Release 0.5.2

What's Changed

• High bit depth fix, doc generation fix by @awxkee in #51

Full Changelog: 0.5.1...0.5.2

Release 0.5.1

What's Changed

• ARM f16 compilation fixes, alpha improvements by @awxkee in #50

Full Changelog: 0.5.0...0.5.1

0.5.0

What's Changed

• Added stride support AR30/RA30
• Reworking f16 by @awxkee in #46
• RGBA1010102, Fuzzer by @awxkee in #47
• AVX2 scaling RGBA u16 by @awxkee in #48
• AR30/RA30 x86 by @awxkee in #49

Full Changelog: 0.4.2...0.5.0

Release 0.4.2

What's Changed

• Stride f32, masked loads AVX-512, Vertical AVX-512 u16 by @awxkee in #26
• Build fix by @awxkee in #27
• Added RGB8 AVX2 by @awxkee in #28
• Dot 8 RGB8 by @awxkee in #29
• Added ubs for RGB8/RGBA8 by @awxkee in #30
• NEON faster u16 pre-multiplication if fp16 available by @awxkee in #31
• Some improvements by @awxkee in #32
• Refactor and improvements by @awxkee in #33
• x86 improvements and bugfixes by @awxkee in #34
• Added CbCr support by @awxkee in #35
• Planar u8, improvements by @awxkee in #36
• NEON bi-planar horizontal pass by @awxkee in #37
• Dropped i8mm by @awxkee in #38
• Accelerate benching by @awxkee in #39
• x86 CbCr8 added by @awxkee in #40
• Added C bindings, some improvements by @awxkee in #41
• F16/F32 small improvements by @awxkee in #42
• Refactor and improvements by @awxkee in #43
• Refactor and improvements by @awxkee in #44
• Clippy by @awxkee in #45

Full Changelog: 0.4.1...0.4.2

Release 0.4.1

Fix AVX2 alpha premultiplication panic

Release 0.4.0

AVX-512 added, changed ImageStore, AVX-VNNI added, added NEON i8mm, common improvements and speeding up

Release 0.3.6

Improvements, fuzzing, fixing bugs discovered by fuzzer

Release 0.3.5

Guard RDM feature

Release 0.3.4

Alpha unpremultiplication bugfix

Release 0.3.3

SSE unpremultiply bugfix