File: fuck_that_took_forever.webp (44 KB, 1361x823)

>Relative to lossless JXL filesize, how close in compression does a given lossless image codec get?
That's what I'm trying to answer here. A dozen common type 1920x1080 PNG images were used.

JXL KB = 100% and BMP KB = 0%. Lossless JXL/Webp/AVIF/oxiPNG/QOI tested. Anything above 100% means JXL was defeated.

cjxl in.png out.jxl -d 0.0 -e 10 --brotli_effort=11
cwebp -lossless -m 6 -q 100 in.png -o out.webp
avifenc -l -s 0 --cicp 1/13/16 in.png out.avif
oxipng -o 6 in.png
qoi "...\in.png" "...\out.qoi"

(BMP KB-JXL KB)=A | (BMP KB-Xcodec KB)/A= lossless compression relative to JXL lossless
ALL files: https://files.catbox.moe/x926hp.zip

Discussion for the Codex page Technology Codex.

The Codex page is written by this board's curators, so other users cannot edit it. This thread is where everyone else talks about it: post a reply if you have a question, spotted something out of date, or want to suggest what to add.

File: DDR5 6000 2x16GB.webp (65 KB, 1176x484)

>DDR5-6000 2x16GB (Average price in USD over last 18 months)
Source: https://pcpartpicker.com/trends/price/memory/#ram.ddr5.6000.2x16384
How long until prices go down?

File: eed1e127689e3681947e81582875dd55.gif (534 KB, 500x362)

Welcome to the Technology board!
For AI discussions please use the https://umigalaxy.com/explore/ai board

File: 4065.webp (101 KB, 800x800)

>uses an anode composite of silicon with a carbon matrix
>silicon anode alone swelled 300% during charging, cracked itself apart
>carbon matrix contain expansion better
>Si-C = ~1,000 charge/recharge cycles (potentially higher in the future)
>higher energy density, ~300 Wh/KG (potentially higher in the future)
>lower heat during charging (potential for faster charging)

https://www.androidauthority.com/samsung-first-phones-silicon-carbon-batteries-3689630/
https://phoneprice360.com/what-is-a-silicon-carbon-battery-phone-best-phones-using-it-in-2026/

File: chart.webp (7 KB, 1259x500)

I took the 73 Mbps 4K NLUUG Tears of Steel encode, scaled it to 720p, and used CRF 10 veryslow x264 to simulate 720p BD.
SSIM cheating was disabled in AV1 with tune 0 and variance boost, 720p was chosen to handicap AV1's 128x128 superblocks.
Even when nearly maxing out x264 compression , preset 4 AV1 = ~30% better compression and encodes ~40% faster as well.
I confidently declare that H.264 is now official boomer trash like flickering CRT displays.

ffmpeg -i tearsofsteel_4k.mov ^
-vf "scale=1280:720:force_original_aspect_ratio=decrease,pad=1280:720:(ow-iw)/2:(oh-ih)/2:black" ^
-c:v libx264 -preset veryslow -crf 10 -pix_fmt yuv420p input.mp4

base_av1=ffmpeg -i input.mp4 -c:v libsvtav1 -crf 28 -preset 4 -pix_fmt yuv420p10le -svtav1-params
psy_av1=keyint=5s:lookahead=120:tune=0:enable-variance-boost=1:variance-boost-strength=3:variance-octile=4

frame=17616 fps= 18 Lsize=  212825KiB
bitrate=2375.4kbits/s speed=0.73x elapsed=0:16:45.14

base_h264=ffmpeg -i input.mp4 -c:v libx264 -crf 18 -preset veryslow -pix_fmt yuv420p

frame=17616 fps= 13 Lsize=  304524KiB
bitrate=3399.0kbits/s speed=0.535x elapsed=0:22:51.14

https://ftp.nluug.nl/pub/graphics/blender/demo/movies/ToS/tearsofsteel_4k.mov

File: x264_crf_ssim_correlation.webp (97 KB, 649x884)

I am 90% sure I have fixed my windows batch SSIM AV1 auto-encoder script.

