Flat Frequency Response

Oleg34

Limp Gawd
Joined
Jul 13, 2014
Messages
484
In our line of work, we get asked about ‘flat’ audio responses all the time: What is flat audio response? Is flat response good? Why is flat good? We also run into: “When I hear a system that has a supposedly flat response, I think it sounds stale,” or the more general comment, “I don’t like the way flat sounds!” We get these questions and comments all the time. http://www.centerpointaudio.com/HowToUnderstandFlatFrequencyResponseGraph.aspx

I have heard $50 speakers sound flat.
 
It's an interesting topic. Harman did some research a few years back, particularly with an interest in headphones. From it, they developed the Harman headphone response curve, and interestingly enough, it's not flat.

http://www.innerfidelity.com/content/headphone-target-response-curve-research-update

The first part of the paper details the calibration to absolutely flat of a reference speaker system in a listening room, and a headphone (Sennheiser HD 800) who's response was carefully compensated in order to match the flat response of that listening room. Then two tests were designed that allowed the listener subjects to adjust carefully designed bass and treble controls for listening preference. In one test, bass was set at one of two different levels and listeners were asked to adjust the treble for best listening pleasure. Then the treble was set to one of two different levels and the bass adjusted. In the second test, subjects were allowed to adjust both bass and treble for maximum listening pleasure. These tests were done with both headphones and speakers in a room. Eight trained listeners and three untrained listeners were used in the tests.
 
Flat frequency response is absolutely dead neutral i.e 20/hz/kz. Nothing out there measures this in raw state. Room treatment is a bigger headfuck then it is. Not everybody likes an absolute flat FR.
 
Don't confuse anechoic chamber measurements to room measurements. An absolutely flat response in the chamber is never even remotely flat in the room.

The room also introduces phase etc. problems that affect the overall sound. Speakers/amps also have differences in dynamic response etc. that make huge differences (power compression in small speakers is a very big problem for example).

A speaker with a flat frequency response does _not_ sound flat. It sounds awesome, lifelike and vibrant. If a speaker 'sounds flat' it probably has problems in both higher and lower frequencies.
 
The guy who does the huge Multi-IEM review on head-fi wrote up a 2015 in-ear earphone buyer's guide recently, and divided up people's preferences in four categories: basshead, warm and smooth, v-shaped and balanced.

He has interesting things to say about each category; in the comment about v-shaped, he notes: "Due to the way the human loudness contour works, at lower volumes a mild v-shape can actually present a fairly balanced experience."

I apparently prefer balanced; I really like the sound of Astrotec AM-90 IEMs and I'm on my third set now. I tried the Philips SHE3590 because they were only $8 on amazon and I wanted to see what v-shaped IEMs sounded like but they had way too much bass for me.
 
If I wanted bass I'd eq a bit of low-end for a balanced sounding headphone. This retains clarity, detail and the bonus of some lower spectrum omph. I'd go as far as U-shaped headphones for bass, v-shaped sounds like turd to my ears. Way too much bass.

I have over 25 headphones so it works out for me either way.
 
Don't confuse anechoic chamber measurements to room measurements. An absolutely flat response in the chamber is never even remotely flat in the room.

The room also introduces phase etc. problems that affect the overall sound. Speakers/amps also have differences in dynamic response etc. that make huge differences (power compression in small speakers is a very big problem for example).

A speaker with a flat frequency response does _not_ sound flat. It sounds awesome, lifelike and vibrant. If a speaker 'sounds flat' it probably has problems in both higher and lower frequencies.

^ All correct. You aren't going to get a near-flat response in any room without investing a lot of time and potentially money into sound deadening, bass traps, diffusers, etc. And even then you'll likely never get it perfect. This is why many of us recommend a decent receiver with Audyssey MultEQ XT.

Also different humans have different hearing (look up "equal loudness curve") and flat speakers mean little. Personal preference is more important. If you have perfectly flat response, you have the benefit of hearing what the mixing engineer wanted you to hear, but ultimately different people prefer different sounds. Many of us, for example, like to crank up the bass way beyond what any mixing engineer gives us.
 
