| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Comments:
<Alconquian> we creative people with beautiful minds are lazy farks
<whats_in_a_name> this would normally be a professor at uni, but it neednt be
<whats_in_a_name> i stopped going to cl***es when it dawned on me all i do is day dream in cl***, and then learn the entire material by reading power point slides like 3 days before the exam
<Alconquian> yar, everything is readily forgotten since it usually isn't used again
<TRWBW> PuppiesOnAcid: d/dx (x-1)e^x = e^x+(x-1)e^x=x*e^x
<kanzure> Alconquian: The trick is forgetting how to forget.
<rixxon> I have a problem with will: can't force myself into things I don't really want to do, which I always have to in school.
<whats_in_a_name> i know the answer to that riddle, never learning it in the first place
<whats_in_a_name> ^_^
<vy> PuppiesOnAcid: You'll need to integrate it partially.
<TRWBW> PuppiesOnAcid: in general, d/dx p(x)e^x=(p'(x)+p(x))e^x, and for polynomials you can solve tht
<rixxon> At home, learning from reading texts on the net, and talking to people online... I learn quickly.
<PuppiesOnAcid> TRWBW: http://home.cognisurf.com/~adrleu/images/Pancake_head.jpeg
<TRWBW> heh
<Alconquian> rixxon: same here, but I get really bad anxiety that drives me to learning when I realize I may do badly
<PuppiesOnAcid> vy: What do you mean partially? With a u-substitution or by parts?
<kanzure> rixxon, Same here -- too bad we can't more quickly jot out words via typing (limit of typing speed for me is 150 wpm, not quick enough!)
<vy> S v*dv = u*v - S u*dv
<PuppiesOnAcid> It doens't look appropriate to do by parts
<Alconquian> also, it helps to realize the difference between you, "stupid people", and "smart people" is just time invested in studying
<TRWBW> PuppiesOnAcid: d/dx (x-1)e^x = e^x+(x-1)e^x=x*e^x. you wanted t*e^(t/2)=2*((t/2)*e^(t/2)). see the relationship to x*e^x?
<vy> PuppiesOnAcid: Yep, with a u-substitution.
<rixxon> Alconquian, I've done badly since seventh grade, though: kinda stuck with the problem. Can't re-do everything...
<PuppiesOnAcid> TRWBW: Why are you doing x*e^x? the problem is x*e^(x/2)
<Alconquian> I go back into the book and read over examples forever, then hit the internet if I have trouble after that
<rixxon> "everything" being six years of study
<vy> PuppiesOnAcid: The problematic t variable will disappear in the second part of the resultant integral of the u-substitution method.
<PuppiesOnAcid> Wiat hang on
<Alconquian> of course, it is possible to get a real crappy course book which is as unhelpful as possible unless you buy the $999 student solution manual
<TRWBW> PuppiesOnAcid: contemplate this on the tree of woe
<PuppiesOnAcid> Ok, hang on
<TRWBW> http://www.microwaves101.com/encyclopedia/images/Switches/treeofwoe.jpg
<PuppiesOnAcid> The first step, as I see it, is to do a u-substitution, with u being t/2
<PuppiesOnAcid> Is this correct?
<rixxon> Alconquian, heh, well, I live in the land of free as in free beer. Education doesn't cost.
<rixxon> Alconquian, with the result of nobody really caring for their education, but hey. :P
<Adamant> TRWBW, steel is not strong, flesh is stronger!
<vy> PuppiesOnAcid: No, v=t and du=(e^t/2)*dt
<PuppiesOnAcid> Oh, crap
<PuppiesOnAcid> We are doing u's and v's
<vy> S v*dv = u*v - S u*dv
<PuppiesOnAcid> I have totally forgotten how to do those
<PuppiesOnAcid> Let me read up on them again.
<PuppiesOnAcid> What is it's textbook title?
<vy> elementary calculus?
<PuppiesOnAcid> No, what is that concept called in a calculus book?
<vy> Ah... Partial integration (if I'd translate it from turkish).
<vy> PuppiesOnAcid: http://mathworld.wolfram.com/IntegrationbyParts.html
<PuppiesOnAcid> Ok, integration by parts.
