House Curve Theory

Hipper

New Member
Jun 12, 2011
20
0
0
Hertfordshire, UK
#1
If I understand the House Curve thread correctly:

http://www.whatsbestforum.com/showt...Curve-What-it-is-why-you-need-it-how-to-do-it

for the relatively small rooms that we listen in, using a typical speaker will result in reduced low frequencies (LF) and increased high frequencies (HF), and therefore to counteract this a house curve sloping downwards from the bass region is suggested.

I'm not clear as to exactly why LF is reduced by the room and HF is increased. Furthermore this does not seem to be the case in my (limited) experience in my home circumstances. Perhaps these home circumstances are peculiar but I do prefer a flat response (with some increase in the 5-8kHz region to deal with age related hearing deficiencies).

Could the theory behind this be explained further please. Are we talking of a typical untreated room, a treated one, speakers with wide dispersion but with a measured flat response at the drivers?
 
Jan 29, 2014
983
0
0
Cape Town South Africa
#2
In a typical room , you have bass gain and hf treble droop
Correcting to FLAT sounds bad ... you need the correction emulating what a room would do .. hence you apply a curve to make it right.
 

Hipper

New Member
Jun 12, 2011
20
0
0
Hertfordshire, UK
#3
In a typical room, you have bass gain and hf treble droop.
You mean for the house curve I presume. Meaning, as the linked explanation said, the room before EQ must have low bass and high treble.

But why?

Let's make some assumptions. We have a full range (say down to mid 30s Hz) speaker which presumably sounds flat at the speaker. We have a typical room with no specialised acoustic treatment.

We know that if we hear only direct sound from the speaker it will be more or less flat, although the higher frequencies will be weakened relatively more by distance to the listener. However what we hear at the listening position is not only direct sound. We also hear the influence of the room.

From what I can gather, below about 200Hz (the Schroeder frequency, which is room dependent) a typical speaker will be omni-directional so we don’t really hear direct sound but the rooms reaction to the speakers’ output. This can both enhance or reduce those frequencies dependent, in our typical room, on size of room and location of speakers and ears.

Above about 200Hz we get direct sound plus reflections; off walls, ceilings, floors. This can enhance those frequencies (or of course damage the sound with echo).

Therefore, why, apparently in most people’s experience, when they listen to music in an unqualised room, are the lower frequencies weakened and the higher frequencies increased? What are the mechanisms that cause this?
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#4
You mean for the house curve I presume. Meaning, as the linked explanation said, the room before EQ must have low bass and high treble.

But why?

Let's make some assumptions. We have a full range (say down to mid 30s Hz) speaker which presumably sounds flat at the speaker. We have a typical room with no specialised acoustic treatment.

We know that if we hear only direct sound from the speaker it will be more or less flat, although the higher frequencies will be weakened relatively more by distance to the listener. However what we hear at the listening position is not only direct sound. We also hear the influence of the room.

From what I can gather, below about 200Hz (the Schroeder frequency, which is room dependent) a typical speaker will be omni-directional so we don’t really hear direct sound but the rooms reaction to the speakers’ output. This can both enhance or reduce those frequencies dependent, in our typical room, on size of room and location of speakers and ears.

Above about 200Hz we get direct sound plus reflections; off walls, ceilings, floors. This can enhance those frequencies (or of course damage the sound with echo).

Therefore, why, apparently in most people’s experience, when they listen to music in an unqualised room, are the lower frequencies weakened and the higher frequencies increased? What are the mechanisms that cause this?
You seem to know almost the entire answer but arrive at the wrong conclusion :).

The curve is a *forced* response by the DSP EQ. Put that aside for the moment and follow your own excellent explanation. Walk around your speaker and listen to amount of high frequency. It reduces relative to low frequencies as you change positions. In addition, just about every type of furnishing from carpets to your cloths and even body can be high frequency absorbers. The end result is that without DSP correction, what you hear is naturally rolled off in highs. The lows have not changed.

Key here is that similar thing likely happened in the speaker were music was produced and approved by the talent.

Now if you apply equalization, "it" (the microphone) would attempt to boost those reduced highs to be flat. Now you have boosted the highs relative to prior to EQ. Your ear immediately hears that change and considers that "bad" thing. Or it may at first like it but over time consider it too bright.

The target response then would have a treble tilt down.

For bass, it is more complicated. Normally your room has a lot of modes without EQ. With EQ, those peaks are pulled down but the results while more "correct," now seems anemic in bass. After all, there is less energy being pumped into the room. The solution is to boost these bass frequencies to compensate for again, what you thought was correct bass prior to correction.

Put the two together and you get a target curve that more or less tilts down from low frequencies to high.

Note that all of this is subjective. Who knows what your room and speakers do in indirect reflections relative to someone else. For that reason, target curves like this are a starting point. You must listen and adjust to taste.
 

Hipper

New Member
Jun 12, 2011
20
0
0
Hertfordshire, UK
#5
Walk around your speaker and listen to amount of high frequency. It reduces relative to low frequencies as you change positions. In addition, just about every type of furnishing from carpets to your cloths and even body can be high frequency absorbers. The end result is that without DSP correction, what you hear is naturally rolled off in highs. The lows have not changed.
I agree with this from my experience and from my simple explanation of the room and speaker interaction.

Key here is that similar thing likely happened in the speaker were music was produced and approved by the talent.

Now if you apply equalization, "it" (the microphone) would attempt to boost those reduced highs to be flat. Now you have boosted the highs relative to prior to EQ. Your ear immediately hears that change and considers that "bad" thing. Or it may at first like it but over time consider it too bright.

The target response then would have a treble tilt down.

For bass, it is more complicated. Normally your room has a lot of modes without EQ. With EQ, those peaks are pulled down but the results while more "correct," now seems anemic in bass. After all, there is less energy being pumped into the room. The solution is to boost these bass frequencies to compensate for again, what you thought was correct bass prior to correction.

Put the two together and you get a target curve that more or less tilts down from low frequencies to high.

Note that all of this is subjective. Who knows what your room and speakers do in indirect reflections relative to someone else. For that reason, target curves like this are a starting point. You must listen and adjust to taste.
From this I understand that a house curve is only used after DSP or room treatment. It's not really related to a typical room or speakers but only after some attempt to correct room problems - is that right?

Sorry then, that is my misunderstanding.

I'm not sure I agree about recordings. I listen mostly to pop music on Red Book CD and the recordings are highly variable. On some I get excellent and adequate bass. On a few they sound anaemic. Some have excellent percussion whilst others I know percussion is there (from putting my ear to the tweeter or listening on headphones) but it doesn't get to my ears.

I do agree about the subjective nature of this and clearly my experience is not how most find things. I have quite a lot of room treatment in a 14' x 13' x 8' room, have spent a lot of time trying to get the best position for ears and speaker, plus I use an equaliser to tidy a few things up. I've always aimed for a flat response and when I've tried say a 3dB or 6dB slope from the bass down I find the bass generally too strong. Perhaps that's my taste coming into it too!
 
Apr 3, 2010
16,022
0
0
Seattle, WA
#6
From this I understand that a house curve is only used after DSP or room treatment. It's not really related to a typical room or speakers but only after some attempt to correct room problems - is that right?
Only the former, i.e. using DSP to create the target response. It is impossible to do that with room treatment because they do what they do :).

In a way target curve is mandatory once you sign up for DSP room correction. You will be smoothing out the response but then you can for free also shape it. Many systems allow for multiple curves so you can have multiple settings to taste.
 

About us

  • Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing