How has this not been posted yet? Retro VGS

[triverse]

Just a heads up, I have been told by a game developer for the Coleco Chameleon that they are to stop talking to the press about the CC. This happened in mid interview, concerning the CC, which ended rather abruptly due to this. Seems interesting they are trying to stop anyone from finding anything out about this thing when you think they would be touting the features, new games, new devs (I broke the news of a new one myself) and other things that actually matter. Instead, they are fighting people that are keeping the lights on for this thing. Interesting.

[jaybird3rd]

Here's a new article from someone who spoke to Mike recently: Coleco Chameleon continues quixotic quest for contemporary cartridge console

Thanks for the link! Very interesting article. There's lots to say about it, but one part in particular stood out:

 

That brings up the specific storage capacity for those cartridges, which is still up in the air. While Retro VGS was originally talking about games up to 1GB, that has been reigned in to focus closer to a 100MB range these days, Kennedy says. That should be plenty of space for 16-bit-style games—SNES titles topped out around 48Mbits in the '90s. But it could be a tough limitation for games built-in inefficient environments like Game Maker and Unity, which can add a lot of bloat to even simple games.

 

Then there's the physical medium used for game storage, which is also still being debated. Adding extra storage capacity isn't tough if you use cheap SD-card storage, but untouched data on those cards lasts only a couple of years before degrading. Kennedy says he's looking for much longer retention times for Chameleon cartridges, which should be able to last as collector's items for decades.

With all the bulky assets created by modern game design tools—JPEG or PNG graphics, MP3 or WAV audio, and so forth, not to mention the code itself—a 100MB cartridge is going to fill up FAST, even with a relatively simple 2D game.

 

If they're anywhere near being ready to crowdfund a prototype, shouldn't they already know what the maximum capacity of the cartridge will be, and what kind of storage it will use internally? I don't know how you'd even decide on the pinout for the cartridge port unless you've already nailed down those specs, and until you do that, you can't choose a suitable cartridge connector or begin work on your board layout.

 

There are some broader questions about exactly how the cartridge will work in this system, but I'll get into those below ...

[Flojomojo]

It's refreshing to read some honest looking words from Mike K, and I appreciate that. That sort of talk needs to happen early and often. We will see if any substance comes out soon.

 

Is it significantly better than a SNES or Genesis or Retropie? I haven't yet seen anything to make it seem remotely equal to any of those fine things.

[108 Stars]

Just a heads up, I have been told by a game developer for the Coleco Chameleon that they are to stop talking to the press about the CC. This happened in mid interview, concerning the CC, which ended rather abruptly due to this. Seems interesting they are trying to stop anyone from finding anything out about this thing when you think they would be touting the features, new games, new devs (I broke the news of a new one myself) and other things that actually matter. Instead, they are fighting people that are keeping the lights on for this thing. Interesting.

Not totally surprising imo. They want to prevent the shitstorm from spreading, at this point it would be hard to get any positive reactions even if something cool was shown. So they will just wait for the storm to clear instead f delivering much info that can be dragged through the mud, and hope to present their product with a bang at that toy fair (I know, it is the wrong venue for that). They want to dry the well of criticism I think, so to say, until the day they believe will be their triumph.

[godslabrat]

Just a heads up, I have been told by a game developer for the Coleco Chameleon that they are to stop talking to the press about the CC. This happened in mid interview, concerning the CC, which ended rather abruptly due to this. Seems interesting they are trying to stop anyone from finding anything out about this thing when you think they would be touting the features, new games, new devs (I broke the news of a new one myself) and other things that actually matter. Instead, they are fighting people that are keeping the lights on for this thing. Interesting.

Nice drunken commentary ya got there.

[jaybird3rd]

The Ars Technica article also has me thinking about some of the technical problems involved in delivering on the promise of a modern cartridge-based system that indie developers will be able to easily port their games to from other systems. This is a little long, so apologies in advance, but here it is anyway ...

 

The article raises the possibility of using SD card storage as a cheaper alternative to program cartridges. This is an idea that I've seen raised before: "Why don't they just use SD cards instead of cartridges? It's all solid-state, so it must be the same thing, and SD cards are cheaper!" At the risk of stating the obvious, that's not how real program cartridges work: they're solid-state, yes, but they function very differently from SD cards.

