2. What Counts As Output?

The magic number for any equipment designed for listening to music, the one buyers really notice, is the number of watts. So it's not surprising that a vast amount of attention is paid to it and that the figure is often wildly exaggerated! What you should remember is that the real issue is RMS wattage. Crest, music and maximum values, and of course the famous PMPO values, are all hype. You can cheat a bit with RMS watts too by allowing quite a lot of distortion (say 10%) or by not measuring with all channels on. But that is no way so bad. You shouldn't set too much store by the exact power output value because what really matters in the end is the top sound level (in dB SPL) you can reach. This value depends on two basic factors: power of amplification and speaker performance. So if you have a speaker with an output of 87 dB per watt and another with 90 dB per watt, the 87 dB will need twice the amplification of the 90 dB to reach the same level of sound.

So amplification power should be considered as just another piece of information and not as a guarantee of high sound and good dynamics.

Now all that has been said, it is a good idea to see what output the power can provide. This is a very interesting way of looking at the output issue, given that the audio output is bound to be lower than the power output. For Creative this would be 67.5 W, which seems consistent with the power level they give as 48 W, for Logitech it would be 39.1 W for 40 W (an amplification performance of over 100%) and for Altec 34.5 W for 31 W.

Test Methods

Proper testing of sound systems requires both objective and subjective tests. For the objective tests we used a DAAS32 (Digital Audio Analysis System) with a dedicated acquisition card and Windows-based software. The test microphone was a 1/4" Neutrik 3382.

This system can be used for all sorts of ratings, but we stuck to frequency response using an overall two-in-one measurement system, meaning the usual one at a distance suitable for the satellites and another close-up one for the bass. The system software relates the two ratings once the operator has chosen a frequency to suit the test being run. The overall rating is combined with one for the subwoofer by itself (using the close-up method) to give a good view of what it can do and what the satellite crossover frequency is.

To plot the curve shown here, we used weighting and a third octave plot. This smoothes out frequency response accidents but makes things much clearer for non-experts. Our measurement system gives a very finely defined raw curve (1/24 octave), which requires knowledge of electro-acoustics and a certain amount of practice to read it with any degree of accuracy. We do of course use this for our analysis work.