From: Michael Niedermayer <michael@niedermayer.cc>
To: FFmpeg development discussions and patches <ffmpeg-devel@ffmpeg.org>
Subject: [FFmpeg-devel] [PATCH 1/2] avcodec/ffv1: flip half of float16
Date: Sat, 18 Jan 2025 21:19:45 +0100
Message-ID: <20250118201946.2389042-1-michael@niedermayer.cc> (raw)
float16 is a sign bit followed by "positive is larger" exponent and mantisse
that way the up direction changes between negative and positive numbers
flipping the exponent and mantisse for positive numbers gives a compression
gain of 0.5% for both RCT and noRCT modes using the
ACES_OT_VWG_SampleFrames testset
Sponsored-by: Sovereign Tech Fund
Signed-off-by: Michael Niedermayer <michael@niedermayer.cc>
---
libavcodec/ffv1dec_template.c | 10 ++++++++++
libavcodec/ffv1enc_template.c | 11 +++++++++++
2 files changed, 21 insertions(+)
diff --git a/libavcodec/ffv1dec_template.c b/libavcodec/ffv1dec_template.c
index e983d1ba648..0be17b54bbb 100644
--- a/libavcodec/ffv1dec_template.c
+++ b/libavcodec/ffv1dec_template.c
@@ -185,6 +185,16 @@ static int RENAME(decode_rgb_frame)(FFV1Context *f, FFV1SliceContext *sc,
b += g;
r += g;
}
+ if (f->flt) {
+ if (!(r&0x8000))
+ r ^= 0x7FFF;
+ if (!(g&0x8000))
+ g ^= 0x7FFF;
+ if (!(b&0x8000))
+ b ^= 0x7FFF;
+ if (transparency && !(a&0x8000))
+ a ^= 0x7FFF;
+ }
if (lbd)
*((uint32_t*)(src[0] + x*4 + stride[0]*y)) = b + ((unsigned)g<<8) + ((unsigned)r<<16) + ((unsigned)a<<24);
diff --git a/libavcodec/ffv1enc_template.c b/libavcodec/ffv1enc_template.c
index bc14926ab95..4d20cb93627 100644
--- a/libavcodec/ffv1enc_template.c
+++ b/libavcodec/ffv1enc_template.c
@@ -180,6 +180,17 @@ static int RENAME(encode_rgb_frame)(FFV1Context *f, FFV1SliceContext *sc,
r = *((const uint16_t *)(src[2] + x*2 + stride[2]*y));
}
+ if (f->flt) {
+ if (!(r&0x8000))
+ r ^= 0x7FFF;
+ if (!(g&0x8000))
+ g ^= 0x7FFF;
+ if (!(b&0x8000))
+ b ^= 0x7FFF;
+ if (transparency && !(a&0x8000))
+ a ^= 0x7FFF;
+ }
+
if (sc->slice_coding_mode != 1) {
b -= g;
r -= g;
--
2.48.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
next reply other threads:[~2025-01-18 20:19 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-01-18 20:19 Michael Niedermayer [this message]
2025-01-18 20:19 ` [FFmpeg-devel] [PATCH 2/2] avcodec/ffv1: Compactify floats Michael Niedermayer
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250118201946.2389042-1-michael@niedermayer.cc \
--to=michael@niedermayer.cc \
--cc=ffmpeg-devel@ffmpeg.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Git Inbox Mirror of the ffmpeg-devel mailing list - see https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
This inbox may be cloned and mirrored by anyone:
git clone --mirror https://master.gitmailbox.com/ffmpegdev/0 ffmpegdev/git/0.git
# If you have public-inbox 1.1+ installed, you may
# initialize and index your mirror using the following commands:
public-inbox-init -V2 ffmpegdev ffmpegdev/ https://master.gitmailbox.com/ffmpegdev \
ffmpegdev@gitmailbox.com
public-inbox-index ffmpegdev
Example config snippet for mirrors.
AGPL code for this site: git clone https://public-inbox.org/public-inbox.git