@echo off & setlocal enabledelayedexpansion & set "crf=60" & set "mincrf=20"

set "base_av1=-c:v libsvtav1 -preset 4 -pix_fmt yuv420p10le -svtav1-params"
set "psy_av1=keyint=5s:lookahead=120:tune=0:enable-variance-boost=1:variance-boost-strength=3:variance-octile=4"
set "opus=-c:a libopus -b:a 128k" & set "ffquiet=-hide_banner -loglevel quiet"

echo Weclome to my auto-SSIM AV1 encoder, only input.mp4 accepted, choose desired SSIM:
echo 1= 0.995 (autism quality) & echo 2= 0.990 (very high quality) & echo 3= 0.980 (high quality)
choice /c 123 /m "Enter number option on keyboard: "
if errorlevel 3 set ssim_target=980
if errorlevel 2 set ssim_target=990
if errorlevel 1 set ssim_target=995
echo:
:loop
ffmpeg !ffquiet! -i input.mp4 -crf !crf! !base_av1! !psy_av1! !opus! -y AV1_CRF-!crf!.webm 2>nul

for /f "delims=" %%A in ('ffmpeg -nostdin -r 30 -i AV1_CRF-!crf!.webm -nostdin -r 30 -i input.mp4 ^
-filter_complex "[0:v]format=yuv420p10le[v0];[1:v]format=yuv420p10le[v1];[v0][v1]ssim" -f null - 2^>^&1 ^| findstr "All:"') do set "a=%%A"

set "a=!a:*All:0.=!" & set "ssim=!a:~0,3!" & echo SSIM: 0.!ssim! @ CRF !crf! & echo:

if !ssim! lss !ssim_target! (if !crf! gtr !mincrf! (del /s /q AV1_CRF-!crf!.webm >nul 2>&1 & set /a crf-=2 & goto loop) )
:end

for %%Y in ("input.mp4") do set /a kb_mp4=(%%~zY+1023)/1024 & for %%Y in ("AV1_CRF-!crf!.webm") do set /a kb_webm=(%%~zY+1023)/1024
set /a saved_pct=(kb_mp4-kb_webm)*100/kb_mp4 & echo MP4= !kb_mp4! KB & echo AV1= !kb_webm! KB
echo **** !saved_pct!%% filesize reduction achieved with AV1 ****

pause

File: ssimnorm-vs-crf-value.webp (35 KB, 1254x800)

OHIO umigalaxy

I noticed this place supports AV1, hope it's cool if I crosspost from 4chan's /g/ board. I present you a script that targets SSIM, only 1 input video file per folder though.

@echo off & mkdir temp
:loop
set /p crf="CRF? (30-50 reccomended): "
for %%f in (*.mkv, *.mp4, *.mov) do (
    ffmpeg -i "%%f" -c:v libsvtav1 ^
    -crf %crf% -preset 4 -pix_fmt yuv420p10le -svtav1-params ^
    keyint=2s:lookahead=120:tune=0:enable-variance-boost=1:variance-boost-strength=3:variance-octile=4 ^
    -c:a libopus -b:a 64k -y "%%f_CRF-%crf%_AV1.webm"

 

echo: & echo SSIM result:
ffmpeg -hide_banner -loglevel error -i "%%f" -c:v libx264 -pix_fmt yuv420p10le ^
-preset ultrafast -crf 0 -y temp/temp_source_2_10b.mkv
ffmpeg -hide_banner -i "%%f_CRF-%crf%_AV1.webm" -i temp/temp_source_2_10b.mkv ^
-lavfi ssim -f null - 2>&1 | findstr "[Parsed_ssim"
echo:
)
goto loop

1 problem: AV1 cheats in SSIM. It intelligently preserves the structure of video but destroys fine detail. However these disable SSIM cheating:
tune=0 Prioritizes perceptual/visual quality
enable-variance-boost=1 allocates more bits to low-contrast areas (such as skin, clouds, or foggy scenes) that typically suffer from blurring and artifacting in fixed-quantization encoding.
variance-boost-strength=3 Applies medium strength variance boost, enough to improve quality but not bloat bitrate.
variance-octile=4 Applies medium strength for the new 8x8 variance algorithm, which is used for adaptive quantization.

All of these not only give you better/accurate SSIM scores but allow AV1 to create subjectively medium to high quality AV1 encodes, not just compete in objective metric pissing contests.

File: great_shrike.webp (525 KB, 5562x3708)

I love birds but high res images of them tend to be pretty bloated and not ideal for sharing.

Noticed Webp uploads are allowed here and aren't further compressed, very nice. Nice step up from ancient JPG.

I don't know too much about quality except that high quality SSIM scores are above 0.95 so I'm aiming for 0.96.

'cwebp -sharp_yuv -m 6 -mt -pass 4 -hint photo -size 540000 in.jpg -o out.webp' size=in bytes not KB

https://commons.wikimedia.org/wiki/File:Great_Grey_Shrike_in_Bhigwan_August_2025_by_Tisha_Mukherjee_02.jpg

File: screen_av1.webm (1.23 MB, 800x450)

It's using Staxrip so not ffmpeg advanced but it does give you access to SVT-AV1-PSY + a few advanced options like maxing out frame lookahead, VQ tune used by default.