5798 Commits (da1b85bee38b4e4ceded1c57d25ac13a2a0e8f22)

Author SHA1 Message Date
Dmitry Pentin da1b85bee3 Final cleanup of the non-simd 420 rgb -> ycbcr conversion code 5 years ago
Dmitry Pentin 078703b595 Added docs, renamed LuT converter for 444 and 420 subsampling methods, added debug guards 5 years ago
Dmitry Pentin 44bae0b79e Made non-simd ycbcr lut converter code more readable 5 years ago
Dmitry Pentin 2e25a3ee34 Optimized non-simd ycbcr lut converter code 5 years ago
Dmitry Pentin 7896e24606 Improved non-simd ycbcr lut converter code 5 years ago
Dmitry Pentin de176b699e Initial 420 subsampling lut conversion implementation 5 years ago
Dmitry Pentin 672da457d3 Finished 444 subsampling converter 5 years ago
Dmitry Pentin cb1acaec78 Finished 420 subsampling converter 5 years ago
Dmitry Pentin 5fc29a2e98 Introduced separate 420 converter 5 years ago
Dmitry Pentin 953095f1b9 420 converter fixes 5 years ago
Dmitry Pentin 12b4b83cb6 444 converter fixes 5 years ago
Dmitry Pentin 13e7cf358f Divided YCbCr converters into 444/420 subsampling categories 5 years ago
Dmitry Pentin 0d94435d65 Simplified LoadAndStretchEdges call logic 5 years ago
Dmitry Pentin 39569866fc Added debug guard checks to LoadAndStretchEdges 5 years ago
Dmitry Pentin d6db6b6be7 Fixed compilation errors for non-intrinsic platforms 5 years ago
Dmitry Pentin d50e255c85 [WIP] Implemented 16x8 420 subsampling convertion 5 years ago
Dmitry Pentin 052ebde3ad Replaced GenericBlocl8x8 with Span in ycbcr converter 5 years ago
Dmitry Pentin 8a7749644a Imporved internal rgb -> rcbcr conversion api for 420 subsampling 5 years ago
Dmitry Pentin 201c5341e6 Fixed HuffmanScanEncoder error 5 years ago
Dmitry Pentin 2ad3ddb036 [WIP] Introduced RgbToYCbCrConverterVectorized 420 sampling 5 years ago
Dmitry Pentin a845c00f6f Simplified RgbToYCbCrConverterVectorized.Convert() method 5 years ago
Dmitry Pentin 6ac2b6660b Added comments to vectorized rgb->ycbcr converter for further code changes 5 years ago
Dmitry Pentin 169e98bbcd Simplified Block8x8F.DivideRoundAll() method 5 years ago
Dmitry Pentin 9c0999e9db Huffman lookup tables are now integers instead of unsigned integers 5 years ago
Dmitry Pentin 16842496be Brought back if check 5 years ago
Dmitry Pentin 81979e0f29 Improved flush logic after main encode methods run 5 years ago
Dmitry Pentin 7fb8feef50 Fixed xml docs 5 years ago
Dmitry Pentin 52e6036268 Reimplemented Emit methods in HuffmanScanEncoder to get rid of unreadable amount of int/uint casts 5 years ago
Dmitry Pentin 70474c8fae Removed redundant enum casting durint huffman encoding 5 years ago
Dmitry Pentin ceb4fdfae0 Replaced unsafe Block8x8F/Vector4<float> -> Vector256<float> casts 5 years ago
Dmitry Pentin d2510036a6 Implemented fallback code for runtimes where BitOperations class is not supported. 5 years ago
Dmitry Pentin 28ea2adb08 Fixed comments, removed todo, updated benchmark results 5 years ago
Dmitry Pentin 0664f298d9 Replaced bit count lookup table to lzcnt implementation, Added MinimimBitsToStore to Numberics.cs 5 years ago
Brian Popow 2e87f7ac9a Add Tiff format to the default Configuration 5 years ago
Brian Popow 787d63000f Rework sanitize and set encoder options: BitsPerPixel should be the primary source of truth 5 years ago
Dmitry Pentin 86abb73799 Made FDCT8x8_Avx(...) method prettier with SimdUtils 5 years ago
Dmitry Pentin a65e50377d Added MultiplySubstract method to the HwIntrinsics 5 years ago
Dmitry Pentin ae85722da6 Simplified WriteDefineHuffmanTables method 5 years ago
Brian Popow ccc3f9b881 Rework setting tiff encoder parameters according to review 5 years ago
Dmitry Pentin 0424d8db71 Codestyle changes 5 years ago
Dmitry Pentin dfb181db8a Combined FDCT and IDCT code into single file 5 years ago
Dmitry Pentin a8f717d781 Made DCT code prettier with SimdUtils, added summary to 8x8 dct methods, added debug assertion 5 years ago
Dmitry Pentin f1886add16 Revert "Block8x8F.TransposeInto no longer uses unsafe casts (partially)" 5 years ago
Dmitry Pentin 980f2d2e7f Revert "Block8x8F.MultiplyInPlace no longer use unsafe casts" 5 years ago
Dmitry Pentin d4fa8b254b Rolled back to initial JpegEncoderCore options implementation. 5 years ago
Brian Popow 11a4e2027b Add Compression, PhotometricInterpretation and Predictor to TiffFrameMetadata 5 years ago
Dmitry Pentin 84a143d095 Moved end of image marker writing code to a separate method 5 years ago
Dmitry Pentin 5b05a0a1da Added QoL throw helper method for jpeg w/h size check before encoding 5 years ago
Dmitry Pentin 1b1d136f8c Fixed unresolved reference this.colorType 5 years ago
Dmitry Pentin 7e0a317461 Moved encode method choice to the JpegEncoderCore 5 years ago