How the power of the PC has changed speaker sound

When I was first exposed to speaker engineering tools around the mid 80’s they involved things like the B&K strip recorders. These would synchronize an audio signal with a pen that would travel the length of the strip, moving up and down as the signal level changed. Another big innovation was the TEF which just became widely known:

http://www.mcsquared.com/tef.htm

and was having a big effect on acoustic spaces, including music halls and theaters. All the measurement tools were priced near that of a new car and unless you were an audio manufacturing professional or associated with a graduate program you would probably never see them in person.

This limitation had a profound effect on individual audio designers and how they made speakers and chose parts. The measurement and testing capability of a JBL or Harman far surpassed that of any single person trying to start a new brand in their garage. Getting to any desired outcome if you have the measurement tools is far easier and faster than if you don’t. Period. So, to clarify, even if you have the knowledge, without the tools, audio engineering work was very difficult.

Today there are a number of very affordable measurement tools and microphones which let the average individual achieve professional results, assuming they know what they are doing. Very good impedance, Thiele/Small parameters and near/far/quasi anechoic measurements can be had together for under a grand that will run on the same PC most of us have. Far less cash than your average table saw, another common prerequisite.

As an example of a fundamental change in speaker design today is the frequency and impedance optimizations. It is super hard to do both of these with say a calculator and pen/paper as was often done before, then a spreadsheet, and finally with simulation tools. Some problems I see in old speakers is the failure to simultaneously optimize impedance and frequency. Some old Infinity speakers are good examples of this. They end up with a frequency response that is very good by modern standards, but with impedance curves that makes them subjugate of all but the heaviest amplifiers. At some point this became a selling point for speakers and amplifiers. Around the time of the 1 Ohm Apogee Scintilla, demanding speakers and obsequious amplifiers became desirable due in no small part to bragging rights. Speakers with terrible impedance curves were thought to be more transparent because they could finally give the giant Krell amps an obvious advantage. Just remember people, violins have no impedance curve.

In other old speakers we see the designer not taking into account a woofer’s impedance curve when designing a low pass filter, leading to interesting response shapes. We cannot today say what was and was not intended in every individual speaker design of the past, that is for sure, nor can we claim one set of outputs is what you, the individual buyer, should prefer. What we can say is that for the most part as the measurement and simulation tools have improved speaker designs have become more normal in terms of smooth frequency and controlled impedance curves. Of course, there are a number of exceptions, I’m not talking about your favorite crazy speaker designer. I’m saying that this shift is palpable. I’m saying that the marketplace from which you are offered a choice in speakers has shifted and that you are the subject of these changes which are driven in large part by affordable PC compute power and measurement tools democratizing who can be a speaker manufacturer and bringing the ability to rapidly iterate through more designs while optimizing in more dimensions than we did in the 1980s.

The point of this is not "you should make speakers this 1 way" but rather that these tools have themselves changed the marketplace. Hope you find this perspective illuminating.