Floating point math accuracy

Would you guys be able to confirm the following bit of information:

The Vex PIC processors is precise and can perform accurate complicated floating point math (divide/multiply/add/subtract/exponent) with minimal error at a greatly reduced speed. The only reason we cannot see this accuracy is because the terminal window does not support floating point output.

Is this correct?

The reason I ask is because floating point math is very important when integrating sensor values and multiplying by sin, cos, and tan values located in array tables. Is there any other way I would go about solving my double integration accuracy problem? maybe scaling?

Thank you,

Technic-R-C
(I am using EasyC Pro)

Honestly a float should be accurate enough to compute data within the scope of vex hardware. You can print the values to screen, you just need to convert them to a long and push the decimal place in your code. The compute speed of the processor is greatly reduced when doing operations on floats, as the processor is only 8-bit and has to do all its math by breaking down the numbers into 8-bit chunks.

examples
.1001 x 10000 = 1001
3.14159 x 100000 = 314159

Side note, the new Cortex is 32 bit and is much better equipped to deal with floating point math.