Hi Crusty, I tried what you advise above. I have a pair of Xiao esp32-s3s with wio 1262 boards on the header connectors. I loaded ping pong sketches into both and was previously getting RSSI values of -100 dBm or lower with antennas very close. I now get RSSIs of -70 dBm or so, which is an improvement, but nothing like the figures you appear to be getting.
Wondering if there is anything else you are doing… I tried the same experiment with a pair of Heltec v3s and got RSSIs of about -10dBm which is what you would expect, and what you were getting…