Re: [Gretl-devel] gretl and openmp

Hi, I know this is very old topic but I have new toy: AMD Phenom II X6 1100T + 6 GB RAM so I've decided to check how it handles with OpenMP. Results: with OpenMP: 40000: 0,16 ( 0,249 Mflops) 160000: 0,31 ( 0,523 Mflops) 640000: 0,85 ( 0,754 Mflops) 2560000: 1,04 ( 2,467 Mflops) 10240000: 2,20 ( 4,662 Mflops) without OpenMP: 40000: 0,08 ( 0,500 Mflops) 160000: 0,23 ( 0,696 Mflops) 640000: 0,78 ( 0,821 Mflops) 2560000: 3,85 ( 0,665 Mflops) 10240000: 12,75 ( 0,803 Mflops) Having in mind Jack's question asked at 2nd Gretl Conference: "what should be done before Gretl jumps to 2.0" in my opinion is/could be to allow (power) user to use OpenMP inside scripts dividing big loops into multiple cores of CPU.