Research notes on every problem, and a library of the papers behind them. Built from the open erdos repository.
Updated
This proves L126. Only the increment bounds are needed of the large path in the finite construction.
One representation at each smaller dyadic scale
Put , and . For a dyadic in the stated range,
Choose an integer of the form in the interval . This is possible because the interval has length , whereas the spacing is at most . This also defines an integer , and gives
In particular is an integer in . Write
The increment bounds give , so and . Our choice of therefore ensures
There is an admissible path with . To construct it, put and . Equation (2) implies . Concatenate copies of the word and copies of the word , in any order. The resulting word has length and sum . It contains neither nor , including at the concatenation boundaries. Extend it by repetitions of and truncate at length . This defines the required admissible .
Use the bare points, with cluster shifts both zero:
Equations (1) and give the exact integer equality
Thus no discarded carry or approximate equality is involved.
Distinct pairs and an infinite family
For these small summands,
The intervals are disjoint at successive dyadic values of : the next lower endpoint is . Hence all are distinct, so (3) gives distinct ordered pairs in the actual sets, even after taking unions. The number of available dyadic scales is
For the infinite assertion choose dyadic scales with . At step , prescribe any admissible large path at and apply the finite construction. Every smaller scale used in this step is larger than , so none of its paths has been assigned earlier. After these assignments choose arbitrary admissible paths on the unused scales, including all . No earlier path is changed, and taking further unions preserves each displayed set of distinct pairs.
For , the target satisfies . Consequently
The tend to infinity, proving the claim.