Twin Shifter mod Prototype build results

[mrbombermillzy]

@sporniket

That looks great. Im quite busy with easter stuff at the moment, but will look into it at some point soon.

Thank you

[sporniket]

No problem, it's also busy on my side.

[mrbombermillzy]

Could you maybe construct some sort of bitwise table/chart of which shifter positions are falsely repeated/missed by each shifter?

Here it is. I hope it is clear enough.

Just to let you know, this is the first chance I have had to have a proper look at this.

There seems to be 2 types of anomaly here.

For the first, which is where some of the bits set on both shifters cause errant unrelated bits to be set: I'm starting to look at RGB colour channel transformation to achieve minimal colour error. (This would be SO much simpler if there were not 5 consecutive bits with erroneous bits set!) Hopefully, once I've figured out the results, I can use the data to intercept bad bit combos and slightly tweak the colour so that (hopefully) it's unnoticeable to the naked eye. Anyway, more later, once I have some numbers to crunch/absorb and results to show.

As for anomaly 2, where one bit is set in only one shifter; I can't remember if you have something on the new board rev. that fixes this? (If not we will have to discuss/recap and I will have then a look at it too).

[sporniket]

For the first, which is where some of the bits set on both shifters cause errant unrelated bits to be set: I'm starting to look at RGB colour channel transformation to achieve minimal colour error. (This would be SO much simpler if there were not 5 consecutive bits with erroneous bits set!) Hopefully, once I've figured out the results, I can use the data to intercept bad bit combos and slightly tweak the colour so that (hopefully) it's unnoticeable to the naked eye.

You can try that for the challenge of doing it, and I believe that you should not lose too much time on that, since it is a hardware bug that MUST be fixed. I plan to fix the schematic next month (now is school holidays), so I hope that a new version should be ready by July. Now that the manual interruptor is not needed anymore, I can repurpose the selector so that when the Shifter /CS is not asserted (Shifter not selected to read/modify internal register), the second shifter always receive the same data than the main shifter.

I also believe that it is best to leave your current board untouched, as it is working enough to allow you to experiment with it.

(edit : fixed typos and missing words)

[mrbombermillzy]

Ok then. With my time constraints, it's maybe better time spent for me to hang on if you are confident that you can get those issues fixed.

I have plenty of challenges elsewhere

Will return to coding the 16bpp/16 col image viewer.