MiniMig with AGA

[mikej]

"So each JAMMA game out there will need an FPGA core written for it to work on the hardware?"

Yes, as the FPGA core models the original hardware.
Different games which ran on the same/similar hardware can use the same core, like frogger/scramble.

MikeJ

[mikej]

I've posted a few more pictures now the boards are fully assembled.
MikeJ

[mikej]

Looking good so far. Board is powered up, all supplies working well.
FPGA configured over JTAG and running so clocks are working.
Next step is to configure the ARM but I need to find my micro-usb cable for that first...
/MikeJ

[mikej]

"It might be a good idea to clearify what memory modules that are suitable for the board."

Doesn't need any, it has onboard 32MByte of RAM.

"what is the power usage of the FPGAArcade?"
5V input, depends on FPGA usage but 1A is more than enough. The polyfuse is 2A to allow power for the expansion boards.

"Will you offer the board in kit form where everything is mounted except for the through-hole parts"
no, fully assembled only. Option for no coded/SVHS to save cost.

"3. Have you determined what the price is going to be, or prices if you are offering it in more than one stage of completion?"
look a few pages back for pricing info.

"4. Have you contacted Colanto and asked them about a discounted bundle price for AmigaForever + FPGAArcade"
No, but it is on my list once the board is up and running
Best,
Mike

[mikej]

"Only 32? :-( "
Sorry, it's 64MByte. The schematic is correct, my memory isn't
/Mike

[mikej]

All the important ports are in the ATX window at the back.
On the side is the micro-usb used for updating the ARM firmware, and maybe connecting USB peripherals in the future. There is a header on the board so you can cable this off to a back panel if you want.

The serial port is on the other side, but this is primarily used for debug and not necessary normally.

Power is on the other side as well. If it is a case you would use either the molex or the two pin 5v pin header for power input.

The holes line up with the ATX case spec (well, the two that can do do, but it's quite solid).

I'm talking to some manufactures about a IO panel for the atx/itx cases.

Best,
Mike

[mikej]

The ARM ADC inputs are connected. Four to the rear "cable" connector and four to the main board stacking expansion connector.

I imagined the rear connector being used for joystick/buttons/analog joypad etc but you could use them for line level audio.

The expansion board could contain a decent mic input if there was demand.
I want to keep the feature set on the main board to a basic level to keep the main board cost down.I imagine a number of expansion boards for different applications could appear.

The schematics and top/bottom component placement are available.
Detailed mechanical drawings will also be available.

I'm not taking any pre-orders just yet - until we have got these boards tested and Yaqube has had a play.
Best,
Mike

[mikej]

For composite/VHS outputs you get real NTSC/PAL exactly like the original machine which will drive 15K linescan monitors. In the same way,the RGB analog out of the DVI can be connected to RGB monitors or SCART connectors.

You can cable the DVI output to HDMI (50/60Hz Progressive or Interlaced) and modern TVs will be fine with this.

For "computer" monitors the best way is for the cores to support higher resolution (1280x720) resolution for example. I'm looking at various frame buffer options, but they will add a frame of latency. This may be fine. T

/Mike

[mikej]

Sorry for the delay, I've been a bit busy.
Board is up and running, no problems as yet.

There will be quite diverse core support, a fair number of people are interested in supporting it.

All the system on fpgaarcade have been ported (including VIC20)
I am working on Star Wars, BBC B and Atari ST.

We also have C64, spectrum, zx81, apple 1, cray 1(!) and lots of home games consoles being worked on.

I'm involved with the visual6502 project who are doing die scans of original chips and I'm converting these into VHDL - so we will get even more accuracy going forward for these retro systems.

Best,
Mike

[mikej]

Hopefully we'll see an Amstrad CPC (+) core too

Somebody is working on that as well.
/Mike

[mikej]

Just to clarify the specs.
The Replay board hardware supports 24 bit and 16 bit video modes. It has a output filter / cable driver to get decent analogue output.

The FPGA is a Spartan3e1600 which is fully supported by the free toolset and around 3x larger than the original Minimig board.

Best,
Mike

[mikej]

"Did you get that video yet?"
No, they haven't sent it to me yet, they wanted to do some editing. I'll ask again.
/Mike

[mikej]

"I hope the boards arrive"
I have 10 boards already on test!

[mikej]

All good so far. One minor issue with the perfomance of one of the linear power regulators (LDO) is not as good as the prototype part (a different one). The reverse leakage is too high and the board will not always turn off. I have modded one board and have some different LDOs on order which should fix it.

Main power is fine, FPGA is booted and running ok so beer all round really.
/Mike

[mikej]

So far yes.

This is sort of complex, but here is the full story.

The main reason for the RevB board is that the DAC has a leakage problem. If you have a DVI/HDMI connection, sufficient current leaks back from the monitor and powers the 3.3V supply.

To fix this, a 5V to 3.3V LDO was added to this board as Chrontel recommend. For the prototype it worked well. I couldn't get the part I wanted for the first 10 boards so I used a microchip part.

It's not a good idea to run the main power through the slide-switch only the 5V supply goes through the switch and the high current DC-DCs are driven from the 5V input with the enable line driven from the 5V switch supply. This is a good simple low cost approach, but has the problem that if there is any leakage from the low voltage supplies back onto the 5V line then the DC-DCs will not turn off.

So, the board is a bit of a zombie. You turn it on. It goes on, You turn it off .... it stays on.

Most LDOs will turn off the FET when the input voltage drops below a certain threshold. This one does not, so I get about 1.6 V fed onto the 5V line. This is enough on some boards to keep the DCDCs running.

The other LDO I have works fine so I just need to fit a different one.

The differences to the RevA board are minor and there are 7 of those running for over a year now perfectly stable.

Best,
Mike

[mikej]

The input voltage must be regulated 5V and cable of supplying 1A or more. The 5V is used directly, the lower voltages are generated internally. There is a polyfuse (2A) and over voltage/reverse protection.
Ideally the 5V would be generated as well, but that would need another DC-DC and increase the cost.

You can also power it through the molex (ata hard disk power) connector from a PC power supply, assuming you have frigged the power on line.
Mike