<vy> After getting an AA in the PDE course and couldn't recall the langevin eq in the statistical mechanics final exam, I tried to solve the PDE by myself. There were two probabilities: Either I'd found the langevin eq. by myself or I'll fail.
<vy> I wish god won't put me again in such a situation.
<PuppiesOnAcid> % Integrate[E^(t / 2), t]
<PuppiesOnAcid> mbot is still offline
<linolium> lol
<TRWBW> PuppiesOnAcid: do you understand integration by parts?
<PuppiesOnAcid> Yeah, I just looked it up
<PuppiesOnAcid> I just get confused on integrating
<PuppiesOnAcid> Like, doing the anti-chain rule
<vy> Just replace du with e^(t/2)*dt and v with t in the integration by parts formula. That's all.
<vy> Sorry but, you should be able to integrate e^(t/2)*dt by yourself to be able to p*** that course.
<PuppiesOnAcid> vy: I get the integration by parts, it's the u-substitution I have trouble with
<TRWBW> PuppiesOnAcid: try this, differentiate 2*(t/2-1)*e^(t/2)
<kanzure> vy: integral of e^(t/2) is (t/2)e^(t/2) + some constant, yes?
<kanzure> erm
<PuppiesOnAcid> TRWBW: Isn't that just 2t * e(t/2) ?
<PuppiesOnAcid> Why did you write (t / 2 - 1) ?
<TRWBW> PuppiesOnAcid: (t/2-1)*e^(t/2)+e^(t/2)=(t/2)*e^(t/2)
<PuppiesOnAcid> Alright, I've got it now.
<TRWBW> PuppiesOnAcid: it's one step away from that x*e^x i gave you above
<PuppiesOnAcid> Right, I've got it now.
<TRWBW> PuppiesOnAcid: if you really nail differentiating, which is easier, chain rule and product rule and inverse rule and substitutions and all, it's becomes a lot easier to see why the integration tricks work
<TRWBW> PuppiesOnAcid: otherwise it's like trying to learn division when you haven't got multiplication down yet
<cimple> How do I map a vector onto a plane which is defined by a normal vector n and a point p?
<TRWBW> cimple: think of it like this, every point in space can be written uniquely as p+x+a*n, where x.n=0 and a is a scalar. make sense?
<PuppiesOnAcid> Ok, so after doing the integration by parts, I've got something even more complicated: 2e^(t/2) - Integrate[(2/t)(e^(t/2)),t]
<PuppiesOnAcid> The integral I have now is more complicated than my original integral
<cimple> yes that makes sense
<TRWBW> PuppiesOnAcid: you did it the wrong way. do it the other way
<TRWBW> cimple: okay, so you have some point q, and the point you want is p+x. right?
<PuppiesOnAcid> What is the other way? I set u = t; dv = e^(t/2);
<TRWBW> PuppiesOnAcid: there is only one other way
<ihope> Did somebody say mbot?
<vy> PuppiesOnAcid: You did the integration in the wrong direction. Replace du and v.
<PuppiesOnAcid> TRWBW: To switch my u value and dv value?
<ihope> % Sum::usage
<mbot> ihope: "Sum[f, {i, imax}] evaluates the sum of the expressions f as evaluated for each i from 1 to imax. Sum[f, {i, imin, imax}] starts with i = imin. Sum[f, {i, imin, imax, di}] uses steps di. Sum[
<mbot> f, {i, imin, imax}, {j, jmin, jmax}, ... ] evaluates a sum over multiple indices."
<oon> JuEnculeLotreJu: hey :)
<ihope> % NimSum::usage
<mbot> ihope: NimSum::usage
<oon> oon: have u seen prosper ?
<PuppiesOnAcid> In the end, I do uv - S v du
<oon> JuEnculeLotreJu: sorry, for u
<PuppiesOnAcid> That is the form I am using
<JuEnculeLotreJu> he has taken a pseudo
<JuEnculeLotreJu> the Ours !
<PuppiesOnAcid> vy: Is that right so far?
