This is the same interface as before, it has 4 unsigned to ascii algos including a new one from brethren that has been tweaked to get its speed up. All algos are proven exhaustively over the full unsigned range and there is a subdirectory that has 2 of the testbeds to do this.
AMD Sempron(tm) Processor 3100+
Results 8 pass average
timing utoa 601 ms
timing utoa_ex 187 ms
timing utoa2 324 ms
timing utoa3 560 ms
P3, 500MHz
Results 8 pass average
timing utoa 2591 ms
timing utoa_ex 1023 ms
timing utoa2 1609 ms
timing utoa3 2189 ms
(http://www.masm32.com/board/index.php?action=dlattach;topic=15438.0;id=8516)
The attached zip file contains only the above displayed image. Please, don't waste your bandwidth by downloading it.
Gunther
Intel(R) Atom(TM) CPU N270 @ 1.60GHz
Results 8 pass average
timing utoa 1451 ms
timing utoa_ex 535 ms
timing utoa2 861 ms
timing utoa3 1451 ms
AMD Turion(tm) 64 X2 Mobile Technology TL-52
Results 8 pass average
timing utoa 674 ms
timing utoa_ex 201 ms
timing utoa2 369 ms
timing utoa3 630 ms
Apparentely everybody is trying to optimize his own bin to ascii PROC. :P
Intel(R) Core(TM)2 CPU 6600 @ 2.40GHz
Results 8 pass average
timing utoa 372 ms
timing utoa_ex 163 ms
timing utoa2 216 ms
timing utoa3 296 ms