 

An SD card, like a floppy disk or an optical disc, is a storage medium, so any game stored on it must be copied into memory (RAM) before it can be executed. This is why non-cartridge systems experience loading times; these delays may be shorter for SD cards than they are for discs, but they're still there. Real program cartridges—at least the ones for classic systems that the RVGS/Chameleon is meant to remind us of, like the SNES and Genesis—contained PROM or EPROM chips, which are already a form of memory. When the cartridge is made, a complete executable image of the game is burned into the chips, and they are automatically "mapped" by the system to a specific location of memory, just like the RAM. There's no need to load the program first, because it's already available to the system the instant you turn it on. Usually, some very simple firmware performed a few startup and initialization tasks, but then it jumped directly into the cartridge. Hence, zero loading times. (Sometimes, the game data needs to be copied or unpacked into RAM before it can be used, but that's a slightly different issue.)

 

If the RVGS/Chameleon team wants to create a system based on real program cartridges, as they've repeatedly said, it seems to me that they've got some difficult design decisions to make. They need to design a system that can be bootstrapped almost instantly from a cold start, and that can execute code directly from the cartridge, without any intermediary loading process. Unfortunately, that's not how modern off-the-shelf operating system architectures like Android, or games created with the likes of Game Maker or Unity, are designed to work.

 

So, how exactly do they intend to deliver on their promise of a cartridge-based system, while still allowing indie developers to easily port their games using the platforms and tools they're already familiar with? They can't take a truly "retro" approach and design a console with no onboard OS at all, because that would mean coding directly on bare metal (or close to it), which means that prospective publishers would have a whole new platform and toolchain to learn, and nobody's going to go to all that trouble for such an obscure platform as this. Perhaps they could bundle an embedded OS image together with the games in every cartridge, but then they'd need something a lot bigger than the 100MB they're reportedly targeting, and there would be significant loading times.

 

Since they're under pressure to get something working quickly and cheaply, I suspect they'll end up cobbling something together out of off-the-shelf technologies, which means that there will be an OS in the console (probably some stock flavor of Android), and their supposed "cartridges" will be nothing but ordinary SD cards or Flash memories in disguise. That goes against their stated goal of "zero load times!" and "long-term storage!", and also takes away the one thing that would have made their system unique. If their system runs the same OS as everyone else's, and if it uses the same cheap Flash chips as everyone else's, then who cares if those chips are soldered onto the motherboard or plugged into a cartridge slot? By moving the program storage into the cartridge, without changing anything else about the architecture, all they've added is another point of failure (the cartridge port) and another expense for the users (having to buy multiple cartridges for multiple games), for no tangible benefit.

[triverse]

Well, the latest article is now live. This one discusses the Arstechnica interview, points out some flaws in Mr. Kennedy's logic and gives a bit more on the 3rd party developer that was told to stop talk with us.

 

http://retrogamingmagazine.com/2015/12/22/mike-kennedy-breaks-silence-about-coleco-chameleon-while-3rd-party-developer-stops-talking-to-press/

[some_canuck]

i wouldn't be surprised if it was gamester who told the guy to stop the interview, because he also works for collectorvision.

[triverse]

i wouldn't be surprised if it was gamester who told the guy to stop the interview, because he also works for collectorvision.

I was talking with him in a separate chat window at this time also. I am not sure who told Jean-Francois to stop talking though. John Lester didn't make mention of it, only Jean-Francois. Mr. Lester and myself were discussing discrepancies in my previous article (I mistakenly said Tiny Knight was guaranteed as coming and Sydney Hunter was not when it was vice versa).

[Keatah]

Many SD cards and similar Flash-types of memory contain warnings on their packaging stating short-term usage only.

 

Consider the Golden Age consoles. Zero load time was a product of the technical limitations of the day. Early consoles were essentially simple computers in which you removed and replaced part of the circuitry.

 

I disagree though that cartridges serve no benefit. They indeed do. There's freaks and junkies that like the smell of fresh plastic and holding petrol distillate containers in the flesh.

 

And it's the proliferation of crap games that dilutes a cartridge collection. Soon you end up with shells of plastic with meaningless games that you never play but have to have for completion purposes. Soon you have a whole wall of stuff, of excess baggage. There's maybe 5 good games out of a hundred.

 

I'm not interested in getting back to that state of affairs.

