Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[RulerOfNothing]

So what is the difference between Node_Base, Node and Seq_Node? Also you can implicitly cast pointers-to-derived-objects to pointers-to-base-objects.

[Anvilfolk]

If you don't use too many classes, you could also have a method like isSeqNode() and isBaseNode(), and then static_cast based on what they return?

[TolyK]

I think I really took the purpose separation thing too far.

Node_Base has the data and operations for making/editing/destroying trees (a Parent pointer, an array of Children pointers, along with helpful operations).
Node holds values for each player (in a sequential game) and functions for finding the Nash equilibrium of a subtree.
Seq_Node is derived from Node and also has a template interface (to the "definition" of the game). It holds a (game) move, as well as an interface to a payoff function, as well as other stuff. It's basically the most complicated class.

If you don't use too many classes, you could also have a method like isSeqNode() and isBaseNode(), and then static_cast based on what they return?

I... don't think so.

I think my best bet is just rewriting them all as one class.

[MorleyDev]

It's my experience that in almost all circumstances, having to use dynamic_cast is just a symptom of a flaw in the overall design. It's a code smell, an indicated that aspects of the design need reconsidering because in all likelihood: There's a cleaner way.

[TolyK]

Putting it all in one class improved performance by ~10%. Whoo!

[Nadaka]

What is the best way to write code in a room with a jackhammer and gasoline fumes?

[cerapa]

Drunk.
Somewhere else.
Wearing a gas mask and headphones.
Wearing the skin of the person with the jackhammer.

Don't know any other methods sadly.

[TolyK]

Of course, you could ask the jackhammer guy to do it on your keyboard, it would make for some interesting code...

And anyways, a True Coder^TM hears only the tap-tappin' of the keyboard and the rotting of food in his keyboard.

[Siquo]

I'm learning LISP while on my honeymoon. Does that count towards my True Coder^TM membership?

[Anvilfolk]

What is the best way to write code in a room with a jackhammer and gasoline fumes?

I'm guessing emacs has a mode for that?

[malimbar04]

Fizzbuzz!

I was reading this: http://www.zoharbabin.com/which-fizzbuzz-solution-is-the-most-efficient

2 questions:
how do they determine speed of a function? More importantly, how do I do this? Not sure if it matters, but I'm using C++ and Ubuntu, using command line and a basic text editor only.

How can it be that bulding a string step by step is faster than using else-if solution? It would do lots of checks for every number, where an else-if could skip over a few if it matches the first case.

[Skyrunner]

To determine the speed of a function, you run it over and over and over then divide the elapsed time by the number of times you've ran it.
You can also guess or roughly get a feel by looking at how it does what you're doing.

if/else being slower than building a string step by step probably has to do with processor trickery. >_>

Someone else probably has a better answer.

[Anvilfolk]

You might be able to find a "profiler", which essentially tells you how much time your code spends in each function. It's pretty nifty to figure out where you need to optimise.

I didn't read the problem through, but it might be that if you're using Strings as identifiers (i.e., you never ever change them), the compiler might be smart enough to convert them to integer identifiers or  so.

[JanusTwoface]

how do they determine speed of a function? More importantly, how do I do this? Not sure if it matters, but I'm using C++ and Ubuntu, using command line and a basic text editor only.

If you want very basic timing in Linux, you can use the time command. Basically:

Code: [Select]

$ time ./some-program
./some-program  0.25s user 0.07s system 16% cpu 1.907 total

You're going to get a lot of noise in that, but for basic timing tests, it's good enough. And you don't have to modify your actual program. And and it works on any type of program, not just C++ or whichever.

How can it be that bulding a string step by step is faster than using else-if solution? It would do lots of checks for every number, where an else-if could skip over a few if it matches the first case.

Mostly: Branch prediction. In most cases, avoiding comparisons will speed up computation. You can get the same benefits if the compiler can guess which way it's going to go, which compilers are getting pretty good at these days.

[Lost in Nowhere]

OK... I'm trying to load an image from the code directory in Java. Nothing that I have tried so far has worked.
Here is my current code:

Code: [Select]

              //Returns the index of the image in imageStorage
public int addImage(String path) {
int index=imageStorage.size();
BufferedImage img=null;
try {
    img = ImageIO.read(new File(Garden.directory+"\\"+path));
} catch (IOException e)  {
System.out.println(e.toString());
index=-1;
if(Garden.abortOnErrors) {
System.exit(0);
}
}
imageStorage.add(img);
return index;
}

Code: [Select]

private LinkedList<BufferedImage> imageStorage=new LinkedList<BufferedImage>();

Code: [Select]

public static String directory=System.getProperty("user.dir")+"\\bin";So, what is wrong with this?
e - If it matters, I'm using Eclipse.