ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let bb be the base and XX the HWM-convergent index. Let NCD(b,X)NCD(b,X) be the number of correct digits and set

number of zeroes⁡=X,\operatorname{number\ of\ zeroes}=X, exponent⁡=NCD(b,X)+X+2.\operatorname{exponent}=NCD(b,X)+X+2.

HWM convergent error formula conjecture. The error of the HWM convergent is

error⁡={1.0×b−(exponent⁡−1)base b≥5, X=0,(b−1).1b×b−exponent⁡base b≥3, X=1,1.012×b−exponent⁡base b=2, X=2,(b−1).00…00⏟number of zeroes⁡(b−1)(b−1)b×b−exponent⁡b=2, X≥3; b≥3, X≥2.\operatorname{error}=\begin{cases} 1.0\times b^{-(\operatorname{exponent}-1)} & \text{base }b\geq5,\ X=0,\\ (b-1).1_b\times b^{-\operatorname{exponent}} & \text{base }b\geq3,\ X=1,\\ 1.01_2\times b^{-\operatorname{exponent}} & \text{base }b=2,\ X=2,\\ (b-1).\underbrace{00\ldots00}_{\operatorname{number\ of\ zeroes}}(b-1)(b-1)_b\times b^{-\operatorname{exponent}} & b=2,\ X\geq3;\ b\geq3,\ X\geq2. \end{cases}

The formula is presented as a unified description of the observed errors, with mantissas written in base bb and exponents in base ten in the examples. The source gives no evidence that the formula has been proved or refuted.

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.