Internet Radio Update
My quest to run an internet radio station is inching closer to reality. I have spent the better part of the last month testing the capabilites of my home network for radio broadcasting, which has included several 24-hour test runs of different formats to accurately gain an idea as to what the demand for this venture would be. Below are the results.
Testing details
I utilized the free Shoutcast servers for the three tests. Based on current broadband limitiations, I had to go with the mp3PRO format and 64kbps bitrate to have 6 simultaneous listeners. For those unfamiliar to mp3PRO, click here to learn more. After each test, an analysis of each days streams statistics was completed. This analysis included the total number of unique IPs logged, length of listen, average listenership, and total number of repeat listeners.
Day 1 - Classic Alternative/Modern Rock
During this first 24-hour period, the playlist was a healthy rotation of classic alternative tracks (early 80s alternative music) and more modern flair heard on most alternative rock stations today. Analysis concludes the following:
Total unique listeners: 24
Total repeat listeners: 5
Average length of listenership: 0:57:32
Longest consecutive stream: 3:24:21
Day 2 - Classic Rock/Hair Metal
This 24-hour period featured rock from the 70s, 80s, and early-90s. Examples of the 70s would be Led Zeppelin, Pink Floyd, The Doors, BTO, etc. 80s consisted of Tom Petty, Def Leppard, and the hair bands, and the early 90s focused on bands like Faith No More, Metallica, Guns N' Roses, Alice in Chains, and Megadeth. Analysis results are:
Total unique listeners: 38
Total repeat listeners: 12
Average length of listenership: 0:46:34
Longest consecutive stream: 1:51:54
Day 3 - Bootleg concerts
This unique format for the last 24-hour period consisted of 10 of my favorite bootlegs that I have gathered over the years. This format was the most structured for an automated process, for I had to actually pick the shows I wanted to broadcast and set their playlist up manually. There was no random function on this stream, so it was straight shows. To alleviate the possibility of stream recording this particular set of shows so that they don't end up being burned onto a CD and sold to unsuspecting kids via ebay, I placed 10-second tags during the show to break it up and destroy the flow of each show. I think it worked rather well. Results:
Total unique listeners: 18
Total repeat listeners: 6
Average length of listenership: 4:27:41
Longest consecutive stream: 13:17:38
According to the results, the most "viewed" format was Classic Rock/Hair Metal, but the format with the most loyal listenership is the Bootleg concert format. For the classic rock format, I assume the lack of quality rock stations that play GOOD rock has caused the mainstream listeners to search for their fix online, so I attribute the larger audience to that. Of course, these listeners also weren't listening for very long, so either they weren't satisfied with what they were hearing, or they got their quick fix and moved on. The modern rock format has more loyalty than classic rock, as more than half the listeners listened for at least 90 minutes, but there were several who listened for a couple minutes, and then dropped out. As for the bootleg concert format, obviously the numbers don't lie. This is probably the most respected and desired format, as everyone digs a live show. Because of bandwidth limits, the numbers are skewed severely in the direction of failure, and had I not scoured over the total listener time, I wouldn't have noticed that the lack of listeners was due mainly to my listener max being MAXED OUT almost 95% of the time. One listener was logged into my stream for over half the entire testing period, which says a lot. So I think what's going to happen is that I am going to start out with the bootleg format, and when that grows tired, I will change it up, mixing choice performances into the normal rotation of rock music. Now, if only I could decide on who to host my content...
Subscribe to:
Post Comments (Atom)

No comments:
Post a Comment