Toby Flenderson
Toby Flenderson
> I have been trying to work on this issue, however, it seems to be a bigger problem than just 1,000,000,000/64487. The calculation always results in zero when any form...
In addition 1000000001/64487 is also 0
In my opinion, the problem can be solved by rewriting one of the three functions intrat, andrat, and flatrat. Rewriting the intrat function should be the easiest. here is my...
I mean, in this loop, sometimes the index 'i' may be equal to x.used, but usually, index 'i' should be less than x.used(Because if i==x.used it may be out of...
> Sorry for the...ahem...short delay but here I am. > > It is possible because it can depend on `x.alloc > x.used` and a zeroed "overhang". > > This fails...
Can it be modified like this: check whether x.used is less than x.alloc before executing the loop, if x.used is equal to x.alloc, grow the memory usage of x, so...
I ran the program you wrote on my computer, and the results of the program show that it seems that KNUTH's algorithm is a little bit faster than SCHOOL on...
> Why is it too big? Any specific problems beside runtime? No, there are no other problems. I change OLOOP and ILOOP to 68 just because it will cost me...
> Well, something to do for another day :-) "another day" means 10 months later. (JK) 😁 The year before last, I accidentally found the book "BigNum Math" more than...
> 第十七章已补充 非常感谢，辛苦了！