^ All correct. You aren't going to get a near-flat response in any room without investing a lot of time and potentially money into sound deadening, bass traps, diffusers, etc. And even then you'll likely never get it perfect. This is why many of us recommend a decent receiver with Audyssey MultEQ XT.

Also different humans have different hearing (look up "equal loudness curve") and flat speakers mean little. Personal preference is more important. If you have perfectly flat response, you have the benefit of hearing what the mixing engineer wanted you to hear, but ultimately different people prefer different sounds. Many of us, for example, like to crank up the bass way beyond what any mixing engineer gives us.

Yes but tuning the sound introduces the inevitable problem that depending on mixing, some records are going to sound much worse than they would with a flat setup.

If you boost bass, recordings that already have plenty of it will sound boomy and heavy. Vice versa with boosting highs - that is of course unless your high hearing has already been gone. Then it might please you and sound horrible to your child :D
 
Anything Creative built will never reach anything resembling a flat response in a room. But generally speaking the sound setup that sounds uneventful is usually the best in the long run.

If it amazes you in some aspect, chances are that same feature is going to start to annoy you in the long run (especially when speaking of cheap desktop stuff). Of course a B&W 801 will also amaze you but it won't start to annoy you lol.
 
Yea. To me hearing flat response does annoys me, but if anything is how it could be than it's OK with me. Just have to accept the fact. Also i have asked my friend "what speakers sound better in your opinion ?." The answer he gave me, was. Logitech speakers.
 
A speaker with a flat frequency response does _not_ sound flat. It sounds awesome, lifelike and vibrant. If a speaker 'sounds flat' it probably has problems in both higher and lower frequencies.

This gets me thinking. I start to cringe every time someone describes studio monitors as flat, harsh and not made for the enjoyment of music. That makes absolutely no sense. Who wants to mix/master for hours on terrible sounding equipment that causes ear fatigue?

I'm not sure where this assumption originated, or why it continues to persist.
 
Because most studio monitors measure as neutral as possible and intend to reproduce what the source sounds like without a drop or bump to other frequencies that sounds "musical" to our ears.
 
Because most studio monitors measure as neutral as possible and intend to reproduce what the source sounds like without a drop or bump to other frequencies that sounds "musical" to our ears.

That's another thing, whether a monitor behaves in a neutral fashion depends on a various factors, environment being one. Neutral isn't the same as a flat frequency response, since I've read accounts of seemingly flat monitors being harsh and fatiguing. The speaker just has to be tonally acceptable. A number of "hi-fi"speakers could be used as monitors. Then there are some monitors that could be argued to intentionally color the sound.

I just find it amusing that some folks seem to have this perception of all studio monitors being boring tools that are terrible for listening on. That sounds like a bad speaker, or a badly placed speaker.
 
I just find it amusing that some folks seem to have this perception of all studio monitors being boring tools that are terrible for listening on. That sounds like a bad speaker, or a badly placed speaker.

I think of this in the exact opposite way. It's funny to me how many people on this forum think that monitors are for quality sound and home/hifi stuff is some sort of low-quality high-distortion crap. Most people here have no real reason to care for monitors but do because it's some sort of fad these days. (NOT saying monitors are all crap - like hifi gear, some is and some isn't.)

MANY monitors aren't nearfield and plenty of hifi gear IS suitable for nearfield, so while I expect someone to give that reason, it really isn't a valid one.
 
I think of this in the exact opposite way. It's funny to me how many people on this forum think that monitors are for quality sound and home/hifi stuff is some sort of low-quality high-distortion crap

Really? I haven't noticed it so much myself. I have seen arguments presented about the superiority of having built in amps over passive/exterior amp combos. This depends.

Most people here have no real reason to care for monitors but do because it's some sort of fad these days. (NOT saying monitors are all crap - like hifi gear, some is and some isn't.)

I disagree, since a number of powered monitors are priced the same as PC speakers, whilst offering superior sound quality.

I don't see it as a fad, as opposed to be people seeking out better/convenient sources for sound. People also seem to prefer them because they are active.

MANY monitors aren't nearfield and plenty of hifi gear IS suitable for nearfield, so while I expect someone to give that reason, it really isn't a valid one.

