Simple calc question

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
If I integrate 1/2x dx...

Just straight integration, I get (ln x)/2

But if I use u-sub where u = 2x, then I get (ln 2x)/2

So what's the problem?
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
u = 2x
du = 2dx

so dx = du/2

So we're integrating 1/2u du which is (ln u )/ 2 which equals (ln 2x) / 2

No?
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
Can you provide the steps please? For u-sub. Or at least point out my error. Thanks.
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
Well...thanks, I guess.

But why is the answer different when using u-sub?

That's my original question. :) Is there some fundamental law of calculus being broken or something?
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: gwlam12
If I integrate 1/2x dx...

Just straight integration, I get (ln x)/2

But if I use u-sub where u = 2x, then I get (ln 2x)/2

So what's the problem?

You're almost right.

a) 1st part should get ln(x)/2+C1
b) 2nd part should get ln(2x)/2+C2 = ln(2)/2+ln(x)/2 + C2 = ln(x)/2 + C3

So they turn out the same if you just lump together the constants.
 

Scrooge2

Senior member
Jul 18, 2000
856
0
0
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: Scrooge2
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.

He forgot the PLUS C part at the end of any indefinite integral. The second integral will turn out to be the same as the first if he just lumps together all the constants.
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
The constant...the constant. Hah. So you're saying with u-sub and without u-sub, the constants are different? Cool. I just graphed it, and you're right, there's a difference by .35 all the way through.

Thanks.

And Scrooge2, no, I am not taking Calc 2. I missed the constant, just like you did.
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
But wait...

If you consider the definite integral..shouldn't they be the same?

<edit>nevermind on this</edit>


BUT

so if you were taking a test and it asked you to find the integral of 1/2X...what would YOU put as your answer?
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
Originally posted by: Scrooge2
Heh, some help i turned out to be :D. What are you taking then?

I'm not taking any math right now. My friend is studying for her GRE and popped that question to me. I tried it out, had the same problem, thought I'd go to geek site, as I like to refer to this place as. :)
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
Originally posted by: TuxDave
Originally posted by: Scrooge2
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.

He forgot the PLUS C part at the end of any indefinite integral. The second integral will turn out to be the same as the first if he just lumps together all the constants.

Hey TuxDave, I'm lost again. I thought the difference between lnx/2 +C1 and ln2x/2 + C2 was constant, but it's not. The difference gets bigger as the numbers get larger. Am I missing something?
 

electricJ

Senior member
Apr 10, 2004
386
0
0
Well, try this:

Let 1/(2*x) = (1/2)*(1/x)

1/2 is pulled outside of the integral, thus you only need to integrate (1/x). This integration leaves you with ln(x), which must be multiplied by the factor you pulled out giving you (1/2)*ln(x) (assuming that it is a definate integral, if not, then it must be (1/2)*ln(x) + C
 

gwlam12

Diamond Member
Apr 4, 2001
6,946
1
71
Thanks...but...

My question is, why is it that when I do u-sub, I don't get 1/2 * lnx :)
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: gwlam12
Originally posted by: TuxDave
Originally posted by: Scrooge2
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.

He forgot the PLUS C part at the end of any indefinite integral. The second integral will turn out to be the same as the first if he just lumps together all the constants.

Hey TuxDave, I'm lost again. I thought the difference between lnx/2 +C1 and ln2x/2 + C2 was constant, but it's not. The difference gets bigger as the numbers get larger. Am I missing something?

uhh... the difference between lx(x)/2 and ln(2x)/2 should be a constant. It should be off by exactly ln2/2
 

dighn

Lifer
Aug 12, 2001
22,820
4
81
Originally posted by: gwlam12
Originally posted by: TuxDave
Originally posted by: Scrooge2
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.

He forgot the PLUS C part at the end of any indefinite integral. The second integral will turn out to be the same as the first if he just lumps together all the constants.

Hey TuxDave, I'm lost again. I thought the difference between lnx/2 +C1 and ln2x/2 + C2 was constant, but it's not. The difference gets bigger as the numbers get larger. Am I missing something?

it is a constant

ln(2x) = ln(2) + ln(x) => off by a constant ln(2)
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: gwlam12
But wait...

If you consider the definite integral..shouldn't they be the same?

<edit>nevermind on this</edit>


BUT

so if you were taking a test and it asked you to find the integral of 1/2X...what would YOU put as your answer?

I would put ln(x)/2 + C because that's the simpliest solution.
ln(x)/2+A+B+C+D+Es also correct, but not as simple.
 

electricJ

Senior member
Apr 10, 2004
386
0
0
Originally posted by: dighn
Originally posted by: gwlam12
Originally posted by: TuxDave
Originally posted by: Scrooge2
TuxDave that one flew right over my head, what did you do? I think he's justt taking calc 2.

He forgot the PLUS C part at the end of any indefinite integral. The second integral will turn out to be the same as the first if he just lumps together all the constants.

Hey TuxDave, I'm lost again. I thought the difference between lnx/2 +C1 and ln2x/2 + C2 was constant, but it's not. The difference gets bigger as the numbers get larger. Am I missing something?

it is a constant

ln(2x) = ln(2) + ln(x) => off by a constant ln(2)

Ding, ding, ding... we have a winner! :beer:
 

TuxDave

Lifer
Oct 8, 2002
10,571
3
71
Originally posted by: gwlam12
Thanks all, it is constant. I feel stupid now, but thanks.

:beer: It's ok. Even I was confused for a second when you first posted the problem.