Demystifying the "flat curve"?

[aphex944]

Quote:

Originally Posted by lindrone I've always had the believe that curves don't say anything about the way any particular equipment really sounds like. In the end it comes down to sound signature that can't be respresented by some frequency measurements alone.

I totally agree. I think it's interesting to see the particular frequency response of a piece of equipment, but in the end the only thing that matters is how it sounds. Simply because something specs out well doesn't mean it's going sound good. I get more satisfaction out of picking equipment for a specific sound instead of always shooting for a flat, completely neutral sound.

 

[WmAx]

Quote:

Originally Posted by lindrone However, if there are so many different situations that one has to account for, flatness in the final playback equipment doesn't seem to matter all that much.

YEs, it does. The reason is consistant behaviour. The job of the source and amplifier is to remain linear. If a modification of the signal frequency balance is desired, use an equalizer. Superb value E.Q.s are on the market such as the Behringer DEQ2496 that give you virtually any tone curve you could desire. Don't have a mindset that we should regress our standards and start hardwiring EQ in the source or amps.

Quote:

I've always had the believe that curves don't say anything about the way any particular equipment really sounds like.

False. Do you think you can take two amplifiers, one with a flat response(within 0.1dB) and one that has a 2dB rise at 10kHz, and have them sound the same?

Quote:

In the end it comes down to sound signature that can't be respresented by some frequency measurements alone.

WHat is their besides frequency response, amplitude, distortion at any given amplitude/frequency and noisefloor? Any competant modern design amplifier or cd player will have an insignificant noisefloor and inaudible distortion.

-Chris

 

[lindrone]

WmAx, I think you misunderstood my post.. when I say "playback equipment" I was referring to only the very last step, the speaker/headphone step... the amp & source step have a definite requirement for flat curve, and that I understand quite well.

Well, some amps perform their own equalization as well... but source should definitely be flat.

 

[WmAx]

Quote:

Originally Posted by aphex944 Simply because something specs out well doesn't mean it's going sound good. I get more satisfaction out of picking equipment for a specific sound instead of always shooting for a flat, completely neutral sound.

True. You might prefer a certain frequency band emphasis or demphasis. However, it is inefficient to do so nor reliable to do so by picking an amplifier that may have a poor frequency response. That's why E.Q.s exist. However, if the effect you are achieving is in part psychological(which probably most situations are) then even two amplifiers with identical internal parts but a different shell will not sound the same to you in sighted listening.

-Chris

 

[WmAx]

Quote:

Originally Posted by lindrone WmAx, I think you misunderstood my post.. when I say "playback equipment" I was referring to only the very last step, the speaker/headphone step... the amp & source step have a definite requirement for flat curve, and that I understand quite well. Well, some amps perform their own equalization as well... but source should definitely be flat.

Ah, thank you for the clarification. Disregard my last post to you.

-Chris

 

[aphex944]

Quote:

Originally Posted by WmAx WHat is their besides frequency response, amplitude, distortion at any given amplitude/frequency and noisefloor? Any competant modern design amplifier or cd player will have an insignificant noisefloor and inaudible distortion.

If these measurements told the story of audio, we'd have a perfect amplifier, source, headphones, speakers, etc.

However, they don't. For example, lets take the Creative Labs Audigy, it's a computer sound card not highly regarded around here. Here are some measurements from it: http://audio.rightmark.org/test/crea...digy-1648.html

On paper, it looks great! Almost completely flat frequency response, extremely low noise and distortion. According to this, this HAS to sound great if these measurements tell exactly what's going on.

 

[WmAx]

Quote:

Originally Posted by aphex944 If these measurements told the story of audio, we'd have a perfect amplifier, source, headphones, speakers, etc. However, they don't. For example, lets take the Creative Labs Audigy, it's a computer sound card not highly regarded around here.

First of all, I'm not aware of a single controlled listening test determining this card is 'not good', if that's your implication. What is 'high regarded' here is irrelevant to the subject of audibility if you you want to progess your claim beyond testimonials and speculations.

Here are some measurements from it: http://audio.rightmark.org/test/crea...digy-1648.html

First, that is a quite worthless distortion measurement. Distortion is that percent at what amplitude? Into what input impedance/load? Wht does the distortion harmonics of one fundamental frequency tell me?

Quote:

According to this, this HAS to sound great if these measurements tell exactly what's going on

I don't know anything form this particular measurment about the distortion harmonics except at 1khz at some unspecified amplitude and load.

This was a big problem in the 60's and 70's when audio opamps came out such as the 709 and it was used in applications such as the Crown IC-150 preamp, btw. People would measure full band at a smaller then realistic amplitude or measure at single frequency such as 1khz and go on to quote low distortion or what not when in reality the high frequency linearity under some loads and at specific amplitudes would create loads of harmonics.

Here is a fun fact: no one has demonstrated an unmeasurable quality to exist that is audible and can be picked out in a double blind test with proper methodology/controls. Actually, if you can, you can 1 million dollars. Refer to http://www.randi.org/ look under the paranormal JREF prize. They recently expanded the test to include unexplained things in hi-fi audio.

