FPGA Based Videogame System

[Keatah]

If you make a bare-board option I'd mount it in tandem with my mini-ITX emulation machine. Could have the best of both worlds in one box. I could sneak it in without the wife ever knowing about it! And if she caught me I'd make up some bullshit that its a patch board because something broke on the mainboard. Or it's an expansion kit.

Edited September 21, 2015 by Keatah

[tcdev]

I could sneak it in without the wife ever knowing about it! And if she caught me I'd make up some bullshit that its a patch board because something broke on the mainboard.

 

To which she'd reply, "If that's so, then why is there no high speed circuitry on your so-called 'patch board'? There can't be anything on there running more than about 200MHz!"

[univbee]

This is sounding pretty close to a dream machine already and I would absolutely pay handsomely if the compatibility is that high. In a perfect world I'd like analog connectivity as an option, but without a doubt HDMI is the priority and if including analog as an option pushes the price outside of most people's ranges and should be removed entirely, do it.

[Keatah]

Would there be any support for different aspect ratios like 4:3, 5:4, 16:9, and 16:10? A way to force them or manually adjust them via slider? Same thing with color palettes. All part of the infrastructure and housekeeping stuff.

[SIO2]

Whatever you come up with is fine. I am more of McDonalds give me a number whatever than a Subway - do you want the black or green olives, spinach or lettuce, green peppers, hot peppers, white or wheat seasoned or plain cheese..... Argg, just give me flipping sandwich already - type of guy.

[+Gemintronic]

I've been waiting for the FPGA Arcade to become readily available for years. Still want one. I think the concept of an FPGA machine with real cart slots is awesome. Real hardware without wearing out our ancient original machines.

[Greg2600]

Honest question, beyond the "cool factor," what does one gain (emulation wise) from this type of system over an Android-base or modded Xbox or modded Wii, etc.? I suppose using original controllers, but that's only going to be realistic for a few systems.

[+Gemintronic]

Honest question, beyond the "cool factor," what does one gain (emulation wise) from this type of system over an Android-base or modded Xbox or modded Wii, etc.? I suppose using original controllers, but that's only going to be realistic for a few systems.

 

It's not emulation. The system actually becomes a hardware clone. So, compare playing an NES game on an Android machine and a real Famiclone/NES.

[Greg2600]

 

It's not emulation. The system actually becomes a hardware clone. So, compare playing an NES game on an Android machine and a real Famiclone/NES.

 

So it would be (using an SD interface) compatible with essentially EVERY game that was developed for the systems covered? Although most emulation already is.

Edited September 21, 2015 by Greg2600

[Keatah]

Maybe someone could address the timing and consistency advantages. I need to blow outta here.

[Draikar]

No thanks, I think I will just get the retron 5 but keep up the good work. Just don't use a ugly shell hackers use but real plastic, like the Neo Geo omega shell but It will cost money. Many be a aluminum shell like Analogue Nt ? If you don't think about that kind of stuff then your just doing like every other hack, selling a $500 board and cheap shell that a retron 5 dose or even better the real system like Atari or Sega dose beacuse they were made to play the games made for them by the company that made the system in the first place ???

[pixelplatforms]

i literally signed up for an account so i could vote on this. i was on board for the retro vgs, but have since retracted my contribution after reading through all these pages on atari age. i will 100% support kevtris, and i know a lot of people would as well! let's make this happen!

[+Gemintronic]

 

So it would be (using an SD interface) compatible with essentially EVERY game that was developed for the systems covered? Although most emulation already is.

 

Same advantages and disadvantages of real hardware systems. So, you get the feel of real hardware and you could potentially use real hardware accessories. The disadvantage is like that of any clone system. Some weird incompatibilities will pop up from time to time. My Retron 3 disliked flash carts but the Retro Trio works with all but my sega genesis everdrive.

[Bill Loguidice]

I love the idea and would pay quite a bit for the convenience of a decked out model, despite having all of the original systems.

 

I almost wonder though if feature-creep is going to be an issue. Perhaps it would make sense to start out with a core system that just has say, HDMI out, SD card slot, DB-9-style controller pins, and maybe a USB or two, get that done, then create a second model down the line that builds off of that with cartridge slot options, additional controller ports, etc. This way both the > < $200 markets are each served.

 

In any case, I'll be interested in seeing how this develops. Good luck.

[cimerians]

Awesome. Would love SD/HDMI (save states if possible), would pay $200+ bucks for it cause it sounds like a dream come true.

[toiletunes]

I will buy one. Don't need cart ports or fancy shell colors. Keep it simple.

