Let me introduce you to ...

Variable length opinions about arbitrary: things, situations and opinions (meta, right!).

Posts tagged plot

0 notes

Bowers & Wilkins P5 (frequency response)

I know, I have already written a post about the Bowers & Wilkins P5 headphones, they are good, no doubt, I totally recommend this to you for further details read the full blog. Now the only thing I have found (quiet often), are wannabe audiophiles who claim this headphones have great mid-ranges or that this headphones have a bad bass and what not. I don’t like to guide things by opinions but by facts, therefore I present you a small analysis of the frequency response of the Bowers & Wilkins P5.

First thing first, if you are not familiar with the terms and all the jib-jab, don’t confuses the frequency response with the frequency range or bandwidth. I have even seen quiet a few audio product related manuals confusing the frequency range with the frequency response. I would probably blame translators and not the companies themselves.

When it came to my mind that I wanted to analyze this headphones’ frequency response, the first place where it occurred to me to look for was at the B&W site and I could not find such thing there. Then I decided to contact their support team, after a day (that was fast), someone came back to me, pointing at this review by PC Magazine. 

In that review, they present a frequency response chart, comparing the B&W P5 with the Grado GS1000 (Fig. 1) and another frequency chart comparing the response of the two channels (Fig. 2).

Figure 1. Frequency response chart comparison between the Bowers and Wilkins P5 and the GRADO GS10000. In the x axis the frequency in Hertz and in the y axis the magnitude in decibels. 

Figure 1. Frequency response chart comparison of the two channels of the Bowers and Wilkins P5 and the GRADO GS1000. In the x axis the frequency in Hertz and in the y axis the magnitude in decibels. 

If you know something about signal processing you are probably wondering, where’s the test signal, what was the sampling frequency of that signal (probably 40K), what was the output impedance of the player used for this test, what’s the frequency response of the transducer used to perform this test, what was the window used to estimate the transform. Wonder no more because there is no record of that data and you know what, probably not even PC Mag will be able to answer you this questions, they probably don’t even know or care about those things. That’s why I got back to the B&W support team and they told me that they didn’t have the data, that I should request the data to PC Magazine.

Now, you would wonder, why would you care about all those things? well because if you don’t care about those things, you aren’t really doing a good frequency response analysis, you are just performing the Fourier transform to some data.

Anyway I kept looking for a more serious analysis and I found out that Inner Fidelity wrote a very complete blog reviewing the B&W P5. There your will find another frequency response chart (Figure 3). I have already requested the data of both of the sites, if any of them provide me with the appropriate files, I’m willing to do a much more exhaustive analysis of the frequency response, I will write the software that’s necessary for such purpose.

Closing statement

What do we learn about this? Do not trust a freaking audiophile telling you shit about headphones, trust the data and trust the controlled data only. In the end what’s more important is not the fancy charts, the real important thing is that you like what you buy.

BTW and if you didn’t already know, Fourier is awesomesauce! 

Filed under fourier frequency response frequency bode plot electrical engineering digital signal processing dsp b&w bowers and wilkins p5 headphones audiophile