David Banks

Sorry, there was an error in v53, here's v54 that fixes that. [YUV_CPLD_v54.zip](https://github.com/hoglet67/RGBtoHDMI/files/4008500/YUV_CPLD_v54.zip) This doesn't seem to work particularly well for me (v52 was better I thinkl, as was v53...

I've also notices those glitches, so yes, some filtering may help. The sync pulse is a bit assymetic, specifically: - the leading (rising) edge is ~50ns (from black to sync)...

Ian, I've just pushed YUV CPLD v56 that is based on v52 (single counter) and adds 2-bit HS deglitching. This should ignore glitches of 1, 2 or 3 samples (i.e....

Ian, One last experiment for this morning... I've just pushed YUV CPLD v57 that uses the leading (rising) edge of HS as the timing reference. With this, I'm seeing far...

I think there's one more uncontrolled factor possibly coming into play here - the quality of the power supply. If I power the Spectrum from my bench power supply, there...

> Does this version still work with the Atom? It should do (the Atom's sync pulse width is 4.61us (16.5/3.579545) which is almost identical to the Spectrum's 4.57us (32/7). But...

One thing I noticed is that the auto calibration doesn't work. I think it's because it's operating directly on the frame buffer, before the palette is applied.

I'm happy to see the calibration summary screen split.

Even if you could find a spare GPIO pin, I very much doubt there is sufficient memory bandwidth on the Pi to deal with copying 24BPP data into memory (the...

> Is this due to the interface between the cpu and the rest of the chip? I mean, the thing is capable of 1080p (which is obviously not needed here),...