[bretthorror]

Will games like Shovel Knight even be able to run on the system at 100MB? These guys, I just don't know. Maybe because I use Steam and actually know a bit what I'm talking about unlike a lot in regards to these digital games, but I just bought one on the Steam sale called Life of Pixel that'd be perfect for the console and it's 90MB. I don't want to see games dumbed down for their console (and if they have to be altered at all, what is the likelihood of people taking the time to do so?). And I don't want just the lower tier of the indie world to appear on it, I'd like to see the legit good retro styled games as well. With all due respect to the people who made these games, I'm not saying they're not a lot of fun or don't deserve to be played, but frankly the games Mike is looking to attract probably don't really deserve/require a physical release in 2015. With the 100 MB limit, you're gonna get the shaft because the best of the best are likely too big to play on the Chameleon and purists aren't going to get their legit retro experience where a NES style game is 16kb or whatever they are. So you're just getting the dregs off the upper echelon of indie Steam titles. Let's not mention a lot of these retro Steam games have online co-op, DLC, etc. you also will not get. Axiom Verge is too big for this system - and that game is the closest thing it'd have to a killer app in terms of retro gameplay.

 

I hate to go back to it, but all this Goddamn thing has is "cartridges". It doesn't even have games because no one wants a watered down port. This thing is gonna run N64 tops, and music on the system is gonna suffer because of it. The best things about modern retro games is they use the retro style, but limitations be damned, they can put whatever colors/graphics/sprites/music they want to. And that's what makes them amazing. And it'll be that kind of thing that'll make Steam games look like Genesis and Chameleon like Game Boy in the old Sega Does What Nintendon't commercials. If only people with half a brain were in the know, they'd have launched this thing as a successor to the retro systems of old that maintains its style, yet bolsters more power. That concept is at least forward thinking and logical. Axiom Verge is the perfect retro title, but the retro systems can't run it. But, the Chameleon can. It's a fine line, but that perception of that way of thinking for a new, retro console is a lot more palatable to me. Instead they focused on sending us to the past and leaving us there to die.

 

And just so everyone here knows, a game that was included in the IndieGoGo, Adventure in the Tower of Flight, is on sale on Steam right now for $1.50. Why pay $1.50 when you can pay $40... at the special backer rate price?

[jaybird3rd]

Many SD cards and similar Flash-types of memory contain warnings on their packaging stating short-term usage only.

 

Consider the Golden Age consoles. Zero load time was a product of the technical limitations of the day. Early consoles were essentially simple computers in which you removed and replaced part of the circuitry.

Exactly. At the time, there was no affordable storage medium that the games could be loaded into memory from (tapes and floppy disks either weren't available or would have greatly increased the cost, depending on what year you're talking about), so they simply preprogrammed the memory at the factory and put it inside a plastic cartridge, and added an inexpensive edge connector and slot so the cartridge could plug in to the bus. That's all. But this is obviously not the way that modern OSes are designed to load software, so how do they intend to make a system that can satisfy both goals ("zero loading time" and "easy indie ports from other modern platforms")? They can provide a simulacrum of the cartridge experience with SD cards and the like, but again, those are NOT real program cartridges of the kind they've been promising, and are not meaningfully different from what everyone already has.

 

I disagree though that cartridges serve no benefit. They indeed do. There's freaks and junkies that like the smell of fresh plastic and holding petrol distillate containers in the flesh.

If that's all they want, they should just start shipping labeled (empty) cartridge shells in pretty colors that the fetishists can rub all over their bodies if they want to ... as long as they don't tell me about it!

 

Don't get me wrong: I still use cartridges all the time and I love the format, but they're not going to be able to make cartridges their primary medium without designing a whole new system around them, one that bears little resemblance to the modern OS architectures that lazy Game Maker and Unity indie developers are used to working with. They're going to have to choose: do they want a system that can attract developers, or do they want one that uses cartridges?

[Shawn]

$250 to $300k sounds a lot more reasonable than $2 to $4 million but I'm still very skeptical that it is an amount of money people will trust the Nostalgia S.H.I.T team with.

[jaybird3rd]

A few reader comments from the Ars Technica article:

 

Even if there were a viable market for the Coleco Chameleon, and it could be produced at a reasonable price, the cluelessness surrounding the original RetroVGS announcement clearly indicates that these chuckleheads are the last people capable of actually making it happen.

You get one chance to make a first impression, and they blew it.

 

I'm not buying this being entirely one person's fault. They talk a good spiel about the previous attempt being a great learning experience, and how they "found out" that they built a machine nobody wanted to pay for. They try to paint it as being all this Carlsen guy doing something dumb, while only briefly 'taking responsibility for letting him do it'.

 

Well, there's the rub, isn't it? The rest of the company bought into charging Xbox One pricing for SNES technology. The price point should have been a no-go decision right there. Why wasn't it?

