Wolfram Mathematica

6:22 PM

Have you used Matlab's symbolic toolbox? What do you think about it?

http://www.mathworks.se/products/symbolic/description1.html

I need to implement this kind of function done in Mathematica but in Matlab:

The mathematica code works in 15 seconds, the Matlab code works in 4 hours...

acl

6:41 PM

Regarding this question. My answer is basically the answer to "how do I compute quantum dynamics in the case of a time-dependent Hamiltonian". I think it's a clean answer, and would be useful to anybody wondering how to do this.

Now the question is: assuming the OP agrees (which he might or might not, and I wouldn't go against it), I would like to change the question to ask something along the lines of "I have a time-dependent quantum Hamiltonian; how do I solve for the time evolution of an arbitrary initial state?"

Obviously this is not a site on physics, but since I put in the effort to write the answer it would make sense to make it actually useful.

Wolfram Blog

Wolfram Blog: The Mathematics of Queues

posted on March 21, 2013 by Devendra Kapadia

Waiting in line is a common, though not always pleasant, experience for us all. We wait patiently to be served by the next free teller at a bank, clear the security check at an airport, or be answered by technical support when we call a phone service provider. At a more abstract level, these waiting [...]

acl

@rm-rf (can't think of who else might know what the canonical answer to my question above would be)

rm -rf

@acl That would make sense and would actually be preferable to the current vague title

And such generalizations/minor change in focus of the questions are welcome and preferred (as long as it isn't drastic and doesn't trample upon several other answers) if it makes the situation better (more often than not, the OPs have no clue what they should be asking)

acl

@rm-rf OK I'll ask him if it's ok

@rm-rf I am mainly wondering how good an idea to have a question like "I want to do quantum dynamics; how?". There are much worse questions around, but...

Michael E2

@halirutan I found a fix for the Dynamic@{x, y other} -- y not updating at the end:

timeConsuming[x_] := (Pause[1]; x);
DynamicModule[{y},
 Manipulate[
  y = timeConsuming[x];
  Dynamic@{x, y, other},
  {x, 0, 1,
   Manipulator[
     Dynamic[x, {0 &, (x = #) &, Update[y] &}], ##2] &}, {other, 0, 1},
  SynchronousUpdating -> False]
 ]

Sorry, but off to a meeting.

rm -rf

6:57 PM

@acl Does the OP's code have any relevance to what you showed in your answer?

You could just reword the body to "I'm trying to do X in the case of a time-dependent Hamiltonian and this is what I've done. How can I fix blah" and the title to something descriptive

Pleo

@caveman yeah, there is also fieldOfView but it should be of no importance as its meaning is overridden by focalLength (at least it's value makes no difference in my acrionscript experiments as long as the focal length is specified) there is also this page help.adobe.com/en_US/FlashPlatform/reference/actionscript/3/…

rm -rf

@Pleo I think you entered the wrong room... you seem to be continuing from your previous conversation thread which was in this room:

$Mathematics$ Mathematics

Associated with Math.SE; for both general discussion & math qu...

11

46

acl

@rm-rf I don't know I did not read it

@rm-rf something like that would be the way yes

rm -rf

7:13 PM

@acl Those are the best kinds of answers (I like writing them) :)

hhh

8:17 PM

cf = Compile[{{x, _Real}}, Sin[x] + x^2 - 1/(1 + x)]

Could someone explain how the compile works?

acl

@rm-rf just realized that the OP of that question never accepted a single answer (and ignore my asking to go to a chat room). guess I'll wait a bit and then just change his question

@hhh could you be more specific?

hhh

What does the term "compile" mean?

Compile function Sin[x]+x^2-1/(1-x) for machine real x?

acl

@hhh did you read the docs?

hhh

This is from docs

but I don't understand it

rm -rf

"Compile" is fairly standard terminology...

hhh

8:20 PM

creats a compiled funcction that evaluates expr assuming numeriacl values of the x_i

rm -rf

Compiler

A compiler is a computer program (or set of programs) that transforms source code written in a programming language (the source language) into another computer language (the target language, often having a binary form known as object code). The most common reason for wanting to transform source code is to create an executable program. The name "compiler" is primarily used for programs that translate source code from a high-level programming language to a lower level language (e.g., assembly language or machine code). If the compiled program can run on a computer whose CPU or operating...

hhh

Why? Does it make code run faster? Is the purpose here to optimize the code somehow?

acl

@hhh start eg here

or for a more thorough discussion, here

rm -rf

@acl Well, 2 deleted, 2 closed/nearly closed (well, one at 4/5)... so this is their first "decent" question.

Might be worth giving them a welcome comment, but it might look awkward coming from you now... :)

acl

@hhh in particular here (see Benefits of compiling etc)

@rm-rf The question is, to be honest, garbage, not decent

if he was my student I'd tell him to come back with an understandable summary of what he's trying, not a mass of non-working code.

rm -rf

8:25 PM

@acl Hence in quotes :) For whatever reason, it didn't get any close votes and even got an answer (of course, because of your exceptional mood these days), so not garbage enough :P

acl

@rm-rf not getting close votes isn't really a good indicator. some of the stuff that gets upvoted is incredible.

rm -rf

Yes yes, I complain about that all the time :)

acl

I mean, upvoting is easy, but I don't think anybody else would have answered. It's nice to be friendly but I don't think encouraging people to dump their half-thought-through code here is good.

rm -rf

Well, the point is: moving forward, it'd hurt your answer, the chances of someone else finding the answer, etc. if it is attached to a crappy question, so the onus is on you (since you grok the stuff) to improve the post.

