User blog comment:Hyp cos/BEAF isn't so strong/@comment-5150073-20130921115609/@comment-10429372-20130921141513

Using my new definition, {X,X / 2}+1&n would be approximately f_LVO*w(n), in fact, if #&n reach f_a(n), #+1&n will reach f_a*w(n), so Ikosarakt1 will be correct, I think, but I'll analyse this later.