fix a mistake in big endian conversion

notaz/pcsx_rearmed#318

diff --git a/plugins/gpu_unai/gpu_inner_quantization.h b/plugins/gpu_unai/gpu_inner_quantization.h
index 4368ddb..8a4e935 100644 (file)
--- a/plugins/gpu_unai/gpu_inner_quantization.h
+++ b/plugins/gpu_unai/gpu_inner_quantization.h
@@ -89,7 +89,7 @@ GPU_INLINE u16 gpuColorQuantization24(u32 uSrc24, const le16_t *pDst)
 {
        if (DITHER)
        {
-               u16 fbpos  = (uintptr_t)pDst - (uintptr_t)gpu_unai.vram;
+               uintptr_t fbpos = pDst - gpu_unai.vram;
                u16 offset = ((fbpos & (0x7 << 10)) >> 7) | (fbpos & 0x7);
 
                //clean overflow flags and add