Mini Comparison - Vibe (1st gen), C700, PK2, RE0, NE-7M, PFE, ER4S, OK1, TF10, UM3X, SE530, IE8 - Page 6

oarnura, I'm not reinventing the wheel here. Nothing I say should be awe inspiring. Sound is waves, just like waves in the ocean.

There is such a thing as too quick. This has to do with the fact that a wave is not an abrupt wall. When the diaphragm moves back and forth, it is supposed to mimic the wave shape. If the diaphragm moved instantly, it would take a long wave and turn it into a very narrow wave. You'd end up with the same amplitude but a shorter duration. You get high impact but you lose the area under the curve. The area under the curve of the wave shape is the body. The variation in shape of this wave is the texture, the subtleties in the music. You compress the duration way down by building a very fast driver, you get a light, crisp sound, but you lose the subtleties. It's like taking an ocean wave, maintain the height, but cut the thickness way down to say, a couple inches instead of a couple feet. What do you feel? You get this smack of initial contact, but there is no mass behind it. You also lose the shape of the wave. Did it have a thick leading edge, was symetric, or maybe it was a double wave inside the large wave. Information gets lost or at least compressed time wise enough that we lose some of the perception. In audio, we can't simply get rid of the body of the note. The complex shape (not a perfect sine) gives us all the small details of perception. We don't want to lose those.

On the other side, you can have too thick and everything just blends together. You can lose the independence of the wave patterns and just end up with one, congealed mess. There is a balance to this, and the balance is to best match the original source.

The driver knows nothing. It is simply there to recreate the electrical signal in the form of air pressure. It's simply a converter. Some converters just more accurately mimic the source data.

No, I have no techinical papers explaining this. I'll leave it up to you to make up your own mind and find your own data. My point is simply that a wave has a time domain and shape to it. It's not an on off switch. You need to recreate the shape in air. Attack and decay time should match the rise and fall time of the wave patterns. If it doesn't, it's messing up the original source.

The goal of a speaker designer is to build the most accurate converter possible.

Yes, a driver should be fast to react, transient, not sluggish, etc..., but it should not move faster then its reproductive source. Yes, physics can make the driver move inappropriately relative to the input signal. It is reactive, i.e. after the input, but it can still move quicker or slower then it should. For example, what happens when you overdampen a spring/damper/mass setup? If you provide a force input, the system will transmit that to the mass. However, the overly high dampening will make the mass stop sooner then what was input initially. If the goal is to exactly reproduce the input force, it stopped too quick.

Body isn't abstract because it can be seen. You can look at a wave form and see the body (area under the curve). If you were able to measure air pressure waves and map out intensity versus time, you should end up with the same curve shape and amount of body as the original electrical wave form.

Servos are seldom used in subwoofers. It has been used only on a couple of occasions. I don't know the science well enough to understand why they are not used more. The concept is neat, but I have a feeling there are underlying limitations of sorts that prevent more companies from building any.

A flat response isn't mic measured flat. Our ears have a different perception of flat. Some headphones like the ER4S and PFE(gray filter) are geared for ear flat or what one would perceive as flat. In my testing I run a pink noise track to test frequency response. Pink noise has the same intensity over the entire frequency spectrum. I run through my EQ and adjust to what I hear as flat. Some headphones need very little EQing, some a lot. In my reviews, I do note some frequency response issues with the various headphones I tested. If there's a 3dB dip at 1kHz or a 6dB spike at 10kHz, I find this out via pink noise and adjusting EQ. This is what I have to fix to end up with something that is reasonably ear flat. I don't go into great detail, but I do make small notes as I deem needed.

I'm knowledgeable enough about the industry and science of audio. I don't claim to know everything, but I know a good bit. Head-fi is new to me but not audio itself.

This is hard to explain but easy to show in pictures, lol. I just don't have a readily easy mean to draw examples. Basically, the speaker is a reactive system. It always operates after the source input. However, it doesn't have to move at the same rate as the source input. For example, a 50Hz sine wave takes 5 milliseconds to go from 0 volts to say a peak amplitude of 5 volts. We feed this 50Hz sine wave through a driver that has a strong motor relative to its dampening. The motor is strong and chucks the cone outward. The dampening is relatively low so it doesn't provide a ton of resistance to the motion. In 3.5 milliseconds it reaches the equivalent output peak that 5 volts translates to. The slope of rise is sharper then the slope of rise of the original source wave. The diaphragm moves too fast.

No, the diaphragm can move more or less depending on the spring/damper setup. For example a woofer with a high loss surround will actually limit cone travel on higher excursion. The side result is an overly tight bass response. It sounds controlled but is constrained also. The driver does not have to match the source wave shape. It's good if it does, but it doesn't have to and won't depending upon design.

It can become compacted because the rise and fall is artificially steep. Amplitude could be high or low depending on design. In most cases, the diaphragm motion will not be an exact copy of the shape of the source wave pattern. It would be nice if it was, but pretty much always it's not. Good hardware follows it closely. Poor hardware barely follows it at all and almost completely doesn't resemble the source wave at all.

I make up what I say, but I base it off what I learn. I am here to be helpful, to teach, and to learn. I am a fan of self-exploration, a hater of spoon feeding, and I really don't have all the time in the world to look up every little example and instance of everything I've ever learned to show others examples or specifically what I mean. This is a hobby. If one has interest, they seek to further that interest. All I can do with my sliver of time interacting with you is to convey my own findings, perceptions, and understandings. Right or wrong or in the gray, that is sometimes debatable.

