Search results

...hlight lang=apl inline>R</syntaxhighlight> along one axis using an integer vector, which is negative to indicate values from the left argument and positive t ...> and [[Boolean]] vector <math>u</math>, all of the same length, to be the vector <math>c</math> whose [[compress]]ions by <math>u</math> and its [[Negate|ne

...Singeli]: a metaprogramming language for building abstractions around CPU instructions ...s://dyalog.tv/Dyalog18/?v=paxIkKBzqBU Sub-nanosecond Searches Using Vector Instructions]

...not have a low-overhead way to dynamically emit and execute native HP 3000 instructions. As a result, performance, especially on those very early HP-3000 models, w ...es III in the 1980s. Only thirty-five years later have the special APL CPU instructions been reverse-engineered, allowing APL to run once again on a simulated HP-3

...taxhighlight lang=apl inline>v</syntaxhighlight> is a non-negative integer vector and <syntaxhighlight lang=apl inline>s</syntaxhighlight> is a singleton. ...ed only by compression with a [[Prefix and suffix vectors|prefix or suffix vector]]. It was included in [[APL\360]], which changed the doubled slash to a bar

...[Maximum]], [[Minimum]], and [[Plus]] [[Reduce|reductions]] using [[vector instructions]] on x86 and POWER platforms. He also developed a technique for transposing ...ky]], and the development of [[tolerant hashing]] for real and [[complex]] vector searches.<ref>[[Roger Hui|Hui, Roger]]. [https://www.jsoftware.com/papers/H

...l inline>3 3</syntaxhighlight> [[matrix]] of grids to obtain a 9-element [[vector]]; otherwise [[Reduce]] would only allow us to sum rows or columns. Here AP ...the last step was a [[scalar]] containing one grid; when paired with the [[vector]] <syntaxhighlight lang=apl inline>3 4</syntaxhighlight> it is extended to

...The type used should also be supported by the hardware (usually, the CPU's vector processor), which for example makes [[wikipedia:IEEE 754|IEEE 754]] floats ...grown in prominence for high-performance computing since then. In APL, CPU vector instruction sets such as Intel's [[wikipedia:Streaming SIMD Extensions|SSE]

* thirty-two vector accumulators (VA1, VA2..., VA32) The IMU is a temporary location for instructions, to "give the programmer a usable memory of 16,384 words." Each cell is a 3

...ctions can be used on them directly, for example by [[sum]]ming a result [[vector]] to find out how many times the comparison was true: ...n working with integers. While most processors now have SIMD registers and instructions, current SIMD implementations don't provide all the bitwise operations that

...ent BQN more effectively. CBQN uses Singeli for primitive code that uses [[vector instruction]]s, and can be built with or without these optimized implementa

...ighlight>), treating as constant functions, improved [[Vector instructions|vector instruction]] usage

...en Taylor. [http://archive.vector.org.uk/art10013790 "How we got here"]. [[Vector journal]] Volume 23 special supplement "Dyalog at 25". 2008-09.</ref> Dyalog makes heavy use of [[vector instructions]] on all platforms, as well as other special instruction sets primarily on

...was to extend Mix with [[Function axis|axis]] to allow the axis to be a [[vector]] indicating the final positions of each axis of the argument elements (rat ...] in many areas. The most important improvements were better use of vector instructions including support for [[wikipedia:NEON (instruction set)|ARM NEON]] and [[w

...[http://archive.vector.org.uk/art10009060 "Dyalog APL 2003 Conference"]. [[Vector journal]] volume 20 issue 2. 2003-10.</ref> ...t10013720 "Dyalog APL Users Meeting: LO-skolen, Elsinore October 2005"]. [[Vector journal]] volume 22 issue 1. 2005-11.</ref>