Skip to content

Commit

Permalink
x86: Increase precision of the final inverse ADST transform stages
Browse files Browse the repository at this point in the history 
16-bit precision is sufficient for the second pass, but the first pass
requires 32-bit precision to correctly handle some esoteric edge cases.
  • Loading branch information
@gramner-twoorioles @Gramner
gramner-twoorioles authored and Gramner committed Sep 4, 2019
1 parent e2702ea commit a9315f5
Show file tree
Hide file tree
Showing 2 changed files with 693 additions and 310 deletions.
Loading

0 comments on commit a9315f5

Please sign in to comment.