, it plays all games as far as I know: http://atariage.com/forums/topic/213827-potential-new-hardware/). Are there any good games to test that have identical attract modes? I could record some footage comparing NES and PC Engine between the real machine, a MiST, and the Retro Freak.

I have about 74 famicom games dumped, so it might be related to system. It dumps a file called "plesam.nes" among the ROMs. The game is very simple: you have 10 seconds to push buttons to make a little blob advance...

Here's the thing with FPGAs, once the code for one is mature it can be turned into an ASIC and sold at low cost. It is also much more power efficient to run a hardware clone than an accurate emulator, as the FPGA chips often need less power than the original machine. The reason I like FPGAs (especially open source efforts) is that they lead to something new: fully compatible clone hardware and possibly extensions to the original. The Everdrives and CD drive replacements done for the Dreamcast and Saturn are examples of things made possible using this technology. A fully-compatible NES clone would allow building new NES or Famicom (compatible with carts and identical in all ways to the original) without cannibalizing older machines. There is no project right now promising that. My point is that the current efforts lead there by having people take over and expand previous projects. Open source has been running for years in software, but I understand it's newer for hardware. Current offerings such as the MiST combine an ARM with a tiny firmware to handle things like SD card load and on-screen menus. HDL projects like the Minimig (runs on MiST but also other boards) also offer a wide range of configuration options that rival good emulators, but this is limited to mature cores aiming faithful reimplementation (rather than just play games).

Fun fact: I was blown away when I learned (not that long ago) that the Nasir that did a bunch of Apple II games I played as a kid is the same Nasir Gebelli that worked on the Final Fantasy series...

The code for these cores is open source, so as long as they respect the license terms it's fair game.There are also open cores for NES and other consoles out there. The problem is that the approach of these FPGA boards is to be completely open for developers, which the CC definitely won't be (or "Coleco" will lose their monopoly on cartridge manufacturing for the machine). Here's a good video for anybody wanting to know more about the MiST or FPGAs in general:

That wasn't targeted to you in particular so sorry if it came out that way. I'm anticipating that if (when?) things go pearshape they will try to victimize themselves personally to make people feel sorry for them, but this is inconsistent with the image they themselves are pushing forward right now in order to get people to fund their (vague) ideas. As somebody else said in this or the older thread, "for 2 million anybody can bring cartridges back"

Don't make this personal; this is a company selling a product that does not yet exist. I think it's fair to say that COLECO has no clue about hardware and the team they assembled to build one is apparently lacking the expertise to do so (in truth COLECO is a licensed name, but I'm making it consistent with the image they want to project as the original company). It's as if Nintendo were pitching a new console but they laid off all their hardware and game designers, and just kept their manufacturing guy and advertising manager, and then they put up ads saying "Help-a me builda my nexto console" with Mario wielding a Virtual Boy. The usual tactic is betting on there being enough passionate geeks out there that might be willing to do work for them for chump change while they reap the rewards for "having the idea" and owning the brand name. And I'm not blaming one guy either - this is a classic corporate trick which does work but also ends up parodied in Dilbert. So let's not talk about nostalgia of how old things were, but look at what is in front of us. If we look at it as a business the whole thing is shaky from lack of clear model. if we look at it from a hardware point of view it's vaporware so far. IMHO this is an ad company trying to get support/captivate an audience so they can monetize it in a way that they either don't know or haven't shared publicly. Or maybe they're betting their margins on manufacturing will be enough to make the KS/IGG campaign profitable in itself.

Interesting they mention having an Amiga core. The code of Minimig (FPGA Amiga) is open source under GPL3, so as long as they respect the license terms I suppose they can use it for free. I'm not aware of any open source INTV cores though.

They have 4 games at launch so that's 64 bits. Soon it will be 128 bits and then the sky's the limit!!!!!!1

That's my point: the bonus minigame is a Famicom game and was the first thing that unlocked for me. So although any 100 carts seem to work for the credits, I think this minigame has a more specific unlock condition. Its nothing special as a game (and very short) but it's neat to have that in there.

Pffft no cartridges. I'd like to see one in person, the quality of the LCD screen makes or breaks handheld systems, in my personal tastes...

I can confirm the credits is at 100 carts. I don't remember which one unlocked the small NES game but it was much less, like maybe 25-30. Maybe they need to be Famicom games...

Attack the claims and maybe ideas, not the people I think many of us see some value in what's going on but strongly disagree in the way it's being carried forward. This won't be the first nor last project of this sort regardless of whether the KS succeeds or not, so just bring out the popcorn and wait...

On second thought I think you're right, because if you add the licensing of 3rd party cores (assuming you couldn't package open source ones republishing modified code) and stuff like advertising etc. you end up with a higher total cost. Unless you have enough volume to save significantly in manufacturing.

I think they're faking the prototype, but looking at the price of the MiST on dragonbox.de, its around 188 USD. It does not run SNES yet but (in theory) it has the FPGA for it. Somebody (chaos) pulled off Amiga AGA, so a SNES seems possible at that price point, barring greed adding a huge margin to it. Granted it doesn't have cartridge ports so maybe we should add the price of a Retrode (60 bucks). That'sa bit over 200 but I'm sure you could save on the case and component costs if they're wired together.

Torlus (Greg Estrade) has also written a fantastic PC Engine FPGA core which is open-source and has been ported to the TC64 and the MiST (with improvements) by AMR (Alistair Robertson). Runs most games with sound except the larger ones and the tiny supergrafx library. I like this way of doing things; if that core was made/funded by a corporation it surely wouldn't be open and any extensions would be unlikely.