Anton A. Karpov wrote: > MY laptop has CPU: Mobile Intel(R) Pentium(R) 4 - M CPU 2.00GHz > (1988.52-MHz 686-class CPU). I've tested tx states with the following > script: > > Here is some output: > > Supported freq levels are: 1991 1742 1493 1244 1196 1046 995 897 746 > 598 497 448 299 248 149 > [=== Results testing [1991] freq level ===] > type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes > blowfish cbc 44067.47k 46774.97k 46975.74k 47349.96k 47353.11k > ... > [=== Results testing [1196] freq level ===] > type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes > blowfish cbc 26456.76k 27924.48k 28192.14k 28336.50k 28410.68k > ... > [=== Results testing [149] freq level ===] > type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes > blowfish cbc 2541.21k 2660.68k 2731.58k 2720.56k 2720.05k > > It seems to work. By the way, what does /-1 mean, in > dev.cpu.0.freq_levels output? The second value is power in milliwatts, the first is frequency in Mhz. If you have only throttling support, there's no way to know what the power consumption is so -1 is "unknown". If you have both throttling and some other driver that knows the power consumption, these values will be accurate. -- NateReceived on Sun Feb 06 2005 - 23:52:01 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:27 UTC