# Talk:Gain

WikiProject Professional sound production (Rated C-class, High-importance)
This article is within the scope of WikiProject Professional sound production, a collaborative effort to improve the coverage of sound recording and reproduction on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C  This article has been rated as C-Class on the project's quality scale.
High  This article has been rated as High-importance on the project's importance scale.

## More Ambiguity

The article defines gain as "the mean ratio of the signal output of a system to the signal input of the same system". Therefore when talking about voltage gain, you would expect this to be G=Vout/Vin. The line "The term gain alone is ambiguous, and can refer to the ratio of output to input voltage, (voltage gain), ..." reinforces this idea. Below, under the voltage gain section, voltage gain is defined as G=(Vout/Vin)^2, which seems to contradict the definition of gain above. Therefore this seems to still be power gain even though it's in terms of voltage. I know this is the standard way of doing it, but maybe someone with more knowledge and experience could attempt to distinguish between the two using better names.
Edit: From the article here (http://www.allaboutcircuits.com/vol_3/chpt_1/5.html), it states "Because the bel is fundamentally a unit of power gain or loss in a system, voltage or current gains and losses don't convert to bels or dB in quite the same way". It seems as if the concept of gain is different when expressing it as a dB number compared to just a ratio. In anycase, I think it should be reworded to make this all clearer.

## Ambiguity

[ EDIT -- The previous paragraph doesn't make sense. In any context where it isn't clear whether it is voltage or power gain that is being referred to, the statement "Gain = 20 dB" is ambiguous. It could mean either "V2/V1 = 10 , P2/P1 = 100" or "V2/V1 = 100 , P2/P1 = 10000". If what was meant was "unless stated otherwise, it's assumed that gain refers to power rather than voltage", then that's what should have been said. Edit to Edit - The original statement is correct because a 10x voltage increase is equivalent to a 100x power increase, and both a 10x voltage increase and a 100x power increase correspond to a gain of 20 dB. Therefore, it is uneccessary to specify if the gain refers to a power increase or a voltage increase as the two are related. ]

## Split

I have split up gain so that now this aritcle only discussed the electornics side of things. See Gain (disambiguation) for where the other info ended up.--Commander Keane 20:51, 9 December 2005 (UTC)

Split looks good as is. The paragraph on antenna gain is probably OK here, especially since the decibel equations above it are essentially correct for antenna gain calculation as well. --ssd 12:54, 24 January 2006 (UTC)
I agree that the antenna gain deserves a place in here. Anyone looking for more information on Antenna Gain in particular can go to the link, those who just want a general overview of gains regarding electronics can only "gain" (heh heh!) from seeing that it can be applied to EM design also. --Diom1982
I think antenna gain should be kept with antennas. Electronic gain requires energy input from a power supply. Antennas dont provided gain in this sense, they just capture more energy than an isotropic reciever would.--Light current 17:44, 9 February 2006 (UTC)
I have moved antenna gain stuff to antenna now.--Light current 17:51, 9 February 2006 (UTC)

## Huah?

This article sucks. It has no preamble to describe what gain is. Maybe you electricians understand it, but it's useless to others. —The preceding unsigned comment was added by 129.97.234.72 (talk) 19:44, 6 December 2006 (UTC).

I agree. It's apparently the result of editors trying to out-math each other. I went here from a link regarding high gain as a distinctive part of the band Dinosaur Jr.'s sound; I still have no idea what high gain sounds like, and I don't want to crunch equations to make sense of the article. Can someone with practical knowledge improve this article a bit? Vbdrummer0 02:40, 14 August 2007 (UTC)
It was deleted by a vandal. — Omegatron 22:07, 6 December 2006 (UTC)

## The example of GAIN

The example is very strange. I think the answer is 20(dB)

Ahem. I was wondering when you would notice that. Sorry, my mistake. --Heron 18:32, 2 May 2007 (UTC)

