Lrsky's comments are valid, but somewhat limited in scope. What he and may others run into is called dynamic compression. That is, the output did not match the input except for a limited spl range. This type of phenomena can take place with the entire speaker system or individual drivers at specific frequencies. On top of that, room acoustics also change as frequency and amplitude are varied, so you have to take that into account. Nearfield measurements of the speaker at various spl ranges will somewhat negate the room response aspects, so that way one can judge what part of the equation the room and / or the speakers / individual drivers are playing in the big picture. Sean
> |
A flat measured response is sometimes fools gold. The output of a loudspeaker in a static input of, say, 85db, or 90db, is one indicator. Yet, when the dynamic signal of real music is applied, there is no emperical proof that the 'dynamic' output will remain anywhere near that first measurement.
The only (fundamental) test is listening.
I have designed speakers which measure reasonably flat, insofar as the static input goes, with tragic results with dynamic input.
Now, you could argue that multiple input volumes starting at very soft, to very loud with sweeps; then overlay the results to compare. That could give some meaningful data, but its still not the 'real' test.
Speaker design is truly 'art and science' and transcends science, and goes into the artform arena. It takes a special person who can do both, then be objective with the results. That is why there are many loudspeakers that are 'technically' correct, yet leave us wanting musically.
So, "How important...."
Somewhat, however....
Best, |
Sean wrote:
As such, a neutral response assures us that our system is running like a computer i.e. "GIGO" ( Garbage In, Garbage Out ). If you have a good recording, you hear what is recorded. If you have a garbage recording, you hear what is recorded. Otherwise, you don't hear what is recorded, but what the system's colourations and distortions have added to the recording. On top of that, you'll hear this same colouration / distortion on every recording that you listen to. While you might like that specific sonic signature, i sincerely doubt that it would be "accurate" in terms of what one would hear at any given recording session or live performance. It might sound "pleasant" or "enjoyable", but accurate would be a long shot.
This is what I experienced using Paul Lam's LW-1 Passive Controller. It is the only way one can accurtely test equipment or to see what is happening when designing something.
My speakers are very good also. There sensitivity to what is happening up stream is very helpful when trying to design cables.
Just thinking! |
You can't recreate a wide variety of instruments with high levels of accuracy when a component ( speakers for example ) introduce a very specific tonal imbalance into the equation. Some instruments may sound relatively natural, but others, due to their tonal balance and harmonic structures possibly being altered, will never sound "right".
As such, a neutral response assures us that our system is running like a computer i.e. "GIGO" ( Garbage In, Garbage Out ). If you have a good recording, you hear what is recorded. If you have a garbage recording, you hear what is recorded. Otherwise, you don't hear what is recorded, but what the system's colourations and distortions have added to the recording. On top of that, you'll hear this same colouration / distortion on every recording that you listen to. While you might like that specific sonic signature, i sincerely doubt that it would be "accurate" in terms of what one would hear at any given recording session or live performance. It might sound "pleasant" or "enjoyable", but accurate would be a long shot.
Having said that, i don't think that the problem of flat frequency response is what you are concerned with Newbee. I think that the approach that your mentioning here is more of an attempt to "band aid" poor mic'ing and recording techniques as used by most engineers and limited bandwidth in the recording chain. This type of system is probably going to sound "warm and smooth", lacking leading edge information and "bite". If we were to view this on a scope, the front of a square wave would show limited rise time with rounded edges. Some would refer to this type of system presentation as "sugar coating", which helps "sweeten" bad recordings. While this can sound very "pleasant", it simply isn't "accurate" to what the music or recording "should" sound like. There's nothing wrong with this, but chances are, a system built like this will work better on specific genres of music and individual recordings than it does universally. If one doesn't listen to a wide variety of musical stylings and is happy with that type of limited presentation, there's nothing wrong with this. Having said that, this is where we get into the divergences between "music lover" and "audiophile". For lack of better terms "audiophiles" want to preserve the music as recorded and "music lovers" want every recording to be "pleasant". Different outlooks with different goals. That's why comparing notes on various components / systems can be so wide-ranging i.e. people are looking and listening from different perspectives.
