I have generated a new batch of low weight k's.
These are much smaller than the k's Thomas11 posted before and hence they can use the Zero-padded FFT and then takes 1/3 of the time to test.
I have generated a new weight formula that is more accurate than extended nash weight.
Feel free to reserve the k not listed below. It would be nice if we can get all of these to 2M.
Code:
k new weight formula extended nash weight Tested upto
355262321784119 334 17 2 4.6M
706809843271963 348 51 2 400K
452565773039167 361 27 2 400K
596571372172469 373 23 2 400K
286565972092003 382 19 1 400K
265685529211859 387 32 1 400K
296990097378209 393 16 1 400K
586824605110333 406 29 2 400K
329604383181013 409 23 2 400K
449348815525481 413 19 2 400K
242081876749393 413 20 2 400K
151649765670817 430 41 2 400K
30402136144241 436 23 2 400K
569227451812807 437 20 2 400K
546088659477761 447 19 1 400K
174421373354771 447 20 2 400K
122586944321669 448 31 2 400K
427662785542799 450 39 2 400K
280418984909561 453 26 2 400K
9703351863493 453 32 2 400K
85877608135057 456 24 2 400K
571267621226129 458 50 2 400K
413338777986037 459 25 2 400K
508015032236651 466 21 2 400K
103432747403021 466 32 2 400K
606564403438897 468 19 1 400K
380289695293271 476 20 2 400K
294408177753337 477 22 2 400K
225184019593957 478 20 2 400K
633114842538613 479 34 2 400K
640520549022929 483 44 1 400K
456234152728799 484 23 2 400K
18047352362383 488 18 2 400K
290388549449203 488 22 2 400K
689369809338823 490 28 2 400K
578664594032099 492 19 2 400K
270003999160183 494 20 2 400K
70904559410129 494 38 2 400K
124855201083911 504 26 2 400K
129859680799207 505 23 2 400K
87131678294021 507 23 2 400K
309363900392467 508 94 1 400K
242539796323031 509 26 2 400K
481134684645673 513 19 2 400K
352406054088929 519 20 2 400K
583299325638817 519 26 2 400K
376161389883883 520 26 2 400K
612472684451819 520 30 2 400K
659819349704111 524 26 2 400K
151550479574083 533 27 1 400K