Yes, this is true.
 
More like ATC and bust :p. I'd love to have their speakers but no way I can afford $7500 for a pair of monitors. I can dream though :).
 
Have heard better then some of ATC's totl big ass studio monitors. Some of the older JBL studio monitors (go for thousands when they do show up) are more realistic and revealing sounding.

The old Yamaha NS1000M's are also worthy counterparts. For 40 year old speakers.
 
I think of this in the exact opposite way. It's funny to me how many people on this forum think that monitors are for quality sound and home/hifi stuff is some sort of low-quality high-distortion crap. Most people here have no real reason to care for monitors but do because it's some sort of fad these days. (NOT saying monitors are all crap - like hifi gear, some is and some isn't.)

MANY monitors aren't nearfield and plenty of hifi gear IS suitable for nearfield, so while I expect someone to give that reason, it really isn't a valid one.

Most monitors are intended to be used in near field. They're designed to give an accurate sound in their intended use. A few catch-22:s exist:

1) Not all monitors are created equal. A 200 dollar Behringer and 3200 dollar Genelec have a huge difference

2) A cheap monitor may sound ok in near field (which is 30-40 cm listening distance in most cases). If their radiating and power response design is not optimized for also regular speaker use, they can sound really overbright, hard and unbalanced due to early reflections messing up the sound. In near field it works, step a few steps back and boom your sound is crap.

If your power response looks like a mountain the sound will be real bad in far field despite your response being flat like a pencil near field.

Studio monitors (albeit high quality ones) are what mixers and master workers use with the performing artists when they tune how the record will sound like. So if someone doesn't how the record sounds near field with a proper studio monitor, they're not looking for high fidelity but some truth of their own.

I know many people who buy eq:s and put a smiley on them. Whatever if it makes them happy :D
 
Just had a bizzare/idiotic/laughable discussion with some saps on "youtube". Apparently, they're of the opinion that speakers/amps were better in the 70' and reached their pinnacle around that time. Anything that came afterwards is "crap."

Naturally, when I requested thorough data ( measurements, blind evaluations, etc) to corroborate such (IMHO) absurd claims, they resulted to insulting me (shocking, I know).

I then mentioned the advent of powered speakers as a sign of progression, and one commentator, who claimed to be 53 with a "Masters in Audio Engineering" said this about some speakers I mentioned :

"How does little tiny 28 pound speakers produce 290 watts on its own? It can't"

This is the speaker in question.

http://www.sweetwater.com/store/detail/KH310L

According to him, the measurements are a "fairy-tale" and Neumann are liars.

I makes me wonder how I keep getting caught up in these ridiculous discussions.
 
Last edited:
Just had a bizzare/idiotic/laughable discussion with some saps on "youtube". Apparently, they're of the opinion that speakers/amps were better in the 70' and reached their pinnacle around that time. Anything that came afterwards is "crap."

Naturally, when I requested thorough data ( measurements, blind evaluations, etc) to corroborate such (IMHO) absurd claims, they resulted to insulting me (shocking, I know).

I then mentioned the advent of powered speakers as a sign of progression, and one commentator, who claimed to be 53 with a "Masters in Audio Engineering" said this about some speakers I mentioned :

"How does little tiny 28 pound speakers produce 290 watts on its own? It can't"

This is the speaker in question.

http://www.sweetwater.com/store/detail/KH310L

According to him, the measurements are a "fairy-tale" and Neumann are liars.

I makes me wonder how I keep getting caught up in these ridiculous discussions.

Heh, if you really want to twist his brain, tell that this is a 3-way active speaker with 360 watts / channel and weighs only 42 pounds :D

Genelec_8351AP_coaxial_monitor_loudspeaker.jpg
 
Just had a bizzare/idiotic/laughable discussion with some saps on "youtube". Apparently, they're of the opinion that speakers/amps were better in the 70' and reached their pinnacle around that time. Anything that came afterwards is "crap."

Along the same lines, many on AVSForum consider weight of a receiver/amp to be one of - if not the #1 - factor in power output / overall quality... even though the heavy gear they like is just heavy because it's inefficient. A lot of people distrust stuff with switchmode power supplies at all even though they can be done very well and contribute no more distortion than a much-less-efficient linear supply.

