ANALOG EXPERIMENTAL ttgf0p3 UART/SPI-controlled ASIC lab for exploring true-random number generation and PUF-style hardware entropy sources. - gojimmypi/ttgf0p3-analog-UART-FSM-TRNG-Lab
✅ MichaelBell reported @urish's USB CDC (Serial) Device as working: Works well. Tested with a loopback on the UART, and with an external UART. I pasted a fair amount of text to check for flow control issues and didn't see any errors.
✅ MichaelBell reported @urish's USB CDC (Serial) Device as working: Works well. Tested with a loopback on the UART, and with an external UART. I pasted a fair amount of text to check for flow control issues and didn't see any errors.
So as you might have been able to guess, I’m looking at integrating this with TinyQV. Specifically, on that one pin row version that should be able to run at the required 48MHz
✅ rohanverma94 reported @RebelMike's TinyQV Risc-V SoC as working: CORDIC peripheral on TinyQV works. Tested sine and cosine for angles from 0 to 2π. Check the results and the MicroPython script here.
Problem: the core only converges within ±99.9° (sum of micro-rotations, Σatan(2⁻ⁱ)). Push it past that and the rotator diverges into garbage.
Fix: fold every angle into [-π/2, π/2], track the quadrant, adjust signs accordingly.
Result: accurate across the whole range. RMS error ~0.23e-3, mean absolute error ~0.16e-3.
Worst case lands right at the fold boundaries (±90°), where the input angle is largest so the rotator runs the most micro-rotations and piles up the most residual, hence peak error 0.57e-3 at 168.75°/348.75°.
Folding checks out: all four quadrants mirror each other cleanly, and the exact angles (0/90/180/270°) come out within 0.6e-4.
1
1
3
1