I have spent years building an extensive movie collection that requires lots of shelf space and is useless when away from my Blu-ray player. Before buying yet another bookshelf I decided to research ways to digitize my collection. What began with a dozen movies occupying a few gigabytes has since grown to include my entire movie collection occupying over 7,000 gigabytes. As usual, the road to where I am today was not easy and included plenty of missteps along the way.

Iteration 1 - External Hard Drive and Poor Quality Movies

The first iteration wasn't even a media server. It was an old 512GB external hard drive connected to a Raspberry Pi (credit card sized Linux computer) running XBMC. It was slow, movies were poorly ripped, and only include a few of my favorite movies compressed with Handbrake. It satisfied the immediate need but left a lot to be desired. I wasn't sure how to solve the speed issue but knew the remaining two issues were well within my control and could be solved by ripping the movies without compression and ripping my entire collection.

Iteration 2 - More External Hard Drives and Blu-ray Quality Movies

The second iteration started with two, brand-new, 3TB external hard drives and ripping movies without compression. Using Handbrake was a great starting point but finding a consistent way to rip movies at Blu-ray quality wasn't easy. Searching for an alternative yielded MakeMKV which converts disc-based media into play-anywhere MKV files. Because MakeMKV is only converting the video format rather than compressing it, ripping only takes a fraction of the time (30 minutes vs 8 hours in Handbrake) and preserves the original quality of the video and audio. This was a huge benefit since I was looking for Blu-ray quality without the disc. Though it took a while, ripping my entire collection resolved the "favorites only" problem and using MakeMKV produced picture-perfect conversions. However, using MakeMKV presented two new issues that needed to be addressed.

Blu-ray Quality Issue 1 - HD Audio

The first issue introduced by MakeMKV (though not the fault of MakeMKV) was that, as of this writing, the Raspberry Pi cannot passthrough high definition audio. Playing a file with HD audio results in choppy video with downgraded audio or a lack of audio all together. This set off the hunt for a small, inexpensive, quiet device capable of handling the HD video and accompanying audio. I tried various streaming media players but all failed to perform as needed. Luckily, Lenovo had recently released the Q190 which is described as an "entertainment hub." Being a full-blown Windows PC it can handle XBMC and HD video/audio. Just to make things a little easier, I installed OpenELEC to completely take Windows out of the equation.

Blu-ray Quality Issue 2 - Storage Needs

The second issue introduced by MakeMKV (again, not the fault of MakeMKV) was the file size of converted movies; upwards of 30GB for a single Blu-ray. This meant that my two 3TB external drives quickly reached capacity. Furthermore, by simply using multiple external hard drives I had no protection against hard drives failure. Though I had "backups" of every movie (my original discs) I did not enjoy the thought of spending months re-converting my entire collection. This drove me to the final iteration and an actual media server.

Iteration 3 - Custom Built Media Server

The third (and current) iteration is a custom built media server. I needed something that could store my entire movie collection as well as provide a fail-safe against drive failure. After exhaustive research I decided on standard desktop hardware and software that would bring everything together. The end result is a media server with 12 terabytes of usable storage (7.5TB in use) with over 1,100 MKVs representing my entire movie and TV collection. Furthermore, there is room to more than double the currently available usable storage which is all protected against single drive failure (multiple drive failure if I so choose). Seemingly, the new media server provides for all of my original needs without any downsides (other than a slight hit to my bank account).

As mentioned at the opening of this article, the current media server is the result of many missteps that cost hundreds of dollars but at least showed me which ideas would not work. The above will certainly not be the answer or even path for every movie/technology aficionado but hopefully it will save someone the trouble and money I expended. With this background out of the way, future articles will explore the current state of specific hardware and software. I say "current state" because even in the course of writing this group of articles some of the above has changed (and some changed back again). This is a constantly evolving project and I will be sure to keep these posts updated as I discover bigger or better alternatives.