Definitely not pretty.
How does the 16MHz line look versus the 32 (as per Exxos' images above)?
BW
Definitely not pretty.
Assuming both your probes are on X10, then press the CH1 button and make sure the PROBE menu is on X10 (Also check for CH2)
Holy about-10ns-near-perfect-out-of-phase transitions, Batman!
Doubt it. I've seen this odd type of skew before. I've even talked about it in the past. If you think about it, a perfect sync on falling edges is almost impossible, you will always have delays. But the clocks ARE in sync. So Atari must have added internal buffers to equal the delays. So they changed manufacture of chips, those delays no longer add up. Considering the skew seems to be a few ns to short, it fits the "newer chips are faster" type of thing, so timings get shifted and things screw up.Badwolf wrote: ↑Wed Jan 12, 2022 10:06 am @exxos does this smell deliberate to you? the peaks and troughs are almost perfectly in phase rather than the edges. Such a stark difference from the older chip makes me wonder if it were introduced to match the requirements of a newer MCU or something?
Dunno. I could be seeing ghosts but seems awfully shonky if it's by accident.
I occasionally see something like that in Kicad 5. Solution is normally to lift the component. Recalculate the flood, place the component again.