As a side note, when i recorded "demo's" for a few local bands, i placed mic's nearfield to the instruments and / or amps and then two mics out away from the "stage". Those two "audience" mics were what the main mix consisted of and i could use the individual instrument mics to highlight certain instruments as needed. This gave an overall "natural" presentation as one would hear from being in the audience in the nearfield. At the same time, it allowed us the flexibility to alter / highlight individual parts of the presentation as we saw fit. I was able to retain the "musical presentation" of the group as a whole while still being able to perform technical changes without detracting / distracting from that presentation.
The "funny" thing about this is that after working with a few bands and making recordings like this, several of them came back to compliment me months / years later. After working in "Professional" studios, they were happier with the sonic results that we had achieved recording "on the fly" in the warehouse that i used. Yes, the "Pro Studio" was able to achieve a higher "gloss factor", but the music lacked the proper feel and presentation that the band was all about / wanted to deliver. Most of the differences probably had to do with heavy handed production techniques, phenomenal amounts of compression, poor microphone techniques and being in a non-friendly environment ( affecting the mind-set and performance of the performers ).
All of these were things that i tried to avoid as i knew them to be detrimental to what i and these specific bands were trying to achieve. That is, i wanted to capture the sound of the band, not turn them and the recording into something that they weren't. Using a system that isn't "flat" does much the same thing as what the heavy-handed recording engineers did i.e. added "distortions" to the music and presentation that wasn't really there to begin with. If one likes that given distortion characteristic(s), it is their system and recordings to do with as they please.
As mentioned above though, flat frequency response is only a small portion of what we hear. High levels of linearity in the frequency domain combined with high levels of linearity in the time domain ( transient response ) are what we should be striving for. At the same time, we should be pushing the recording industry to get their act together. While we can build our systems to achieve the first two thirds of the equation, i think that we will always be limited by the quality of the source recordings. That is, so long as "engineers" are doing the recording and "audiophiles / music lovers" have no input into the procedure. Sean
> |
Sean, If the reason we have stereo systems is to be able to listen to music and if we agree that we would like it to be as realistic as possible why would we want "flat frequence response" at the listening position? Most recordings are close multi miked - the end result of having FFR would be the equivilent of having your ear a few feet or in some case inches from the instrument(s) or voice, something that never happens live and sounds most unrealistic (have you ever heard a singers sibilance in an unmiked live performance?). Not at all realistic! The only exception would on those recording where the mikes are placed at an appropriate distance or very simple techniques are used. Any system set up will be static so some recording will sound more realistic on some systems and not on others. I would tune a system to sound as realistic as possible based primarily on the types of recordings I listen to most. FWIW while I accept your use of the terms "audiophile" and "music lover" I think it is perhaps more descriptive to say "audiophile" and "technophile". :-) |
Flat frequency response is only a portion of what we strive to achieve in a high linearity system. It is quite possible to achieve something that one finds to be "quite enjoyable" without being anywhere near "flat". This is where personal preference comes into play. Some people like a specific sound and tailor their system to achieve that whereas others want the system to try its' best at reproducing what is on the recording without introducing a lot of its' own colourations. This is the difference between a "music lover" and an "audiophile". Due to the amount of highly coloured speakers being marketed and devoured by consumers, i would say that most people, even many that consider themselves to be "audiophiles", fall more into the "music lover" camp. This is not meant to be a "put-down" to "music lovers", only to show that there is a difference. Sean
> |
I would like to hear Rives Audio's input on this one!
Rives?...
Also, I know a little about at least seeing pictures of what room modes look like, and I have a question. if you use an equalizer to eliminate a frequency mode, does that remove the dip at that frequency? If so can't you simply remove all the bass modes with an equalizer, dips, peaks and all? I'd like an asnwer to this. Or, can you simply eliminate the peaks in a room with an equalizer, like the Rives PARC. Maybe I should email Rives myself.
