Reply to Updates on compression woes
If you don't have an account, just leave the password field blank.
Made some progress parallelizing SharpCompress's BZIP2 algorithm, but it still feels quite slow. Taking about 3 minutes to compress the Periculo tar on the highest compression setting (block size 9). It is better than before, when it took forever (tm), but it is still horrible.
I'll see if I am able to optimize anything else about it and if not, next week I'll probably just continue working on the dinky using my current modifications...
Edit:
Well well well, I found some other bzip2 C# implementation that seems to be faster than SharpCompress multithreadded out of the box getting about 2 minutes in single thread mode for the Periculo tar. Modified it to also be multithreadded, and now I'm getting about 20 seconds to bzip2 the tar.
I should be able to put all of this together and have a Dinky update next weekend with the dmod packing feature (working fast as lightning too) and also partial* linux support.
* by partial I mean that currently it won't have the automated Dink game installer part, since that is a bit more complicated on linux.
I'll see if I am able to optimize anything else about it and if not, next week I'll probably just continue working on the dinky using my current modifications...
Edit:
Well well well, I found some other bzip2 C# implementation that seems to be faster than SharpCompress multithreadded out of the box getting about 2 minutes in single thread mode for the Periculo tar. Modified it to also be multithreadded, and now I'm getting about 20 seconds to bzip2 the tar.
I should be able to put all of this together and have a Dinky update next weekend with the dmod packing feature (working fast as lightning too) and also partial* linux support.
* by partial I mean that currently it won't have the automated Dink game installer part, since that is a bit more complicated on linux.
