A few times I’ve seen people complaining about unacceptably low volume being provided by one of the Gigaset handsets. Most recently this was being said about the S67H handset, but I can recall a similar complaint about the A58H as well. We don’t feel that the handsets we’ve used have a problem in this area.
However, when the complaint occurs more than once it makes me think that perhaps a little investigation is worthwhile. Could this be production variance in the handsets? Is there a difference between the various models of handsets? How do they compare to the output of some respected desktop phones?
So the past two weekends I set about a little informal testing.
I started by gathering some sample test tones from online sources. For the purposes of confirming earpiece volume a little pink noise is all that’s required. There’s a nice source of such material at AudioCheck.Net
For this little experiment I used a pink noise wave file in combination with V-emotion and Eyebeam v1.5. This turned my laptop into a reliable signal source.
V-emotion hooks into the computers audio subsystem below the soft phone. It can inject an audio file into the call path causing it to be heard on the call. Since this is all file-based the audio level is absolutely consistent from one playback or call to the next.
With a reference grade signal source secured, I turned my attention to devising a way to record the output of each handset in a consistent fashion. From gear around the office I was able to gather the following:
- Behringer ECM8000 Reference Condenser Microphone
- Spirit by Soundcraft, Notepad Folio audio mixer
- Zoom H2 Recorder
These three items along with suitable cables would allow me to make reasonably accurate recordings at the earpiece of each of the various handsets.
The mixer is necessary because the microphone requires a source of 48vdc “phantom power.” The H2 has a line level input to take the mixer output, and its record level can be set and essentially locked down.
In recording the output of each handset I joined it to a test call, then placed the handset so that the earpiece was 3-4 mm distant from the microphone. The other end of the test call was my desktop Eyebeam/Vemotion suite with the pink noise source.
I would have preferred a Neuman K100 Dummy head microphone like the one pictured to the left. Such a microphone would simulate the way that the handset couples to the outer ear, so more accurately reflect the sound level delivered into the ear canal.
Alas, given my budget, such equipment was not possible. Luckily, it wasn’t truly required for my purposes.
I verbally slated each call so that I could tell the recordings apart. All calls were IP-to-IP and used G.722 to provide the best possible signal path under the circumstances.
Each handset was turned up to maximum volume, and the volume setting in the Gigaset DECT base was set for “High.”
|Source||Direction||Target||Ave RMS Power||Difference|
|-18.75 dB||-5.34 db|
|-21.45 dB||-8.04 dB|
|-26.83 dB||-13.42 dB|
|-29.3 dB||-15.89 dB|
|-30.82 dB||-17.41 dB|
Let’s cast this in a practical light. It’s generally accepted that 3 db the smallest change in level that most people can detect. We have used all of the Gigaset handsets around our house over the past two years. We’ve not have a problem with any of them producing sufficient volume at the earpiece. I’ve also consulted with various people I know who use the Gigasets, and none have complained of inadequate output volume.
However, there is very clearly some difference between the various models. If you are especially sensitive to the output level of the handset then you should select the C59H or S79H when choosing expansion handsets. As I’ve said before, I find that the S79H offers the best combination of features & performance in the entire line.