FPGA Based Videogame System

[TPR]

He doesn't need to support the whole F18a core to allow non-flickering sprites, I think he already did something similar (no flicker aka more sprites per line) on the NES core so he knows what to do assuming he cares enough to do it.

 

Right I know. I was just saying that as the F18a issue did come up previously and it would be one solution, not the only solution. I'm all for no sprite flicker on the ColecoVision regardless of how he does it because it looks great!

[rezb1t]

I rescind my previous report about Final Fantasy 3 and Recca. I tested them out more extensively and it turns out the NT Mini is doing the correct thing, after all. I'm pretty sure it is just filtering making it sound different

[RevQuixo]

Looks like Kev just worked on a new firmware release for HI-Def NES:

 

https://www.game-tech.us/

[Andromeda Stardust]

Looks like Kev just worked on a new firmware release for HI-Def NES:

 

https://www.game-tech.us/

 

Kevtris

 

Using a powerpak, though (with powermappers) DPCM is kind of broken. Samples end earlier than they should, like the length isn’t being programmed right. The analog audio coming out of the CPU is broken in the same way, and watching the CPU bus the sample fetching just stops! I am at a loss as to what’s going on here. Needless to say, it only happens with one specific CPU chip out of the dozen or so test samples, so I am not sure what’s going on. I am suspecting the chip might be flaky somehow.

Sounds like an early revision Famicom chips that did that.

[retro_fan]

Looks like Kev just worked on a new firmware release for HI-Def NES:

 

https://www.game-tech.us/

It's great that he did the update, I heard people were having issues. I really hope that when Kevtris will finally set up the bug-tracker the Hi-Def NES mod will be there as well (to report bugs and not to suggest new features, the FPGA is full!). Cross-reporting seems like a good idea because it looks like Kevtris' projects share codebase (at least the NES portion), and more users reporting means more fixed bugs.

[kevtris]

It's great that he did the update, I heard people were having issues. I really hope that when Kevtris will finally set up the bug-tracker the Hi-Def NES mod will be there as well (to report bugs and not to suggest new features, the FPGA is full!). Cross-reporting seems like a good idea because it looks like Kevtris' projects share codebase (at least the NES portion), and more users reporting means more fixed bugs.

I don't think I will update the hi def any more. It's about as mature as it's going to get. So no bug tracker needed :-)

[retro_fan]

I don't think I will update the hi def any more. It's about as mature as it's going to get. So no bug tracker needed :-)

Are you sure the users won't find any bugs?

[Wolf_]

Are you sure the users won't find any bugs?

[Great Hierophant]

Are you sure the users won't find any bugs?

 

[kevtris]

Are you sure the users won't find any bugs?

Maybe, but there's a lot of them out there, and I fixed all the known bugs now in this update so hopefully that will be it. I certainly can't add anything new so there shouldn't be any new bugs anyways.

[retro_fan]

Maybe, but there's a lot of them out there, and I fixed all the known bugs now in this update so hopefully that will be it. I certainly can't add anything new so there shouldn't be any new bugs anyways.

If you say so, I guess that's OK. Well, nobody's expecting that you will be working as much on your older projects as on newer ones (everyone can get bored with something, even you ), however, just take into consideration that the situation when a developer fixes even occasional 1 or 2 bugs, years after the release, is really appreciated by the community, and it can drive the sells of new products because people appreciate such continuous (even if very small) support. Another option is releasing the source code (if you don't sell a product anymore), so that people can tweak and fix some small stuff for themselves.

Edited April 30, 2017 by retro_fan

[RevQuixo]

If you say so, I guess that's OK. Well, nobody's expecting that you will be working as much on your older projects as on newer ones (everyone can get bored with something, even you ), however, just take into consideration that the situation when a developer fixes even occasional 1 or 2 bugs, years after the release, is really appreciated by the community, and it can drive the sells of new products because people appreciate such continuous (even if very small) support. Another option is releasing the source code (if you don't sell a product anymore), so that people can tweak and fix some small stuff for themselves.

 

Dude he literally just rolled out an update, lecturing him on the importance of supporting the hardware at this point is a little premature. If something major pops up, he'll probably fix it...he just fixed basically all the bugs that are known..so hopefully that is that.

[kevtris]

If you say so, I guess that's OK. Well, nobody's expecting that you will be working as much on your older projects as on newer ones (everyone can get bored with something, even you ), however, just take into consideration that the situation when a developer fixes even occasional 1 or 2 bugs, years after the release, is really appreciated by the community, and it can drive the sells of new products because people appreciate such continuous (even if very small) support. Another option is releasing the source code (if you don't sell a product anymore), so that people can tweak and fix some small stuff for themselves.

umm, the last update was 14 months ago, so I rolled out an update over a year after the last one. The device itself was designed at the end of 2014 and the start of 2015, so this is technically 2 years later.

