- Feb 28, 2025
-
-
BRONES Romain authored
-
- Nov 28, 2024
-
-
BRONES Romain authored
-
- Nov 20, 2024
-
-
BRONES Romain authored
-
BRONES Romain authored
* This add the documentation initiated in main
-
- Nov 13, 2024
-
-
BRONES Romain authored
-
- Oct 23, 2024
-
-
BRONES Romain authored
* Avoid breaking compatibility...
-
BRONES Romain authored
* Port Integrate and Dump to corr_iir
-
BRONES Romain authored
-
BRONES Romain authored
* Proper rounding on rounded output * Fixed point output has the same dimension as input
-
BRONES Romain authored
-
- Oct 22, 2024
-
-
BRONES Romain authored
-
BRONES Romain authored
-
BRONES Romain authored
-
- Jul 18, 2024
-
-
BRONES Romain authored
* Port Integrate and Dump to corr_iir
-
BRONES Romain authored
* Trying to simulate the StopToCurrent going to mean value * Simulation time is very very long... probably not a good idea
-
BRONES Romain authored
* Correction output is integrated and dumped every 1024 values * when corrector is disabled, switch output to averaged value
-
- Jul 11, 2024
-
-
BRONES Romain authored
-
BRONES Romain authored
-
BRONES Romain authored
* fix writting mistakes * add a non-rounded output for chaining * use that non rounded output as input of second filter
-
BRONES Romain authored
* It is the same corr_ll layout, but with fixed point * Also align matmul output to DAC units (provided that matrix coeffs are in DAC/nm, signed fixed point with 30 decimals)
-
- Jul 05, 2024
-
-
BRONES Romain authored
* Also I had a nasty GHDL error that was solved by this, not sure why...
-
- Jul 04, 2024
-
-
BRONES Romain authored
-
BRONES Romain authored
* align memory access * adapt to new rounding in filter
-
BRONES Romain authored
-
BRONES Romain authored
-
BRONES Romain authored
* add round of 8 bits on mult A * Add a pipeline stage
-
BRONES Romain authored
* increase coeff +2 bits * add saturation and proper rounding on result (sat 7 bits, round 15 bits) * In scale with Libera implementation of MatMult
-
- May 30, 2024
-
-
BRONES Romain authored
* Add constant for number of BPM * Change the register addresses
-
- May 29, 2024
-
-
BRONES Romain authored
* Not sure, but maybe there was a sign issue there
-
BRONES Romain authored
* for debug purpose
- May 22, 2024
-
-
BRONES Romain authored
* Allow ID up to 255 * This changes dimension of memories, registers
-
- Feb 02, 2024
-
-
BRONES Romain authored
-
- Jan 08, 2024
-
-
BRONES Romain authored
* X and Y corrector controls are split * Signed representation of coefficients
-
- Dec 14, 2023
-
-
BRONES Romain authored
* The memory length depends on the serializer counter
-
BRONES Romain authored
* Matrix lines are numbered from 0 to 99 * X from 0 to 49 included, Y from 50 to 99 included
-
- Nov 17, 2023
-
-
BRONES Romain authored
* Use DESYRDL specific property
-
- Nov 07, 2023
-
-
BRONES Romain authored
-
- Nov 06, 2023
-
-
BRONES Romain authored
* Two different register for enable and reset * Duplicate coeffs * Based on PSCID **table index**. > 50 is Y * Threshold autostop is kept on both plane. Control moved to a dedicated register. When reached, stop both planes.
-
- Nov 02, 2023
-
-
BRONES Romain authored
feat(average): Add orbit error and correction command moving average fix(sim): Fiw simulation to match two correctors, add traces for moving average
-