I found a *useful* method to calculate log(a+b), check it out

  • Context: Undergrad 
  • Thread starter Thread starter guifb99
  • Start date Start date
  • Tags Tags
    Log Logarithm Method
Click For Summary

Discussion Overview

The discussion revolves around a proposed method for approximating the logarithm of the sum of two numbers, Log(a+b). Participants explore the utility and limitations of this method, which is based on approximations rather than exact calculations. The conversation includes mathematical reasoning and notation clarification.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Mathematical reasoning

Main Points Raised

  • One participant presents a method for approximating Log(a+b) as Log(2) + Log(SQRT)(a*b), emphasizing its usefulness for practical applications.
  • It is noted that the method provides approximations rather than precise results, with conditions on the values of a and b affecting accuracy.
  • Another participant questions whether this method effectively approximates the arithmetic mean by the geometric mean.
  • A participant expresses unfamiliarity with the concept of geometric mean, indicating a range of mathematical backgrounds among contributors.
  • Clarifications are made regarding notation, with suggestions for clearer representation of square roots in mathematical expressions.
  • Some participants discuss the assumptions necessary for the method to hold, particularly regarding the positivity of a and b.
  • There is a mention of the importance of the difference between a and b being small for the approximation to be valid.

Areas of Agreement / Disagreement

Participants express varying levels of understanding and agreement on the method's effectiveness and its mathematical foundations. There is no consensus on the validity of the approximation or its implications, as some participants seek clarification while others provide alternative perspectives.

Contextual Notes

Participants highlight the need for careful notation and the conditions under which the approximation is valid, such as the positivity of a and b and the closeness of their values.

Who May Find This Useful

This discussion may be of interest to individuals exploring logarithmic approximations, students learning about logarithmic properties, and those interested in mathematical reasoning and notation clarity.

guifb99
Messages
10
Reaction score
1
So, I found this method, I don't think I was the first to, though, but I don't see any post related to this anywhere on the internet, so maybe there's a slim chance I was the first? Anyway, it doesn't really matter. The method does not give the precise result, only approximations, but I find it really useful, since when I have to use logarithms, I only work with approximations. Here it is:

Log(a+b) = Log(2) + Log(SQRT)(a*b)

It works with any base you choose, as long as it is the base for Log(a+b), Log(2) and Log(SQRT)(a*b) at the same time

See the "equals to" symbol as an "is approximate to" symbol

Also, "(SQRT)" means "square root of", quite obvious

And since Log(SQRT)(a*b) is (Log(a*b))/2, then

Log(a+b) = Log(2) + (Log(a*b))/2

And since (Log(a*b))/2 = (Log(a) + Log(b))/2, then

Log(a+b) = Log(2) + (Log(a) + Log(b))/2

In base 10, Log(2) is approximately 0,3, so

Log(a+b) = 0,3 + Log(SQRT)(a*b) in base 10

***IMPORTANT***

The bigger (a+b) is, the closer the results will be to the real values

If a>b

The smaller (a-b) is, the closer the results will be to the value

If b>a

The smaller (b-a) is, the closer the results will be to the value.

Example:
If you use this to find Log(11) and you want to use only natural numbers, you should use "6" and "5" for "a" and "b", it will not work if you use, instead, "10" and "1" for "a" and "b"

Also, "a" and "b" can be any REAL number to make this work, I don't know if this works with complex numbers, but I'm guessing it doesn't.

Anyway, I hope this helps you guys, any questions just ask me :)
 
Last edited:
  • Like
Likes   Reactions: Battlemage!
Physics news on Phys.org
I'm brazilian so we use commas in cases like Log(2) = 0,3

But for americans, you can see the 0,3 as a 0.3 ;)
 
So you basically approximate the arithmetic mean by the geometric mean?
 
  • Like
Likes   Reactions: fresh_42
micromass said:
So you basically approximate the arithmetic mean by the geometric mean?

I don't even know what a geometric mean would be, I only know the basics of logarithms... I'm 15 years old
 
Google it.
 
guifb99 said:
Log(a+b) = Log(2) + Log(SQRT)(a*b)
Your notation in the last term on the right is confusing. Most people would normally write a square root of a*b as SQRT(a*b), not make it look like (SQRT) "times" (a * b).

The reason behind your IMPORTANT notes is this:

Suppose ##\frac{a + b} 2 = \sqrt{ab}##
This equation is equivalent to the one I quoted above, assuming that a + b > 0 and ab > 0 for the logs to be defined.
Square both sides: ##\frac 1 4 (a^2 + 2ab + b^2) = ab##
Multiply both sides by 4: ##a^2 + 2ab + b^2 = 4ab##
Add -4ab to both sides: ##a^2 - 2ab + b^2 = 0##, or ## (a - b) = 0##

With the assumptions I listed above, all of the steps are reversible, so if a = b, then ##\frac{a + b} 2## is exactly equal to ## \sqrt{ab}##, and the logs of both expressions would be equal as well.
 
  • Like
Likes   Reactions: Battlemage!
Mark44 said:
Your notation in the last term on the right is confusing. Most people would normally write a square root of a*b as SQRT(a*b), not make it look like (SQRT) "times" (a * b).

The reason behind your IMPORTANT notes is this:

Suppose ##\frac{a + b} 2 = \sqrt{ab}##
This equation is equivalent to the one I quoted above, assuming that a + b > 0 and ab > 0 for the logs to be defined.
Square both sides: ##\frac 1 4 (a^2 + 2ab + b^2) = ab##
Multiply both sides by 4: ##a^2 + 2ab + b^2 = 4ab##
Add -4ab to both sides: ##a^2 - 2ab + b^2 = 0##, or ## (a - b) = 0##

With the assumptions I listed above, all of the steps are reversible, so if a = b, then ##\frac{a + b} 2## is exactly equal to ## \sqrt{ab}##, and the logs of both expressions would be equal as well.

Sorry, I'm new to this forum, didn't even know we could use these "uncommon outside the mathematic language" characters such as √ here, but thinking about it now, it makes sense that we would have these as this forum is basically about physics and mathematics. Anyway... thanks for explaining that, now I know why (a–b) must be close to zero in order to make this approximation works

Writing it again...

a>b ⇔ Lim(a–b)→0Log(a+b) ≈ Log(2) + Log√(a⋅b)
 
guifb99 said:
Sorry, I'm new to this forum, didn't even know we could use these "uncommon outside the mathematic language" characters such as √ here
See https://www.physicsforums.com/help/latexhelp/.

I used \frac{a + b} {2} to make the fraction and \sqrt{ab} to make the square root. Put a pair of # characters at the beginning of the expression, and another pair at the end, and your browser will turn them into nicely formatted math.
 
This is really cool. Thanks.
 

Similar threads

  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 12 ·
Replies
12
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 8 ·
Replies
8
Views
9K
  • · Replies 3 ·
Replies
3
Views
3K
Replies
8
Views
2K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 11 ·
Replies
11
Views
6K