Previous
Floating Point Calculations and the ANSI C, C++, and Fortran Standard
White Papers These can result in differences in rounding and small variations in floating point results, which may be more or less accurate than the ANSI-conformant result. High application performance is an important goal of the Intel compilers, even at...
[May 31, 2008, 1:01]
Decimal Floating Point Computations in SAP NetWeaver 7.10
White Papers Most processor architectures support binary floating point in hardware. However, binary floating point is not suitable for financial or humancentric applications. SAP NetWeaver 7.10 ABAP kernel introduces a new SAP ABAP data type called "DECFLOAT...
[April 9, 2008, 1:02]
Decimal Floating-Point in z9: An Implementation and Testing Perspective
White Papers Use of the newly defined Decimal Floating-Point (DFP) format instead of binary floating-point is expected to significantly improve the performance of such applications. Although decimal arithmetic is widely used in commercial and financial...
[March 7, 2007, 0:00]
FIR Coefficient Finder
Downloads FIR Coefficient Finder is a digital signal processing tool which quickly generates floating point and fixed point coefficients for Finite Impulse Response filters. Comparison plots between floating point and fixed point coefficients allow you to...
[October 26, 2006, 8:00]
Intel banks on Pentium III - Part II
News New instruction sets that will complement the usefulness of MMX technology - particularly the floating-point operations that are key to the delivery of 3-D images. All told, 70 new instructions have been incorporated into Pentium III to streamline...
[February 22, 1999, 9:20]
Intel shows teraflop chips
News Intel chief executive Paul Otellini caused an early stir at the chipmaker's Developer Forum on Tuesday when he briefly showed off a silicon wafer containing chips capable of a trillion floating point operations per second.
[September 26, 2006, 18:20]
Intel 'jumping all over' new bug scare
News Reported in Robert Collins' x86 Monthly Digest, a well-known Web site for observers of Intel activity, the brouhaha refers to an arcane problem allegedly experienced by Pentium II and Pentium Pro chips converting floating point to integer numbers.
[May 6, 1997, 10:45]
Five Years Ago: Intel 'jumping all over' new bug scare
News Reported in Robert Collins' x86 Monthly Digest, a well-known Web site for observers of Intel activity, the brouhaha refers to an arcane problem allegedly experienced by Pentium II and Pentium Pro chips converting floating point to integer numbers.
[May 3, 2002, 7:01]
China launches first supercomputer
News Named the Legend Deepcomp 1800, Legend said its first supercomputer was able to reach a speed of 1.06 teraflops, or one trillion floating-point operations per second. A teraflop is a trillion floating-point operations per second.
[September 2, 2002, 8:57]
Pentium Pro, Pentium II bug warning hits Intel
News The bug concerns the way floating point numbers are converted into integer numbers. According to Collins, "The host software is supposed to be warned by the microprocessor when such a floating point conversion error occurs; a specific error flag is...
[May 6, 1997, 9:40]
Floating Access Enables Wi-Fi Connectivity to Over 700 Boat Slips, Multiple Boathouses, and Local Businesses With a D-Link Business Class Wireless Access Point Solution
White Papers Floating Access' most recent project, a large-scale wireless networking implementation at Eagle Point Marina at Lake Lewisville, Texas proved the typical marina challenge. Floating Access is a Texas-based solution provider specializing in...
[April 3, 2009, 1:22]
Met Office buys IBM petaflop supercomputer
News The supercomputer is due to come online in 2009, and will initially run at a maximum speed of 125 trillion floating point operations per second. By 2011, the Met Office supercomputer is expected to operate at speeds approaching one quadrillion...
[August 5, 2008, 12:19]
One Year Ago: Pentium Pro, Pentium II bug warning hits Intel
News The bug concerns the way floating point numbers are converted into integer numbers. According to Collins, "The host software is supposed to be warned by the microprocessor when such a floating point conversion error occurs; a specific error flag is...
[May 5, 1998, 6:00]
Microsoft opens 64-bit Windows
News The Itanium chip, aimed at heavy-duty workstations and servers for businesses and engineering, supports greater amounts of system memory and offers stronger floating-point, or mathematical, capabilities than current 32-bit desktop processors.
[August 28, 2001, 17:16]
Cyrix to add 3D, MMX to MediaGX
News The MXi will be based on a new core that Cyrix calls Cayenne and will provide fully pipelined floating point and an MMX unit with 15 new multimedia floating point instructions; Cyrix has often been criticised in the past for the MMX and, more...
[October 16, 1997, 14:39]
Palm Basic
Downloads Palm Basic supports integer, floating point, and string variables, as well as single and multi-dimensional arrays. It contains built-in procedures for formatted printing and inputing, basic graphic operations, as well as a full set of floating...
[June 16, 2008, 8:00]
Intel toolkit brings 3D to PDAs
News The GPP toolkit gets around many of these limitations -- such as a lack of floating point arithmetic, division, square root and trigonometric functions -- using shortcuts, said Intel senior software engineer Gopi Kolli.
[August 30, 2002, 14:05]
AMD makes bold quad-core claims
News The quad-core chip will also outperform AMD's current dual-core Opterons on "floating point" mathematical calculations by a factor of 3.6 at the same clock rate, he said. A faster floating-point engine performs mathematical calculations — long an...
[January 24, 2007, 6:17]
Sun and Fujitsu to boost Sparc speed
News And for floating-point processing, which is used in many mathematical calculations, its performance is 2.5 times that of Sparc64 V. Single-thread performance improves, too, by a factor of 1.4 on integer processing and by a more dramatic factor of 5...
[October 11, 2006, 9:20]
UK Tech Week: TI predicts major growth in mobile comms
News Richardson told ZDNet today that the company has a world first with the launch of a new device -- its DSP 320C6 family -- that performs 2 billion operations per second and has common instruction set for floating and fixed point maths.
[October 7, 1998, 12:07]
