Originally Posted by kapone
So, then it all boils down to cost benefit, time, your technical abilities, money...the usual suspects.
I completely agree.
In this regard, and at the risk of being overly quantitative, it may be helpful to try to articulate the feature dimensions' that have to be implicitly or explicitly addressed in order to successfully build a media server. I know we all have different skills, goals, and budgets. My thinking is that if we can use this list to quantify our priorities, it may simplify our decision process. Some of these are obvious and are only listed for completeness sake; some of them can simply be builder options; but some of them are pivotal and will significant affect how useful the profiles defined in this thread are. Anyway, here's the list, please feel free to offer improvements
Media Server Feature Dimensions:
A. Runtime model (Bit streamer vs. Power HTPC; power up on demand vs. 24/7 operation; HW RAID vs. SW RAID)
B. Fault Tolerance (Quality components, minimize runtime corruption, backup plan, battery backup. (IMHO, there's little point in building this server without very solid robustness.))
C. Operating Costs (Assuming good MTBF components and relatively straightforward array extensions, this is primarily a power consumption issue.)
D. Initial costs (Both money and setup time)
E. Capacity (Execution and network bandwidth / storage size)
F. Maintainability (Primarily issue is extending array, hopefully little else to worry about)
G. System Complexity and User Technical Skills (Is our target user an IT Engineer or motivated AV hobbyist)
H. Software (OS choice, RAID, Ripping tools & Process, Converters, Playback SW)
I'd love to say that I want all of these, but unfortunately some of them oppose each other and force tradeoffs. (E.g. Skill vs. costs vs. capacity) Ultimately, these are odditory's call and will be addressed in the profiles he defines, but how he addresses them will ultimately determine how many people benefit from this thread. So my thinking is that if people chime in now and let him know what they value, it should help the overall value of the thread. I have previously stated that I am primarily interested in a low energy, robust, bit streamer, and am willing to subvert other features' for these but clearly there are other points of view, and I think it would be great for people to register them before we get too much inertia going