[Keatah]

Through the magic of software emulation and hardware recreation we can continue to enjoy all the classic games we grew up with. We can even reach new audiences. I much prefer a standard x86 emulation computer and fpga host board combo over all the original systems because of aesthetics and clutter control. Between the two methodologies I think we have comprehensive coverage that can last many years.

 

People complain about inaccuracies. But the advantages and additional features of a rich hosting environment gained outweigh the subtle problems. Not forgetting accumulating piles of consoles and controllers, disk drives and peripherals, cartridges and disk images, adapters and cables.. Pfaggh that's so tedious!

Edited September 21, 2015 by Keatah

[roland p]

No cart ports for me either:

- No parts to wear out. (only if you use them though...)

- No extra costs for the ports

- Smaller casing

- Would it even be possible to have cart ports for every system for a feasible price?

 

For me the main appeal would be the lack of an operating system and accuracy (50hz/60hz without fiddling). I would like to have the 'it just works' experience

[Asbrandt]

I'm also in the "keep it simple" camp regarding hardware features.

 

I'd guess that most who are interested in this are ok with the "feel" of emulation, lacking real carts and in many cases real controllers, but dislike the lag-causing technical limitations, such as the framebuffer and the OS input system, both of which can be overcome but as of now have not been.

 

However, I also understand there is a crowd that strives for accuracy including the video output framerate, even when not using original hardware, such as the GroovyMAME + CRTEmuDriver users. Until AdaptiveSync or similar becomes within your reach, that's only really possible with analog output unless you break HDMI spec, so I would favor the "optional analog addon" even if it only does say 15 or 31khz RGB from a VGA port.

[sunteam]

This thread seems to be weighted heavily in terms of just a clone system for playing ROMs but I'd be much more interested in this suggestion that kevtris floated:

 

A few more thoughts about some kind of new "retro" videogame system I thought I would throw out there:

 

I was thinking the other day if you really wanted to make some kind of "new" retro videogame system, the best approach might be to start with a retro-style video interface, including sprites, background layers, and all the usual trappings of say the SNES, genesis, etc. but the twist is integrating a modern CPU such as an ARM with it. Think 256+ sprites per scanline, any size, 8 background layers with 'mode 7' style rotation/scaling, and with 32 channel 16 bit stereo digital audio.

 

This would allow for retro-style games, done in a more original form but with a modern CPU to make coding it so much easier and faster. The sprite and background engines would be fairly decent, and more capable of either SNES or genesis, while still operating in a similar same manner. That would kind of "enforce" the look and feel of the games while giving it most of the modern day conveniences like fast CPUs and large storage capabilities and being able to write your code in i.e. C.

 

This would be especially great if there was a universal palette for sprites & backgrounds, or perhaps one palette for each. I do a lot of art for the PCE/TG16 and sorting palettes is just one of those headaches that makes development more painful than it should be.

 

 

If there was a new console that was somewhere between the NES and PCE in terms of resolution and colours, but had the extra bells and whistles mentioned and was easy to code for, I would be very interested (much more so than just a clone machine). Maybe it could have several 'modes' so that game devs could release a game for (for example) 'Mode 1' which would look like early 8-bit consoles or 'Mode 2' which looks more like the later 8-bits.

Stick a SID chip in there and you can take all my money.

Edited September 21, 2015 by sunteam

[cimerians]

 

Stick a SID chip in there and you can take all my money.

 

Oh yes. The almighty SID, I'd pay extra

[Pixelboy]

Alright, so here are my random thoughts on the subject:

 

1) The main appeal for me is to be able to play the older 8-bit games on a modern TV, via HDMI, with a perfect picture quality. I want to see the pixels clearly on the screen, with no ugly color bleed or any kind of "wavy" interference which I often see when using the original consoles on old CRT televisions. The systems I'm most interested in are ColecoVision, NES and Game Boy + Game Boy Color (and Game Boy Advance if possible, in the future). The Atari 2600, Sega Master System and Game Gear are interesting too, but only for a few specific games I like to play on those systems.

 

2) Every time I turn off my ColecoVision I wonder if it will turn on properly the next time. If it doesn't, then I have to send it to Yurkie for repairs, and cross my fingers that said repairs are actually possible. That problem would go away with a newer system with newer electronics, assuming the newer machine reproduces the original machine perfectly (and from what I've seen in your YouTube videos, you've certainly got that covered). I'm also not very interested in purchasing an NES because of its infamous cartridge connector problems, even though there are tons of NES games I enjoy playing. The NES toploader is not an option for me because it only offers crappy RF output.