[Andromeda Stardust]

If you say so, I guess that's OK. Well, nobody's expecting that you will be working as much on your older projects as on newer ones (everyone can get bored with something, even you ), however, just take into consideration that the situation when a developer fixes even occasional 1 or 2 bugs, years after the release, is really appreciated by the community, and it can drive the sells of new products because people appreciate such continuous (even if very small) support. Another option is releasing the source code (if you don't sell a product anymore), so that people can tweak and fix some small stuff for themselves.

I think it is understandable that as a product matures, it contains fewer bugs if any, so if EOL as far as firmware updates go happens a couple years out, then that is that. It does not surprise me that the HiDef NES may become depreciated when options like AVS and NT Mini exist. I'm sure there's still a market for purists to buy the HiDef kit, but that market is shrinking compared to current options. So the HiDef is a mature product, it doesn't need daily updates, unless you are *cough* Microsoft *cough*.

 

NT Mini will reach maturation at some point but the unofficial updates will likely continue for some time as more consoles are added. Currently it's like 12, 14, however many products rolled into one with the jailbreak firmware. By the time Zimba 3000 rolls around, many of the "cores" will likely be mature. And that's not a bad thing.

[retro_fan]

umm, the last update was 14 months ago, so I rolled out an update over a year after the last one. The device itself was designed at the end of 2014 and the start of 2015, so this is technically 2 years later.

Wow, the time goes by so quickly! I did not want to sound like I am not happy with support or anything, that was a general piece of advice for supporting older products and not a complaint about the Hi-Def NES (I'm not a native speaker of English, so if I did not get my point through, I'm sorry )... Anyway, all the updates so far are awesome, keep up good work Kev!

[retro_fan]

Kevtris, have you considered some kind of expandability for the Z3K? I'm currently watching the ZX Spectrum Next FPGA project, and they want to let the users expand the device with RAM, Wi-Fi, and RTC clock. Also, you would be able to use the Raspberry Pi Zero as an accelerator (how something based around the ARM CPU will communicate with the FPGA, I have no idea..)!

[Keatah]

Kevtris, have you considered some kind of expandability for the Z3K?

 

Along that line of thinking, I cannot emphasize it enough, that the z3k have some spare elements and reserve capacity to do things not thought of when it rolls off the factory assembly line. If this adds $100 to the price tag, then so be it.

 

Think along the lines of the Apple II and its bank of expansion slots. I don't believe SW and SJ envisioned the computer being used 40+ years later being equipped with the likes of modern Flash memory - of capacities millions or billions of times what was available in 1976. Or just simply all the thousands of expansion cards - with new ones being made even as we speak.

 

So, yes, some reserve capacity and/or expandability is a must.

Edited May 2, 2017 by Keatah

[Pixelboy]

Along that line of thinking, I cannot emphasize it enough, that the z3k have some spare elements and reserve capacity to do things not thought of when it rolls off the factory assembly line. If this adds $100 to the price tag, then so be it.

 

Think along the lines of the Apple II and its bank of expansion slots. I don't believe SW and SJ envisioned the computer being used 40+ years later being equipped with the likes of modern Flash memory - of capacities millions or billions of times what was available in 1976. Or just simply all the thousands of expansion cards - with new ones being made even as we speak.

 

So, yes, some reserve capacity and/or expandability is a must.

I can agree that expandability would be a nice-to-have feature, but let's keep in mind that the Zimba 3000 will be a gaming device first and foremost. If you run a Commodore 64 core on it, you will probably do so to play C64 games, not to run a spreadsheet program, or whatever else that's not a game. So if the C64 game(s) you want to play can run on a stock Zimba 3000 with the standard C64 core, expandability is a moot point. Same goes for the games of any other home computer.

 

On the other hand, there is one area where I see a possible need for expandability, and that is to support arcade machine cores: Support for specialized controllers, support for several types of screens (not just HDMI) to run in TATE mode, extra RAM to support some of the later arcade machines that can almost run on a stock Z3K but not quite, etc.

 

Also, let's be realistic for a moment: When the Zimba 3000 becomes reality, Kevtris is not going to be the only one developing cores for it. Even if Kev wants to keep core development to himself, some people are going to pop the hood of this machine as soon as it lands in their hacky hands, and document their findings, the same way that the NES Classic Mini was hacked all the way to the moon.

