Auto comment: topic has been updated by YogeshZT (previous revision, new revision, compare).

Auto comment: topic has been updated by YogeshZT (previous revision, new revision, compare).

Auto comment: topic has been updated by YogeshZT (previous revision, new revision, compare).

If I am not mistaken, the TLE approach is looking for an upper bound of x in the block of memory delimited by $$$[m.begin, m,end()) $$$, while the AC code actually makes use of multiset::iterator to find the answers.

The TLE approach may work with vectors, but be careful when using it with other data structures.

What i have found is that multiset.upper_bound() is faster than upper_bound(multiset.begin(),multiset.end()) and the reason attributed is the fact that multiset.upper_bound is optimized to take advantage of the internal structure of a multiset, which is typically implemented as a balanced binary search tree (e.g., a Red-Black Tree). The member function can directly leverage the tree's properties to perform the lookup efficiently as compared to the general upper_bound function which is from algorithm header file and it is not designed to take advantage of the properties of multiset , which the member upper_bound function does.

You can check this out too if you want : https://stackoverflow.com/questions/64536493/difference-between-set-upper-bound-and-upper-boundset-begin-set-end-stl

Sorry if there is any grammatical error.

lower_bound(b.begin(), b,end(), val) is $$$O(n)$$$,

while b.lower_bound(val) is $$$(\log n)$$$