Description
The range of values in at least one of the inputs exceeds the 32-bit integer value range. This could occur when using floating-point rasters, which can store a larger range of values than integer rasters of comparable bit depths. Note that, aside from this situation, the values in floating-point input rasters are truncated to integers.
Solution
If possible, reduce the range of input floating-point values to fall within the 32-bit integer range (-2,147,483,648 to +2,147,483,647).