Welcome to the Piano World Piano Forums Over 2.9 million posts about pianos, digital pianos, and all types of keyboard instruments
Join the World's Largest Community of Piano Lovers
It's Fun to Play the Piano ... Please Pass It On!
Interesting vid for those considering new monitors for you keyboards. From a producer.... "Every speaker [monitor] from every manufacture sounds different..."
Interesting commentary given that virtually every response curve I looked at for powered monitors is ruler flat.
I find it interesting how studio producers/engineers and audiophiles speak exactly the same, but choose very different equipment. The words and desires are almost the exact same... but the choices are vastly different. Interesting...
BTW, I was in both Sam Ash and Guitar Center in Manhattan last weekend and listened for... quite a long time... to various products.... Adam, Eve, Focal, Yamaha... et al.... Finally I had a sales rep who was not a child... a 56 year old producer and he played uncompressed/lossless files for me. Really fabulous. Finally!!!
What did I learn about their powered monitor selection? My beloved brand ... Focal (I own a very expensive pair as my main home audio system) did not sound all that great. In fact, I was very disappointed in their sound. Yamaha sounded very very good... worth considering for your keyboards. The Adams were also very good at first but something about the upper midrange was not correct... (I was surprised to hear the engineer in the vid above make a similar comment... I heard the speakers before I saw this vid). The best I heard that day? Eve. About $2K for a pair. Eve is a newer company started by the engineer who developed Adam speakers.... the difference is he added digital sound processing.
Digital sound processing in a powered monitor shocked me.... first, because it is anathema to accuracy. You are changing the analog signal to digital, dicking with it, then converting it back. It can't sound good... but they did. I actually heard depth and clear separation. I still am a bit upset by this as audiophiles and engineers want accuracy and do no harm... but they sounded so darn good.
Anywho.... lesson to be learned: Go hear them before you buy. They all sound and present sound very very differently. Opinions on these boards vary from entertaining to informative but you must make your own decision. Go listen and be educated. Also, frequency response curves are irrelevant... they all are fabulous... and useless.
As a corollary to the above, I would add that good speakers are the ones that sound good to me. They might sound s**t to someone else, but I don't care. That someone else can buy some other speakers. I buy what I like.
Are my speakers flat? I don't know. And I don't care. They sound really good.
"I don't know. And I don't care." Wow, them's audiophile words.
Oh, just throwing this out there. Given what I heard and on a budget, I would seriously consider the Yamaha brand for my keyboard.
As some of you may know, I had real trouble buying monitors for my RD 2000. I could not get a good demo. Everywhere had some dopey teenager with ultra-compressed MP3 sources material of music that immediately induced my gag reflex. So, I leaned back on what I knew and purchased a used pair of B&W home audio speakers and powered them with a quality Adcom amp I modified with an improved internal power supply. I knew what the B&Ws sound like... all their speakers regardless of price have a particular "voice" that I know (I am not a big fan of this voice but I know the speakers will sound very very good nonetheless). Doing it again, I would give Yamahas and Eves another listen.
Interesting commentary given that virtually every response curve I looked at for powered monitors is ruler flat.
I'm not aware of any powered monitors that are ruler flat, can you give some examples? Regardless, there are other variables, like dispersion characteristics (i.e. the difference between how they sound head-on vs. off-axis), how clean the sound remains as you push them harder (along with possible changes in the frequency response as you push them), and the headroom available for clean dynamic peaks.
Regarding on off axis: these are "monitors"... they are meant to be listened to on-axis from a control seat or the seat at your piano. Off axis is irrelevant for the most part... however they do fill a room with sound and therefore your acoustics are a factor in perception (and here is where dispersion is a factor)... but they are designed to be listened to on-axis and in the near field. When you audition them, I recommend listening to them as you would be using them.... looking very closely to see they are positioned perfectly symmetrical to each other on the shelf. Then place your ears at exactly tweeter level. Then move left-right, from back. They will "snap in" and when they do... it is really cool. You should "see into" the sound field. Even the cheaper ones will do this as is the function of the monitor. However with just a bit of listening, you will hear their flaws and differences as you compare.
Regarding how hard you "push them"... they like your piano monitors, are expected to play at a common volume level. You control the volume with your playing. An acoustic piano has no volume control. Note in the video where the engineer notes engineers have a reference volume to do their work. With few exceptions, I play my RD-2000 at a single volume setting.
Monitors are not designed to be used like home audio speakers.
I was just puttering around the Adam website where I got that frequency response link in my prior post and found another really cool analysis. This shows dispersion. Note, the speaker (Monitor) is highly directional... remember, you are supposed to be sitting in one spot (on axis). This is one aspect where monitors are designed with very different design objectives from home audio speakers. Home audio speakers are typically designed to be highly dispersive and not directional.
Here is a link to a random Adam monitor.... this is typical and incredibly flat (in my casual reviewing experience)
An excellent monitor... but seriously, saying "virtually every response curve I looked at for powered monitors is ruler flat" and then picking a $3,500-per-speaker monitor as your example indicates that maybe you're looking at monitors that are beyond what most of us are probably looking at.
Though even then, as great as that curve looks, it will have some sonic character influenced by the fact that the peak at 1500 Hz and the valley at 2500 Hz indicates a 3 dB variation over a small range in a very audible area... and I would expect it to sound different from a just-as-flat speaker whose variations were elsewhere. A good enough monitor for anything, yeah, I'd say so. But not perfect, and someone might prefer something else whose flatness is about equal overall but whose specific variations from perfect flatness differ.