The example is still confusing, the rest of the article talks about gain in terms of decibels, and the example has gain in terms of V/V. Shouldn't I see a logarithm in there somewhere? Is this equivalent to 10 db (voltage) gain? —Preceding unsigned comment added by 157.127.124.14 (talk) 16:58, 1 July 2008 (UTC)

## Gain measured in ohms

I sometimes see references to gain being measured in units of resistance. I admit to not understanding this, or whether there are only certain circumstances when this might be appropriate (e.g., transimpedance gain, which appears not to have its own article or mention in this one). My best guess is that it refers to a voltage gain per some input current current gain per some input voltage, per Ohm's law. Can any experts explain why gain might be measured in units of resistance, and under what condiitons it ought to be? Robert K S (talk) 15:00, 16 June 2008 (UTC)

A transimpedance amplifier is one that converts a current input to a voltage output. The units of its gain, volts per ampere, is ohms. Dicklyon (talk) 17:48, 16 June 2008 (UTC)
After your strikeout and rewrite: the current out per voltage in is what you get from a transconductance amplifier. Its gain is reciprocal ohms, mhos, or siemens. Dicklyon (talk) 17:55, 16 June 2008 (UTC)

Gain is a mathematical concept, and this is an article about its APPLICATION. can you move this to gain (electronics) or something, and leave the maths/signals here? I'm trying to link in from runaway climate change and it doesn't make sense.Andrewjlockley (talk) 01:56, 11 February 2009 (UTC)

I don't know of it as a mathematical term, and there is no math page on the disambig. Are you talking about finance, compounding interest, something? And even accepting that there is something in mathematics called gain, I think you're wrong stating this as an application of it, because I don't think there's a specialized term for the log of a ratio. Awickert (talk) 02:10, 11 February 2009 (UTC)

## Congratulations

It is clear that the definition of decibel in this article is based on actual practice; as a result it turned out one of the very few on Wikipedia that contains no obfuscation and is also mathematically rigorous.

Having said this, I would suggest three things for which a change is needed.

The first is disambiguating log by writing either $\mathrm{log}_{10}$ or lg; unfortunately the latter can be confused with the base 2 logarithm.

Second,

$G_{dB}=10 \log G_{W/W}=10 \log 100=10 \times 2=20\ \mathrm{dB}.$

should be

$G_{dB}=10 \log G=10 \log 100=10 \times 2=20.$

Indeed, according to the definition of power gain at the start, 20 dB = 100, and so the last part of the chain as it currently stands amounts to $10 \times 2 = 100$.

Also, since the chain starts with $G_{dB}$, standing for "G expressed in decibel", it is wise to follow the usual good practices for handling units. If the length of something is 3 ft, then its length in feet is 3 (a dimensionless number), not 3 ft. In formula form, $\ell_\mathrm{ft} = 3 \Leftrightarrow \ell = 3\,\mathrm{ft}$ and $\ell_\mathrm{ft} = 3\,\mathrm{ft}$ is wrong. Similarly for decibel: $G_\mathrm{dB} = 20 \Leftrightarrow G = 20\,\mathrm{dB}$.

A third point is the V/V and W/W notation. This seems like a comment, since, mathematically, these units cancel out and V/V = 1 = W/W. Formalizing the intended effect may not be worthwhile. Better keep this "null information" out of the formulas and provide real information in the text. Simply using fitting identifiers like Pgain and Vgain is also an option. Boute (talk) 04:14, 7 August 2010 (UTC)