Rives?... |
A judiciously treated room will transform your system from a "bar sound" to that of a "concert hall". It will not fix the response of a flawed component. |
It depends how important it is to you. Achieving flat response could be part of the learning process,and it is doubtlessy important.However, I do not think I would recognize it. I think I assign it importance because it is held as one of the markers of an accurate system and a pathway to fidelity. I value that very much, even if only in an ideal.And I would be pleased to be able to descern the difference, which perhaps I will someday. But as it stands, my system colors recordings unbelievably and I go happily along.I imagine the degree of its value to your listening would arrive with your exerience of it, so that you had some way to compare. |
Of course equipment can influence response. Much of the peaking comes from the speakers, speaker-room interaction, and amp-speakers coupling (usually less so). You have good speakers.
Look into the room & speaker placement to reduce modes.
You want to trim the peaks. Leave the "valleys" alone for now:)
However, wait till you get your new amp, to see what's going on there. |
I am using a panasonic digital receiver and philips 963 dvd player with Piega p10 speakers using home depot cables. Could the equipt. have any effect on the response? I am getting a Rowland concerto in 2 weeks and better cables. I will test it out when i get them hooked up. Any body used the parc device? How dramatic was it? I thought i heard a review and it said it was effective but "subtle". |
I second what Ptmconsluting wrote. Ethan Winer is a noted acoustician and here's a link to a discussion forum he moderates. Notice the graph of the bass response in his personal room. It's definitely not flat. Room modes make it virtually impossible to attain flat response. |
What are we trying to do here?
We are trying to experience the spark in the composer's mind expressed by the music made by an orchestra (or Jimi's Strat). Having a flat system response tells us one thing; How well the recording engineer was able to put the performance onto recordable media.
This should be ALL a system is required to do (and look pretty too). |
I've had the pleasure to listen in a Rives designed dedicated room. Wow! I'm not quite sure how to explain the benefits beyond the pure, flat frequency response except, other things are allowed to happen that otherwise wouldn't. Things are....life sized. I hope that makes some sense. Headphones don't measure up to or equal the overall effect either, IMO. Those that say the room is the most important system component have a point that I can't argue with. After saying that however, I spent the last four hours in my non-professionally designed room and enjoyed it immensely. |
It is possible to love a system that is not musically accurate. Just because a person likes the sound of their system doesn't mean it's accurate, and just because a system is accurate doesn't mean you will like the sound of it!
Some people like exaggerated bass, or midrange; that's their personal preference. It all depends what a person is trying to accomplish when establishing their system. Since this is a subjective hobby, anything goes, but there are many to whom tonal accuracy is everything. To reproduce a tone accurately a system has to have an accurate frequency response.
Sdcampbell is correct with his assertion that the room will adversely (usually) affect the sound of a system. A good system in a compromised room will not be capable of good sound. Fortunately it costs less to treat a room than it does to buy new gear. You will not have to sink tons of money into fixing some of the anomolies you are experiencing. |
Most people won't have a clue how flat or at what frequencies they have peaks & nulls, even with some kind of oscilloscope or Rives-like device.
Just mode the mike up, down and sideways by 6 inches and you will see significant changes in the response. It's an inexact science at best.
But .. from what I understand from experts they advise leaving the nulls alone and simply try to trim the peaks. First off, it's far easier to do this. And secondly, something about artificially boosting nulls that gets real tricky and causes all sorts of other problems. This is probably why room treatments tend to be sound absorbing (in general terms, of course). The best way to fill nulls at your listening position might be to find out where they are strongest in the room and try some kind of deflection technique to spread them a bit more eavenly.
Enjoy the process - it's definitely daunting.
Bob |
While it is not necessary to have ruler-flat frequency response in a room to have enjoyable music reproduction, you DO want to avoid having pronounced peaks or dips in the response curve. Large peaks, particularly in the bass frequencies, can significantly color how the mid-range sounds. Likewise, significant valleys in the response curve makes the adjoining frequencies sound too pronounced.
As most long-time audiophiles know, the room is itself a "component" in the audio chain -- perhaps the most important. The Rives product is an intelligent, cost-effective way to address the problem of response peaks in the lower frequency range -- it does NOT, however, act like a graphic equalizer to address the problem of valleys, as its system is designed to reduce response peaks, not boost sagging frequencies.
To answer your question directly, you are NOT hearing your audio system at its best performance if your room has a lot of peaks. You may be very surprised how much better your audio system sounds if you can tame some of those room anomalies. |
