FileForums - Best Compression Methods for 'Specific' Games. Q&A

FileForums (https://fileforums.com/index.php)

- Conversion Tutorials (https://fileforums.com/forumdisplay.php?f=55)

- - Best Compression Methods for 'Specific' Games. Q&A (https://fileforums.com/showthread.php?t=99554)

Quote:

Originally Posted by KaktoR (Post 506169)

No problem, and have fun

Code:

FreeArc 0.67 (March 15 2014) creating archive: Conversion_Output\Data1a-01.bin

Compressed 1 file, 494,902,306 => 1,014,185,526 bytes. Ratio 204.93%

Compression time: cpu 0.27 sec/real 107.59 sec = 0%. Speed 4.60 mB/s

All OK



FreeArc 0.67 (March 15 2014) extracting archive: Conversion_Output\CONVERSION\Data1a-01.bin

Extracted 1 file, 1,014,185,526 => 494,902,306 bytes. Ratio 204.93%

Extraction time: cpu 0.70 sec/real 93.77 sec = 1%. Speed 5.28 mB/s

All OK

Thanks :)

Horizon Zero Dawn Remastered

Same as Forbidden West

Quote:

Forza Horizon 5

Quote:

Version 656.386

Quote:

xtool:c256mb:mpreflate:mzlib+xtool:dd3+LOLZ_ULTRA

Without 4K movies

143GB --> 85.4GB

Quote:

Originally Posted by IRAQIGHOST (Post 506227)

Without 4K movies

143GB --> 85.4GB

fitgirl repack 73.5 gb. How ?

Quote:

Originally Posted by demon964 (Post 506243)

fitgirl repack 73.5 gb. How ?

Using LOLZ_ULTRA usually give worse results rather than using lolz. Plus she has more experience.

+ you can compress video files and audio banks too.

^^ Oh yeah, that also

Plus FG uses 2047m:mt1.

Quote:

Originally Posted by shazzla (Post 506248)

Plus FG uses 2047m:mt1.

No, I don't :)

Quote:

Originally Posted by FitGirl (Post 506250)

No, I don't :)

Then something like that. nevermind.. :D

Incursion Red River

Code:

srep+lolz

Game size: 53.4GB
Compressed size: 13.6GB

Paks are not compressed and are not encrypted.
Compression took ~21 hours on 4c/8t CPU

Spanish (ES) language files ripped

srep+lolz+oggre

This title was an interesting one as the language files (stored as fsb containers) do not have any consistent ordering within the Unity resource archives, majority. An AI/ML model, specifically an ASR model was used in conjunction with some Python scripting to determine whether a given input contained English or Spanish dialogue. Approx. 200 files had to be processed manually as they contained only a single word, or had a heavy post-processing filter applied, making it difficult to get a definitive result, overall success rate was approx. 94%. English was preferred in analysis to reduce likelihood of English files being accidentally misidentified and ripped.

Final total size of archives: 965 MiB

Based on v1.042, GOG Release

I estimate there might still be a few Spanish files I was not able to identify, with more aggressive packing techniques I think it's definitely possible to reduce overall size to ≤900 MiB.

Attached is a checksum file with all ripped files, you're welcome.

Assassin's Creed Mirage (English and Arabic Only)

*.*.forge files

Quote:

xtool:manvil:core_2.8.4:dd3+LOLZ_NORMAL

Sound files

Quote:

xtool:dd3+bcm

rest files

Quote:

srep_new:393b+4x4:b96mb:lzma:ultra:bt4:fb273:lc8:m c1000000:96mb

38.6GB --> 20.5GB

Dragon's Dogma 2

what's the best method ? thx!