Герундиальный оборот

my, his, her, our, your, their

сущ. в притяжательном падеже + герундий

сущ. в общем падеже

Обычно переводится придаточным предложением с союзом что/чтобы с предшествующим ему местоимением то. Герундий становится сказуемым придаточного предложения, а местоимение или существительное – подлежащим.

Pavlov's having devoted all his life to science is known to everybody. To, что Павлов посвятил всю свою жизнь науке, известно каждому. Bad weather prevented them from finishing their work in time. Плохая погода помешала им закончить работу вовремя.

XII. Переведите предложения с герундиальным оборотом на русский язык.

1. Everybody knows of his being awarded the Nobel Prize.

2. His solving this difficult problem shows that his method is perfectly correct.

3. It was told of their having finished their experiment in time.

4. Mendeleyv's having noticed a periodicity in the chemical properties of elements enabled him to lay down his famous Periodic Table.

5. The engineer insists on those devices being a new step in the development of a technique.

6. The students' knowing mathematics well helped them in choosing the appropriate algorithm for programming.

7. Their using the new program for the computer has improved the results of the last series of the tests.

8. Scientists' researching in the field of automatic translation resulted in improving machine programmes.

4. Прочитайте текст, перечислите, какие специальные величины .можно представлять форматами данных 8087.

Special Values

Besides being able to represent positive and negative numbers, the 8087 data formats can be used to describe other entities. These special values provide extra flexibility but most users do not need to understand them in detail to use the 8087 successfully.

The value zero can be signed positive or negative» in real and decimal integer formats; the sign of a binary integer zero is always positive. The fact that zero can be signed, however, is not apparent, to the programmer.

The real number formats allow for the represent at ion of t he special values + and –. The 8087 can generate these values as its built-in response to exceptions such as division by zero, or try to store a result that exceeds the upper range limit of the destination formal. Infinities can participate in arithmetic and comparison operations, and in fact the processor provides two different conceptual models for handling these special values.

If a programmer attempts an operation for which the 8087 cannot deliver a reasonable result, it will at the programmer's discretion, either request an interrupt, or return the special value indefinite. Taking the square root of a negative number is an example of this type of invalid operation. The recommended action in this situation is to stop the computation by trapping to a user-written exception handler. If, however, the programmer elects to continue the computation, the specially coded indefinite value propagates through the calculation and thus flags the erroneous computation when it is eventually delivered as the result. Each format has coding that represents the special value indefinite.

In the real formats, a complete range of special values, both positive and negative, is designated to represent a class of values called NAN (Not A-Number). The special value indefinite is a reserved NAN coding, but all other codings are made available to be defined in any way by application software. Using a NAN as an operand raises the invalid operation exception, and can trap to a user-written routine to process the NAN. Alternatively, the 8087's built-in exception handler simply returns the NAN itself as the result of the operation; in this way NAN's, including indefinite, can be propagated through a calculation and delivered as a final, special valued, result. One use for NAN's is to detect uninitialized variables.

As mentioned earlier, the 8087 stores non-zero real numbers in normalized floating-point form. It also provides for storing and operating on real

numbers that are not normalized, those significance contain one or more leading zeros. Non-normals arise when the result of a calculation yields a value too small to be represented in normal form. The leading zeros of non-normals allow smaller numbers to be represented, at the cost of some lost precision (the number of significant digits is reduced by the leading zeros). In typical algorithms, extremely small values are most likely generated as intermediate, rather than final results. By using the 8087's temporary real format for holding intermediates, values as small as ±3.4•10^-4932 can be represented; this makes the occurrence of non-normal numbers a rare phenomenon in 8087 applications. Nevertheless, the 8087 can load, store and operate on non-normalized real numbers.

The real number formats used by the 8087 are optimized for the processing of those numbers in the 8087 hardware, This format is sometimes called the IEEE format. The 8087 can only use real numbers that are in the 8087 real number format, therefore, all real number data to be used by the 8087 must reside in memory in the proper 8087 format prior to use for input into the 8087. The real number output from the 8087 into memory are in the 8087 real number format.

The real number formats used by the 8088 are optimized for the processing of those numbers with software. This format is the Microsoft Binary real format.

The data formats are described reading from the byte with the highest address or most significant byte to the byte with the lowest address or least significant byte.

The short real number format is:

• Sign bit in the upper bit of the first byte.

• Exponent (8 bits) in the remaining 7 bits of the first byte and the upper bit of the second byte.

• Significand (23 bits) in the remaining 7 bits of the second byte and all bytes 3 and 4.

• Exponent bias: 127 (7FH).

• The long real number format is:

• Sign bit in the upper bit of the first byte.

• Exponent (11 bits) in the remaining 7 bits of the first byte, and the upper 4 bits of the second byte.

• Significand (52 bits) in the remaining 4 bits of the second byte and all bytes 3 through 8.

• Exponent bias: 1023 (3FFH).

Notes:

entity объект; сущность; категория

flexibility гибкость

real вещественный; действительный

infinity бесконечность

discretion усмотрение; дискретность

interrupt прерывание

square root квадратный корень

application software прикладное программное

обеспечение

trap ловушка

trapping заход