*The FAQ section at caraudioforums.com contains several omissions,
glaring errors ...*Point them out, they will be corrected.

Ok. Corrections (apologies in advance for any stepped-on toes):

Part 1: Definitions
I may address other sections later.

"1.2 Frequency response

"The frequency response of a device is the range of frequencies over which
that device can perform in some fashion. The action is specific to the
device in question. For example, the frequency response of the human ear is
around 20Hz-20kHz, which is the range of frequencies which can be resolved
by the eardrum."

**Not quite. The ear drum is responsive over a much broader frequency
range. The frequently-quoted 20-20kHz is a figure limited primarily by the
mechanical transduction process of the inner ear. Also, a more realistical
number would be somewhere between 30Hz and 18kHz.

**Moreover, the phrase "frequency response of the human ear" is a bit faulty
because it's not really a frequency response (the transfer function is a lot
more complex, and it's modulated by other factors). It's a sensitivity
function. Anyway, I'd take that line out.

**This, by the way, also brings up a glaring ommission. There's nothing
written anywhere about the most important part of the audio chain: the human
observer!

"The frequency response of an amplifier may be 50Hz-40kHz, and that of a
certain speaker may be 120Hz-17kHz. In the car audio world, frequency
responses should usually be given with a power ratio range as well, such as
(in the case of the speaker) 120Hz-17kHz +/-3dB. What this means is that
given an input signal anywhere from 120Hz to 17kHz, the output signal is
guaranteed to be within an "envelope" that is 6dB tall. Typically the
extreme ends of the frequency range are the hardest to reproduce, so in this
example, the 120Hz and 17kHz points may be referred to as the "-3dB points"
of the amplifier. When no dB range is given with a frequency response
specification, it can sometimes be assumed to be +/-3dB."

**An important point, but it seems buried in the numbers. I think it should
be stressed that the frequency response must be accompanied by the
"attenuation" numbers, and a sample freq response graph can even be provided
to illustrate how freq response can change if you use different values as
your threshold (a picture says a thousand words). It should also be pointed
out that improving freq response beyond human detection capability is
fruitless.

"1.1 What do all these acronyms mean
snipped reasonable definitions

"W is for watts, a measurement of electrical power. One watt is equal to one
volt times one amp, or one joule of energy per second. In a DC circuit, the
power is calculated as the voltage times the current (P=V x I). In an AC
circuit, the RMS power is calculated as the RMS voltage times the RMS
current (Prms=Vrms x Irms)."

**What's quoted above is not "RMS power", but rather "average power". "RMS
power" is something entirely different, and in fact has no relevance to
anything in audio.

"dB is for decibel, and is a measurement for power ratios. To measure dB,
you must always measure with respect to something else. The formula for
determining these ratios is P=10^(dB/10), which can be rewritten as
dB=10log(P). For example, to gain 3dB of output compared to your current
output, you must change your current power by a factor of 10^(3/10) = 10^0.3
= 2.00 (that is, double your power). The other way around, if you triple
your power (say, from 20W to 60W) and want to know the corresponding change
in dB, it is dB=10log(60/20)=4.77 (that is, an increase of 4.77dB). If you
know your logarithms, you know that a negative number simply inverts your
answer, so that 3dB corresponding to double power is the same as -3dB
corresponding to half power. There are several other dB formulas; for
instance, the voltage measurement is dB=20log(V). For example, a doubling of
voltage produces 20log2 = 6.0dB more output, which makes sense since power
is proportional to the square of voltage, so a doubling in voltage produces
a quadrupling in power."

**What a convoluted explanation!

"SPL is for sound pressure level and is similar to dB. SPL measurements are
also ratios, but are always measured relative to a constant. This constant
is 0dB which is defined as the smallest level of sound pressure that the
human ear can detect. 0dB is equal to 10^-12 (ten to the negative twelfth
power) W/m^2 (watts per square meter). As such, when a speaker is rated to
produce 92dB at 1m when given 1W (92dB/Wm), you know that they mean that it
is 92dB louder than 10^-12W/m^2. You also know than if you double the power
(from 1W to 2W), you add 3dB, so it will produce 95dB at 1m with 2W, 98dB at
1m with 4W, 101dB at 1m with 8W, etc."

**SPL is not "similar to dB". That line should be removed. The "constant"
needs to be redefined. Test frequency should also be introduced here to
describe these sensitivity measurements.

"THD is for total harmonic distortion, and is a measure of the how much a
certain device may distort a signal. These figures are usually given as
percentages. It is believed that THD figures below approximately 0.1% are
inaudible. However, it should be realized that distortion adds, so that if a
head unit, equalizer, signal processor, crossover, amplifier and speaker are
all rated at "no greater than 0.1%THD", together, they could produce
0.6%THD, which could be noticeable in the output."

**That 0.1% number is rubbish. It should be removed. Though a discussion
of threshold is warranted.

Also, as Lizard said, there are many omissions to this list.