As for pushing them, I have no idea if these (or any other particular monitor) cleanly play as loud as a 9' Steinway D (or how many people care whether or not it does), but regardless, monitors are used for more than piano, so I'm not sure that how a monitor's max output compares to that of any particular acoustic piano is necessarily always representative of the max you may want out of your monitors.
What makes you think this is not an accurate representation? And if they wanted to be dishonest and fake it, why even show the 3 dB difference between 1500 Hz and 2500 Hz? And what makes you think anyone is "smoothing" these things?
From what I've read smoothing is commonplace. And by "smoothing" I do not mean "flattening". They just average the measurement from successive intervals. If so, then that 3 dB difference, if unsmoothed, could be a big roller coaster.
Anyway ... in that link there's **nothing** said about the measurement technique other than "measured at 4 meters".
I cannot **know** that smoothing has been done ... but the nearly smooth-as-glass appearance of that chart should raise eyebrows.
Here's the aforementioned Adam S3H, along with the Adam T7V. Has the S3H curve been more smoothed than the less expensive T7V? Or have they been measured under different test conditions? Or is the S3H really flatter? Can we ever really know?
An ideal speaker would measure ruler flat on axis... ...BUT.... ...any speaker, even a cheap piece of rubbish, can be artificially EQ'd until it's amplitude vs freq graph looks "flat", and that won't magically make it a better quality speaker - you'll just a have all your distortions, resonances and noises coming out at matched volume.
The Adam graphs above do not show phase response vs freq graph. (Not to mention needing polar graphs for dispersion, THD graphs, waterfall graphs, etc. all would give a much better picture.)
They are not "flat" anyway, not by a long shot. +/-3dB is pretty poor and AUDIBLE colouration. If <3dB error was inaudible, people wouldn't ever bother making EQ adjustments smaller than 3dB steps! In fact, people mostly EQ most things in steps of +/-0.1dB or at least +/-0.5dB. (Some fussy engineers may even EQ in steps of +/-0.01dB, but nobody will go as far as +/-0.001dB because that's obviously inaudible.) But +/-3dB errors will definitely be heard.
The Adam's amplitude response drops like a stone below 40Hz on that graph. What about the bottom octave????! It's hardly flat response if there's a whole octave missing. It's gross non-flat behaviour, clearly audible! An Earthworks M50 measuring mic for example extends its linear response (+/-1dB) from 3Hz - 50kHz which is beyond audible ranges, so that it can give you enough quality accurate performance to study the audible 20Hz - 20kHz area of interest without colouration. There's no excuse not to be able to measure a speaker down to 20Hz at least and clearly see what's going on.
If you have two dis-similar sounding speakers, and you use digital FIR correction to bring their amplitude and phase responses into line, they will start to sound more similar to each other. The differences will be between power handling / SPL before distortion, and different dispersion characteristics due to physical designs like 2-way vs 3-way vs 4-way, coaxial vs separate drivers, separate subwoofers, size and shape of cabinet and edges, vents, size and shape and reflectivity of surrounding room environment, etc.
If you compare speakers which all use FIR correction but are also of very similar size / woofer size / power / cabinet shape / dispersion pattern, they should sound fairly similar due to the FIR correction bringing them towards a common trend. If you compare "audiophile" speakers with a stupid minimalist philosophy where they do NOT believe in using any DSP for FIR correction, then they'll be at the mercy of their different driver's erratic nonlinear behaviours and despite being similar cabinet size / design styles will likely sound very different from each other.
To summ up my whole point of all of this.... you need to go hear a pair of monitors before you buy. You can not tell if a monitor sounds good or not (to you) by these graphs. All monitors nowadays look good on paper.
As the engineer in the video said, all speakers sound different from one another. If frequency response graphs were so important, then all speakers should be trending to one perfect sound indistinguishable from each other... this could not be further from the truth. My opinion? Stop reading and go listen for yourself. Then post your experiences here.
Regarding bumps being audible.... So here is where I, as many audiophiles, will contradict myself. First, yes, tiny differences mean a ton to those of us who care about sound but they may not matter all that much.
I have never been able to correlate what I hear (sound quality) to frequency graphs... except for low frequencies being obviously chopped off or extended. After we work hard, spend money and time finding just the right amp or speaker or whatever, I then toss the minute differences out. What happens... and this is the contradiction... is your ear/brain system adapts. It "smoothes" out these aberrations much of the times (if small). So these small aberations are just not all that important for quality. There are a ton of other design/engineering aspects to speaker performance than just response curves. Coherance... or the ability of different frequencies to arrive at your ear at the same time they entered the microphone... now that is a really big one and (just one ingredient for imaging) is not represented at all on these frequency graphs (this one is affected big time by electronics too thus muddying up the signal). Damping factor of the internal amp is incredibly important to controlling the woofer cone.... "Q" the affect of the cone suspension system.... so many more....
Anyway, engineering of monitors start with paper, experience, then prototypes using tons of measurements, then finally, tuning by ear. It is just the way it is done. You must purchase using your ears, not your eyes.
There is so much unsubstantiated (un-verified) opinions in this thread that I'm going to present some different perspectives.
1. Bruce, you can train yourself to identify different frequency bands: http://harmanhowtolisten.blogspot.com/ How cool would it be for you to be able to say that "I think this speaker emphasizes 1khz-3khz band"?
2. Musicians don't sing or play the violin (say) differently based on the room. And people enjoy live performances regardless if they are inside an echo-y hall or outdoors. Humans are perfectly capable of hearing sound "through" the room.
3. There is decades of scientific research on audio reproduction. You don't have to agree with this research or the methodology or whatever, but if you are going to discuss audio, at least be familiar with this research: