NVIDIA Quadro 4000 Driver Version 6.14.12.9573 MATLAB2011b (32 bit version) Corsair CSSD-F60GB2 Windows xp Professional x64 Edition Intel(R) Xeon(R) CPU E5405 @ 2.00GHz (8 cores) i = 100 CPU Time = 0.058 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.001 sec i = 200 CPU Time = 0.002 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.000 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.000 sec GPU Time for matmul and all data transfers including K = 0.001 sec i = 300 CPU Time = 0.003 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.000 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.001 sec i = 400 CPU Time = 0.006 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.002 sec i = 500 CPU Time = 0.006 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.002 sec i = 600 CPU Time = 0.001 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.002 sec i = 700 CPU Time = 0.008 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.002 sec i = 800 CPU Time = 0.007 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.003 sec i = 900 CPU Time = 0.009 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.004 sec i = 1000 CPU Time = 0.009 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.000 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.005 sec i = 1100 CPU Time = 0.003 sec GPU Time for only matmul = 0.000 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.006 sec i = 1200 CPU Time = 0.009 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.007 sec i = 1300 CPU Time = 0.006 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.008 sec i = 1400 CPU Time = 0.010 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.009 sec i = 1500 CPU Time = 0.010 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.010 sec i = 1600 CPU Time = 0.004 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.011 sec i = 1700 CPU Time = 0.009 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.013 sec i = 1800 CPU Time = 0.011 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.014 sec i = 1900 CPU Time = 0.010 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.015 sec i = 2000 CPU Time = 0.011 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.018 sec i = 2100 CPU Time = 0.007 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.019 sec i = 2200 CPU Time = 0.009 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.001 sec GPU Time for matmul and all data transfers including K = 0.020 sec i = 2300 CPU Time = 0.009 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.022 sec i = 2400 CPU Time = 0.014 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.001 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.024 sec i = 2500 CPU Time = 0.009 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.026 sec i = 2600 CPU Time = 0.016 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.028 sec i = 2700 CPU Time = 0.016 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.030 sec i = 2800 CPU Time = 0.017 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.031 sec i = 2900 CPU Time = 0.018 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.034 sec i = 3000 CPU Time = 0.018 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.036 sec i = 3100 CPU Time = 0.019 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.039 sec i = 3200 CPU Time = 0.025 sec GPU Time for only matmul = 0.001 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.001 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.040 sec i = 3300 CPU Time = 0.021 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.043 sec i = 3400 CPU Time = 0.025 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.045 sec i = 3500 CPU Time = 0.022 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.048 sec i = 3600 CPU Time = 0.021 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.002 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.002 sec GPU Time for matmul and all data transfers including K = 0.051 sec i = 3700 CPU Time = 0.024 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.054 sec i = 3800 CPU Time = 0.025 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.057 sec i = 3900 CPU Time = 0.024 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.060 sec i = 4000 CPU Time = 0.025 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.062 sec i = 4100 CPU Time = 0.026 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.066 sec i = 4200 CPU Time = 0.034 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.069 sec i = 4300 CPU Time = 0.033 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.072 sec i = 4400 CPU Time = 0.033 sec GPU Time for only matmul = 0.002 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.002 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.075 sec i = 4500 CPU Time = 0.033 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.079 sec i = 4600 CPU Time = 0.039 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.082 sec i = 4700 CPU Time = 0.038 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.086 sec i = 4800 CPU Time = 0.038 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.003 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.003 sec GPU Time for matmul and all data transfers including K = 0.088 sec i = 4900 CPU Time = 0.037 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.093 sec i = 5000 CPU Time = 0.042 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.097 sec i = 5100 CPU Time = 0.045 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.101 sec i = 5200 CPU Time = 0.050 sec GPU Time for only matmul = 0.003 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.003 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.104 sec i = 5300 CPU Time = 0.044 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.109 sec i = 5400 CPU Time = 0.047 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.006 sec GPU Time for matmul and all data transfers including K = 0.112 sec i = 5500 CPU Time = 0.049 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.116 sec i = 5600 CPU Time = 0.049 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.004 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.004 sec GPU Time for matmul and all data transfers including K = 0.120 sec i = 5700 CPU Time = 0.057 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.125 sec i = 5800 CPU Time = 0.058 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.129 sec i = 5900 CPU Time = 0.059 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.006 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.006 sec GPU Time for matmul and all data transfers including K = 0.134 sec i = 6000 CPU Time = 0.056 sec GPU Time for only matmul = 0.004 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.004 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.138 sec i = 6100 CPU Time = 0.060 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.006 sec GPU Time for matmul plus transfer of result = 0.006 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.143 sec i = 6200 CPU Time = 0.064 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.006 sec GPU Time for matmul plus transfer of result = 0.006 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.006 sec GPU Time for matmul and all data transfers including K = 0.148 sec i = 6300 CPU Time = 0.061 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.006 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.153 sec i = 6400 CPU Time = 0.065 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.005 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.005 sec GPU Time for matmul and all data transfers including K = 0.156 sec i = 6500 CPU Time = 0.066 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.163 sec i = 6600 CPU Time = 0.071 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.006 sec GPU Time for matmul plus transfer of result = 0.006 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.167 sec i = 6700 CPU Time = 0.069 sec GPU Time for only matmul = 0.007 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.172 sec i = 6800 CPU Time = 0.073 sec GPU Time for only matmul = 0.005 sec GPU Time for transfer of X plus matmul = 0.006 sec GPU Time for matmul plus transfer of result = 0.005 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.006 sec GPU Time for matmul and all data transfers including K = 0.176 sec i = 6900 CPU Time = 0.074 sec GPU Time for only matmul = 0.007 sec GPU Time for transfer of X plus matmul = 0.008 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.008 sec GPU Time for matmul and all data transfers including K = 0.183 sec i = 7000 CPU Time = 0.080 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.187 sec i = 7100 CPU Time = 0.081 sec GPU Time for only matmul = 0.008 sec GPU Time for transfer of X plus matmul = 0.008 sec GPU Time for matmul plus transfer of result = 0.008 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.008 sec GPU Time for matmul and all data transfers including K = 0.194 sec i = 7200 CPU Time = 0.087 sec GPU Time for only matmul = 0.006 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.006 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.197 sec i = 7300 CPU Time = 0.081 sec GPU Time for only matmul = 0.008 sec GPU Time for transfer of X plus matmul = 0.009 sec GPU Time for matmul plus transfer of result = 0.008 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.009 sec GPU Time for matmul and all data transfers including K = 0.205 sec i = 7400 CPU Time = 0.085 sec GPU Time for only matmul = 0.008 sec GPU Time for transfer of X plus matmul = 0.008 sec GPU Time for matmul plus transfer of result = 0.008 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.008 sec GPU Time for matmul and all data transfers including K = 0.210 sec i = 7500 CPU Time = 0.088 sec GPU Time for only matmul = 0.009 sec GPU Time for transfer of X plus matmul = 0.009 sec GPU Time for matmul