**What is logarithm?**

The **logarithm** of a number is -more or less- how many digits it has.

So the **logarithm** of \(8347238\) is about \(7\) (actually it's \(6.921....\)). The **logarithm** of \(9\) is about \(1\). So it is a convenient way to measure the scale of a number.

Now this was a very rough idea. Let's make it more precise. If you take any number and multiply it by \(10\),it gains an additional digit, so according to what I said above, the **logarithm** increases by one. So what actually happens is, the **logarithm** of a number is how often I have to multiply \(10\) to get that number. So \(\log_{(100)}=2\) because \(10 \times 10 = 100\), so I multiplied two \(10s\) together. Similarly, \(\log_{(10000)}=4\) because \(10 \times 10 \times 10 \times 10=10000\), and so on. Multiplying four \(10s\) together can also be written as \(10^4\). And now we can generalise the idea:

\(\log_{(x)}= y\) is another way of saying that \(10^y = x\).

The final question you may have is: what's so
special about the number \(10\)? Why does it count how many \(10s\) I have to multiply together? Well actually, logarithm is used with other base numbers as well: base \(2\) and base \(e\) are also very common. But for each choice, you get **logarithm** function that measures the scale of the number you put in.

**What are logarithms used for? Are decibels a good example of the usefulness of a logarithm? Are logarithms calculus?**

**Logarithms** can be used to talk about things that can be both tiny and gigantic, such as in

*- earthquake magnitudes,*

*- noise levels in* **decibels,** and

*- acidity (pH).*

A big earthquake can be millions of times bigger than a tiny one. If you tried to make a bar graph where the bars has sizes \(10, 100,\) and \(10 000 000\), it would look stupid. The bars of size \(10\) and \(100\) would be too small to see, and you won't be able to tell that one of them is ten times bigger than the other. If you instead take the logarithm of each number, you get \(1, 2,\) and \(7.\) That makes a bar graph you can understand.

Keep that in mind when you hear about earthquake magnitudes. A \(7.0\) earthquake is ten times bigger than a \(6.0\) earthquake, which is ten times bigger than a \(5.0\) earthquake. Taking **logarithms** lets us put an earthquake caused by a stick of dynamite \((1.2)\) on the same scale as the \(2011\) earthquake in Japan \((9.0)\).

**Logarithms** can also be used to measure how long it will take something to grow *exponentially* or decay *exponentially*,such as

*- money growing with a fixed interest rate*,

*- bacteria growing in a petri dish,*

*- radioactive decay, for example in Radiocarbon dating*, and

*- the sound made by a bell*.

If you have bacteria that divide every \(30\) minutes and are currently taking up \(0.1\)% of the petri dish, you can use **logarithms** to estimate how long it will take them to fill up the entire dish. The same goes for \($5000\) in an account with a \(2\)% interest rate. If you leave the interest in the account, **logarithms** will tell you when you'll have \($6000\).

**Logarithms** can also be used in calculations by
turning multiplication into addition.
If I gave you the option between multiplying twenty numbers together by hand or adding twenty numbers together by hand, you'd pick the second option. If you need to multiply twenty numbers, you can instead take the base \(10\) **logarithm** of each number, add the results, and then raise \(10\) to that power. Finding the **logarithm** might seem hard, but, in the past, people could just look it up in a **logarithm** table or use a slide rule. Finding the answer using **logarithms** was way faster.

The problem of multiplying lots of numbers was the original reason **logarithms** were developed. This method is now obsolete thanks to computers, which are pretty fast at multiplying. In the meantime, though, we've discovered tons of uses of **logarithms**, most of which I haven't even listed here.

