The input and output of an on-line computation involving redundant numbers must generally be compatible with standard nonredundant formats. When rounding of a result is required in the last computation step, the process can be combined with on-the-fly conversion of the redundant output into nonredundant form and thus introduces no extra delay. However, if multiple on-line operations are to be performed before obtaining a final result, the preceding method would be inapplicable to the intermediate results. We present a solution scheme where rounding (subtraction of ulp, no action, addition of ulp) is represented by an additional digit attached to the number, thus obviating the need for modifying any of the previously produced digits. This ad...
A class of on-line algorithms for complex number arith-metic is presented. These algorithms adopt a ...
Although the speed advantage of using signed – digit number systems seemed to have been reduced sign...
We present algorithms for accurately converting floating-point numbers to decimal representation. Th...
The original publication is available at www.springerlink.comDigit recurrence operations mainly use ...
In this thesis we investigate arithmetic subroutines and round-off procedures. An error analysis of ...
: We develop a formal account of digit serial number representations by describing them as strings f...
A number system that is well-designed can affect the computational time and the hardware implementat...
A detailed investigation of three different rounding rules for multiplication and division is presen...
International audienceDuring any composite computation there is a constant need for rounding interme...
(eng) The purpose of this paper is to treat digit set conversions and digit recodings in terms of pr...
Redundant number representations are generally used to allow constant time additions, based on the f...
Double rounding occurs when a floating-point value is first rounded to an intermediate precision bef...
A property of the original Booth recoding is that the non-zero digit following --1 is necessarily --...
(eng) This article shows that IEEE-754 double-precision correct rounding of the most common elementa...
Half-Unit-Biased format is based on shifting the representation line of the binary numbers by half ...
A class of on-line algorithms for complex number arith-metic is presented. These algorithms adopt a ...
Although the speed advantage of using signed – digit number systems seemed to have been reduced sign...
We present algorithms for accurately converting floating-point numbers to decimal representation. Th...
The original publication is available at www.springerlink.comDigit recurrence operations mainly use ...
In this thesis we investigate arithmetic subroutines and round-off procedures. An error analysis of ...
: We develop a formal account of digit serial number representations by describing them as strings f...
A number system that is well-designed can affect the computational time and the hardware implementat...
A detailed investigation of three different rounding rules for multiplication and division is presen...
International audienceDuring any composite computation there is a constant need for rounding interme...
(eng) The purpose of this paper is to treat digit set conversions and digit recodings in terms of pr...
Redundant number representations are generally used to allow constant time additions, based on the f...
Double rounding occurs when a floating-point value is first rounded to an intermediate precision bef...
A property of the original Booth recoding is that the non-zero digit following --1 is necessarily --...
(eng) This article shows that IEEE-754 double-precision correct rounding of the most common elementa...
Half-Unit-Biased format is based on shifting the representation line of the binary numbers by half ...
A class of on-line algorithms for complex number arith-metic is presented. These algorithms adopt a ...
Although the speed advantage of using signed – digit number systems seemed to have been reduced sign...
We present algorithms for accurately converting floating-point numbers to decimal representation. Th...