 

Potential backers for this thing really need a good answer to this question. What killed Ouya? Lack of developer support. Do I see these guys getting a lot of developer support? No. Devs will look at their first attempt, which makes it look like they have no idea what they're doing. And this explanation won't satisfy their concerns. I can't see many developers wanting to bet the future of their company on content aimed primarily at the Chameleon. And without that, it's toast.

This is going to be the dumbest thing on the market, you are basically paying an increased price for steam 3rd party games on a different physical media. I'm sure a few hipsters that want to look like they are into retro without having any kind of retro experience itself will buy them but good lord how did this thing get backed?

It didn't get backed, the original Kickstarter was an enormous failure. They are going to try again with more sane goals.

 

This is a company that didn't even have a proper prototype for the first try. They got the jaguar molds first, came up with an idea and shoved it onto kickstarter with no real plan on how to even produce the thing. Everything about the project makes it seem like it's run by idealistic idiots.

[108 Stars]

The best things about modern retro games is they use the retro style, but limitations be damned, they can put whatever colors/graphics/sprites/music they want to. And that's what makes them amazing.

I don't even want that. Fake retro aesthetics are really a trend I find annoying personally. I want a system to game to get the best out of a hardware. It is a great exception for me to buy a game like that, I only do it if the gameplay is really so great I can look beyond the visuals.

 

The thing is, Chameleon can never deliver in that regard. Ports of games made in Unity etc are probably impossible on tiny carts like that. Actually it is getting more and more difficult for devs to port games even to the Vita, because the retro graphics are just a facade. Yes, you could do games that look very much like them on old hardware; you could do them on the Vita and the Chameleon as well. But the high level languages used, the ressource demanding engines, they are not compatible. You'd have to code again specifically for the hardware like jabird3 said.

 

What Chameleon will be left with is the few games that maybe enthusiasts will actually code specifically for the system, and (if that is still a thing which I doubt) the homebrews developed for MD or SNES that could also run on the Chameleon. But why a Chameleon for that? And I doubt we will ever see games actually pushing the hardware of the Chameleon.

[Newsdee]

 

I'm assuming that this system will also support Game Gear, SG-1000, and Mark III ROMs (...)

 

I'll be curious if the Xbox 360 wireless adapter (or Xbox One for that matter) works with it.

 

I have one of these consoles. SMS/MKIII ROMs are recognized from the SD card; the adapter is only if you want to run/dump original carts.

 

It does not run flash carts, but it would be redundant since you can dump roms in the SD card.

 

Xbox 360 controllers do not work but I'm hoping they will release a patch for it (they release firmware updates from time to time).

 

Also you can switch the menu to English. The XML file for cheats and the webpage for firmware updates are still in Japanese but you can fiddle with a translator to figure things out.

 

Edit: the adapter for legacy controllers is a bit pricey but is very good. It works on Windows and each controller comes up as a separate device, meaning you can use it as a "multitap" for several players.

Edited December 23, 2015 by Newsdee

[bretthorror]

I don't even want that. Fake retro aesthetics are really a trend I find annoying personally. I want a system to game to get the best out of a hardware. It is a great exception for me to buy a game like that, I only do it if the gameplay is really so great I can look beyond the visuals.

 

So was Sin City a fake 30s movie to you because it was partially black & white? It's called style, mon frere.

 

I had more, but my head exploded and I quit out of not being able to even comprehend hating on style... like the guy who invented the first color of paint made everyone else stick with one color because they wanted to get the most out of that color and see how far it could go.

[108 Stars]

You don't really want to discuss "style" now, do you?

If you like that stuff, that's great. For me it is just a relatively cheap way to get around investing time to do more detailed graphics. I'd say that nostalgia and style for the 16-bit systems is as big as for 8-bit, yet few go through the trouble of recreating that "style" for their games. Imo that is because it is substantially more work. For every one who does this truly because it is style he likes I bet 10 others do it because it is easy.

 

If I want 8-bit style, I play games on 8-bit systems. That makes 8-bit games or any other generation games awesome to me, when they make the most of their limitations.

Edited December 23, 2015 by 108 Stars

[Raticon]

I kind of like the retro-pixely style when done right, and the gameplay merge with it and i have no problem with that.

 

Thing is, it is very easy to see when its done out of love, true enthusiasm and with attention and for me it shows directly. "Papers Please" is a game that comes immediately to mind for me that ticks all my boxes in that regard.