 

3) I'm personally torn on the issue of SD versus cartridge port. On one hand, I absolutely want to use my collection of ColecoVision carts on your proposed FPGA system, so if you end up not offering some kind of ColecoVision cart adaptor, that will be a total deal-breaker for me. On the other hand, given that NES carts are getting quite expensive and that I'm not really interested in collecting carts for that system, I would gladly settle for playing NES ROMs off an SD card (again, as long as the games look and sound the same as they do on the original hardware, inaccurate colors/sounds ruin the experience for me). For Game Boy, I could go either way: The carts are small so I could see myself building a small collection of my favorite GB/GBC games on cartridge and playing them on your FPGA system via a cart adaptor, or else I could play those same games from an SD card. I love playing Game Boy games on a big-screen TV (I bought a Game Cube years ago mainly for the Game Boy Player add-on).

 

4) You should be aware that a good number of the latest homebrew ColecoVision games require Opcode's Super Game Module (which essentially adds RAM and an MSX sound chip to the ColecoVision via the front expansion port). So I can only hope that you will develop a new ColecoVision core that includes SGM support, or else the homebrew games that depend on the SGM will be unplayable on your FPGA system.

 

5) As stated above, the main draw for me are the 8-bit systems, but I would like to be able to play my favorite 16-bit titles on your FPGA console, such as Super Metroid, Mega Man 7, Rockman & Forte, Mega Man X/X2/X3, Zelda - A Link to the Past, Super Castlevania IV, Demon's Crest, Street Fighter II, Super Mario World, Super Mario Kart, (...and on Genesis...) Castlevania Bloodlines, Aladdin, Golden Axe, Gunstar Heroes, Landstalker, Mega Man - The Wily Wars, Truxton, Wonder Boy in Monster World, and many others I'm forgetting. All this to say that if 16-bit cores are in your plans, I would be interested in them.

 

6) I'd like to use this opportunity to add my vote for a Vectrex core. It's not one of my favorite classic systems, but it would be really cool to play those games with anti-aliased vectors on a big TV screen.

 

7) I think there should be two distinct versions of your FPGA system: One HDMI-only (with perhaps some extra audio jacks for those who want to output sound through their big living room sound system) and one with the "older" set of audio/video connectors. The non-HDMI system would be there for those who want it, with the understanding that they would have to pay extra for it. I expect the majority would prefer to have a system designed for HDMI-only, which as you stated would cost less to manufacture.

 

8 ) One feature I would absolutely LOVE to have in the black-&-white Game Boy core is the ability to freeze the game and set custom colors any time I want, exactly like it was done in the Super Game Boy cartridge adaptor on the Super-NES. The numerical "color password" system of the Super Game Boy was a really neat idea, as it allowed me to reuse my favorite color schemes without having to memorize exactly what colors I selected. In the Game Boy core, it would be nice to have a similar password system, or to be able to somehow save color schemes on an SD card for later use.

 

 

With all this said, I would like to request some clarifications:

 

1) The FPGA can only contain/run one core at a time, right? In that case, how does running ROMs from an SD card work? Does the user need to load a core manually into the FPGA before running a certain game, or is there a system in place for updating the FPGA automatically if I jump from ColecoVision to NES to SMS to Atari on a rainy day?

 

2) Certain games cannot be finished in one sitting and require a savegame feature (The Legend of Zelda on NES comes to mind) so does your NES/Famicom and Game Boy cores support saving via an SD card?

 

3) How do controllers work on your FPGA system? I suppose the disc on the Intellivision controller translates to a modern analog controller, while the ColecoVision's keypad extends the need for buttons. How do you plan to deal with those issues?

 

4) Do you envision cores capable of running arcade ROMs? I would imagine that would be moderately difficult to implement, and it wouldn't be a feature I would be personally interested in, but I'd like to hear your thoughts on the subject.

 

 

Thanks.

[Keatah]

Stick Pokey, Antic, GTIA, Denise, Agnus, Paula, TIA, Maria, and every damned custom chip in there for the biggest incoherent mashup imaginable. Feature creep.. Keep it under tight control.

 

Don't start adding in physical machine-specific chips. That goes against the whole philosophy of the board.

Edited September 21, 2015 by Keatah

[S1500]

I like this. More focus on the inside(where it counts), rather than the outside(case molds).

[Asbrandt]

Feature creep.. Keep it under tight control.

The SID was mentioned because it's been notoriously difficult to nail down a reimplimentation of it on FPGA but that's being worked on still, if I recall correctly.

 

I could be mistaken, but it appears to me that Kevtris is focusing on console hardware, leaving home computers to projects such as the MiST, so the SID's difficulty would be irrelevant here anyway.

 

EDIT: In my opinion, this separation of goals is a good thing, even if you're paying more by buying two separate projects, there's more focus on the quality of each core this way.

 

I do hope he can pull off 16bit stuff, although I'd imagine the CD addons might complicate things.

Edited September 21, 2015 by Asbrandt