It's really just acceleration and time relative between the source wave shape and the diaphragm acceleration/deceleration rates. It is as simple as that.

When? I was responding to you describing it like an on/off switch stating it is not but rather a rate of change over a period of time. This:

It sort of half is. The motor pushes both ways, yes, but the mass/spring/damper system is a free moving system that is based on inertia. So yeah, you do sort of push the cone out with the motor. The problem is the motor can actually push too hard and move the cone out faster then it should to match the input source signal shape.

Well, yeah. You can be over or under dampened, and this will affect relative motion. It can mess with acceleration/deceleration rates and overall amplitudes. You can make the cone accelerate or decelerate too fast. By doing so, you make for quicker attack and decay. You get a shorter lighter and crisper sounding note. It sounds clean. However, if done too much, you may less accurately reproduce the smaller variations in the source wave shape.

There is very much a thing as over dampening. It's like running a car with really high dampening. How well does it ride? Dampening is relative to the spring rate and moving mass. There is one ideal for proper movement. You can be too high or too low from this ideal and compromise the reactive accuracy of movement of the diaphragm.

Two separate things. Yes, the diaphragm will move in a reactive manner to the source wave. This means that the source could be at 0 volts and the diaphragm is still moving forward, backward, or even oscillating around center.

Body is a separate thing. Body is the fatness of the wave shape. The source wave has a certain fatness to the wave. When the driver reacts to the input signal and moves the diaphragm, it should create the same shape and same fatness. Depending on setup, it's possible to not do so and make a thinner or thicker wave pattern, all depending on the relative acceleration/deceleration rates in relation to the source wave.

There is, as was said above already.

Budget is one concern. I assume there are other limitations. For example, why only subwoofers? Can a servo setup even support say 1kHz rates, 10kHz rates? It relies on an accelerometer and feedback circuitry, and they need to operate fast enough to control diaphragm movement. Does it work? Sure. Is it really needed when considering effectiveness to a higher quality design? Maybe not. If you look at a standard driver that have flat BL, CMS, and Le curves, has a broad and flat frequency response, low distortion over the intended bandwidth, and lacks mechanical noise and any resonance issues, how bad can the classical way be?

Me too. I feel like I need to write a book, in great detail, for you to be content with my words. I assume nothing about your intelligence. I simply express in "jargon and verbosity" as I would normally to describe my ideas to most people. I do it for generalized purposes. For most folks, the gist is enough. You want more, and I'm not sure why. I'm not willing to spend the time and go into great detail making graphs, exact examples with numbers and specific data. I could develop that, but it's more time and effort then I'd want to invest in such a manner. I really don't have the interest in this to put forth that much effort. This is not my job, and I am not building drivers. I've been a hobbyist in audio for 15 years, more heavily in the last 5. I know physics and am in engineering. I have no focus on audio, but I am familiar with the mechanics behind it from an electrical standpoint, driver design, and sound waves. I don't just make up random stuff to act cool or sound cool. I am a fellow hobbyist and helper in the community. This is where I come from.

I try to explain the same things in a variety of generic, descriptive ways, i.e. why I repeat myself. If that comes across as me just babbling crap. Sorry. Without going into actual detail and full out examples, that's all it can be. Either the descriptions are understood or they are not. If they are not, I fail to convey my ideas in a context that matches your understanding. There's not much I can do about it within the level of my willingness on this subject.

Just hitting on end points:
I invent nothing, just repeat what is already known.

I simply mention one peak value, and you automatically assume I'm talking binary. Waves are shapes. Voltage varies relative to the time domain. The slope, rate of change, of the wave is the "acceleration." The diaphragm should move at the same acceleration rate, but it is able to move faster or slower because the driver is not linear. This means the equivalent shape of the pressure wave emitted from the diaphragm can be shaped differently then the source, wider, narrower, with a higher peak amplidute, or shorter peak amplitude. If it's narrower (not saying frequency changes), then acceleration is higher, higher slope, higher acceleration. You can have a wave pattern that repeats every 1/100 of a second, and the wave shape can be narrow and pointy or fat and rounded over. This is attack and decay, the shape of the curve. The shape of the curve doesn't change the period of the oscillation of the wave pattern, but it does change the meaning of the wave pattern and in audio, the end sound.

The shape of the wave pattern is not a smooth sine wave. It's a complex jumble of peaks and valleys of various frequencies and amplitudes. This roughness of shape contains all the details of the sound. Attack is the rise of the wave pattern, decay the fall of the wave pattern. If you skew the shape of the wave pattern, smooth it out, shorten it, or whatever, you're messing up the information. In this sense, proper attack and decay is important. The driver needs to recreate the whole wave pattern in full detail and shape.

The road is the motor, the force provider. A speaker is simply a floating mass attached to a spring damper system with a motor attached to one end. It really isn't much different from a car in context although physically the systems are different. It's a little bit of a stretch, but the concepts are similar.

Really? Most setups favor under damped systems. Like a sub when you put it in a box and gear towards a final Qtc of 0.707. 0.5 is critically dampened. 0.3 is over dampened. 0.7 is under dampened. Most setups are desgined to be critically dampened or under dampened to get the lowest F3. Very few setups are of an over dampened orientation, although some folks prefer it to get a tight, controlled type of sound.