Warning: "update" procedure in the first post is wrong!
The four download sites for the complete new version took forever, so I searched this thread for a torrent. Because I'm not the only one I found tons and started with this one. There are two patches which can be downloaded from Nopy which gives a sha1sum to check if everything is OK. Hashes are also used in torrents. So now I've the following verified sources:
The version in the name of the first patch suggests that the data in the first post is wrong. However if the version in the torrent is correct, both should work. The second patch has a README and again this suggests that the data in the first post is wrong. In theory this patch could work, but there are scenarios in which it isn't.
In the meanwhile after 2,5 hours the full version of the anonfile was downloaded. Now I could test if the complex update procedure which doesn't completely match the intermediate versions is still correct. Well it isn't. There are many differences including in for example images from earlier versions. Again images are probably not a problem, but I also found numerous differences in the code files.
In my signature a game where the final version is over 27 GiB. Early on I proofed that binary patches are possible using the Linux bsdiff command. Another player created a very good Windows executable and Bash -- for macOS and Linux -- using a similar command xdelta3. It also uses hashes to check the original version and hashes to check if the patch procedure correctly works. This convinced the developer to give access to the version so clean patches could be created. This way everybody knows that there aren't any version mishaps. Or different sources which could -- maybe unknowingly -- have altered there version.
The four download sites for the complete new version took forever, so I searched this thread for a torrent. Because I'm not the only one I found tons and started with this one. There are two patches which can be downloaded from Nopy which gives a sha1sum to check if everything is OK. Hashes are also used in torrents. So now I've the following verified sources:
Code:
CollegeKings-10.0.2-pc.zip
=> sha1sum: a1403aa775f95ba87e3da237dced3770c0304eb9
CollegeKings_v10.0.2_Patch.zip
=> Nopy sha1sum: 04216a4b560d223712f68ea4a0dac300d605de88
=> post #1 says 10.0.0 -> 10.2
CollegeKings_v11.0.1_Patch.zip
=> Nopy sha1sum: a1f83e0ecb07c481ff8184df2d9d8ece5d3500a8
=> post #1 says 10.2 -> 11.0.1
=> README.txt says 10.1.1 -> 11.0.1In the meanwhile after 2,5 hours the full version of the anonfile was downloaded. Now I could test if the complex update procedure which doesn't completely match the intermediate versions is still correct. Well it isn't. There are many differences including in for example images from earlier versions. Again images are probably not a problem, but I also found numerous differences in the code files.
In my signature a game where the final version is over 27 GiB. Early on I proofed that binary patches are possible using the Linux bsdiff command. Another player created a very good Windows executable and Bash -- for macOS and Linux -- using a similar command xdelta3. It also uses hashes to check the original version and hashes to check if the patch procedure correctly works. This convinced the developer to give access to the version so clean patches could be created. This way everybody knows that there aren't any version mishaps. Or different sources which could -- maybe unknowingly -- have altered there version.
