Sorry for the little ascii emoticon in the title, but I coudln't help myself.
Two things first:
My O-scope is out of calibration, but for this instance it doens't matter as I dont technically need it for anything other than a relative reference.
I dont know if my picaxe is actually running at an accurate 16mhz (probably not) because my scope is out of cal.
Initial trials are proving successful for my little experiment. The subjects are 24FC512 EEPROM's and MCP23017 IO expanders. The memory can run at up to 1mhz, and the IO chip at up to 1.7mhz.
At 16mhz, using the keyword i2cfast8, the memory runs anywhere from ~1.5-1.8 times the speed (technically 800khz). The reason it's not exactly twice as fast could be from a number of factors beyond my control, but it seems that theoretically it's possible to double the speed with a good design.
As for the IO chips, using the keyword i2cfast at 16 mhz should give a theoretical 1.6mhz. Well..pardon my french, but holy hell. Compared to using proper keywords and an MCP23016, the speed is seemingly increased from ~3 to over (tha'ts right I said OVER) 4x. Again, dont take that as a truly accurate assessment because of what I said above, but I'm impressed, and it would seem our beloved picaxes jsut got a whole lot faster =D.
--Andy P
Two things first:
My O-scope is out of calibration, but for this instance it doens't matter as I dont technically need it for anything other than a relative reference.
I dont know if my picaxe is actually running at an accurate 16mhz (probably not) because my scope is out of cal.
Initial trials are proving successful for my little experiment. The subjects are 24FC512 EEPROM's and MCP23017 IO expanders. The memory can run at up to 1mhz, and the IO chip at up to 1.7mhz.
At 16mhz, using the keyword i2cfast8, the memory runs anywhere from ~1.5-1.8 times the speed (technically 800khz). The reason it's not exactly twice as fast could be from a number of factors beyond my control, but it seems that theoretically it's possible to double the speed with a good design.
As for the IO chips, using the keyword i2cfast at 16 mhz should give a theoretical 1.6mhz. Well..pardon my french, but holy hell. Compared to using proper keywords and an MCP23016, the speed is seemingly increased from ~3 to over (tha'ts right I said OVER) 4x. Again, dont take that as a truly accurate assessment because of what I said above, but I'm impressed, and it would seem our beloved picaxes jsut got a whole lot faster =D.
--Andy P