Why do audiophiles care about flat measuring gear when most rooms are not flat?

[caesar]

Why is that? Is it irrational?

 

[thedudeabides]

I haven't checked manufacturer published frequency response on any of my gear for probably two decades or more.

Maybe with the vast majority of gear and, in the end, it doesn't really matter.

What matters is how it sounds to the end user in their room. Period.

 

[Bruce B]

Why is that? Is it irrational?

For me, I need a reference point.

 

[amirm]

Here is the thing about the room. It adds a constant distortion to what we are hearing. The cognitive part of the brain adapts and dials out fair bit of this. This is why speakers don't all sound the same in the same room. And how we forget after a bit that the room impacted the sound of the speaker when we first turn it on in a new room.

Where this doesn't happen is the low frequency response and that should be corrected. Having the equipment not be flat there just adds to the complexity of the problem.

 

[treitz3]

I hear ya' Bruce on the reference point but wouldn't that be more for the recording end of things and not the playback end that a normal (read not a recording engineer) consumer would use? I have heard a couple of almost completely flat playback systems and the end result IMO had much to be desired compared to other systems that measured worse but sounded oh, so much better.

Tom

 

[RBFC]

In scientific terms, the issue involves confounding variables. When more than one non-linearity is in play, it can become much more difficult to sort things out. If you generate a corrective filter, the equipment may distort the corrected signal and corrupt the intended result.

Lee

 

[microstrip]

Caesar.

Not all audiophiles care about flat measuring gear - anyone having tube amplifiers knows he his not getting flat responses.

And there are many definitions of flat room. Can we have yours?

 

[Bruce B]

I hear ya' Bruce on the reference point but wouldn't that be more for the recording end of things and not the playback end that a normal (read not a recording engineer) consumer would use?
Tom

I need it in the playback end, for mastering, at least for me, because I need to know I am making changes because the music needs it and not because my equipment/room is deficient.

 

[caesar]

Caesar.

Not all audiophiles care about flat measuring gear - anyone having tube amplifiers knows he his not getting flat responses.

And there are many definitions of flat room. Can we have yours?

Microstrip,

Your scientific knowledge of this hobby never ceases to amaze me. I was just plainly thinking of plotting the Hz along the abcissa and dB SPL on the ordinate. What other definitions are there?

 

[JackD201]

It's far easier to curve than to flatten. So give me flat to start with and let me manipulate things to where I like the results.

 

[caesar]

Here is the thing about the room. It adds a constant distortion to what we are hearing. The cognitive part of the brain adapts and dials out fair bit of this. This is why speakers don't all sound the same in the same room. And how we forget after a bit that the room impacted the sound of the speaker when we first turn it on in a new room.

Where this doesn't happen is the low frequency response and that should be corrected. Having the equipment not be flat there just adds to the complexity of the problem.

Interesting, Amir. The reality most audiophiles live with is that they don't buy into correction and have huge peaks and dips. (For the record, I don't mind correction in the lower register where we "feel the music", but I am leery of it in midbass and upper bass where we "hear the music".)

Do the best correction products today completely flatten the room response in the entire room or just a certain radius around the sweet spot?

 

[caesar]

Maybe I just don't get your point, but if your gear does not replicate the recording then its just another musical instrument that adds its tone to each and every recording blending them into something you want and not what the recording engineer thought you wanted...ahahahahah

My thinking is that, for many audiophiles, a deviation of +/-10 db in the frequency range, if not greater, is the norm for their rooms. Seems to me it's irrational for guys who are proud of their flat measuring speakers and electronics to be worrying about the gear measurements when their rooms are all over the place.

 

[amirm]

Interesting, Amir. The reality most audiophiles live with is that they don't buy into correction and have huge peaks and dips. (For the record, I don't mind correction in the lower register where we "feel the music", but I am leery of it in midbass and upper bass where we "hear the music".)

Do the best correction products today completely flatten the room response in the entire room or just a certain radius around the sweet spot?

Depends on the technology and how it is used. Some will use multiple mics or single mic and multiple measurements. In either case you can choose where you put the mic.

 

[Rodney Gold]

The flatness of , lets say , a speaker is the grounding of how it will sound..
Your room is a fixed entity and if you have a component or speaker that is not flat anechoically , you will hear the colouration.
Rooms mainly have a big effect on the low bass , and with DSP it's easily curable at listening position , mids and treble , even with some reflection will sound the same mainly , the room wont mangle it as badly as bass.
You cannot use DSP to correct full freq at listening position , for anything under 300 hz , DSP will do , above that you need physical treatment

 

[dallasjustice]

I've used full frequency DSP on and off for several years. I don't know anyone who uses a fully flat target curve. It doesn't sound right, no matter the recording. So flatness is not the objective of DSP. DSP is mostly useful at fixing SOME LF problems. It cannot fix ALL LF problems in a 2CH system. DSP can also work very well at HF, when it's done right.

