Commits

Branch selector

 3561a46

User selector

All users

Datepicker

All time

Commit History

Commits on Apr 2, 2020

• Fix Sphere/LineSegment intersection not returning the intersection count

  merwaaan

  committed
  8cddc15

  Copy full SHA for 8cddc15

• Remove accidental test noinline attribute

  juj

  committed
  9f2dca0

  Copy full SHA for 9f2dca0

• MATH_COLMAJOR_MATRICES fixes

  juj

  committed
  2529d53

  Copy full SHA for 2529d53

Commits on Mar 29, 2020

• Add SinCosU16ScaledRadians() function to avoid slow float->int conversion in Emscripten LLVM Wasm backend

  juj

  committed
  848a507

  Copy full SHA for 848a507

• Added build option MATH_COLMAJOR_MATRICES, which if enabled, will use a column-major layout for all matrices, instead of row-major.

  juj

  committed
  2b4ef8e

  Copy full SHA for 2b4ef8e

• Fix OPTIMIZED_RELEASE builds to kill off all variants of assert&assume&mathassert macros

  juj

  committed
  c70a51f

  Copy full SHA for c70a51f

Commits on Mar 27, 2020

• FORCE_INLINE float3x4 operator[], LLVM 10 does not want to inline it in -Oz builds otherwise.

  juj

  committed
  e0ba8c7

  Copy full SHA for e0ba8c7

Commits on Feb 23, 2020

• Optimize SAT2D algorithm implementation. Total effect is +90% faster to detect when collision is present (although no observable performance difference when collision is not present)

  juj

  committed
  2103ced

  Copy full SHA for 2103ced

• Fix float4d handling

  juj

  committed
  27e417e

  Copy full SHA for 27e417e

Commits on Jan 10, 2020

• wip float2_ConvexHullContains

  juj

  committed
  d510a2e

  Copy full SHA for d510a2e

Commits on Jan 8, 2020

• Fix bug in SATCollide2D found by @scilari.

  juj

  committed
  8e59677

  Copy full SHA for 8e59677

Commits on Jul 30, 2019

• Optimize float3x3::RotateFromTo corner case

  juj

  committed
  35058b7

  Copy full SHA for 35058b7

• Optimize float3x3::RotateFromTo()

  juj

  committed
  b3a3c6b

  Copy full SHA for b3a3c6b

Commits on Jul 28, 2019

• Disable LCG::Int() and D3D style projection matrices on Emscripten due to code size issues (TODO: figure out a way to resolve these without code size or performance impact)

  juj

  committed
  7822bf7

  Copy full SHA for 7822bf7

• Use pixar Orthonormal basis code in scalar path for orthonormalized basis generation

  juj

  committed
  fe6c63a

  Copy full SHA for fe6c63a

• Update orthonormal basis test

  juj

  committed
  8c788ae

  Copy full SHA for 8c788ae

Commits on Jul 27, 2019

• Add testing for orthonormal basis and fix SSE build

  juj

  committed
  298c3d2

  Copy full SHA for 298c3d2

• Micro-optimize basis_ps()

  juj

  committed
  a91d898

  Copy full SHA for a91d898

Commits on May 23, 2019

• Fix #56 by pre-translating AABB-Capsule intersections to a common reference frame. (TODO: consider doing the same for other types of collision tests as well)

  juj

  committed
  db162d3

  Copy full SHA for db162d3

Commits on May 22, 2019

• Update test case for #55 that is not supposed to collide. Closes #55.

  juj

  committed
  0f992be

  Copy full SHA for 0f992be

Commits on May 21, 2019

• Add some double precision operations to help debug issue #55. Fixes assertion failure, however the collision test in the test case still reports wrong result (no collision - whereas the AABB and Ca…

  Show description for ad32adf
  juj

  committed
  ad32adf

  Copy full SHA for ad32adf

Commits on Apr 4, 2019

• Cleanup minimum Circle computation.

  juj

  committed
  ae6dc5e

  Copy full SHA for ae6dc5e

Commits on Apr 3, 2019

• Optimize optimal circle computation code size - performance is about equal

  juj

  committed
  408c9e0

  Copy full SHA for 408c9e0

• Common subexpression elimination

  juj

  committed
  f9a5eaa

  Copy full SHA for f9a5eaa

• Drop unuseful inline

  juj

  committed
  3dd4cfc

  Copy full SHA for 3dd4cfc

• Micro optimize square roots from 2D minimum circle computation

  juj

  committed
  67411d4

  Copy full SHA for 67411d4

• Micro optimize square roots from 2D minimum circle computation

  juj

  committed
  1896419

  Copy full SHA for 1896419

• Retain smallest and simplest code size version of minimum circle computation, even though it results more search restarts and iterations

  juj

  committed
  4e29f13

  Copy full SHA for 4e29f13

• XXX save temp commit, using a bring to front list for minimum circle computation heuristics

  juj

  committed
  06dff7e

  Copy full SHA for 06dff7e

• Micro-optimize 2D smallest circle computation to perform less swapping

  juj

  committed
  bbd4e33

  Copy full SHA for bbd4e33

• Optimize 2D enclosing circle computation

  juj

  committed
  19f8da8

  Copy full SHA for 19f8da8

• Optimize performance of Circle2D::OptimalEnclosingCircle

  juj

  committed
  5ca7bc2

  Copy full SHA for 5ca7bc2

• Add benchmark for optimal enclosing circle computations, and a micro-optimization

  juj

  committed
  fe22d9a

  Copy full SHA for fe22d9a

• Fix bug with 2D enclosing circle radius computation

  juj

  committed
  789e663

  Copy full SHA for 789e663

Commits on Apr 2, 2019

• Avoid std::sort when using ContainerLib

  juj

  committed
  0c929d1

  Copy full SHA for 0c929d1

Pagination

Previous

Next