plus transfer of result = 0.009 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.009 sec GPU Time for matmul and all data transfers including K = 0.216 sec i = 7600 CPU Time = 0.090 sec GPU Time for only matmul = 0.007 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.007 sec GPU Time for matmul and all data transfers including K = 0.219 sec i = 7700 CPU Time = 0.096 sec GPU Time for only matmul = 0.009 sec GPU Time for transfer of X plus matmul = 0.009 sec GPU Time for matmul plus transfer of result = 0.009 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.009 sec GPU Time for matmul and all data transfers including K = 0.227 sec i = 7800 CPU Time = 0.093 sec GPU Time for only matmul = 0.008 sec GPU Time for transfer of X plus matmul = 0.009 sec GPU Time for matmul plus transfer of result = 0.008 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.009 sec GPU Time for matmul and all data transfers including K = 0.232 sec i = 7900 CPU Time = 0.100 sec GPU Time for only matmul = 0.009 sec GPU Time for transfer of X plus matmul = 0.010 sec GPU Time for matmul plus transfer of result = 0.009 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.010 sec GPU Time for matmul and all data transfers including K = 0.241 sec i = 8000 CPU Time = 0.096 sec GPU Time for only matmul = 0.007 sec GPU Time for transfer of X plus matmul = 0.007 sec GPU Time for matmul plus transfer of result = 0.007 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.008 sec GPU Time for matmul and all data transfers including K = 0.243 sec i = 8100 CPU Time = 0.100 sec GPU Time for only matmul = 0.009 sec GPU Time for transfer of X plus matmul = 0.010 sec GPU Time for matmul plus transfer of result = 0.010 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.010 sec GPU Time for matmul and all data transfers including K = 0.252 sec i = 8200 CPU Time = 0.096 sec GPU Time for only matmul = 0.010 sec GPU Time for transfer of X plus matmul = 0.010 sec GPU Time for matmul plus transfer of result = 0.010 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.010 sec GPU Time for matmul and all data transfers including K = 0.256 sec i = 8300 CPU Time = 0.113 sec GPU Time for only matmul = 0.011 sec GPU Time for transfer of X plus matmul = 0.012 sec GPU Time for matmul plus transfer of result = 0.011 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.264 sec i = 8400 CPU Time = 0.103 sec GPU Time for only matmul = 0.009 sec GPU Time for transfer of X plus matmul = 0.010 sec GPU Time for matmul plus transfer of result = 0.009 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.010 sec GPU Time for matmul and all data transfers including K = 0.268 sec i = 8500 CPU Time = 0.106 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.012 sec GPU Time for matmul plus transfer of result = 0.012 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.277 sec i = 8600 CPU Time = 0.108 sec GPU Time for only matmul = 0.011 sec GPU Time for transfer of X plus matmul = 0.012 sec GPU Time for matmul plus transfer of result = 0.011 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.282 sec i = 8700 CPU Time = 0.110 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.013 sec GPU Time for matmul plus transfer of result = 0.012 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.013 sec GPU Time for matmul and all data transfers including K = 0.291 sec i = 8800 CPU Time = 0.127 sec GPU Time for only matmul = 0.010 sec GPU Time for transfer of X plus matmul = 0.010 sec GPU Time for matmul plus transfer of result = 0.010 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.010 sec GPU Time for matmul and all data transfers including K = 0.294 sec i = 8900 CPU Time = 0.129 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.013 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.013 sec GPU Time for matmul and all data transfers including K = 0.305 sec i = 9000 CPU Time = 0.132 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.012 sec GPU Time for matmul plus transfer of result = 0.012 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.309 sec i = 9100 CPU Time = 0.130 sec GPU Time for only matmul = 0.013 sec GPU Time for transfer of X plus matmul = 0.013 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.013 sec GPU Time for matmul and all data transfers including K = 0.317 sec i = 9200 CPU Time = 0.133 sec GPU Time for only matmul = 0.010 sec GPU Time for transfer of X plus matmul = 0.011 sec GPU Time for matmul plus transfer of result = 0.011 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.011 sec GPU Time for matmul and all data transfers including K = 0.321 sec i = 9300 CPU Time = 0.131 sec GPU Time for only matmul = 0.013 sec GPU Time for transfer of X plus matmul = 0.014 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.014 sec GPU Time for matmul and all data transfers including K = 0.331 sec i = 9400 CPU Time = 0.145 sec GPU Time for only matmul = 0.013 