ESSL_PRB Tests for the ESSL library. TEST01 DSRIS is an iterative solver for a linear system. Number of iterations taken = 9 Relative accuracy = 0.292208E-16 Computed Exact Error Solution Solution 1.00000 1.00000 0.222045E-15 1.00000 1.00000 0.111022E-15 1.00000 1.00000 0.333067E-15 1.00000 1.00000 0.222045E-15 1.00000 1.00000 0.00000 1.00000 1.00000 0.00000 1.00000 1.00000 0.00000 1.00000 1.00000 0.00000 1.00000 1.00000 0.444089E-15 TEST02 SCSINT sets up cubic spline interpolation. The number of data items is N = 7 The (X,Y) data: 1 -4.999922 0.001597 2 -4.529554 0.001946 3 -3.684622 0.002938 4 -2.810408 0.005039 5 -0.413499 0.189590 6 0.327672 0.271427 7 2.556053 0.006085 Call SCSINT. X, interpolated Y, Runge(X): 1 -4.923018 0.001623 0.001648 2 -4.654279 0.001787 0.001843 3 -4.465384 0.002057 0.002002 4 -4.331578 0.002332 0.002127 5 -4.080351 0.002852 0.002397 6 -1.165844 0.103086 0.028588 7 -1.164979 0.103171 0.028629 8 -0.825140 0.139362 0.055490 9 0.194163 0.261895 0.514805 10 0.269287 0.267726 0.355506 11 0.297002 0.269579 0.311988 12 0.889766 0.269059 0.048095 13 1.711494 0.170215 0.013472 14 1.788647 0.156911 0.012349 15 1.792964 0.156151 0.012290 16 1.867727 0.142771 0.011337 17 3.309653 -0.142307 0.003638 18 3.461668 -0.168198 0.003327 19 4.304365 -0.261167 0.002154 20 4.346929 -0.262899 0.002112 TEST03 SGEEV computes the eigenvalues and eigenvectors of a real, general matrix. Computed eigenvalues: 1 121.000 0.00000 2 -4.00000 0.00000 3 -4.00000 0.00000 4 -4.00000 0.00000 5 -4.00000 0.00000 6 -4.00000 0.00000 7 -4.00000 0.00000 8 -4.00000 0.00000 ...... .............. 25 -4.00000 0.00000 Eigenvectors #1 and #N: 1 -0.200000 0.00000 0.00000 0.00000 2 -0.200000 0.00000 0.00000 0.00000 3 -0.200000 0.00000 0.00000 0.00000 4 -0.200000 0.00000 0.00000 0.00000 5 -0.200000 0.00000 0.00000 0.00000 6 -0.200000 0.00000 0.00000 0.00000 7 -0.200000 0.00000 0.00000 0.00000 8 -0.200000 0.00000 0.00000 0.00000 9 -0.200000 0.00000 0.00000 0.00000 10 -0.200000 0.00000 0.00000 0.00000 11 -0.200000 0.00000 0.00000 0.00000 12 -0.200000 0.00000 0.00000 0.00000 13 -0.200000 0.00000 0.00000 0.00000 14 -0.200000 0.00000 0.00000 0.00000 15 -0.200000 0.00000 0.00000 0.00000 16 -0.200000 0.00000 0.00000 0.00000 17 -0.200000 0.00000 0.00000 0.00000 18 -0.200000 0.00000 0.00000 0.00000 19 -0.200000 0.00000 0.00000 0.00000 20 -0.200000 0.00000 0.00000 0.00000 21 -0.200000 0.00000 0.00000 0.00000 22 -0.200000 0.00000 0.00000 0.00000 23 -0.200000 0.00000 0.00000 0.00000 24 -0.200000 0.00000 -0.707107 0.00000 25 -0.200000 0.00000 0.707107 0.00000 TEST04 SGEF factors a general matrix; SGES solves a linear system; System matrix A: 1.00000 2.00000 3.00000 4.00000 5.00000 6.00000 7.00000 8.00000 0.00000 Right hand side b: 14.0000 32.0000 23.0000 Solution X: 1.00000 2.00000 3.00000 TEST05 SGEMM multiplies two real general matrices. Matrix order N = 5 Product matrix C (Should be the identity matrix) 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000 0.0000 0.0000 0.0000 0.0000 1.0000 TEST06 SGETRF factors a general matrix; SGETRS solves a linear system; Computed solution: 1 1.000000 2 1.000000 3 1.000000 4 1.000000 5 1.000000 6 1.000000 7 1.000000 8 1.000000 ...... .............. 25 1.000000 TEST07 SGLNQ integrates a function F(X) from A to B. We will integrate log(x)/sqrt(x) from 0 to 1. The exact answer is -4. We will increase N, the quadrature order. N Result Abs(Result+4) 1 -0.980258 3.01974 2 -1.824280 2.17572 3 -2.277225 1.72277 4 -2.562461 1.43754 5 -2.760170 1.23983 6 -2.906104 1.09390 8 -3.108496 0.891504 10 -3.243250 0.756750 12 -3.340038 0.659962 14 -3.413242 0.586758 16 -3.470725 0.529275 20 -3.555559 0.444441 24 -3.615453 0.384547 32 -3.694968 0.305032 40 -3.745764 0.254236 48 -3.781226 0.218774 64 -3.827806 0.172194 96 -3.877641 0.122359 128 -3.904215 0.957854E-01 256 -3.947364 0.526361E-01 TEST08 SNRAND generates normal random numbers. Initial seed = 123456789. Number of values to generate = 500 Random vector: 1 1.242957 2 0.232724 3 -0.134458 4 -0.688826 5 -0.307862 6 -0.495332 7 -0.430708 8 0.126488 ...... .............. 