Skip to content
Branch: master
Commits on May 15, 2019
Commits on May 19, 2018
  1. Range adjustments for 8-bit noise functions, now preserving a (very) …

    kriegsman committed May 19, 2018
    …slight bit more precision, too.
Commits on May 17, 2018
Commits on May 16, 2018
  1. Fix discontinuties in noise functions, both 8- and 16- bit versions. …

    kriegsman committed May 16, 2018
    …If you need the old behavior, you can edit fastled_config.h and change FASTLED_NOISE_FIXED to 0. Fixes #416.  Thanks to everyone who reported this and helped track it down.
Commits on Mar 24, 2018
  1. Merge pull request #570 from per1234/relative-include

    kriegsman committed Mar 24, 2018
    Use relative paths for internal includes
Commits on Jan 13, 2018
Commits on Jan 12, 2018
  1. MEK: bumping version number to do a release picking up numerous commu…

    kriegsman committed Jan 12, 2018
    …nity contributions -- thank you!
Commits on May 29, 2017
  1. Added AVR asm version of ease8InOutQuad which is a little faster, and…

    kriegsman committed May 29, 2017
    … a little smaller than the C version; it also preserves one additional low-order bit of precision so while it may not give exactly the same numeric results as the C in all cases, it presumably gives 'better' (smoother) ones.
Commits on May 28, 2017
Commits on Mar 24, 2017
  1. Force gradient palettes to be aligned in memory on ARM. Only needed o…

    kriegsman committed Mar 24, 2017
    …n some ARM platforms, but rather than play whack-a-mole with future chipsets and their requirements, we'll just be a little extra careful here. Fixes #428.
Commits on Feb 22, 2017
Commits on Feb 2, 2017
  1. Added AVR asm version of blend8: smaller, faster. Marked blend8 as 'a…

    kriegsman committed Feb 2, 2017
    …lways inline', as it probably should have been previously, since it results is smaller and faster code in almost every case, since the body is only about 12 AVR instructions. ATtiny uses 'C' version since the AVR assembly code uses the hardware multplier, which the 12-years-old-now ATtiny doesn't have.
Commits on Jan 20, 2017
  1. Fixed loss-of-precision bug in pixel blending. Pixel blending should …

    kriegsman committed Jan 20, 2017
    …be a little smoother now and free from error noise in the low bits. This includes palette colors returned from ColorFromPalette, which are now slightly smoother. Also introduced a new lib8tion function, blend8( byte, byte, fract8). The new pixel blending should have negligible effect on code size, and it is ever so slightly faster than the old version with the error. Fixes #404.
Commits on Jan 15, 2017
  1. Updated scale16 and scale16by8 for AVR when FASTLED_SCALE8_FIXED is 1…

    kriegsman committed Jan 15, 2017
    …. scale16(foo,65535) now equals foo. Updated DemoReel100 to stay within range of NUM_LEDS on non-AVR platforms. Fixes #368.
Commits on Jan 8, 2017
Commits on Dec 9, 2016
  1. Merge pull request #379 from jake-b/master

    kriegsman committed Dec 9, 2016
    Fix for crash caused when loading large palettes (Issue #378)
Commits on Feb 20, 2016
Commits on Feb 17, 2016
  1. Change a bunch of 171s to 170s to prevent overflow with the 'fixed' s…

    kriegsman committed Feb 17, 2016
    …cale8. Don't ask my why these are preprocessor constants. (No reason.)
  2. hsv2rgb_rainbow: remove unneeded (and overflowing) +1 to color channe…

    kriegsman committed Feb 17, 2016
    …ls in the inlined, deconstructed version of nscale8_video.
  3. In ColorFromPalette(s), skip redundant (and overflowing) increment of…

    kriegsman committed Feb 17, 2016
    … values after scale8 if scale8 is 'fixed'
Commits on Jan 22, 2016
  1. Speedup in basic HSV to RGB conversion. On AVR this takes us from fro…

    kriegsman committed Jan 22, 2016
    …m about 68,000 conversions/second to about 105,000 conversions/second.
Commits on Jan 7, 2016
  1. 12% speedup in noise on AVR (net with previous commit) by using inlin…

    kriegsman committed Jan 7, 2016
    …e avg15 on AVRs with MUL. Code size goes up 8 bytes for 16-bit, 3-D noise.
  2. 3% speedup in noise on AVR by eliminating a shift-loop. Math is ident…

    kriegsman committed Jan 7, 2016
    …ical, output results tested and found identical in every case tested.
  3. scale16 faster and smaller - no need to initialize 'result' to zero s…

    kriegsman committed Jan 7, 2016
    …ince we're going to overwrite it. A little gcc asm shenanigans required to get this all right in terms of '=r' versus '+r' on the variable usage tag in the asm block. Basically the = versus + don't want to think about multi-byte variables. Coincidentally, neither do I.
Commits on Dec 29, 2015
  1. Noise behavior can be reverted for backward-stability of animation de…

    kriegsman committed Dec 29, 2015
    placed BEFORE the #include of FastLED.h
You can’t perform that action at this time.