sec GPU Time for transfer of X plus matmul = 0.013 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.013 sec GPU Time for matmul and all data transfers including K = 0.338 sec i = 9500 CPU Time = 0.140 sec GPU Time for only matmul = 0.014 sec GPU Time for transfer of X plus matmul = 0.014 sec GPU Time for matmul plus transfer of result = 0.014 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.014 sec GPU Time for matmul and all data transfers including K = 0.345 sec i = 9600 CPU Time = 0.143 sec GPU Time for only matmul = 0.011 sec GPU Time for transfer of X plus matmul = 0.011 sec GPU Time for matmul plus transfer of result = 0.011 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.349 sec i = 9700 CPU Time = 0.146 sec GPU Time for only matmul = 0.014 sec GPU Time for transfer of X plus matmul = 0.015 sec GPU Time for matmul plus transfer of result = 0.014 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.015 sec GPU Time for matmul and all data transfers including K = 0.359 sec i = 9800 CPU Time = 0.151 sec GPU Time for only matmul = 0.013 sec GPU Time for transfer of X plus matmul = 0.014 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.014 sec GPU Time for matmul and all data transfers including K = 0.366 sec i = 9900 CPU Time = 0.139 sec GPU Time for only matmul = 0.015 sec GPU Time for transfer of X plus matmul = 0.015 sec GPU Time for matmul plus transfer of result = 0.015 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.015 sec GPU Time for matmul and all data transfers including K = 0.374 sec i = 10000 CPU Time = 0.150 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.012 sec GPU Time for matmul plus transfer of result = 0.012 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.012 sec GPU Time for matmul and all data transfers including K = 0.379 sec i = 10100 CPU Time = 0.167 sec GPU Time for only matmul = 0.015 sec GPU Time for transfer of X plus matmul = 0.015 sec GPU Time for matmul plus transfer of result = 0.015 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.016 sec GPU Time for matmul and all data transfers including K = 0.389 sec i = 10200 CPU Time = 0.172 sec GPU Time for only matmul = 0.014 sec GPU Time for transfer of X plus matmul = 0.015 sec GPU Time for matmul plus transfer of result = 0.014 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.015 sec GPU Time for matmul and all data transfers including K = 0.396 sec i = 10300 CPU Time = 0.173 sec GPU Time for only matmul = 0.016 sec GPU Time for transfer of X plus matmul = 0.016 sec GPU Time for matmul plus transfer of result = 0.016 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.016 sec GPU Time for matmul and all data transfers including K = 0.405 sec i = 10400 CPU Time = 0.168 sec GPU Time for only matmul = 0.012 sec GPU Time for transfer of X plus matmul = 0.013 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.013 sec GPU Time for matmul and all data transfers including K = 0.409 sec i = 10500 CPU Time = 0.152 sec GPU Time for only matmul = 0.017 sec GPU Time for transfer of X plus matmul = 0.017 sec GPU Time for matmul plus transfer of result = 0.017 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.017 sec GPU Time for matmul and all data transfers including K = 0.420 sec i = 10600 CPU Time = 0.171 sec GPU Time for only matmul = 0.015 sec GPU Time for transfer of X plus matmul = 0.016 sec GPU Time for matmul plus transfer of result = 0.015 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.016 sec GPU Time for matmul and all data transfers including K = 0.427 sec i = 10700 CPU Time = 0.178 sec GPU Time for only matmul = 0.017 sec GPU Time for transfer of X plus matmul = 0.017 sec GPU Time for matmul plus transfer of result = 0.017 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.017 sec GPU Time for matmul and all data transfers including K = 0.437 sec i = 10800 CPU Time = 0.182 sec GPU Time for only matmul = 0.013 sec GPU Time for transfer of X plus matmul = 0.014 sec GPU Time for matmul plus transfer of result = 0.013 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.014 sec GPU Time for matmul and all data transfers including K = 0.441 sec i = 10900 CPU Time = 0.184 sec GPU Time for only matmul = 0.017 sec GPU Time for transfer of X plus matmul = 0.018 sec GPU Time for matmul plus transfer of result = 0.018 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.018 sec GPU Time for matmul and all data transfers including K = 0.452 sec i = 11000 CPU Time = 0.169 sec GPU Time for only matmul = 0.016 sec GPU Time for transfer of X plus matmul = 0.017 sec GPU Time for matmul plus transfer of result = 0.016 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.017 sec GPU Time for matmul and all data transfers including K = 0.465 sec i = 11100 CPU Time = 0.192 sec GPU Time for only matmul = 0.018 sec GPU Time for transfer of X plus matmul = 0.018 sec GPU Time for matmul plus transfer of result = 0.018 