.

[ChanceKJ]

.

[SavannahLion]

HDMI to CGA? Is that possible without an FPGA or complex logic and circuitry? A six second Google search only turns up HDMI->VGA or better.

[ChanceKJ]

.

[yotsuya]

That's probably part of the problem. There's gotta be a way we can't get CGA from a NUC or a pi.

Your biggest problem is trying to find a way to keep the costs down. Production costs, development costs, and then sales. It's a pretty niche product in a niche hobby.

[ChanceKJ]

.

[SavannahLion]

There's gotta be a way we can't get CGA from a NUC or a pi.

Huh? I guess you mean to say: There's gotta be a way we can get CGA from a NUC or a pi.

I'm sure that there is, but that's an awful lot of hardware to down convert from an HDMI signal. From a simple users point of view you'll have to get a HDMI->VGA converter then a VGA->CGA converter if you can't find a true HDMI->CGA converter. That's an added expense that is going to outstrip the cost of the Pi right out of the gate. In the high end gaming market, I wouldn't bat an eye buying a top of the line GPU card that costs more than the MB+CPU+RAM combined. In the low end MCU market, it's hard to swallow spending that ratio of money to get such microscopic improvement.

A better solution might be to output VGA then use a $30 converter to convert from VGA-CGA and eschew the whole HDMI nonsense entirely. But I would gamble that might be unsatisfactory as per this thread: http://forum.arcadecontrols.com/index.php?topic=121893.0

I'm just guessing here so someone else might have more information. But it looks like you'll have to leverage the GPIO to create the CGA signals you need. The video controller doesn't even list CGA as a video option and I don't know the specs of that GPU to even take a guess as to whether it's capable of generating CGA directly at all. That puts a greater burden on the software to generate the necessary signal timing. But it does free up the GPU for other tasks if it's possible to use if for other tasks. I imagine it's doable but probably undesirable. Really depends how much work comes out of that.

I can't venture as to anything about the NUC as I haven't looked at it much.

You've also got some crazy ass projects like this: http://hackaday.com/2011/07/02/controlling-a-cga-monitor-with-an-arduino/
I think someone is building a MCU GPU but I can't recall the name intended to drive some really low resolution monitors over USB. Could look into that since something like the Pi or NUC would just shove data down the pipe and let another controller do the lifting. 

[SavannahLion]

What about even just a blank PCB with edge connection and solder points with lables. Then you mount your own pi, cga converter, Kade and be done with it.

That requires a bit of electrical knowledge which would shrink any sort of market you have even further.

[clewis]

Sorry to bump such an old thread, but I wired a pi up to my red tent this week to test this concept. I can confirm that the edge connector has everything you need.

I'm not sure if anyone has done this, but I used an RF board to wire everything up to the existing original harness.

Mounting everything to a sheet of plexi to slide in like a PCB.

Rgb works through GPIO via a VGA 666 or a Retrotink.

5v from harness works to power the pi.

Sound from rpi3 phono straight to the sound pin.

Controls via usb encoder. I-PAC etc...

What about inverted video? The Sharp monitors in a red tent have a simple toggle switch on the chassis to invert the video.

I ordered a second rpi3, and I should have everything wired up next week and testing netplay between the two sides.

[clewis]

The red tent is still 100% original and not modified in any way. Everything is wired up although I have yet to spend much time with netplay. If anyone wants me to detail how I went about it let me know.