500 2.214809 Current seed = 0.134701336E+10 Number of values to generate = 500 Random vector: 1 -1.491459 2 -0.283386 3 1.089546 4 -0.204865 5 -1.359305 6 -1.983673 7 -2.183466 8 0.645490 ...... .............. 500 -0.258203 Now restore seed to initial value, and recompute the two sets of data at one call. Current seed = 123456789. Number of values to generate = 1000 Random vector: 1 1.242957 2 0.232724 3 -0.134458 4 -0.688826 5 -0.307862 6 -0.495332 7 -0.430708 8 0.126488 ...... .............. 1000 -0.258203 Maximum value = 3.09255 Minimum value = -2.92059 Average value = 0.215986E-01 TEST09 SPINT sets up polynomial interpolation. The number of data items is N = 7 The (X,Y) data: 1 0.474645 0.457023 2 2.624530 0.494329 3 4.159994 -0.851270 4 6.539189 0.253217 5 7.011906 0.665916 6 7.621980 0.973208 7 9.103209 0.316056 Call SPINT. X, interpolated Y, SIN(X): 1 0.726859 0.700815 0.664526 2 2.377744 0.712183 0.691706 3 2.470389 0.634242 0.621929 4 2.727100 0.395661 0.402726 5 2.749068 0.374078 0.382522 6 3.282342 -0.165832 -0.140285 7 3.653387 -0.509787 -0.489742 8 4.364114 -0.931415 -0.939963 9 4.777318 -0.976780 -0.997893 10 6.316347 0.035705 0.033156 11 6.326385 0.045583 0.043186 12 6.515185 0.230112 0.229924 13 7.226604 0.810675 0.809569 14 7.360818 0.882297 0.880840 15 7.533558 0.950013 0.949102 16 7.564105 0.958930 0.958279 17 7.664948 0.981557 0.982186 18 8.847071 0.509279 0.546105 19 9.825502 0.245431 -0.390085 20 9.910374 0.339693 -0.466735 TEST10 For real fast Fourier transforms, SRCFT computes the complex FFT of real data. The number of data items is N = 10 The original data: 1 0.359265 2 0.166507 3 0.486517 4 0.897656 5 0.909208 6 0.060564 7 0.904653 8 0.504523 9 0.516292 10 0.319033 Initialize the FFT routine. Compute the FFT coeficients. The Fourier coefficients: 1 5.12422 0.00000 2 -0.899346 -0.258600 3 -0.815293 0.397972 4 1.03227 0.354306 5 -0.697243 -0.309880 6 1.22765 0.00000 Reinitialize the FFT routine for inverse operations. Recover the data from the FFT coeficients. The retrieved data: 1 0.359265 2 0.166507 3 0.486517 4 0.897656 5 0.909208 6 0.060564 7 0.904653 8 0.504523 9 0.516292 10 0.319033 TEST11 SSORT sorts a vector of real data. The number of data items is N = 1000 The original data: 1 0.986642 2 0.493977 3 0.266144 4 0.090733 5 0.947764 6 0.073749 7 0.500707 8 0.384142 ...... .............. 1000 0.501854 Call SSORT. The sorted data: 1 0.000040 2 0.000595 3 0.000879 4 0.001657 5 0.003000 6 0.003229 7 0.004254 8 0.004816 ...... .............. 1000 0.999458 TEST12 SURAND generates real random numbers. Initial seed = 123456789. Number of values to generate = 500 Random vector: 1 0.218418 2 0.956318 3 0.829509 4 0.561695 5 0.415307 6 0.066119 7 0.257578 8 0.109957 ...... .............. 500 0.681473 Current seed = 0.146345281E+10 Number of values to generate = 500 Random vector: 1 0.522061 2 0.284688 3 0.750461 4 0.994292 5 0.068474 6 0.850280 7 0.653327 8 0.474202 ...... .............. 500 0.819771 Now restore seed to initial value, and recompute the two sets of data at one call. Current seed = 123456789. Number of values to generate = 1000 Random vector: 1 0.218418 2 0.956318 3 0.829509 4 0.561695 5 0.415307 6 0.066119 7 0.257578 8 0.109957 ...... .............. 1000 0.819771 Maximum value = 0.997908 Minimum value = 0.183837E-02 Average value = 0.503040 ESSL_PRB Normal end of execution.