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.019 sec GPU Time for matmul and all data transfers including K = 0.469 sec i = 11200 CPU Time = 0.190 sec GPU Time for only matmul = 0.014 sec GPU Time for transfer of X plus matmul = 0.014 sec GPU Time for matmul plus transfer of result = 0.014 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.015 sec GPU Time for matmul and all data transfers including K = 0.474 sec i = 11300 CPU Time = 0.208 sec GPU Time for only matmul = 0.018 sec GPU Time for transfer of X plus matmul = 0.019 sec GPU Time for matmul plus transfer of result = 0.019 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.019 sec GPU Time for matmul and all data transfers including K = 0.487 sec i = 11400 CPU Time = 0.211 sec GPU Time for only matmul = 0.018 sec GPU Time for transfer of X plus matmul = 0.018 sec GPU Time for matmul plus transfer of result = 0.018 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.018 sec GPU Time for matmul and all data transfers including K = 0.494 sec i = 11500 CPU Time = 0.204 sec GPU Time for only matmul = 0.020 sec GPU Time for transfer of X plus matmul = 0.020 sec GPU Time for matmul plus transfer of result = 0.020 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.020 sec GPU Time for matmul and all data transfers including K = 0.504 sec i = 11600 CPU Time = 0.200 sec GPU Time for only matmul = 0.015 sec GPU Time for transfer of X plus matmul = 0.015 sec GPU Time for matmul plus transfer of result = 0.015 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.016 sec GPU Time for matmul and all data transfers including K = 0.508 sec i = 11700 CPU Time = 0.205 sec GPU Time for only matmul = 0.020 sec GPU Time for transfer of X plus matmul = 0.021 sec GPU Time for matmul plus transfer of result = 0.021 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.021 sec GPU Time for matmul and all data transfers including K = 0.522 sec i = 11800 CPU Time = 0.212 sec GPU Time for only matmul = 0.019 sec GPU Time for transfer of X plus matmul = 0.020 sec GPU Time for matmul plus transfer of result = 0.020 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.020 sec GPU Time for matmul and all data transfers including K = 0.530 sec i = 11900 CPU Time = 0.211 sec GPU Time for only matmul = 0.022 sec GPU Time for transfer of X plus matmul = 0.022 sec GPU Time for matmul plus transfer of result = 0.022 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.022 sec GPU Time for matmul and all data transfers including K = 0.544 sec i = 12000 CPU Time = 0.198 sec GPU Time for only matmul = 0.016 sec GPU Time for transfer of X plus matmul = 0.017 sec GPU Time for matmul plus transfer of result = 0.016 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.017 sec GPU Time for matmul and all data transfers including K = 0.543 sec i = 12100 CPU Time = 0.225 sec GPU Time for only matmul = 0.022 sec GPU Time for transfer of X plus matmul = 0.022 sec GPU Time for matmul plus transfer of result = 0.023 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.023 sec GPU Time for matmul and all data transfers including K = 0.558 sec i = 12200 CPU Time = 0.235 sec GPU Time for only matmul = 0.021 sec GPU Time for transfer of X plus matmul = 0.021 sec GPU Time for matmul plus transfer of result = 0.021 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.021 sec GPU Time for matmul and all data transfers including K = 0.566 sec i = 12300 CPU Time = 0.234 sec GPU Time for only matmul = 0.023 sec GPU Time for transfer of X plus matmul = 0.024 sec GPU Time for matmul plus transfer of result = 0.023 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.024 sec GPU Time for matmul and all data transfers including K = 0.578 sec i = 12400 CPU Time = 0.242 sec GPU Time for only matmul = 0.017 sec GPU Time for transfer of X plus matmul = 0.018 sec GPU Time for matmul plus transfer of result = 0.017 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.018 sec GPU Time for matmul and all data transfers including K = 0.578 sec i = 12500 CPU Time = 0.231 sec GPU Time for only matmul = 0.023 sec GPU Time for transfer of X plus matmul = 0.024 sec GPU Time for matmul plus transfer of result = 0.023 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.024 sec GPU Time for matmul and all data transfers including K = 0.595 sec i = 12600 CPU Time = 0.256 sec GPU Time for only matmul = 0.022 sec GPU Time for transfer of X plus matmul = 0.022 sec GPU Time for matmul plus transfer of result = 0.022 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.022 sec GPU Time for matmul and all data transfers including K = 0.602 sec i = 12700 CPU Time = 0.258 sec GPU Time for only matmul = 0.024 sec GPU Time for transfer of X plus matmul = 0.024 sec GPU Time for matmul plus transfer of result = 0.024 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.025 sec GPU Time for matmul and all data transfers including K = 0.615 sec i = 12800 CPU Time = 0.242 