**What is the intuition behind the logarithm?**

**The logarithm counts the number of groupings.**

Suppose a bakery puts \(12\) cookies in a package, and places \(12\) of these packages in a larger box for transport:

Then a box can be seen as cookies which have been grouped twice: one box contains \(12^2 = 144\) cookies. Inversely, when ordering \(144\) cookies, and knowing that this bakery works with base \(12,\) the **logarithm** will return the number of groupings: \(\log_{12}{144} = 2\) groupings.
Suppose the transport company also works in base twelve; \(12\) boxes are wrapped in plastic, \(12\) plastic units are stacked onto a wooden pallet, \(12\) pallets are transported in a van, which makes \(12^3=1728 \) boxes per ride:

Now the total number of cookies in one transport, is obtained by multiplying the number of cookies per box with the number of boxes per ride: \(144 \times 1728 = 248832\) cookies.

However, if we look at this on a grouping scale (**logarithmic** scale) then we must use addition instead: the total number of groupings is \(2 + 3 = 5\). Stated otherwise: \(\log_{12}{144} + \log_{12}{1728} = \log_{12}{248832}\).
Now suppose \(20736\) packages of cookies were stolen. Then we know that this involved \(\log_{12}{20736} = 4\) groupings. So counting upwards from these packages...

...we know that this monster learned how to drive:

No vote yet

1 vote

×

Problem Loading...

Note Loading...

Set Loading...

Easy Math Editor

`*italics*`

or`_italics_`

italics`**bold**`

or`__bold__`

boldNote: you must add a full line of space before and after lists for them to show up correctlyparagraph 1

paragraph 2

`[example link](https://brilliant.org)`

`> This is a quote`

Remember to wrap math in \( ... \) or \[ ... \] to ensure proper formatting.`2 \times 3`

`2^{34}`

`a_{i-1}`

`\frac{2}{3}`

`\sqrt{2}`

`\sum_{i=1}^3`

`\sin \theta`

`\boxed{123}`

## Comments

Sort by:

TopNewestHow do I setup a logarithm for calculating the increase from interest rate or that petri dish thing you mentioned? I've always wondered that.

Log in to reply

For the petri dish example, at

\(t=0\) the petri dish is \(0.1\)% full,

\(t=30min\) the petri dish is \(0.1\)% \(\times 2\) full,

\(t=60min\) the petri dish is \(0.1\)% \(\times 2^2\) full,

\(t=90min\) the petri dish is \(0.1\)% \(\times 2^3\) full,

\(t=120min\) the petri dish is \(0.1\)% \(\times 2^4\) full.

You can see that a general formula for how full the petri dish at time \(t\) is \(0.1\)% \(\times 2^{\left(\frac{ t }{ (30min)}\right )}\). For example, for \(t=120min, \frac{t }{(30min) }= 4,\) and indeed the petri dish is \(0.1\)% \(\times 2^4\) full.

Assuming this growth rate continues until the petri dish is full, to find when that happens, we want to find the time t so that \(0.1\)% \(\times 2^{\left(\frac{t} {(30min)}\right)} = 100\)%. We begin by dividing both sides by \(0.1\)% to get \(2^{\left( \frac{t }{ (30min) }\right)} = 1000.\) Then we take the base two logarithm of both sides. \(\frac{t}{ (30min)} = log_2(1000)\) Then we multiply both sides by \(30min\) to get \(t = (30min) \times log_2(1000)\), or about \(5\) hours.

Log in to reply

Thanks! It's easier than any logarithmic equation exercise they teach us, and yet no one says anything about how useful this is in real life. Too bad high school algebra is all like that.

Log in to reply

Using logarithms is still a useful way to multiple numbers. While for small numbers, it's probably faster to just multiply - as the computer probably has a dedicated circuit for it - for larger numbers (numbers with a lot more bits than the word size), using logarithms to change it into an addition problem is probably going to be faster. This is especially true the less we care about the precise accuracy, and our numbers are too large to represent as floating point numbers (which has good hardware support).

Log in to reply

" This is especially true the less we care about the precise accuracy, and our numbers are too large to represent as floating point numbers (which has good hardware support)"

I didn't get it?

Log in to reply

Floating point doubles top out around 10^300, and even quads top out around 10^5000. Doing calculations on larger numbers requires other methods.

Log in to reply