Is a 0-day tracker worth it compared to specialized trackers?

[Rart]

I'm actually curious as while I am vaguely familiar with the concept of compression for encoding/transcoding things such as music and video, I have no idea how the compression of files in general for actually making files...you know, smaller, actually works. I'm assuming the two concepts could be somewhat similar, although obviously things such as varying bitrates during certain lulls in the music/videos wouldn't work for something like software.

The reason I bring this up is I was wondering would it be possible for trackers to you know, use winrar/whatever archiving program for actually... compressing the file? Making less strain on the servers and the users, and allowing files to be downloaded more quickly? Is this feasible? Is there some technical reason for why trackers haven't done this?

[megabyteme]

For most of us, bandwidth is abundant. The work involved to reduce a file is not worth the effort to do so anymore. There was a time when file size mattered, that was when everyone used dialup connections. While what you are suggesting may make some difference, nobody would really notice it. Can you tell the difference between downloading a 1GB file and an 800MB? Is this going to change your mind about what you are downloading? I expect most people, since we are not paying by the MB, do not care. We want what we want, and we don't want complications. None seems to be the standard tolerance. We are spoiled by cheap, abundant bandwidth and a never ending supply of media. We can also look for this media on a number of trackers. We'll grab from the one that causes us the least amount of resistance.

[mjmacky]

I'm actually curious as while I am vaguely familiar with the concept of compression for encoding/transcoding things such as music and video, I have no idea how the compression of files in general for actually making files...you know, smaller, actually works. I'm assuming the two concepts could be somewhat similar, although obviously things such as varying bitrates during certain lulls in the music/videos wouldn't work for something like software.

Many executable files are already compressed (also called self-extracting archive), though that's not to say they cannot be compressed further, you would just get a marginal reduction in file size. In other words, they're like a zip file that doesn't require another program to extract it.

General compression is similar to what you're already familiar with, where repeated information/sequences/code gets indexed, which is contained in the archive, and the index is used to rebuild the file(s) when extracting (or rebuild the media file). There are many algorithms that achieve this, mp3s use a Fourier transform-type algorithm, H.264 codes frames as differential to a reference frame, etc. The entire point is to get rid of redundancy.

The different algorithms do not stack because once you have reduced redundancy with one method, the subsequent algorithm would not be able to calculate that much more. In fact, if you tried to compress a file that was already compressed using a superior method, you could end up with a file that is slightly larger than the original (on account of the index).

In case this seemed to pleasant and helpful a disposition, suck a dirty dick, bitch.