After seeing a problem by @Hector Ricardez , I got interested in a certain function: If I zoomed out on its graph , then it would seem to be linear.
If the linear approximation can be represented as where and are integers, then what is
Bonus: Why?
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Note that for x < 0 the series diverges, so for the problem, we will assume x ≥ 0 . Then y = f ( x ) = n = 1 ∑ ∞ ⌊ 5 n x ⌋ = n = 1 ∑ ⌊ lo g 5 x ⌋ 5 n x − { 5 n x } = 4 x − 4 x ⋅ 5 − ⌊ lo g 5 x ⌋ − n = 1 ∑ ⌊ lo g 5 x ⌋ { 5 n x }
Zooming out the graph is a uniform scaling of both axes, which means we should consider substitute ( r x , r y ) for ( x , y ) and let r → ∞ . Doing this gives r y = 4 r x − 4 r x ⋅ 5 − ⌊ lo g 5 r x ⌋ − n = 1 ∑ ⌊ lo g 5 r x ⌋ { 5 n r x } Dividing by r and subtracting 4 x from both sides before applying the absolute value: ∣ ∣ ∣ y − 4 x ∣ ∣ ∣ = ∣ ∣ ∣ ∣ ∣ ∣ 4 x ⋅ 5 − ⌊ lo g 5 r x ⌋ + r 1 n = 1 ∑ ⌊ lo g 5 r x ⌋ { 5 n r x } ∣ ∣ ∣ ∣ ∣ ∣ ≤ 4 x ⋅ 5 − ⌊ lo g 5 r x ⌋ + r ⌊ lo g 5 r x ⌋ r → ∞ 0 , for any fixed x
It follows that the apparent graph by scaling is y = 4 x , ⟹ m = 4 , b = 0 so that the answer is m − b = 4
Remark: Note that this is really the only reasonable interpretation, where we scale all our variables by a scale factor r and then redefine our unit along the axes to be that same scale factor r . Otherwise, we would have to deal with the fact that all values of b actually look like zero after enough scaling, which sounds like a good thing but it actually means that the value of b in the problem can be considered to be undefined unless we redefine the unit