sec GPU Time for only matmul = 0.018 sec GPU Time for transfer of X plus matmul = 0.018 sec GPU Time for matmul plus transfer of result = 0.018 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.019 sec GPU Time for matmul and all data transfers including K = 0.617 sec i = 12900 CPU Time = 0.271 sec GPU Time for only matmul = 0.024 sec GPU Time for transfer of X plus matmul = 0.025 sec GPU Time for matmul plus transfer of result = 0.024 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.025 sec GPU Time for matmul and all data transfers including K = 0.634 sec i = 13000 CPU Time = 0.259 sec GPU Time for only matmul = 0.023 sec GPU Time for transfer of X plus matmul = 0.023 sec GPU Time for matmul plus transfer of result = 0.023 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.023 sec GPU Time for matmul and all data transfers including K = 0.643 sec i = 13100 CPU Time = 0.254 sec GPU Time for only matmul = 0.025 sec GPU Time for transfer of X plus matmul = 0.025 sec GPU Time for matmul plus transfer of result = 0.025 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.026 sec GPU Time for matmul and all data transfers including K = 0.654 sec i = 13200 CPU Time = 0.240 sec GPU Time for only matmul = 0.019 sec GPU Time for transfer of X plus matmul = 0.019 sec GPU Time for matmul plus transfer of result = 0.019 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.020 sec GPU Time for matmul and all data transfers including K = 0.658 sec i = 13300 CPU Time = 0.237 sec GPU Time for only matmul = 0.025 sec GPU Time for transfer of X plus matmul = 0.026 sec GPU Time for matmul plus transfer of result = 0.026 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.026 sec GPU Time for matmul and all data transfers including K = 0.675 sec i = 13400 CPU Time = 0.268 sec GPU Time for only matmul = 0.024 sec GPU Time for transfer of X plus matmul = 0.025 sec GPU Time for matmul plus transfer of result = 0.024 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.025 sec GPU Time for matmul and all data transfers including K = 0.680 sec i = 13500 CPU Time = 0.283 sec GPU Time for only matmul = 0.027 sec GPU Time for transfer of X plus matmul = 0.028 sec GPU Time for matmul plus transfer of result = 0.027 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.028 sec GPU Time for matmul and all data transfers including K = 0.693 sec i = 13600 CPU Time = 0.283 sec GPU Time for only matmul = 0.020 sec GPU Time for transfer of X plus matmul = 0.021 sec GPU Time for matmul plus transfer of result = 0.021 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.021 sec GPU Time for matmul and all data transfers including K = 0.698 sec i = 13700 CPU Time = 0.283 sec GPU Time for only matmul = 0.027 sec GPU Time for transfer of X plus matmul = 0.028 sec GPU Time for matmul plus transfer of result = 0.028 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.028 sec GPU Time for matmul and all data transfers including K = 0.713 sec i = 13800 CPU Time = 0.284 sec GPU Time for only matmul = 0.025 sec GPU Time for transfer of X plus matmul = 0.026 sec GPU Time for matmul plus transfer of result = 0.026 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.026 sec GPU Time for matmul and all data transfers including K = 0.722 sec i = 13900 CPU Time = 0.309 sec GPU Time for only matmul = 0.028 sec GPU Time for transfer of X plus matmul = 0.029 sec GPU Time for matmul plus transfer of result = 0.028 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.029 sec GPU Time for matmul and all data transfers including K = 0.734 sec i = 14000 CPU Time = 0.289 sec GPU Time for only matmul = 0.021 sec GPU Time for transfer of X plus matmul = 0.022 sec GPU Time for matmul plus transfer of result = 0.022 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.022 sec GPU Time for matmul and all data transfers including K = 0.737 sec i = 14100 CPU Time = 0.307 sec GPU Time for only matmul = 0.029 sec GPU Time for transfer of X plus matmul = 0.029 sec GPU Time for matmul plus transfer of result = 0.029 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.030 sec GPU Time for matmul and all data transfers including K = 0.755 sec i = 14200 CPU Time = 0.323 sec GPU Time for only matmul = 0.026 sec GPU Time for transfer of X plus matmul = 0.027 sec GPU Time for matmul plus transfer of result = 0.027 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.027 sec GPU Time for matmul and all data transfers including K = 0.764 sec i = 14300 CPU Time = 0.277 sec GPU Time for only matmul = 0.029 sec GPU Time for transfer of X plus matmul = 0.030 sec GPU Time for matmul plus transfer of result = 0.029 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.030 sec GPU Time for matmul and all data transfers including K = 0.775 sec i = 14400 CPU Time = 0.281 sec GPU Time for only matmul = 0.023 sec GPU Time for transfer of X plus matmul = 0.023 sec GPU Time for matmul plus transfer of result = 0.023 