I remember one post from long ago when I used to edit like crazy — it was at -5 or so and after the edit, it slowly climbed to +5 and got good answers (I didn't answer). So yeah, if you have a good answer, fix the question too so that people don't skip your answer just because the question is crappy

acl

1

Q: Solving a system of ODEs (Schroedinger equation for a two-level system)

I want to solve the time-dependent Schroedinger equation: $$ i\partial_t \psi(t) = H(t)\psi(t) $$ for matrix, time-dependent $H(t)$ and vector $\psi$. What is an efficient way of doing this so that it efficiently scales to high-dimensional spaces?

brutal, but...

rm -rf

8:31 PM

lol.. that works too :)

hhh

@acl Thank you, that is clear now -- now getting to next line in your code :D

acl

@rm-rf it's more a question of finding it. if someone searches for "time-dependent schroedinger equation", they should find this (OK I need to edit again to add some keywords, I guess)

@hhh no problem

hhh

9:24 PM

What are the benefits of using local variables with Module?

Speed?

acl

@hhh to localize variables. almost all languages have scoping mechanisms.

the docs explain it

@hhh behold

Hi!

Hello!

hey

Welcome :)

9:29 PM

Does anyne know if it is possible to install a Mathematica "Application" in a non-standard location so that the documentation will be integrated as well? Without the documentation, it should be enough to change $Path

rm -rf

@Szabolcs I thought it should be possible if the documentation is placed with the currect directory tree and it has a pacletinfo.m

I haven't tried it though

Package -> Documentation -> English -> ReferencePages -> Symbols

Szabolcs

I'll try ...

I should read the WorkBench docs

rm -rf

@Szabolcs If you just create a new application, it'll create the documentation folder automatically

Szabolcs

Yes, I just did that ... I think I really need to read the docs on this in detail. It's not for MATLink, it's for some package I need to share with a colleague.

rm -rf

9:47 PM

@acl yay, more latex

acl

@rm-rf was just admiring that...

ha

hm, my good mood is being rapidly used up

Szabolcs

10:04 PM

@rm-rf Did you manage to change the indentation distance in Worbench? 4 spaces are a bit too much. I think I'll soon go back to Vim, I don't like WOrkbench that much

rm -rf

@Szabolcs I don't use the WB for anything other than the kernel connection it provides for testing.

I never use spaces, only tabs. The good thing with tabs is that you can change the tab width to your tastes (which is easy in vim)

However, I think I remember seeing such an option somewhere in eclipse... let me check

Szabolcs

ah don't bother

acl

@Szabolcs can't you use tabs and then just change how they're displayed?

that is what I always do, everywhere

like @rm-rf says

Szabolcs

@acl Does Workbench have that flexibility? To force it to always use tabs and to get it to display it with as many spaces as I like?

acl

@Szabolcs I think so, let me check

Szabolcs

10:07 PM

With C code I also use tabs, but with Mathematica I didn't want to

rm -rf

@Szabolcs yes, go to prefs and type "tab" in the search box. Under "Editors > Text editors", you'll find the options

Change the tab width and uncheck the "use spaces instead of tabs" option

acl

@Szabolcs it has a setting, but it appears to just insert as many spaces as you specify (rather than tabs). could be wrong though, don't really use it

@rm-rf ah

rm -rf

FWIW, I also use markers to indicate a tab and EOL, so if there isn't one, it's most likely a space

a bit faint, but you can see the » and ¶ if you look hard... (shows up nicely on my screen)

Szabolcs

It should be faint for usability.

rm -rf

Exactly

Dominic Michaelis

10:31 PM

hi

Adam Dreaver

11:04 PM

I wish mma had CSS layouts!

halirutan

11:44 PM

Trick-Question day:

Without entering or viewing the following in a Mathematica notebook, try to guess the semantics (or Fullform's) of the following 3 constructs:

Hold[1.1.1] // FullForm
Hold[1.1.a] // FullForm
Hold[a.1.1] // FullForm

Adam Dreaver

This isn't a question for newbies :(

FullForm on my last entered f[x] is a bit verbose, god knows what daemons lie behind the doors of your trick question.

halirutan

@AdamDreaver If you don't tell here, you can of course cheat and simply look it up.

Nothing bad will happen, but you may be surprised how it is interpreted.

Matariki

I guess it will interpret one or more of the .'s as separators giving something like 1.1 and 0.1? I can't think in FullForm

@halirutan I have scanned the Wagner book but have found about 17 pages with ink scribblings on it. Could you provide me with scans from your copy of the books?

halirutan

@Matariki Yes, I may be able to scan it on the weekend.

Can you write me a mail which pages do you need?

Matariki

@halirutan Great. Can I email you the page numbers?

halirutan

11:57 PM

31 secs ago, by halirutan

Can you write me a mail which pages do you need?

@Matariki Hehe

@Matariki Do you see my mail address in my profile?

Matariki