 

So Kev will be faced with a choice: He can adopt Nintendo's way of thinking and make hacking his machine as hard as possible to discourage other people from messing around with it (hint: it may not work very well for an FPGA machine designed to be a multi-system device from the start) or he can offer the hardware docs openly and say "Make your own cores if you want, but I'm not responsible if you mess something up". If he does end up offering the docs, I can see some core developers having fun adding cores for some arcade machines (the same way it was done with MAME) while Kev keeps dibs on his main cores (Super-NES, Genesis, Neo-Geo, C64, MSX, etc.).

 

And that brings us back to expandability, because some arcade games may require some resources that the stock Z3K will lack. A little planning in advance will go a long way in helping the Z3K stay relevant and interesting for core developers, and all the Z3K owners will keep an eye on new developments and benefit from them (in a more-or-less legal fashion). In the end, it will extend the lifespan of the Z3K.

 

After all, who wouldn't want to play Star Wars Arcade in all its vector glory on his HDMI television, via a plug-and-play game console? But we cannot expect Kevtris to make all the cores for all the arcade games we want to play, there are simply too many. That's why I believe other core developers will want to step in just for the bragging rights of making the most desired arcade games work on the Z3K.

[+Random Terrain]

Stop making me drool!

[phoenixdownita]

C'mon there's no need to have the Z3K designed as a tank that needs to support all sorts of whacky stuff people want to do on an FPGA say 5 years from now. There can always be a Z3K 2 if needs be.

There's plenty of FPGA boards as it is, the Z3K should focus on supporting consoles up to and including the 16bits kevin mentioned (NeoGeo being the long pole as well as the SA-1/SuperFX special chips for SNES).

 

Because that's what we are talking about here a board with one or two FPGA, some memory (hopefully enough CellularRAM), IO connections (SD card, joypad/joystick, power. A/V/ HDMI) ..... if you take away kevtris cores there's literally dozens of boards like that (aside the second FPGA for video processing but one can argue to just use a bigger single FPGA), so in this regard the Z3K is no different, what is different about it is the cores that kevin puts out .... that's really it, they work extremely well (aside when rushed by his own admission) and are as close to the real deal as possible but support digital out as well as analog (for who still cares about it) and that is what makes the difference. it's about him and not the Z3K per se, although it will be designed to perform the task excellently rather than with a more generic design to allow for future expandability (honestly for that to happen just expose a connector that brings out as many FPGA IO pins as possible and you're technically done at the electrical level [maybe add some protection or voltage translators if need be]).

[Pixelboy]

C'mon there's no need to have the Z3K designed as a tank that needs to support all sorts of whacky stuff people want to do on an FPGA say 5 years from now. There can always be a Z3K 2 if needs be.

There's plenty of FPGA boards as it is, the Z3K should focus on supporting consoles up to and including the 16bits kevin mentioned (NeoGeo being the long pole as well as the SA-1/SuperFX special chips for SNES).

 

Because that's what we are talking about here a board with one or two FPGA, some memory (hopefully enough CellularRAM), IO connections (SD card, joypad/joystick, power. A/V/ HDMI) ..... if you take away kevtris cores there's literally dozens of boards like that (aside the second FPGA for video processing but one can argue to just use a bigger single FPGA), so in this regard the Z3K is no different, what is different about it is the cores that kevin puts out .... that's really it, they work extremely well (aside when rushed by his own admission) and are as close to the real deal as possible but support digital out as well as analog (for who still cares about it) and that is what makes the difference. it's about him and not the Z3K per se, although it will be designed to perform the task excellently rather than with a more generic design to allow for future expandability (honestly for that to happen just expose a connector that brings out as many FPGA IO pins as possible and you're technically done at the electrical level [maybe add some protection or voltage translators if need be]).

Of course, there's no actual need for Kev to add anything more than some kind of generic expansion port on the Z3K, but on the other hand, I'm having a very hard time imagining Kev creating a Z3K-2 (or should that be Z4K?). If I were in his shoes, I'd be happy just to release the Z3K to the masses, and then give a "been-there-done-that" response to anyone who wants to persuade me to do a new iteration of my hardware and cores. That's the main difference between a hobbyist releasing a commercial product and an actual company like AtGames releasing a similar product. The hobbyist can rest on his laurels with the knowledge of a job well done (and possibly move on to something else afterwards) while a company will want to milk the market for all it's worth and continue to draw profits from it.

 

And yes, phoenix, there's no doubt Kevtris' excellent cores are the main draw of the Z3K, but that won't stop other FPGA core experts from trying to add their own cores to the mix. As soon as you put hardware out into the wild, it stops being yours and hackers will want to have fun with it. That's just a fact of life in this day and age.