I do like some of the 70s receivers for what they are, and those that still have good capacitors are still fine to use if you don't need surround or digital inputs, but thinking they were better-sounding is absurd. In some cases the build quality can be better as they are metal and all the switches have great feel, etc., but that stuff doesn't affect sound quality.

And when it comes to vintage gear, with few exceptions, what I don't tend to like is speakers. One of the exceptions, however, is that I would like (and will probably buy) a nice set of one of Apogee's larger ribbon speakers. Mmm.
 
Heh, if you really want to twist his brain, tell that this is a 3-way active speaker with 360 watts / channel and weighs only 42 pounds :D

Genelec_8351AP_coaxial_monitor_loudspeaker.jpg

He might have a heart-attack, but not before accusing Genelec of witchcraft.

Along the same lines, many on AVSForum consider weight of a receiver/amp to be one of - if not the #1 - factor in power output / overall quality... even though the heavy gear they like is just heavy because it's inefficient. A lot of people distrust stuff with switchmode power supplies at all even though they can be done very well and contribute no more distortion than a much-less-efficient linear supply.

Ironically, that was one of the arguments being made; particularly when I mentioned the wattage of the light-weight, tri-amplified Neumann speaker.

It seemed unfathomable.

I get the impression that some guys (particularly older ones) can't comprehend, or aren't even aware that powerful powered speakers exist.

My dad was shocked by the performance of the M-Audio Bx5' I go him, and turned them into hi-fi speakers. LOL.

I do like some of the 70s receivers for what they are, and those that still have good capacitors are still fine to use if you don't need surround or digital inputs, but thinking they were better-sounding is absurd. In some cases the build quality can be better as they are metal and all the switches have great feel, etc., but that stuff doesn't affect sound quality.

LOL. The AMP discussed was a vintage McIntosh 2105. I told him a cheap Onkyo TX-NR636 would probably be better suited for modern day purposes. He started hinting at the"musical amp" phenomena, ranted that the amp was made of plastic (incorrect), didn't have a transformer (doubly incorrect) and was cheap chinese junk.

And when it comes to vintage gear, with few exceptions, what I don't tend to like is speakers.

One of the speakers he brought up was the JBL 100. The response for the speaker was all over the place, and most had been modified. I didn't say anything though, since how they sound matters most.
 
Studio monitors (albeit high quality ones) are what mixers and master workers use with the performing artists when they tune how the record will sound like. So if someone doesn't how the record sounds near field with a proper studio monitor, they're not looking for high fidelity but some truth of their own.

Well put.



I know many people who buy eq:s and put a smiley on them. Whatever if it makes them happy :D

I have a dual 31-band eq. I set it up with pink noise, sine sweeps and white noise through a spectrum analyzer. LOL, now I need a smiley for it... :D
Never knew how good Huey's Thinline Tele sounded on GG&HT until I added some 8200hz to flatten the highs, or how much tighter Ducky's bass sounds in I'll Play The Blues For You when I took that 63hz wall rumble out. Now I know that Albert's humbuckers are supposed to be so sharp they cut flesh, and that Kurt is "essing" on purpose on the Unplugged take of The Man Who Stole The World, which adds to my enjoyment of the music.

Science :cool:
 
LOL.

Another guy just commented, asking me if I've ever seen an "acoustic suspension" speaker? I assume he was going to insinuate the fact that they are superior and no longer produced......

Apparently, he has never heard of "sealed" speakers. Easily found and purchasable. For shame.
 
LOL.

Another guy just commented, asking me if I've ever seen an "acoustic suspension" speaker? I assume he was going to insinuate the fact that they are superior and no longer produced......

Apparently, he has never heard of "sealed" speakers. Easily found and purchasable. For shame.

I take it that guy was bragging about his old KLH speakers, then? (Though, to be fair, they actually were pretty good decades ago.)
 