-Chris

 

[rabbitsfoot]

Microphones are to the recording engineer what paint is to the artist. As in speakers, there is no microphone that has a perfectly flat frequency response. The wide variety of microphone types: ribbon, condenser, , dynamic along with widely varying frequency responses and pickup patterns gives the sound sculpter brushes to start with. Even if the on axis frequency response is relatively flat, the off axis response is usually varied, and this effects the sound, such as how much ambience/room sound or other instruments playing in the same area is allowed to be included in the pickup/recording . Microphones are picked for various instruments based on these criteria. Some models have a revered status for their success based on an ability to capture that perfect female vocal or perfect drum overhead or tom sound. . When digital recording first came out, the flat frequency response of the medium caused sound engineers to rethink and redo old methods that didn't work with a medium capable of flat frequency response. The old system of magnetic tape required more highs to make it to the end product of the LP, whereas in digital-what goes in-comes out. Because speakers and microphones are imperfect in frequency response and other, possibly more meaningfull measures, I'm sure we're going to have many different opinions on the sound coming from these devices. What still comes through though is the passion of the artist. It IS moving.

 

[Swampwalker]

This is a how many angels arguement in the ultimate analysis, but KurtW and rabbitsfoot have said it all, from a practical standpoint. Ya gots ta listen. We may not even know what to measure or how to measure it. There are lots of parameters we know of that affect sound quality besides frequency response, and one can assume that there are plenty of others we don't know. Even assumming it is important, truly flat response as measured at the transducer (can or speaker) is irrelevant unless you have a truly flat room or pinnae and have suffered no differential hearing loss due to aging or illness, or too much R&R. The real question is, does the music that you hear move you? Its art, not specs!!!

 

[WmAx]

Quote:

Originally Posted by rabbitsfoot As in speakers, there is no microphone that has a perfectly flat frequencyresponse.

As in speakers, their ARE microphones that have a nearly flat response within a dB throughout the main audible passband.

Quote:

Even if the on axis frequency response is relatively flat, the off axis response is usually varied, and this effects the sound, such as how much ambience/room sound or other instruments playing in the same area is allowed to be included in the pickup/recording .

Yet, the most precision mics, such as those used for measurement, are omnipolar with nearly the same off axis as on axis response.

Quote:

Microphones are picked for various instruments based on these criteria. Some models have a revered status for their success based on an ability to capture that perfect female vocal or perfect drum overhead or tom sound. .

I won't debate differnt pickup patters; thier is legitimate use for this in various situations. However, as for non linear amplitude response. Their just is no good reason for this today. Why distort the captured signal purposely from the start? Equalization can be applied to achieve these transfer functions later in production, and with more flexibility/control/accuracy then using a microphone with a poor frequency response. But, using mics as 'EQ' is an old practice, sort of a tradition, perhaps this will die out eventually.


Quote:

Because speakers and microphones are imperfect in frequency response and other, possibly more meaningfull measures, I'm sure we're going to have many different opinions on the sound coming from these devices. What still comes through though is the passion of the artist. It IS moving.

This last statement is interesting. In my opinion, the artists' 'passion' comes through less than ever on most major recordings, due to the standard practice of excessive dynamic compression contests that the lables seem to have with each other. THis, ironically, is a move backwards IMO.

-Chris

 

[WmAx]

Quote:

Originally Posted by Swampwalker There are lots of parameters we know of that affect sound quality besides frequency response, and one can assume that there are plenty of others we don't know.

One can assume that little invisible elves with bad breath live under one's bed. However, it's not rational, or substantiatable. WHat one 'can ' assume and what one 'should' assume are two different things.

-Chris

 

[rsaavedra]

Quote:

Originally Posted by breez Music performed -> flat response mic -> flat response reproduction -> ears with their non-flat response Music performed -> ears with their non-flat response Obviously flat measured response from sound reproduction is closer to actual music performance than sound reproduction with non-flat response.

Ditto, that's the reason why flat freq. response is so important in the whole chain between performance and transducers, at least for speakers. So that our ears will perceive the same soundfield that existed around the microphone at recording time, e.g. as if our ears were where the microphone was at recording time.

 

[rsaavedra]

Quote:

Originally Posted by lindrone I've always had the believe that curves don't say anything about the way any particular equipment really sounds like.

I disagree with that though. I've you've played with any equalizer you would be aware of how minor deviations in the response produce very noticeable effects in the sound.

 

[lindrone]

Well, I'm not saying that sound signature aren't effected by equalizers or from the source & amp. I'm just saying that you can't really tell what anything sounds like just by reading the curve alone... neither do I think that reading by simply reading the curve on the headphone/speakers would we be able to tell the performance of a product without having auditioned it. It just shouldn't be the only criterium such as companies like Ultimate Ears made them to be, that's all.

Sorry if I was unclear, as I noticed I wasn't thinking much or writing clearly for several posts in this thread

 

[crazyfrenchman27]

Call me a minimalist and an assassin of mumbo-jumbo jargon, but what sounds good is good.

And yes, terrifying though it might be, the Sony V6s might sound "good" to some people *gasp*!

I'm a heretical post-modern, relativistic solipsist, though, so don't mind my own thoughts.

There is no objective reality! There is no truth! There is no reality that exists external to my perception or imagination!