I feel like I groked floating point numbers easier when looking at 16bit floating point numbers. The values are much smaller and easier to juggle in your head.

I feel like I groked floating point numbers easier when looking at 16bit floating point numbers. The values are much smaller and easier to juggle in your head. The fact that the fraction part literally translated to x / 1024 suddenly became clear to me when you're staring at just 10 bits https://en.wikipedia.org/wiki/Half-precision_floating-point_format#Half_precision_examples https://en.wikipedia.org/wiki/Half-precision_floating-point_...