 

I do, however, have grown to despise and loathe when developers use that 8 or 16 bit "retro" style just "because retro", or just to sell more shit because pixels are the shit these day, or when developers are too lazy to make good textures or sprites and just make them a pixelated mess and call it "retro". A whole slew of low tier indie titles and flash games come to mind.

 

As many excellent posters above remarked, it is still a big question of how the carts will work and frankly without more elaboration from the devs we will have to assume they will either go for some stupid flash solution making your carts into plastic bricks in a few years when all the ones and zeros on them has degraded, thus invalidating the collecting point or go for a whole own type of hardware thst will be a bitch to work on and discouraging devs from making ports when the financial end doesnt justify the added work anymore.

[InnovaX5]

Makes sense that Collectorvision / Acclaim was told to stop talking about it. Seems like they are the only ones bound by a contract. Which is sad, John Lester is a good man and I've never been disappointed with a Collectorvision purchase that I made.

 

John's interview with Mike Kennedy about the RVGS on John's Gamester81 channel makes even MORE sense now, too. This was of course before Triverse's atomic bomb of truth interview with the team.

 

I swear that this whole RVGS / CC ordeal like the best B-Moive script ever. We just need Gary Busey to play Mike Kennedy and slap this thing on Lifetime. Give up on the console mike, just sell the movie rights and move on.

[Raticon]

Even better, make it a series on HBO. "Halt and catch fire" doesn't got shit on this silicon-infused drama.

 

"Retro Hard: with a cartridge", except the hero-console is dead on arrival instead of just hungover.

 

 

I will fix the Beatles licence so the theme can be "Mike in the sky with cartridges".

[+Random Terrain]

A few reader comments from the Ars Technica article:

 

I'm sure a few hipsters that want to look like they are into retro without having any kind of retro experience itself will buy them . . .

 

Springtime for hipsters in burgundy.

 

 

 

 

 

 

[Andromeda Stardust]

Thanks for the link! Very interesting article. There's lots to say about it, but one part in particular stood out:

 

With all the bulky assets created by modern game design tools—JPEG or PNG graphics, MP3 or WAV audio, and so forth, not to mention the code itself—a 100MB cartridge is going to fill up FAST, even with a relatively simple 2D game.

 

If they're anywhere near being ready to crowdfund a prototype, shouldn't they already know what the maximum capacity of the cartridge will be, and what kind of storage it will use internally? I don't know how you'd even decide on the pinout for the cartridge port unless you've already nailed down those specs, and until you do that, you can't choose a suitable cartridge connector or begin work on your board layout.

 

There are some broader questions about exactly how the cartridge will work in this system, but I'll get into those below ...

Capacity does not rely on bus pinout like the old days. You don't need to add extra pins for extra bits of address space. USB uses GND, VCC, Data In, and Data Out for instance. A mass storage device could be on megabyte or ten terabytes. Maybe a few extra pins for timing or other use. A parallel bus could speed data transfer, but with flash tech and serial bus, you don't have any artificial limit on storage like old school EP or Mask ROM.

 

EDIT: Excellent points people have brought up regarding flash storage. I imagine the solution will be obvious: The Chameleon will be a Raspberry Pi type device with the SD slot being the cartridge bay. Carts will have 7-9 pins matching SD tech or using some proprietary equivalent. The flash chip controller will be designed in such a way as to protect the majority of storage space from overwrites. The remaining space could be used to update the program or store save data. I know Mike said zero updates, but come on, every indie game out there gets updates and DLC, or bug fixes. They are shooting themselves in the foot to not allow an update mechanism. Updates could be applied through a USB flash drive plugged into the system or something, assuming there's no network connectivity.

[StopDrop&Retro]

If they keep up the silence they should get a nickname for it. I suggest "North Coleca"

[jaybird3rd]

Capacity does not rely on bus pinout like the old days. You don't need to add extra pins for extra bits of address space. USB uses GND, VCC, Data In, and Data Out for instance. A mass storage device could be on megabyte or ten terabytes. Maybe a few extra pins for timing or other use. A parallel bus could speed data transfer, but with flash tech and serial bus, you don't have any artificial limit on storage like old school EP or Mask ROM.

True, and that would simplify the cartridge interface quite a bit. But can you execute code directly from a memory device on a serial bus, or does it have to be loaded into RAM? My understanding was that if you have (for example) an SD card with a large binary on it, that binary can't run in place from the SD card; it has to be copied into RAM first, which means that there's going to be some loading times involved.