Trammell Hudson @th@v.st

well that's not quite right.

We have hsync and vsync!

Amsterdam’s underwater bike share.

Acquired a better hdmi adapter and now the 5x PLL on the TMDS clock input into the fpga LVDS looks solid.

3D printed puzzle based on Lee Swallow's "Geomagic Squares". The nine pieces form three 3x3 cubes in three different ways: https://www.thingiverse.com/thing:5846405

Linux guests are at least usable. Windows guests just go into a boot-loop with this BSOD. Rebooting isn't going to help -- those ACPI tables are in read-only firmware.

dear lazyweb: is there any way to pass through the subset of the ACPI tables for devices in addition to the PCI resources? Linux guests are oops'ing when trying to find the Intel HDA DMIC entries.

With a few extra pull-ups the hdmi cml signal is finally starting to lvds.

keep scrolling

Scientists have discovered a truly cursed bit endianess.

Press to reset your nun.

Moon shadows

The panel is properly mapped now, so maybe the remaining glitches are due to the IT6604 HDMI decoder configuring a YCbCr 4:2:2 mode rather than RGB.

Signs of life! Video input on hdmi isn’t mapped right, although it responds in reasonable ways.

LED display from the train still works. Now to graft the hdmi interface onto it.

I am so bad at this.

Hsync, Vsync, Pclk, and data confirmed.

I like to take as close to orthorectified photos of the PCB as possible and then align the pinout diagrams from the datasheets to see if anything matches. So far no luck on any of the HDMI interfaces in QFP-128 packages from the usual sources.

It's not an SII1127AC or ADV7619 based on the NC pins and TDMS inputs.

That is just rude.