I think some people also do not know, or fail to take in to account, the improvements that have become possible in speaker and amp design with the use of computers. It is amazing how much designs can be improved and optimized with good computer simulation and measurement. These days designers will use advanced shit like laser interferometry to measure the breakup patterns of drivers and so on. It's lead to some pretty amazing speakers that just weren't possible a few decades ago.
 
I take it that guy was bragging about his old KLH speakers, then? (Though, to be fair, they actually were pretty good decades ago.)

I wish. He owns a Logitech Z906 and claims that if a speaker costs over a $1000, it had better be able to produce "thunderous" bass.

Apparently, he didn't realize the KH310 I posted about in the discussion was sealed, or he wouldn't have posed a redundant question.

I think some people also do not know, or fail to take in to account, the improvements that have become possible in speaker and amp design with the use of computers. It is amazing how much designs can be improved and optimized with good computer simulation and measurement. These days designers will use advanced shit like laser interferometry to measure the breakup patterns of drivers and so on. It's lead to some pretty amazing speakers that just weren't possible a few decades ago.

LOL.

I actually explained this, and even used computers as an analogy/example ofincreasingly efficient tech. This went completely over the "Audio Engineer's" head, and he started ranting about musical amps, deceptive advertising; and the impossibility of self-powered speakers putting out 290 w RMS. He also went into the atypical "Made in America" territory as well.
 
Last edited:
I have a dual 31-band eq. I set it up with pink noise, sine sweeps and white noise through a spectrum analyzer. LOL, now I need a smiley for it... :D
Never knew how good Huey's Thinline Tele sounded on GG&HT until I added some 8200hz to flatten the highs, or how much tighter Ducky's bass sounds in I'll Play The Blues For You when I took that 63hz wall rumble out. Now I know that Albert's humbuckers are supposed to be so sharp they cut flesh, and that Kurt is "essing" on purpose on the Unplugged take of The Man Who Stole The World, which adds to my enjoyment of the music.

Science :cool:

That's the old way of EQing. The new way is a measurement mic, REW (Room EQ Wizard), and a PEQ. Also EQing should be used to subtract, never boost. Dips are phase and room problems, boosting makes them worse. After that there's programs like rePhase to generate convolution filters to correct phase. There's also programs like Acourate, AudioLense, and Dirac Live that take it all to the next level. Worth noting: some people don't like EQing above the Schroeder frequency, personally for nearfield and line array style systems I don't think it's as big as a problem with systems where the room influences the response messing with the phase.

An example of REW auto EQ'ing one of my JBL LSR305's with my house curve (green line before EQ, light green line predicted response):
eq.png
 
That's the old way of EQing. The new way is a measurement mic, REW (Room EQ Wizard), and a PEQ. Also EQing should be used to subtract, never boost. Dips are phase and room problems, boosting makes them worse. After that there's programs like rePhase to generate convolution filters to correct phase. There's also programs like Acourate, AudioLense, and Dirac Live that take it all to the next level. Worth noting: some people don't like EQing above the Schroeder frequency, personally for nearfield and line array style systems I don't think it's as big as a problem with systems where the room influences the response messing with the phase.

An example of REW auto EQ'ing one of my JBL LSR305's with my house curve (green line before EQ, light green line predicted response):
eq.png


Thanks for the good info and informative links.


REW seems almost worth downloading JAVA to have a place to drop the .jar...But all I have for mics at the moment is a "singer/songwriter kit" of a condenser and a cartoid pencil through a tube pre>mixer>usb to pc. I'd be afraid that particular chain would introduce enough artifacts to have me chasing my tail in short order. I don't think I'd be out of line supposing that the quality of sound measurements are quite dependent upon the quality of the microphone, would I? :p
 
That's the old way of EQing. The new way is a measurement mic, REW (Room EQ Wizard), and a PEQ. Also EQing should be used to subtract, never boost. Dips are phase and room problems, boosting makes them worse. After that there's programs like rePhase to generate convolution filters to correct phase. There's also programs like Acourate, AudioLense, and Dirac Live that take it all to the next level.

How good are Yamaha's YPAO and similar systems?
 
Thanks for the good info and informative links.


