Впрочем толку мало - увидишь "хваленую" оракловскую доку

NUMBER Datatype
The NUMBER datatype stores zero, positive, and negative fixed and floating-point numbers with magnitudes between 1.0 x 10-130 and 9.9...9 x 10125 (38 nines followed by 88 zeroes) with 38 digits of precision. If you specify an arithmetic expression whose value has a magnitude greater than or equal to 1.0 x 10126, Oracle returns an error.

Specify a fixed-point number using the following form:

NUMBER(p,s)


where:

p is the precision, or the total number of digits. Oracle guarantees the portability of numbers with precision ranging from 1 to 38.

s is the scale, or the number of digits to the right of the decimal point. The scale can range from -84 to 127.

Specify an integer using the following form:

NUMBER(p) is a fixed-point number with precision p and scale 0. This is equivalent to NUMBER(p,0).

Specify a floating-point number using the following form:

NUMBER is a floating-point number with decimal precision 38. Note that a scale value is not applicable for floating-point numbers.

See Also: "Floating-Point Numbers"




Scale and Precision
Specify the scale and precision of a fixed-point number column for extra integrity checking on input. Specifying scale and precision does not force all values to a fixed length. If a value exceeds the precision, Oracle returns an error. If a value exceeds the scale, Oracle rounds it.

The following examples show how Oracle stores data using different precisions and scales.

7456123.89
NUMBER
7456123.89

7456123.89
NUMBER(9)
7456124

7456123.89
NUMBER(9,2)
7456123.89

7456123.89
NUMBER(9,1)
7456123.9

7456123.89
NUMBER(6)
exceeds precision

7456123.89
NUMBER(7,-2)
7456100

7456123.89
NUMBER(-7,2)
exceeds precision




Negative Scale
If the scale is negative, the actual data is rounded to the specified number of places to the left of the decimal point. For example, a specification of (10,-2) means to round to hundreds.

Scale Greater than Precision
You can specify a scale that is greater than precision, although it is uncommon. In this case, the precision specifies the maximum number of digits to the right of the decimal point. As with all number datatypes, if the value exceeds the precision, Oracle returns an error message. If the value exceeds the scale, Oracle rounds the value. For example, a column defined as NUMBER(4,5) requires a zero for the first digit after the decimal point and rounds all values past the fifth digit after the decimal point. The following examples show the effects of a scale greater than precision:

Actual Data Specified As Stored As
.01234
NUMBER(4,5)
.01234

.00012
NUMBER(4,5)
.00012

.000127
NUMBER(4,5)
.00013

.0000012
NUMBER(2,7)
.0000012

.00000123
NUMBER(2,7)
.0000012




Floating-Point Numbers
Oracle allows you to specify floating-point numbers, which can have a decimal point anywhere from the first to the last digit or can have no decimal point at all. A scale value is not applicable to floating-point numbers, because the number of digits that can appear after the decimal point is not restricted.

You can specify floating-point numbers with the form discussed in "NUMBER Datatype". Oracle also supports the ANSI datatype FLOAT. You can specify this datatype using one of these syntactic forms:

FLOAT specifies a floating-point number with decimal precision 38, or binary precision 126.

FLOAT(b) specifies a floating-point number with binary precision b. The precision b can range from 1 to 126. To convert from binary to decimal precision, multiply b by 0.30103. To convert from decimal to binary precision, multiply the decimal precision by 3.32193. The maximum of 126 digits of binary precision is roughly equivalent to 38 digits of decimal precision.