IEEE 754

(redirected from IEEE floating-point)
References in periodicals archive ?
Its key features include: Support for four 32x32-bit multiplier-accumulators (MACs) per cycle with 72-bit accumulators, more than double the performance of other audio DSPs for computationally intensive functions such as fast Fourier transform (FFT) and finite impulse response (FIR), Support for eight 32x16-bit MACs per cycle under specified conditions, Four very long instruction word (VLIW) slot architecture capable of issuing two 64-bit loads per cycle, Optional vector floating-point unit available, providing up to four single-precision IEEE floating-point MACs per cycle, Software compatible with the existing HiFi DSP family consisting of over 140 HiFi-optimized audio and voice codecs and audio enhancement software packages.
A special feature of the text is its detailed coverage of number storage formats and IEEE floating-point arithmetic.
The board s now feature full 128-bit IEEE floating-point precision that allows for a higher level of accuracy and the ultimate visual quality.
IEEE floating-point libraries also include support for features such as features such as [+ or -][infinity], 0 and NaN (not a number).On the Blackfin 16-bit fixed-point DSP, IEEE-754, floating-point functions are available as library calls from either C/C++ or assembly language.
Whereas guaranteeing 100% correctness of an IEEE floating-point implementation is hardly feasible, as the famous Intel Pentium bug clearly demonstrated, several good but unrelated tools exist to test different aspects of a floating-point implementation for compliance with the IEEE-754 standard.
UCBTEST is a whole set of programs for "testing certain difficult cases of IEEE floating-point arithmetic" [Hough et al.
However, since the publication of the IEEE floating-point standards, algorithms for correctly rounded conversion between decimal strings and binary floating-point formats have become available which incur little time penalty in common cases [Gay 1990].
There are three fundamental datatypes: integer, IEEE floating-point [6], and VAX floating-point [2], each available in 32-bit and 64-bit forms.
Within the IEEE floating-point function field, code points are nominally reserved for double-extended precision (128-bit) arithmetic.
In addition, even if standardization contributions are acceptable, the IEEE floating-point standardz is, I believe, a scientific and technical mistake in the sense that it hinders growth in mathematical algorithms in order to help protect non-experts from errors.
The version presented here operates correctly on a large number of different floating-point systems, including those implementing the new IEEE Floating-Point Standard.

Full browser ?