REW seems almost worth downloading JAVA to have a place to drop the .jar...But all I have for mics at the moment is a "singer/songwriter kit" of a condenser and a cartoid pencil through a tube pre>mixer>usb to pc. I'd be afraid that particular chain would introduce enough artifacts to have me chasing my tail in short order. I don't think I'd be out of line supposing that the quality of sound measurements are quite dependent upon the quality of the microphone, would I? :p

Yeah that mic wouldn't be good for that. The recommended ones are a Dayton UMM-6 or miniDSP UMIK-1 (neither is better than the other, the main difference is the UMIK-1 is a requirement for miniDSP's Dirac Live products). Regular ones with factory calibration files are about $75. Ones that have been measured against a reference mic and individually calibrated (instead of batch calibrations) are about $100 (Cross Spectrum Labs).

How good are Yamaha's YPAO and similar systems?

Never used YPAO but looked at Yamaha receivers a few years ago. It appears to use PEQ's and I believe if you have REW and a measurement mic you can manually enter them but I think it's limited on the bass frequencies unless you have on of the higher end models. Bass is where most of the improvement is with PEQs. From what I gather from reading AVSforum and HomeTheaterShack over the years is Acourate, Audiolense, and Dirac Live are the cutting edge of room correction and EQ. Audyssey is the mainstream so to speak. Yamaha YPAO and Pioneer MCACC are more the entry level room correction, though I've heard most people prefer YPAO over MCACC.
 
Last edited:
I recently read a review of the Denon AVR-X7200W that uses Audussey correction and it was really impressive. It flattened basically all frequencies nearly totally in the room.
 
How good are Yamaha's YPAO and similar systems?

I haven't been that happy with YPAO. I have a Yamaha receiver in my living room and I'd say YPAO is overall not an improvement. I leave it off, just use it to auto set speaker levels.

Audyessy I love, I have a receiver with Audyssey MultEQ XT in my computer room and that sounds great. It can't make a bad system and room good, but it can improve what you have. Take a lot of time to do all the measurements, but if done right I think it makes a nice improvement.

Dirac Live is amazing. Haven't heard it on my stuff, but I've heard it on another system and it is probably the best room correction out there.

I haven't heard Pioneer's MCACC.
 
I haven't been that happy with YPAO. I have a Yamaha receiver in my living room and I'd say YPAO is overall not an improvement. I leave it off, just use it to auto set speaker levels.

Audyessy I love, I have a receiver with Audyssey MultEQ XT in my computer room and that sounds great. It can't make a bad system and room good, but it can improve what you have. Take a lot of time to do all the measurements, but if done right I think it makes a nice improvement.

Dirac Live is amazing. Haven't heard it on my stuff, but I've heard it on another system and it is probably the best room correction out there.

I haven't heard Pioneer's MCACC.

The Dirac looks interesting. The price is a bit stiff for a software based solution though.
 
Never used YPAO but looked at Yamaha receivers a few years ago. It appears to use PEQ's and I believe if you have REW and a measurement mic you can manually enter them but I think it's limited on the bass frequencies unless you have on of the higher end models. Bass is where most of the improvement is with PEQs. From what I gather from reading AVSforum and HomeTheaterShack over the years is Acourate, Audiolense, and Dirac Live are the cutting edge of room correction and EQ. Audyssey is the mainstream so to speak. Yamaha YPAO and Pioneer MCACC are more the entry level room correction, though I've heard most people prefer YPAO over MCACC.

I think Trinnov is still tops but they need it to come out on better receivers than the Sherwood Newcastle R-972... But it takes so much DSP processing power. Most receivers come with old SHARC DSPs that can't handle any of these algorithms very well. Even Audyssey MultEQ XT which has been in receivers for years is a really cut-down version because of the crappy DSPs almost all receivers use.
 
Last edited:
The Dirac looks interesting. The price is a bit stiff for a software based solution though.

You can get it in a few pieces of hardware too if you like. Emotiva uses it in their new XMC-1 processor. The MiniDSP guys that sell little general purpose DSP units have ones with Dirac Live on them too.

I'd love to see a pro sound card with room correction, but I don't think such a thing is likely to happen. Too many pros don't like it and you wouldn't want to waste the money on another processor in your unit for a feature many would hate on.
 
Back
Top