<TRWBW> cimple: if that wasn't clear, q=p+x+a*n (you don't know the x and a, you just know that like every point it can be written this way). you want to find p+x
<oon> JuEnculeLotreJu: bear
<JuEnculeLotreJu> oon: beark
<PuppiesOnAcid> vy: ?
<ihope> % NimberSum::usage
<mbot> ihope: NimberSum::usage
<cimple> I'm trying to follow TRWBW
<ihope> % Xor::usage
<mbot> ihope: "Xor[e1, e2, ... ] is the logical XOR (exclusive OR) function. It gives True if an odd number of the ei are True, and the rest are False. It gives False if an even number of the ei are True,
<mbot> and the rest are False."
<TRWBW> PuppiesOnAcid: what every you did, switch them.
<ihope> Bwah.
<PuppiesOnAcid> Switch _what_
<PuppiesOnAcid> There are multiple things to switch
<PuppiesOnAcid> I set u = e^(t/2); dv = t?
<TRWBW> cimple: you have q and you want to project it into the plane given by p and n. that means finding some point p+x in the plane such that q=p+x+a*n (for some a)
<PuppiesOnAcid> That seems kind of heinous
<kanzure> PuppiesOnAcid, Perhaps this shows that we need more firm names for certain parameters and numbers so that we can easily address what we're all talking about, rather than just communicating in new strands of variables and equations, heh.
<TRWBW> PuppiesOnAcid: i'm not sure, my guess is you had the names right but you integrated the wrong one.
<PuppiesOnAcid> I don't know, I'm just following the form in the text book.
<PuppiesOnAcid> Ok, here are my values
<cimple> yes TRWBW
<TRWBW> PuppiesOnAcid: well i don't know what your textbook says, but your mistake is clear. you integrate one and differentiate the other, and you picked the wrong ones. pick the other ones.
<Kasadkad> You integrated e^(t/2) wrong; the integral is 2e^(t/2), not (2/t)e^(t/2)
<vy> PuppiesOnAcid: http://rafb.net/p/iUqdPp42.html
<PuppiesOnAcid> u = t; du = dt; dv = e^(t / 2); v = (2 / t)(e ^ (t / 2))
<TRWBW> cimple: so q=p+x+a*n. using "." for dot product, q.n=p.n+x.n+a*n.n. but x.n=0. q and p and n you know. that gives you a. then p+x=q-a*n
<PuppiesOnAcid> ****.
<PuppiesOnAcid> That is all I have to say
<PuppiesOnAcid> I integrated one item wrongly.
<PuppiesOnAcid> Wait no
<PuppiesOnAcid> vy: You made a mistake
<PuppiesOnAcid> S e^(t/2) != 2e^(t/2)
<cimple> okay
<PuppiesOnAcid> Wouldn't it be (2/t)(e^(t/2)) ?
<vy> PuppiesOnAcid: No!
<PuppiesOnAcid> vy: How do you get that then?
<vy> Diff[e^(t/2)] = (dt/2)*e^(t/2)
<PuppiesOnAcid> dt/2?
<PuppiesOnAcid> What does that mean?
<PuppiesOnAcid> Why are you bringing differentials into this?
<TRWBW> cimple: okay you see or okay you see so far but don't know the next step?
<vy> PuppiesOnAcid: Can you differantiate e^k*t wrt t for me?
<PuppiesOnAcid> wrt?
<vy> with respect to
<cimple> don't know the next step
<PuppiesOnAcid> ke^(kt)
<TRWBW> PuppiesOnAcid: you really need to master differentiating like it was second nature before you can integrate.
<vy> PuppiesOnAcid: No. It is (k*dt)*e^(k*t)
<TRWBW> cimple: q.n=p.n+x.n+a*n.n , you understand what this means? as in the "." is dot product
<xaaax> hey guys, if I wanted to model mathematics from scratch(axioms and the like) in a programming language, what would you recommend?
<PuppiesOnAcid> TRWBW: It just gets confusing seeing all this in ASCII
<spx2> anyone here got any courses on ODE ? anyone ?
Return to
#math
or
Go to some related
logs:
hangvidia
pearl strftime
#perl
#web
#perl
#bash
udev swap dsp1
#css
#perl
gentoo broadcom 4400 modprobe
|
|
