Samsung G9 won't display — HDMI switch fixes it

The AVI InfoFrame and control period fixes (PRs #35 and #38) worked. Every flat-panel monitor I tried now displays the A2FPGA output cleanly — including monitors that previously showed nothing at all.

Except the Samsung Odyssey Neo G9. Plugging the A2FPGA directly into the G9 produces a blank screen. No image, no “no signal” banner, nothing. The monitor clearly receives something — it doesn’t go into power-saving mode — but whatever it receives, it refuses to display.

Tried the obvious things: different HDMI cables, different HDMI ports on the monitor, power cycling both devices in different orders. No change.

On a whim, ran the A2FPGA into an HDMI switch, then from the switch to the G9. Everything displayed immediately and has been stable since.

The switch isn’t doing anything visible — no upscaling, no OSD showing it doing format conversion. It’s a passive-ish device with its own signal drivers. My best guess is that the Tang Nano 20K’s HDMI output is electrically marginal: signal levels or TMDS pre-emphasis that falls within tolerance for most consumer displays but outside whatever the G9 requires. The switch regenerates the signal with its own output stage and that’s enough.

The G9 is a high-end gaming monitor with unusually aggressive HDMI compliance requirements. It makes sense that it would be less forgiving than a generic flat panel. The workaround — HDMI switch in the chain — is reliable and costs almost nothing. But the root cause is almost certainly in the A2FPGA’s PCB HDMI output stage or the FPGA I/O drive configuration, not in the protocol fixes already applied.

Noting this here because it will affect anyone who tries the A2FPGA with a strict gaming monitor and gets a blank screen.