Busy, busy, busy. Spending this week working in Oregon (with a very slow connection from the hotel) and getting ready for a week in Mexico (for work) later this month.
Here's how the judges tabulated points awarded for this problem set:
Intel Technology Used. Same as in previous problems; 25 points for TBB and 10 points for each other Intel software tool (compiler, MKL, IPP, etc.) incorporated.
Elegance. A subjective rating of 0-20 points on the documentation included with and in the program. Ten points were given for each of requirements given in the problem statement (the sorting of the smallest 20 path lengths from low to high, the sorting of the largest 20 path lengths from high to low, and using the file "path.out") for 30 more points. A maximum of 50 points were possible in this category. Up to 10 bonus points were awarded to several entries that included some form of input checking (even though the judges had guaranteed input would be problem free). As one entry stated, "Trust, but verify."
Execution & Time Score. Three input data files were used containing 3000, 4500, and 6000 node graphs. The total execution time for the three runs was added together. The minimum total was awarded 100 points. A base-10 logarithmic function was used to award points for the other entries. Essentially, the number of digits in the difference in an entry's total time from the minimum would knock off ten points. A total time in the range of 100-999 results in a score in the 70's.
Other. Five points awarded for each forum post, up to the maximum 50 points.
For the three data sets, the entry from "dgeld" was the fastest code, but only by a single second difference. The judges had created data sets with 15000 and 24000 nodes, but found that many entries took much too long to process these files. of those entries that were able to handle the two larger data sets in a timely fashion, the entry from "klampi" ran fastest.
--clay
"It's all very complicated and would take a scientist to explain it." -- MST3K