The gear is the starting point. The more linear the starting point, the easier the rest will be when using DSP. Speakers can be made more linear with DSP on axis. I have very linear speakers and they still benefit from gentle >1khz correction. It relaxes the music and increases 3D clarity. A better matched R/L speaker response will produce a better stereo image.

I am never surprised to hear nice improvements in listening from nonlinear gear like tubes. IMO, many folks are simply using nonlinear gear to tailor their preferred target curve sans DSP. There's nothing wrong with that. However, there are better (and much cheaper) ways to do it.

 

[Orb]

Why is that? Is it irrational?

Same could be said about headphones, they (most) measure pretty appallingly over 20hz-15khz, and then there is the decision on what correction should be applied (not all apply the same).

Cheers
Orb

 

[Rodney Gold]

Target curves at listening position should never be flat , it should be a sloping curve with bass lift and treble droop..
In any room at normal listening distances , there is room gain in the bass and as distance increases , treble droop.
I too have used room correction for many years , sigtech , tact , krk ergo , lyngdorf room perfect , acourate , dirac etc...
Never liked to mess with mids and hf too much.. there are good reasons why higher mids and HF dont work well with RC.
If I am doing RC , I do it under 200hz and if I want to tweak the rest , I use a parametric eq to do so.
Essentially , having to define a target curve with full range correction , means it's a matter of taste , as there is no real "correct" solution to the target curve.

 

[Atmasphere]

Caesar.

Not all audiophiles care about flat measuring gear - anyone having tube amplifiers knows he his not getting flat responses.

And there are many definitions of flat room. Can we have yours?

I didn't know tube amps were not flat. Our amps are 1/2 db down from 2Hz-200KHz. But you can't please everyone...

Seriously though the issue has to do with how much feedback the amplifier has and if the speaker is designed for that (some are not, such as ESLs). But when I look at loudspeaker curves it does make me wonder why anyone with any kind of amplifier would think that they have flat frequency response. They don't.

I am never surprised to hear nice improvements in listening from nonlinear gear like tubes. IMO, many folks are simply using nonlinear gear to tailor their preferred target curve sans DSP. There's nothing wrong with that. However, there are better (and much cheaper) ways to do it.

Funny- tubes can be very linear, much more so than transistors. This has been well-known for decades. But tubes are always getting nicked for being 'nonlinear'. Its not always the tubes that are nonlinear, its often the designs. Tubes are so linear that you can build zero feedback amplifiers with them fairly easily. That's pretty hard with transistors. But interestingly, some of the better sounding if not best sounding transistor amplifiers are zero feedback (Nelson Pass First Watt for example). This has caused me to really wonder what is meant by 'linear' and 'nonlinear' when it comes to audio. Certainly not frequency response, and likely not simple distortion specs on paper either! Its obvious that our ears are not linear (logarithmic seems closer to the mark)- they seem to not care about certain kinds of distortions, and then care a lot about others (higher ordered harmonics are a good example of the latter).

This is what I mean about what is really linear: if our ears care more about trace amounts of a certain type of distortion, shouldn't we weight that form of distortion to a greater degree? If we did (and since our ears really are the only thing we can use to listen to anything), the landscape of what is linear and what is not would look a lot different!

 

[KeithR]

funny, was reading a Jeff Fritz review and was puzzled by the flat frequency response in his vault. are his measurements taken from the speaker or the listening position? i can't imagine having flat sound as treble would drive you out of the room. see here for an example:

http://www.ultraaudio.com/index.php...d=158:magico-vs-the-world-the-q3-loudspeakers

Keith

 

[LL21]

I didn't know tube amps were not flat. Our amps are 1/2 db down from 2Hz-200KHz. But you can't please everyone...

Funny- tubes can be very linear, much more so than transistors. This has been well-known for decades. But tubes are always getting nicked for being 'nonlinear'. Its not always the tubes that are nonlinear, its often the designs. Tubes are so linear that you can build zero feedback amplifiers with them fairly easily. That's pretty hard with transistors. But interestingly, some of the better sounding if not best sounding transistor amplifiers are zero feedback (Nelson Pass First Watt for example).

...Its obvious that our ears are not linear (logarithmic seems closer to the mark)- they seem to not care about certain kinds of distortions, and then care a lot about others (higher ordered harmonics are a good example of the latter)...

Very interesting post...I have read (I think from Mike Lavigne) that certain SET amps can be extremely linear within that first watt...one piece of the puzzle that forms to make SETs magical for many people. I also have read LOADS of tube are not linear.

Could you actually explain a bit more about where/how/under what circumstances tubes can actually be equally linear or perhaps more so than their SS counterparts (transistors, I suppose)? In 'dumbed down' terms for us non-techies? Very, very interesting.

Thanks for any 'idiot's guide' guidance here. This could be very educational for me (and hopefully for others).