[phoenixdownita]

... but that won't stop other FPGA core experts from trying to add their own cores to the mix. As soon as you put hardware out into the wild, it stops being yours and hackers will want to have fun with it. That's just a fact of life in this day and age.

What's stopping them now to create/add them to existing powerful development boards?

 

https://www.arrow.com/en/products/deca/arrow-development-tools

65US$ for 50K LE. HDMI connector already in, SD card reader already in, 512MB of SDRAM in etc...etc....

 

Is it the aluminum milled case? Is it the fact that they can design and implement a perfect arcade core (to the transistor level no less) but can't design a PCB to interface a joypad? What is it?

 

I mean seriously the only thing I think those "experts" seems to be waiting for is for someone else to prepare the hardware (ports and all) that resembles something users may want in their home but then again some are fine with an RPi in plastic case, sell said HW to as many people as possible with an initial set of kick-ass cores (build an audience), open up the architecture (sort of document/prepare an "SDK" for the HW) so they can jump in the winner bandwagon.

 

Nothing bad with that but as I said there's plenty of boards the only thing that would speed up the process is if kevin releases his HDMI IP block as such those "expert" do not need to rebuild it, same for all the controller reading/translating logic etc..... so basically we are asking kevin to build the Z3K as a console/arcade core development board where all the peripherals are already being handled and the grunt work done ... maybe he'll do it not sure. MiST seems to be working like an open systems and there are good third party cores on it, at the same time if kevin himself says he wants to build the SNES, MD and NeoGeo cores (given he has done already all of the 8bits more or less) I don't see why he would need to complicate his design further to support "arcade core experts development". His project, his hardware, his cores .... I don't see a problem with that and I will gladly buy it as it will do what he says it will do almost to a fault

[NE146]

I could (happily) be proved wrong but mentally, I don't expect anything about the Z3k for a long time. I've been watching kev chat about his cores since like 2010 (or maybe earlier) while we could only drool about it on IRC.. so for all this to finally come out in recent times still feels like christmas after a long decade wait finally. I may be off, and don't mean to speak for the guy but it obviously is worked on when he has the feeling and when he enjoys it. You don't push hobbyists or make demands or even "suggestions". You just have to sit and wait and see what happens when they do as they do.. and hopefully maybe sometime you'll get to reap the benefits.

 

[kevtris]

Kevtris, have you considered some kind of expandability for the Z3K? I'm currently watching the ZX Spectrum Next FPGA project, and they want to let the users expand the device with RAM, Wi-Fi, and RTC clock. Also, you would be able to use the Raspberry Pi Zero as an accelerator (how something based around the ARM CPU will communicate with the FPGA, I have no idea..)!

 

The original board I did had a so-called 'pony connector' on it for expansion. I did really like the famicom port on the nt mini as a method of expansion though, at least for cart adapters. The original board has pcie connectors for expansion capabilities. Of course it is not using pcie, I just used the connector since it's cheap, high density and awesome.

 

I can agree that expandability would be a nice-to-have feature, but let's keep in mind that the Zimba 3000 will be a gaming device first and foremost. If you run a Commodore 64 core on it, you will probably do so to play C64 games, not to run a spreadsheet program, or whatever else that's not a game. So if the C64 game(s) you want to play can run on a stock Zimba 3000 with the standard C64 core, expandability is a moot point. Same goes for the games of any other home computer.

 

On the other hand, there is one area where I see a possible need for expandability, and that is to support arcade machine cores: Support for specialized controllers, support for several types of screens (not just HDMI) to run in TATE mode, extra RAM to support some of the later arcade machines that can almost run on a stock Z3K but not quite, etc.

 

Also, let's be realistic for a moment: When the Zimba 3000 becomes reality, Kevtris is not going to be the only one developing cores for it. Even if Kev wants to keep core development to himself, some people are going to pop the hood of this machine as soon as it lands in their hacky hands, and document their findings, the same way that the NES Classic Mini was hacked all the way to the moon.

 

So Kev will be faced with a choice: He can adopt Nintendo's way of thinking and make hacking his machine as hard as possible to discourage other people from messing around with it (hint: it may not work very well for an FPGA machine designed to be a multi-system device from the start) or he can offer the hardware docs openly and say "Make your own cores if you want, but I'm not responsible if you mess something up". If he does end up offering the docs, I can see some core developers having fun adding cores for some arcade machines (the same way it was done with MAME) while Kev keeps dibs on his main cores (Super-NES, Genesis, Neo-Geo, C64, MSX, etc.).

