FPGA Replay Board

[JimS]

who would make these cores? and could they do that with a replay board or do they need a full dev kit?

You'd need a PC with the Xilinx development software installed. It's a free DL, but huge - about 4GB. I tried to download it at the library last week, but their security blocked the download manager Xilinx wanted to send first.

[JimS]

Free from memory. The Action Replay uses an ARM cpu. It's a simple matter of loading the core of choice. Cores are written in VHDL or Verilog.
Cores designed for other board designs should be easy to adapt (wishbone?). So a massive amount of cores should appear soon after the release.
Macintosh 68k and 80386+VGA based demos ought to be quick. All 8-bit variants should pose no problem.

It's unclear weather the default firmware for the ARM will allow programs to be loaded into it and then executed. (mikej?)

If you look at the demo video mike posted on youtube, you can see him selecting demo disk menus from an onscreen display.

[JimS]

The FPGA camp doesn't have a fixed color. It's loaded at boot time from another camp.
;-)

[JimS]

OK, I just reread that, and had an image of a bumper sticker withe the word "Amiga" using rainbow colors.

How about the word "Replay" using letters from the fonts associated with the major systems to have cores?

[JimS]

I don't see the need for a zorro slot either... A video slot for the Toaster might be slightly sensible. ;-) Or at least something in the "neat hack" category.

[JimS]

You can connect another FPGA to the expansion port. The base FPGA is wired directly to the expansion port (no FSB).
(So picasso96 etc.. could be implemented)

Asfaik, lest mikej says something else.

The Spartan FPGA on the Replay is larger than the minimig... wonder if there would be enough room to do a picasso96 card in there as well.... maybe make it share the buffer with the delacer so they could overlay each other on screen.

[JimS]

There's a lot of room. The FPGA is four times as big. And an RTG card is in the works.

Cool. Thanks for the update. Kudos to you and MikeJ for a job well done.. ;-)

[JimS]

JimS, I was referring to implementing another RTG in a FPGA daughterboard. Maybe soft-68060 is doable as well

A cool thing would be multiple video output RTG card. Like running 8 monitors from the same Amiga..

Ah.... well, there's already a soft 030 in there right... so why not an '060 ;-)

How about doing a Toaster stub so that those of us with a Toaster in our real Amigas could run Lightwave and render to a simulated Toaster frame buffer? ;-)  Lots of wacky ideas out there. ;-)

[JimS]

While you can never have enough memory, we also need to keep in mind that machine specs are always a tradeoff.  Getting a working product out now is better than having the promise of a product in the future.

This seems to be one of the problems facing Natami.  Changing specs keeps it from ever making to to production.  I will happily take 128M memory on the FPGAArcade, and if that turns out to be too little in a few years, I will happily pay to buy a new FPGAArcade with more memory at that time.

This way I can play today, AND play tomorrow.  As opposed to hoping that some day I will get to play with the perfect board.

Babbage Syndrome.... keep thinking "wouldn't it be cool if.....", then never actually finishing the project before running out of funds.

My "real" Amiga never had more than 13MB, and I was never pressed for space. Well, except in chipram, which won't be a problem for either the Natami or Replay.

[JimS]

Absolutely perfect description, mind it I use it?

Be my guest.... I probably borrowed it from somewhere myself.

[JimS]

@MikeJ

I've got a new toy on my desk. It's on loan, but it's a Virtex-7 2000T. This baby would take a full Amiga Chipset + CPU (many times over) and early runs with Synplify + ISE have MiniMig 68k meeting timing easily constrained at 100MHz

Shame it cost over $100k

Xilinx has announced the first shipments of its Virtex-7 2000T Field Programmable Gate Array (FPGA). The Virtex-7 2000T is the world’s highest-capacity programmable logic device – it contains 6.8 billion transistors, providing customers access to 2 million logic cells. This is equivalent to 20 million ASIC gates, which makes these devices ideal for system integration, ASIC replacement, and ASIC prototyping and emulation.

6.8 billion transistors don't come cheap, eh? ;-)

[JimS]

That is good news, as I also have a Miracle Piano keyboard that I would like to be able to use with the Replay board.  Of course I would need instructions on how to make an adapter to get it connected and working, if something more complicated than a simple 9pin to 23pin Serial port adapter is required.

Maybe someone else with mechanical ability and electronic smarts will make a few of these adapters for dummies like me to purchase in the future.

What about a Parallel port that is compatible with the original Amiga's parallel port for devices like the DCTV, DSS8+, or other peripherals that need a connection to an Amiga parallel port?  Would that be possible with some extra code using the available extra i/o pins?

You might have to rig up an external power supply for some of those devices. Remember that the Amiga supplied +-12v through the serial port for external devices. The parallel port had 5v... There's another problem trying to do the parallel port. The FPGA uses 3.3v logic levels. It may not be 5v tolerant, so you'd need some kind of level shifting in between the old school devices and the FPGA. At least that's what I've read so far... just got the Xilinx ISE13 last week.

[JimS]

The FPGA I/O level depend on the Vcco voltage (asfair) that can vary between 1.2 - 3,3 V. Which voltage that is used depend on the designer (mikej). Any other voltage has to be dealt with using a voltage conversion. The standard trick is a zenerdiode to "Vcc" to limit voltage and a resistor to limit current. The trick has the drawback that it MIGHT overload the supply line that the zenerdiode it's wired to and thus fry most things there (see Xilinx appnote).

A better approach is a dedicated level conversion chip.

I'm going to have to look at the Xilinx site for some of those appnotes. The docs for the simple Spartan 3 dev board I'm looking at mentions the interface problems but doesn't go into much detail on dealing with it. That dedicated chip is going to have to be bidirectional on the parallel port.

[JimS]

Many of those people own peripherals that they once used on their original Amiga computers and might want to use them again on a modern replacement, since the FPGA clones are able to run almost all Amiga software, it would also be good if they could use some of the old Amiga third party hardware expansions, like the Miracle piano keyboard, a DCTV paint and image capture device, and other devices that connected to the Serial, or Parallel ports.

After all, isn't the Amiga community motto, "We did it just because we can"?

I would think the Miracle piano keyboard should work with nothing other than a simple DB9 to DB25 adapter. The trick is whether or not the serial port in the FPGA minimig core can hit the MIDI serial baud rate.

[JimS]

The point is that there are usually more efficient ways to accomplish what those peripherals do. Especially using I/O efficient interfaces like I2S etc. That usually also does away with any messy voltage level conversion in massive parallel arrays. Direct connection to the FPGA also imply the risk to fry the main chip.

You can do messy line interfacing and use lots of I/O to connect a sampler. Or you could use a small chip that does it all directly. Just because something is possible doesn't make it the right solution.

Oh, I agree with you totally, it makes a lot more sense to build some new custom hardware replace whatever external devices you might have on the serial & parallel ports. But there is the software involved. If it were relatively easy to do the Amiga's ports, then you wouldn't have to write a new sampler program for example.

Baud is signaling rate, bits per cycle. The bit rate is possible either by dividing some other clock or by using a PLL (DCM). It might require a update to the FPGA binary (core) however.

That's what I meant to say. To drive the Miracle piano, the FPGA core would have to hit the MIDI bit rate. I guess that would depend on the design of that core.