Unitized Data
2004-04-28 by grantrichter2001
Hi All, It looks like the internal math processor on the Basic Atom Pro is 32 bits wide. According to Nathan, Longs and Floats process faster than words, byte, nibs or bits. The nice thing about floating point is you have the same bit resolution, no matter what decade you are in. So it makes sense to use unitzed data. That is data that ranges between +1 and -1. An example is reading the input ADC and transfering output to the DACs. In THEORY, you can read the ADC into a Long, convert it to a Float, divide by 1024 to form a unitzed floating point number between 0 and 1. Then you can multiply by 4096 and convert to integer for output to the DAC. The nice thing is you can also multiply by 4095.555 or what ever to give exact voltage matching between the two. Any thoughts?