Both the V/V and W/W ratios are nondimensional; yet to get to decibels, you need to know which it is. In the case of voltage ratios, the dB equivalent is 20log(ratio). That's one problem with your attempt to rationalize dB into a numeric operator. As to your 10 x 2 = 100 comment, that doesn't make sense within the usual interpretation of these equations, which doesn't include the homomorphism that you're basing your interpretation on. Dicklyon (talk) 05:14, 7 August 2010 (UTC)
You must indeed know whether V or W is involved, but writing V/V kills that information. So this does not work, which is why I suggested keeping that information in the (con)text or use proper identifiers. By the way, I am not rationalizing dB into a numeric operator any more than already follows from the NIST standard Special Publication 330, as the derivation on the decibel talk page shows. Note that the definition in the gain article writes $\text{Gain}=10 \log \left( {\frac{P_{\mathrm{out}}}{P_{\mathrm{in}}}}\right)\ \mathrm{dB}$ and not $\text{Gain}_{dB}=10 \log \left( {\frac{P_{\mathrm{out}}}{P_{\mathrm{in}}}}\right)\ \mathrm{dB}$. The rest of the article has to be consistent with that. Finally, it follows from the nearly universal practice of working with units (most often already taught in high school) that $\ell_\mathrm{ft} = 3\,\mathrm{ft}$ is wrong, and since dB is considered a unit, so is $G_\mathrm{dB} = 20\,\mathrm{dB}$. The correspondence $\ell_\mathrm{ft} = x \Leftrightarrow \ell = x\,\mathrm{ft}$ is also nearly universal practice, up to notational variants. In fact, it is not "my interpretation" but a consequence from the common definition of measurement in units $\ell = \ell_\mathrm{ft}\,\mathrm{ft}$. Why make exceptions for decibel?
Afterthougt: I admit there is a difference with the decibel situation. Whereas the practice I mentioned for units is nearly universal, for decibel the number of variations in the ways in which people write up things is extremely large (see the references in my paper), so the "usual interpretation" to which you refer does not exist, only chaos. Many authors and spec sheets write things correctly (sometimes bypassing the poor definitions they presented at first!), but I cannot quote percentages (which would require a larger sample). As regards a list of references showing this, I think you said you had my paper, otherwise I will send you a copy. Boute (talk) 06:32, 7 August 2010 (UTC)
I just found that the NIST Guide to the SI applies exactly the aforementioned universal convention for units also in its treatment of dB, Neper etc.: $\{L_F\}_\mathrm{B} = 2\,\mathrm{lg}(F/F_0)$ and $L_F = 2\,\mathrm{lg}(F/F_0)\,\mathrm{B}$. So abuse of notation and ambiguity appears to be avoided by (some) standards bodies. It is the ONLY way to do things properly. Boute (talk) 12:37, 7 August 2010 (UTC) 21:55, 9 August 2010 (UTC)
Note (earlier typo corrected in the preceding): F is a subscript indicating that the formula pertains to field quantities, F is a variable designating a field quantity. For the {LF}B = 2 lg(F/F0) <=> LF = 2 lg(F/F0) B issue, this is unimportant, but for the formula itself it is obviously significant. Boute (talk) 13:43, 7 August 2010 (UTC)
Warning: in NIST Guide to the SI - html version on which the preceding "correction" was based, italics and normal fonts are known to be displayed incorrectly at various places depending on the browser. The only correct (reference) version is | NIST Guide for the Use of the International System of Units - pdf version, which indicates that $\{L_F\}_\mathrm{B} = 2\,\mathrm{lg}(F/F_0)$ and $L_F = 2\,\mathrm{lg}(F/F_0)\,\mathrm{B}$ as shown first were the correct formulas (in $L_F$ the variable is $F$, not $L$ as in some other standards). Boute (talk) 21:55, 9 August 2010 (UTC)

## Search for oxymoronic "input gain" came up empty

There is plenty of usage of the seemingly contradictory term "input gain," presumably (?) as the OUTPUT of a preamp or whatever. If gain is the increase in amplitude from input to output, how, without "cheating" like this, can there be an ill-named "input gain." By failing to address this usage, even if it's gibberish, the article falls down on the job. The same goes for the redundant term "output gain." — Preceding unsigned comment added by 74.93.212.105 (talk) 14:30, 13 June 2013 (UTC)

In my work in live audio I see many people and manufacturers use the term "input gain" to refer to a passive attenuator—usually a potentiometer—that controls the amount of input signal sent to the preamp. In that case, input gain is not gain at all, it is attenuation prior to gain. Binksternet (talk) 16:24, 13 June 2013 (UTC)