Ln -sf /System/Library/Frameworks/amework/Frameworks/amework/Versions/Current/libBLAS.dylib libRblas.dylib cd /Library/Frameworks/R.framework/Versions/4.1-arm64/Resources/lib Version.string R Under development (unstable) ( r80062)ģ,500,000 Fibonacci numbers calculation (vector calc): 0.127 (sec). Comparing my numbers with yours, I see much slower times on "2,500 x 2,500 cross-product matrix (b = a' * a)" and "Inverse of a 1,600 x 1,600 random matrix". I'm not sure that anyone here noticed how much faster your M1-native numbers are. # Matrix calculation benchmarks (5 tests):Ĭreation, transp., deformation of a 5,000 x 5,000 matrix: 0.25 (sec).Ģ,500 x 2,500 normal distributed random matrix^1,000: 0.105 (sec). Macbook Pro M1 (16 GB RAM) using R version 4.0.3 using the platform arm-apple-darwin20.2.0 (64-bit) version from homebrew:ģ,500,000 Fibonacci numbers calculation (vector calc): 0.111 (sec). Matrix function benchmarks (5 tests): Cholesky decomposition of a 3,000 x 3,000 matrix: 4.37 (sec).ĭeterminant of a 2,500 x 2,500 random matrix: 5.53 (sec).Įigenvalues of a 640 x 640 random matrix: 0.793 (sec).įFT over 2,500,000 random values: 0.321 (sec). Matrix calculation benchmarks (5 tests): Creation, transp., deformation of a 5,000 x 5,000 matrix: 0.568 (sec).Ģ,500 x 2,500 normal distributed random matrix^1,000: 0.139 (sec). MBP16 2019 (RAM 16GB CPU 2.6GHz 6-Core intler Core i7) Programming benchmarks (5 tests): 3,500,000 Fibonacci numbers calculation (vector calc): 0.236 (sec). Matrix function benchmarks (5 tests): Cholesky decomposition of a 3,000 x 3,000 matrix: 3.97 (sec).ĭeterminant of a 2,500 x 2,500 random matrix: 1.94 (sec).Įigenvalues of a 640 x 640 random matrix: 0.435 (sec).įFT over 2,500,000 random values: 0.126 (sec). Matrix calculation benchmarks (5 tests): Creation, transp., deformation of a 5,000 x 5,000 matrix: 0.399 (sec).Ģ,500 x 2,500 normal distributed random matrix^1,000: 0.147 (sec). Grand common divisors of 1,000,000 pairs (recursion): 4.62 (sec).Ĭreation of a 3,500 x 3,500 Hilbert matrix (matrix calc): 0.18 (sec).Ĭreation of a 3,000 x 3,000 Toeplitz matrix (loops): 1.31 (sec).Įscoufier's method on a 60 x 60 matrix (mixed): 65.5 (sec). Mac mini m1 (RAM 8GB) Programming benchmarks (5 tests): 3,500,000 Fibonacci numbers calculation (vector calc): 0.15 (sec).
0 Comments
Leave a Reply. |