Tandy video mod

[Trixter]

I did take me a couple of hours, on and off, to get the colors right.

I have tested two of the bit-c128 boards and found they output a low green level by design. Post-conversion adjustment of green (you need to boost it 20%) is necessary to get things matching.

If you would like a nice pattern generator, I wrote a few test plates into the CGA Compatibility Tester. It's the last menu, and provides a good RGBI test plate for adjusting your bit-c128/gonbes setup. The "Contrast" test is also good for adjustment.

The J. Alex boards are *not* the same design. Both boards have the same root design, but the J. Alex board outputs correct levels.

[GHiero]

I wish J. Alex posted PCB layouts so other people could have boards made from OSH Park. The through hole design is easy to build and if it works correctly, why bother with something you need to adjust?

Was there any particular reason given why the green is off on the bit-c128 board? I would never buy anything from that particular seller even if it were flawless, but it was claimed that this board was "improved" over the J. Alex design to display more accurate colors : http://atariage.com/forums/topic/240783 ... -v11-2015/

Microbee Technology also sells a converter board for the Gombes, http://www.microbeetechnology.com.au/st ... m-kit.html, but you have to build it yourself. I don't know how good it is compared to the bit-c128 board.

[alanh]

I was thinking of doing a converter board with CGA/MDA/HGA/EGA in/out, VGA out - for some modes as it would only include a line doubler and not a scaler, and USB in/out (USB video class capture). Is there that much demand for something like that? Would be in the $50-60 price range. Could be cheaper with just RGBI to VGA though.

[KenG]

I would think there would be plenty of demand for such a device. Especially if the colors were right without adjustment. CGA and EGA monitors are getting hard to find and get repaired. I would think that most vintage computer hobbyists would want one or more. If we could also do a case for it or design it to fit in an existing case, that would be great.

Should this discussion be moved to a more appropriately named thread?

[alanh]

Yes, but will be a while before I can get to it. I want to get the new JR stuff done first. I've been derailed again for the past few days. But I'm getting back to it today.

[GHiero]

There would be interest for sure, especially because there is no known device that can translate 18KHz MDA/Hercules or 21KHz High-Res EGA to VGA. To be fair, VGA is usually very backwards compatible with EGA, so a VGA capture card can usually do justice to EGA output. I know that VGA does not line double or scale 350-line EGA or emulated MDA, I believe it just adds more vertical blanking to compensate.

[alanh]

Damn this forum for encouraging rabbit holes....

On the input side, I believe I've come up with a scheme - at least on paper - that would auto-magically sync to pretty much any non-standard timing rather quickly. It would also lock and track any crystal drift and make sure the captured image is always stable. Basically the 'auto sync' function on most VGA monitors but a little more advanced.

For output, I looked at the USB Video Class (UVC) spec some more. That is looking like a long-shot due to lack of low bit per pixel support in the uncompressed payload modes. The RGB data itself isn't huge. Worst case EGA 640 x 350 x 6 bpp x 60 fps = 10 MBytes/s which a Cypress FX2LP can easily handle. Even if 6 bpp was possible, there is no LUT/palette support in UVC. Being able to tweak color renditions w/o using trim-pots is a feature I want to keep. Using more than 6 bpp blows up USB bandwidth (realistically ~33 MBytes/s tops). But not using UVC means writing a custom video input driver or application. I could handle this for Linux but I wouldn't know the first place to start in a Windows world.

I do have an alternative for output. I thought about and researched pretty much all the information I need to make a MIPI CSI-2 output. This would allow for direct connection to a Raspberry Pi. From there you could just open the v4l video player on your desktop if you are, for example, a vintage computer enthusiast giving a presentation at a VCFestival with a connected HDMI projector. Or you could encode to h.264 and stream it to a net player or file recording. Or in the future one might write an app to propagate each frame applying things like HQnX or nXBG scalers. /shrug. Just brainstorming.

[Trixter]

I wish J. Alex posted PCB layouts so other people could have boards made from OSH Park.

He shared the design; I believe this is the link: https://oshpark.com/shared_projects/9MPsJlrJ
It is on my to-do list to build once I build some smaller projects first to get my skills up, something I will be doing tomorrow afternoon.

Was there any particular reason given why the green is off on the bit-c128 board?

In my discussions with the designer, he designed his board by measuring the post-conversion voltage values of his actual Commodore 1084S-D monitor. It's my theory that his monitor was faulty/damaged/non-calibrated when he did his measurements.

I am not intentionally trying to paint him in a negative light, only reporting my experience. I tested two boards, including one certified by him as 100% functional before he sent it to me, and got identical wrong green levels with multiple capture devices and monitors. I even took the entire setup, Gonbes and all, to a vintage computer gathering so that its output could be compared to a real C128 hooked up to a 1084S-D, and no amount of Gonbes POT adjustment could make it look right.

Microbee Technology also sells a converter board for the Gombes, http://www.microbeetechnology.com.au/st ... m-kit.html, but you have to build it yourself. I don't know how good it is compared to the bit-c128 board.

I have one that is built, but it needs power and a few components, and will test it when I gain the experience to understand how to get it to work.

[Trixter]

I was thinking of doing a converter board with CGA/MDA/HGA/EGA in/out, VGA out - for some modes as it would only include a line doubler and not a scaler, and USB in/out (USB video class capture). Is there that much demand for something like that? Would be in the $50-60 price range. Could be cheaper with just RGBI to VGA though.

I can't speak to general demand, but I would find a way to have your first-born child if you could create such a board. However, I'm worried by your intention to make it a USB video class capture device. Would it be transparent? For example, CGA RBGI is ~240 lines at 59.92 Hz (not a typo). Would the output of such a board provide the same framerate linedoubled? Any 4:4:4 colorspace at that frame size and rate exceeds USB 2.0's available bandwidth.

I'm much more interested in conversion than capture; I already have dedicated VGA capture devices.

As for cheaper RGBI to VGA, see the link to the previous board design. It seems very inexpensive.

[alanh]

I've already corrected my thoughts on USB in the post above. See the notes about making a conversion device to a MIPI CSI-2 camera input for use with a Pi. Or a proprietary USB format.

CGA - what I call 'Mode 1' - is 59.9227510135 Hz to be more exact. 912 total horizontal pixels (64 sync, 140 FP, 640 ACT, 68 BP) x 262 total lines (3 sync, 34 FP, 200 ACR, 25 BP) from a 14.318 (18 repeating) MHz reference. All other CGA modes are just pixel doubling and quading.

Just looking at the latest incarnation from the C128 guy (the SMT version), the component selection is very obvious. It's a discrete logic sync combiner, input Schmidt's for the RGBI signals, and an output R2R ladder.

VGA DAC quality can vary a lot. I've built more than my fair share of boards professionally. Just 3 op amps on the final drive stage of an R2R ladder can improve signal rendition tremendously as the transmission impedance won't load down the ladder. But if quality is the final motivation, use a VGA specific DAC. My go to part is a TI THS8135 as I've used it on quite a few boards so far and was going to reuse a layout here.