I’ve found over the years that house curves are something that are almost as universally misinterpreted as the Fletcher-Munson curves. The biggest problem is that things like this tend to get oversimplified and then someone latches on to one aspect without fully understanding it. And as we all know, on the internet, whoever has the loudest voice usually wins.
Let’s start with the basic idea of a house curve. In the most general sense, a house curve is an EQ setting to make your speakers sound better in your room. If you do a quick Google search on the internet, you will see most people tend to come to a general consensus that a proper house curve equalizes your speakers to measure with a downward sloping response, which is then claimed to be the most pleasing to the ear.
This is really the point that becomes misunderstood and misused by many around the internet. Many people take this to mean that your speakers should measure with a downward sloping response on axis but that is not the case. The general concept of the downward sloping response came from research conducted by Olive and Toole. However, they didn’t claim that your speakers should have a downward sloping response. They claimed that a smooth and downward sloping power response was the most pleasing to listeners.
Power response is technically something that we can’t directly measure but approximate by taking spatially averaged responses from a speaker over several points on and off-axis and relates closely to polar response. If you have a speaker that has a polar response that is even and smooth, you should also have a fairly smooth power response. So what this means is that a well-designed speaker which has a smooth polar response, like our Criton 1TD or Model P215, will have a smooth and downward sloping power response. This means that as you move off-axis, you will see a downward sloping frequency response.
Now a well designed speaker, when measured around the listening position and averaged (unless it’s directly on the tweeter axis), will already have a downward sloping frequency response. This is what you are really looking for in a house curve. Coupled with some typical room gain, you'll get a slight boost to the lower frequencies. So in essence, a well-designed speaker shouldn’t need a house curve. A well designed speaker in a normal, slightly toed in position and measured around your listening position, should result in a response that looks like a house curve.
So what happens if you design a speaker to measure with the downward slope on-axis to incorporate the house curve? Now at your listening position, you end up with a top end that is even more rolled off. Given some of the house curves I’ve seen on the internet, you could be looking at over a 20-30 dB difference between the low end and top end response.
On the flip side, this is why you wouldn't want your room correction to adjust to flat at the listening position. If you were to EQ to flat, you would now have an upward sloping response on axis, which would make your speakers sound bright and thin. But this goes back to understanding the difference between on and off-axis measurements and what they represent. If all of your speakers were pointed directly on-axis to your ears, flat might be OK. But in most households, this won't be the case, so your response should end up downward sloping at the listening position.
Some argue that the movie industry has a house curve (called the X-Curve) that they set as standard and we need to at least match that or it will sound bad. Unfortunately this is still misguided. The X-Curve is for soundstage dubbing and theater playback. Home movies are not mixed the same. A home movie mixed for Blu-Ray is completely different from a theater mix, often times not even being done by the same person.
My personal “house curve” has zero adjustments above 100 Hz on my main speakers. I do like a little extra emphasis on the very low end for movies so I have my sub set about 3-5 dB higher than my mains and crossed at 60 Hz. Is this the flattest? No. But I also don’t like this setting for music, where I feel the sub stands out too much. On music I usually just run my mains full range without the sub. Overall, when I measure at the listening position, I get what looks like a typical house curve.
As with all things audio, there are always compromises in everything we do and you should always do what makes you happy, but you should also strive to understand why you like what you like or the real facts behind something before jumping on an audio bandwagon. Hopefully this article gives you a bit more information to make an informed decision on whether or not you need or want a house curve.