vc2enc_dwt: use 32 bit coefficients by default

The problem is that with particularly complex images and especially at
high bit depths and 5-level transforms the coefficients would overflow,
causing huge artifacts to appear. This was discovered thanks to the fate
tests, which will have to be redone as this fixes a multitude of
problems and increases PSNR.

There is a slight performance drop associated with this change, making
the encoder slower by 1.15 times, however this is necessary in order to
avoid undefined behavior and overflows.

It would be worth to template the transforms to keep the performance for
8 bit images as 32 bit coefficients are unnecessary for that case, but
the primary use of the encoder is to encode video at 10 bits.

Reviewed-by: Christophe Gisquet <christophe.gisquet@gmail.com>
Reviewed-by: Michael Niedermayer <michael@niedermayer.cc>
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>

diff --git a/libavcodec/vc2enc_dwt.h b/libavcodec/vc2enc_dwt.h
index af6fe3e2a2088bd9245aead201f0659486cf1a7c..7fbbfbe0ed9cfc1168b8116e9d130787a86b6230 100644 (file)
--- a/libavcodec/vc2enc_dwt.h
+++ b/libavcodec/vc2enc_dwt.h
@@ -25,7 +25,7 @@
 #include <stddef.h>
 #include <stdint.h>
 
-typedef int16_t dwtcoef;
+typedef int32_t dwtcoef;
 
 enum VC2TransformType {
     VC2_TRANSFORM_9_7    = 0,   /* Deslauriers-Dubuc (9,7)  */