Steven Ross wrote:
From my reading, it actually has a larger L1 cache (and smaller L2 cache), and for that purpose a MAX_SPLITS as high as 13 might be better. Increasing LOG_MIN_SPLIT_COUNT might also speed up your test a little.
I'm afraid adjusting those makes it slower.
OK, I've put a 14 Mbyte file called placenames.dat.bz2 in the directory http://chezphil.org/tmp/ (I've not written the URL in full to stop search engines from wasting their time on it). When un-bziped you'll find a tab-delimited 3-column text file. I got basically the same results using the latitude and longitude columns.
What's the best way to read data like that from a file? fscanf("%f %f %s\n")? Won't that have problems if there is a space? I don't do much file parsing in C/C++.
Nor me. awk it into something else first. You can probably use fscanf("%f\t%f\t%[^\n]") or try cin >> lng >> lat; getline(cin,name);. Phil.