Commander X16

On 8/27/2020 at 6:40 AM, martinot said:

You could make more use of the FPGA, and make the X16 cheaper by having less discrete components and move more logic to the FPGA (including the CPU).



 

Thing is, whether a kit is ever provided, the through hole kit buildable system is the system reference design.

Tinkering with and tweaking the hardware can easily become a slippery slope, which leaves people scattered all along the slope depending on where they personally dug in their heels, and no sustainable community developing for the platforms because every system is incompatible in some way with every other system.

The challenge of getting a board made of mostly ASIC through hole parts including a real 8bit microprocessor is what keeps the temptation to endless hardware tweaking in check. Now, that is "to the extent that that is practical", and with no suitable VGA tile/sprite chip, an FPGA equivalent had to be created. But "to the extent practical" still disciplines the system reference design, and then the system reference design gives the implementation target for the systems that rely on a larger FPGA or additional CPLD (as the case may be).

Then, hypothetically, people who huy a Vera and are able to organize a clean room ROM alike that doesn't use CX16 originated or licensed code might be ready to go ... except good luck finding enthusiastic 65C02 programmers who haven't had a look at either C64 ROMs or the CX16 code, and without those enthusiastic but provably "never seen that code" set of people, you don't have a clean room and really should be working on something else. Of course, the way the original PC BIOS clone clean rooms were populated was by finding competent minicomputer programmers who had never worked on x86 chips before and PAYING THEM to learn how to program 8086s and then write the clean room code. So unless some nostalgic high tech millionaire wanted to fund the whole thing, that specific strategy is not available.

 

There is another potential. You just do. It provide ready to use ROMs but just Diffs to the cloanto ROMs . That way every one need to purchase a copy of the cloanto ROMs and then patch them. 

 

That should be possible. 

 

and yes you can put the complete X16 on a DE10 Nano. Also the DE10 Nano comes with quite a few GPIO pins. You do not need any of the expansions and can use them as expansion port. Still I do not see any point in doing so. If you want hardware you go for certain reasons the X16 way. If you do. It need the hardware ... why then bother with an FPGA? Run vice with a C128 or Frodo for a CPC or even Amiga or ST in an Emulator. You get everything you need there. X16 does not give you anything better or more. 

45 minutes ago, SerErris said:

Run vice with a C128 or Frodo for a CPC or even Amiga or ST in an Emulator. You get everything you need there. X16 does not give you anything better or more. 

There is NO BASIC computer on the market, now or ever, that's quite like the CX16. The 128 is much, much slower, and the Amiga and ST are not BASIC computers. 

The closest thing might be the MSX computer running one of the faster CPUs... but it's awkward and hard to use in emulation, and setting up and running it properly in MiSTer has proven to be a bit of a challenge. I've done so, but it's... complex. 

 

On 8/26/2020 at 6:47 AM, Fnord42 said:

Plus, I want a real 65*02 CPU.

I know that's not really rational, but I also know that I'm not alone with this. ?

Rationality is not about what you want, it's about how you go about getting it -- including accepting you can't always get EVERYTHING you want.

Now "reasonable"? Obviously since I also want a real 65xx processor., I'll accept it as PERFECTLY reasonable.

 

23 hours ago, TomXP411 said:

There is NO BASIC computer on the market, now or ever, that's quite like the CX16. The 128 is much, much slower, and the Amiga and ST are not BASIC computers. 



The closest thing might be the MSX computer running one of the faster CPUs... but it's awkward and hard to use in emulation, and setting up and running it properly in MiSTer has proven to be a bit of a challenge. I've done so, but it's... complex. 



 

That was not the point ... and even there you already have an emulator. It is there is no need to run it on an FPGA. Simply cannot see any reason to do this. Either you want hardware - go with X16 - you do not want hardware - take the emu or any other available option.

9 hours ago, SerErris said:

That was not the point ... and even there you already have an emulator. It is there is no need to run it on an FPGA. Simply cannot see any reason to do this. Either you want hardware - go with X16 - you do not want hardware - take the emu or any other available option.

But I expect the CX16e FPGA version of the CX16 will fit inside CX16c case with room to spare and have at least block pin headers for an expansion slot, so it solved the connectivity problem with the hardware that is made to work with the CX16p/CX16c ... even if I am only getting the CX16e if I can't afford the CX16c.

As far as something like the DE-10 Nano kit, the loss leader aspect of that board may make it much less expensive than it would be on a cost-plus-margin basis, but whether or not the Vera design is eventually released as open source it's perfectly reasonable to expect that the project will be given a chance to get its footing before that happens, and without a cycle exact Vera design, I'm not sure I see the actual benefit over the smallest NUC that can run the emulator at full speed.

On 8/27/2020 at 7:22 AM, Cyber said:

 



 





That's exactly what X16 team is planning to do at stage 3. So just wait patiently. )



 

Perfect, as it makes a lot of sense for cost reasons.

Looking very much forward to it! ?

Isnt Vera a FPGA logic anyhow? So cycle exact is just releasing the rtl code of it and compile it?

7 hours ago, SerErris said:

Isnt Vera a FPGA logic anyhow? So cycle exact is just releasing the rtl code of it and compile it?

Yes, though "just" is doing a lot of work in that sentence. After the CX16 has hit the market, it would not be surprising if there eventually was a release of the code for Vera at some future point.

As far as the ROM goes, I don't see why it would be too difficult (relatively—obviously it would take a fair amount of work) to simply write another one using the same entry points for the routines, similar to what Compaq did to avoid licensing the PC BIOS from IBM, who probably wouldn't have agreed to it. Cloning the whole BASIC interpreter would probably be a lot of work, yes, but if compatibility with BASIC programs isn't a priority then a clone could just focus on the KERNAL routines and run assembly programs, and let the user load a BASIC interpreter from disk. After all, PC clones generally left out the PC's ROM BASIC interpreter for the same reason.