Large Hardon Collider

[Kaylia]

My bad, I knew I was misunderstanding something x_x

It's not "your" bad, it's mine. I should have realized the proof was working for complex number, I just ignored it and tried to find a more...complex derivation.

Life on Mars Theory Boosted by New Methane Study

I definitively like where it is going. It's really hard to not get your hope up with study like this, even if it could very well be something else.

[Evensong]

Scientists analysing data from telescopic observations and unmanned space missions have discovered that methane on Mars is being constantly replenished by an unknown source and they are keen to uncover how the levels of methane are being topped up.

The mystery deepens...2018 is pretty far off though, but I'm guessing the extraterrestial enthusiasts will be kept happy with theory number one until then. Well, with only two plausible options left, little wonder I suppose.

[Max™]

It's not "your" bad, it's mine. I should have realized the proof was working for complex number, I just ignored it and tried to find a more...complex derivation.

"The use of words expressing something other than their literal in~ten~tion!
Now, THAT... IS... I~ron~y!" ~ Bender

I definitively like where it is going. It's really hard to not get your hope up with study like this, even if it could very well be something else.

I farted on the meteorite, sorry to get your hopes up.

[Woozie]

Kaylia/Eliseos, how do I use exponents in java? I don't mean like e^5, I mean like 3^5.

[Kaylia]

Kaylia/Eliseos, how do I use exponents in java? I don't mean like e^5, I mean like 3^5.

int x = java.lang.Math.pow ( 2, 3); // 2^3


Might need to import maths in the header of your program

[Woozie]

Thanks. I looked for it on the java website thingy but somehow I missed that method.

[Woozie]

Now I need to make a matrix with 3 columns and k rows.

I'm used to matlab, where everything is already a matrix. Here I'll have to use arrays since it's java.

So would I say

double[][] coil = new double[3][k];

or

double[][] coil = new double[k][3];

?

[Eliseos]

I think the dimensions go from left to right for arrays, so the first brackets would be rows, second is columns. Let me look up the documentation to be sure.

EDIT: I was wrong, it's the other way around. http://leepoint.net/notes-java/data/...rays-2D-2.html Also, I hate how they teach arrays and vectors, but they don't do much with maps or lists unless you take a data structures course. I've never once used an array or vector in my internship, it's all about the google collections API.

EDIT: I wasn't actually wrong, I just can't read. Rows first, then columns.

[Kaylia]

Now I need to make a matrix with 3 columns and k rows.

I'm used to matlab, where everything is already a matrix. Here I'll have to use arrays since it's java.

So would I say

double[][] coil = new double[3][k];

or

double[][] coil = new double[k][3];

?

It doesn't matter, as long you are coherant with the rest of your algorithm. So, just go for the usual line/column to avoid confusion.

When you use [][], you basically create an array inside every array elements. As long you understand what you're doing, and read it correctly, it doesnt matter.

That said, certain function from Math. might ask for a specific format, but I don't remember using any.

[Kaylia]

Actually, if you care about performance, you need to know which one come first (it doesnt matter for small matrice though).


For a 2-d array M[m][n], M[i] would return a single array, M[i][j] would return an element. M[nothing][j] would crash (you won't get a collumn)


If you were scanning a large matrice for example, you would need to retrieve a M[i] array first, then read every element in it.

[Woozie]

So since I need column matrices, I should use M[3][k] so that I can get column matrices returned to me when I need them by using M[0][i] to get the 0th column, for example?

[Kaylia]

So since I need column matrices, I should use M[3][k] so that I can get column matrices returned to me when I need them by using M[0][i] to get the 0th column, for example?

Well, java is starting to be far behind, and I don't have anything on this pc to test quickly. I'm not exactly sure how it would react, and your guess is probably as good as mine.


Normally, the first time you make a program that read/use array, you should use 2 "for" loop to read your matrice

For example, if you want to read the 3rd column (don't forget index start at 0)


j = 2;
double[] column = new double[n];

for ( int i=0; i<m; i++ ) // For every row from 0 to m-1
column= M[i,j]; // read the value in the big matrice and asign it to the array



[edit]
Took out the [code], since it's being homo and limit it to a tiny window

[Kaylia]

Simlarly, if you want to read the whole matrice



for ( int i=0; i<m; i++ )
for ( int j=0; i<n; i++ )
System.out.println(M[i,j]);



Your loop will read every j for a given row, then move to the next row.

[Woozie]

One more question. In matlab, I have something in place that tells me how long my program took to finish. My very fist line of code is:

ttt0 = clock;

and my very last line of code is

etime(clock, ttt0)

(the semicolon is omitted from the second one on purpose).

Is there anything similar in java?

[Kaylia]

One more question. In matlab, I have something in place that tells me how long my program took to finish. My very fist line of code is:

ttt0 = clock;

and my very last line of code is

etime(clock, ttt0)

(the semicolon is omitted from the second one on purpose).

Is there anything similar in java?

Yes, I'm pretty certain there is a way to clock the time it takes. I've no clue how to do it though, I will check online and try to find a way.

[Kaylia]

Meh, I couldnt find a simple solution using default java classes. I really hate how java is documented, I always have harder time finding what im looking for. Anyway, easiest way I found is to download their package, import it

http://www.koders.com/java/fid8EE455...67F0C44E4.aspx


package opale.tools;

msBegin = System.currentTimeMillis(); //start the time
//insert your code here//
msEnd = System.currentTimeMillis(); // end timer


System.out.println(msEnd-msBegin);


[edit]
I'm sure you could do it as easily using the default class, but i can't find how to get the time in millisecond using the clock function.

[edit2]
nvm, it's with timer

[Eliseos]

At work we use Joda Time, its relatively easy to use. http://joda-time.sourceforge.net/

A solution similar to Kaylia is to use miliseconds. It looks like the entire standard Java date class is deprecated, and they use the Calendar class. The Joda API has a lot of comparison functions to automatically convert from one type to another, like seconds to days, etc.

[Woozie]

Wow, Java is definitely faster than matlab for this. I thought it would be the other way around.

Edit: Crap, my results slightly contradict each other. Both graphs are shaped how they're supposed to be shaped so I can't even tell which one is wrong.

[Kaylia]

Wow, Java is definitely faster than matlab for this. I thought it would be the other way around.

Edit: Crap, my results slightly contradict each other. Both graphs are shaped how they're supposed to be shaped so I can't even tell which one is wrong.

You're probably reading your matrice in the wrong direction or something stupid like that. Iterator are a pain.


If you want to post your code, I can look at it (assuming i understand the maths behind it in the first place)

[Woozie]

I figured it out. It's because Java passes arrays by reference. I was doing a calculation that used the values of an array, but I wanted the original array to not be changed due to the calculation. So what I ended up doing was copying the array after passing it, and then only manipulating the copy.

Also, there was a discrepancy due to the fact that matlab goes from 1 to k, whereas java goes from 0 to k-1. So I went changed my matlab file so that the calculations use the index -1, instead of just the index. For example, my first calculation has cos in it. So matlab was giving me 1.86 something, because it was evaluating r*cos(n*1) + x0, whereas java would give r*cos(n*0) + x0 = 2. My entire calculation depends on where it starts, so starting at 1.86 and at 2 makes a noticeable difference.