60f5d0e34a
The value range supported by the 'Animated_rect' utility is constrained to 2^11 by the used 32-bit integer type. The interpolation is performed on fixpoint numbers with a fractional precision of 10 bits. The non-linear interpolation as performed by the 'Lazy_value' utility involves squaring the values. Hence, the interpolated integral values must not exceed 2^11 (2*11 bits for the square plus 10 bit for the fractional part require 32 bits). This range is too small for high-resolution displays. Hence this patch replaces the 32-bit integer type by a 64-bit integer type. Fixes #3627 |
||
|---|---|---|
| .. | ||
| aes_cbc_4k | ||
| depot | ||
| gems | ||
| nano3d | ||
| polygon_gfx | ||