Compression using single threaded process at Intel(R) Core(TM)2 Quad CPU Q9550 @ 2.83GHz :
real time is the execution time and user time is processortime*processes
time xz -9 -k pwr-ninlnd.wud
real 165m26.508s
user 164m26.633s
sys 0m43.435s
time zpaq a pwr-ninlnd.wud.zpaq pwr-ninlnd.wud...
I did some compression testing on a couple of the releases and as previosly noted they are pretty compressible. Given that big parts of the data is encrypted they should be able to make even small using a strategy similar to the wia filetype.
25025314816 (24G) pwr-ninlnd.wud
3056920574 (2.9G)...
This is quite interesting since it means that the wiiu discs are not filled with random junk like the wii since it is highly compressible. It also means that the probable compressible junk on the disc is not encrypted if did not need any decryption before compressing it.
I have tested some additional algorithms for post compression of the wia
format. The most reasonable was to apply it to uncompressed wia files. The
results below is on the DATA partition only. Lossless compression only.
4699979776 RMCE01.iso
2776629248 RMCE01.wbfs
2713317692 RMCE01.wia (no...
Thank you for keeping developing wit. It is a great tool, and will soon even be in debian stable.
I've decided to archive my wii dumps with the wia format and I've repeated your testing for one of my dumps which seem to give the same conclusion:
raw: 4699979776
wbfs: 1738539008
wbfs + xz -9...