 

And that brings us back to expandability, because some arcade games may require some resources that the stock Z3K will lack. A little planning in advance will go a long way in helping the Z3K stay relevant and interesting for core developers, and all the Z3K owners will keep an eye on new developments and benefit from them (in a more-or-less legal fashion). In the end, it will extend the lifespan of the Z3K.

 

After all, who wouldn't want to play Star Wars Arcade in all its vector glory on his HDMI television, via a plug-and-play game console? But we cannot expect Kevtris to make all the cores for all the arcade games we want to play, there are simply too many. That's why I believe other core developers will want to step in just for the bragging rights of making the most desired arcade games work on the Z3K.

 

I seriously doubt anyone will take the time to figure out how the hardware works enough to write a new core for it. There's a ton of hard work that would need to be done- the PCB layout needs to be traced to see what pins of the FPGA connect to the RAM, video, audio, etc. Also, you'd need to spend a bunch of time REing how the FPGA and CPU intercommunicate, then how the clocking works, etc. It is not very easy and would take a lot of time to do.

 

The NES mini that nintendo released is different- that was basically a gift-wrapped linux machine with a debug serial port right on it. It didn't take long to hack since everything on it was standard pretty much. You could just sit back and poke it with some linux knowledge and make it do your bidding. The FPGA isn't so easy- it does not have an OS or anything standard on it. Everything is custom made or written, and this will make life a lot more difficult for anyone attempting to figure it out. Also, the cores would need to be added somehow to the menu system, and without all the tools I wrote to do this it'd be tough going. There's no debug serial either to hook into :-)

 

There's a huge amount of "back end" code and processing that goes on which no one gets to see; there's a few dozen pieces of "middleware" I wrote to massage the data and process things to generate the firmware, files, and do debugging.

 

So while it would be possible to use the system as basically a videogame based FPGA dev board, it'd be pretty hard to re-integrate your cores with the menu system and other software.

 

As for WHY, I don't have time to document everything, make the middleware bug free/easy to use, and help people get it working. That's a huge amount of work, and I don't have time to do it. Then there's all the support I'd have to do... that's just a time burgling rathole I don't want to go down. Sorry about that.

 

Honestly, even if I had all this stuff open enough to let people write cores, I seriously doubt anyone would bother. The people who are interested in it aren't the people who can write cores, and those that can write cores don't have the time/inclination to make it work on other hardware.

[Pixelboy]

What's stopping them now to create/add them to existing powerful development boards?

 

https://www.arrow.com/en/products/deca/arrow-development-tools

65US$ for 50K LE. HDMI connector already in, SD card reader already in, 512MB of SDRAM in etc...etc....

 

Is it the aluminum milled case? Is it the fact that they can design and implement a perfect arcade core (to the transistor level no less) but can't design a PCB to interface a joypad? What is it?

 

I mean seriously the only thing I think those "experts" seems to be waiting for is for someone else to prepare the hardware (ports and all) that resembles something users may want in their home

Hey everybody! I think he's got it!

 

 

but then again some are fine with an RPi in plastic case, sell said HW to as many people as possible with an initial set of kick-ass cores (build an audience), open up the architecture (sort of document/prepare an "SDK" for the HW) so they can jump in the winner bandwagon.

Let's not mix things up here. An RPi is different from an FPGA board, even though they can technically produce the same end result using different methods.

 

 

Nothing bad with that but as I said there's plenty of boards the only thing that would speed up the process is if kevin releases his HDMI IP block as such those "expert" do not need to rebuild it, same for all the controller reading/translating logic etc..... so basically we are asking kevin to build the Z3K as a console/arcade core development board where all the peripherals are already being handled and the grunt work done ... maybe he'll do it not sure.

That's really up to Kevtris. I figure he'll want to keep his cores close to home, especially while he's developing the Genesis and Super-NES cores (and probably a few others) but once those cores are done, why wouldn't he want to help other developers with their own "third-party" FPGA cores? If Kev feels that some parts of his work is highly proprietary, then that's all fine, I guess. Ohterwise, the Z3K can be a party for all.

 

 

MiST seems to be working like an open systems and there are good third party cores on it, at the same time if kevin himself says he wants to build the SNES, MD and NeoGeo cores (given he has done already all of the 8bits more or less) I don't see why he would need to complicate his design further to support "arcade core experts development". His project, his hardware, his cores .... I don't see a problem with that and I will gladly buy it as it will do what he says it will do almost to a fault

So what you're really saying is that Kev has no business letting other people poke around his hardware?