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.023 sec GPU Time for matmul and all data transfers including K = 0.781 sec i = 14500 CPU Time = 0.329 sec GPU Time for only matmul = 0.031 sec GPU Time for transfer of X plus matmul = 0.032 sec GPU Time for matmul plus transfer of result = 0.031 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.032 sec GPU Time for matmul and all data transfers including K = 0.799 sec i = 14600 CPU Time = 0.328 sec GPU Time for only matmul = 0.029 sec GPU Time for transfer of X plus matmul = 0.030 sec GPU Time for matmul plus transfer of result = 0.029 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.276 sec GPU Time for matmul and all data transfers including K = 1.094 sec i = 14700 CPU Time = 0.346 sec GPU Time for only matmul = 0.289 sec GPU Time for transfer of X plus matmul = 0.290 sec GPU Time for matmul plus transfer of result = 0.289 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.291 sec GPU Time for matmul and all data transfers including K = 1.124 sec i = 14800 CPU Time = 0.294 sec GPU Time for only matmul = 0.236 sec GPU Time for transfer of X plus matmul = 0.243 sec GPU Time for matmul plus transfer of result = 0.236 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.245 sec GPU Time for matmul and all data transfers including K = 1.081 sec i = 14900 CPU Time = 0.351 sec GPU Time for only matmul = 0.296 sec GPU Time for transfer of X plus matmul = 0.298 sec GPU Time for matmul plus transfer of result = 0.297 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.299 sec GPU Time for matmul and all data transfers including K = 1.150 sec i = 15000 CPU Time = 0.354 sec GPU Time for only matmul = 0.280 sec GPU Time for transfer of X plus matmul = 0.282 sec GPU Time for matmul plus transfer of result = 0.281 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.283 sec GPU Time for matmul and all data transfers including K = 1.145 sec i = 15100 CPU Time = 0.299 sec GPU Time for only matmul = 0.303 sec GPU Time for transfer of X plus matmul = 0.304 sec GPU Time for matmul plus transfer of result = 0.305 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.315 sec GPU Time for matmul and all data transfers including K = 1.181 sec i = 15200 CPU Time = 0.301 sec GPU Time for only matmul = 0.256 sec GPU Time for transfer of X plus matmul = 0.256 sec GPU Time for matmul plus transfer of result = 0.256 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.259 sec GPU Time for matmul and all data transfers including K = 1.144 sec i = 15300 CPU Time = 0.316 sec GPU Time for only matmul = 0.318 sec GPU Time for transfer of X plus matmul = 0.312 sec GPU Time for matmul plus transfer of result = 0.309 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.310 sec GPU Time for matmul and all data transfers including K = 1.209 sec i = 15400 CPU Time = 0.309 sec GPU Time for only matmul = 0.294 sec GPU Time for transfer of X plus matmul = 0.298 sec GPU Time for matmul plus transfer of result = 0.295 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.296 sec GPU Time for matmul and all data transfers including K = 1.208 sec i = 15500 CPU Time = 0.328 sec GPU Time for only matmul = 0.325 sec GPU Time for transfer of X plus matmul = 0.324 sec GPU Time for matmul plus transfer of result = 0.324 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.327 sec GPU Time for matmul and all data transfers including K = 1.251 sec i = 15600 CPU Time = 0.323 sec GPU Time for only matmul = 0.262 sec GPU Time for transfer of X plus matmul = 0.264 sec GPU Time for matmul plus transfer of result = 0.264 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.265 sec GPU Time for matmul and all data transfers including K = 1.198 sec i = 15700 CPU Time = 0.323 sec GPU Time for only matmul = 0.332 sec GPU Time for transfer of X plus matmul = 0.333 sec GPU Time for matmul plus transfer of result = 0.333 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.338 sec GPU Time for matmul and all data transfers including K = 1.281 sec i = 15800 CPU Time = 0.337 sec GPU Time for only matmul = 0.315 sec GPU Time for transfer of X plus matmul = 0.314 sec GPU Time for matmul plus transfer of result = 0.315 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.315 sec GPU Time for matmul and all data transfers including K = 1.276 sec i = 15900 CPU Time = 0.399 sec GPU Time for only matmul = 0.339 sec GPU Time for transfer of X plus matmul = 0.339 sec GPU Time for matmul plus transfer of result = 0.338 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.339 sec GPU Time for matmul and all data transfers including K = 1.314 sec i = 16000 CPU Time = 0.408 sec GPU Time for only matmul = 0.276 sec GPU Time for transfer of X plus matmul = 0.278 sec GPU Time for matmul plus transfer of result = 0.278 sec GPU Time for transfer of X plus matmul plus transfer of result = 0.280 sec GPU Time for matmul and all data transfers including K = 1.263 sec >>