51
edits
SIMD/Uses/DCT: Difference between revisions
→Code examples
Maikmerten (talk | contribs) ("Uses" section) |
Maikmerten (talk | contribs) |
||
| Line 18: | Line 18: | ||
The Theora video codec features a DCT and IDCT using integer arithmetics. SSE2 implementations are available can be found in [http://svn.xiph.org/trunk/theora/lib/x86/sse2fdct.c sse2fdct.c] (the DCT) and [http://svn.xiph.org/trunk/theora/lib/x86/sse2idct.c sse2idct.c] (the IDCT). | The Theora video codec features a DCT and IDCT using integer arithmetics. SSE2 implementations are available can be found in [http://svn.xiph.org/trunk/theora/lib/x86/sse2fdct.c sse2fdct.c] (the DCT) and [http://svn.xiph.org/trunk/theora/lib/x86/sse2idct.c sse2idct.c] (the IDCT). | ||
The reference DCT for JPEG is defined using double precision floating values. A SIMD version of this can be found, e.g., in [https://github.com/mozilla/mozjpeg/blob/master/simd/jfsseflt-64.asm mozjpeg's jfsseflt-64.asm]. However, for performance reasons, the JPEG DCT is often approximated with integer arithmetics, as in [https://github.com/mozilla/mozjpeg/blob/master/simd/jfss2fst-64.asm jfss2fst-64.asm]. | |||
=== Involved data types === | === Involved data types === | ||
| Line 24: | Line 26: | ||
The IDCT of the Theora video compression format can implemented using 16-bit arithmetics, according to [http://www.theora.org/doc/Theora.pdf the Theora specification], section 7.9.3. Thus the natural SIMD data type to use there is [[SIMD/Types/int16x8 |int16x8]]. | The IDCT of the Theora video compression format can implemented using 16-bit arithmetics, according to [http://www.theora.org/doc/Theora.pdf the Theora specification], section 7.9.3. Thus the natural SIMD data type to use there is [[SIMD/Types/int16x8 |int16x8]]. | ||
For full-precision floating point DCTs, [[SIMD/Types/float64x2 |float64x2]] is an option. Much more common are integer approximations, though. | |||