LabVIEW
What do you think about this snake game code?
Solved!
Hi alhebu,
@alhebu wrote:
How can I improve it? Wait for your reply
Why is there a 1000ms wait function in parallel to your event structure with a 100ms timeout event?
Your "tablero 2" arrays shows the result of the last iteration instead of showing the result of the current iteration…
Why do all your controls have a " 2" in their label?
Great first effort! Congratulations! 😄
Now try to redo it with cleaner and less code. Some ideas:
- You don't need an event structure because the loop is spinning anyway and you can just poll the controls.
- Your inner loop sounds very dangerous, for example if all fields are green, it would never stop. (unlikely, but still something to keep in mind)
- Your random init coordinates are not fair because not all values occur a the same probability
- Try to operate on the board "in place". There is no need to delete and insert.
- Instead of buttons, try to read e.g. the cursor keys.
- Create some subVIs for simple operations. The big code on the right could be a subVI with cluster in, cluster out, and boolean out (stop?). Same for the init code.
- Make your cluster a typedef to make changes easier later, especially once you create subVIs.
- Don't stop the code at the end. Just wait for a reset to start over.
- ...
See ow far you get!
Hi alhebu,
@alhebu wrote:
Thank you for your reply! 😄
1.But if I would like to use the cursor keys I would use an event structure right?
2.Okay so I would have to write some code in order to get to a winning message if there´s no enough space for a red field right? Or why is it risky? Should I change this part?
3. I don´t understand... could you explain why not ocurr at the same probability?
4. Okay, I understand, LabView is slower if I insert and delete it many times right? Or just for do a cleaner code?
5. ...
6, 7, 8. I´ll do it thank you!
Thank you for taking some time to check my code and I would be grateful if you help me with my questions 😄
1. Yes.
2. It's risky because your code may end up in a loop running forever: bad UX for the players…
3. Because the RandomNumber function outputs the numbers within a certain range. And there are certain rounding rules applied when converting from DBL to I32…
4. Cleaner code (most often) results in less buggy code: why do you even need to delete/insert when you can replace?
Ideas on "simpler code":
Why do you need 2 comparisons and an OR function to check the postion when there is a CoerceAndInRange function doing the same operation?
Why do you need to duplicate that check for the other coordinate? (To reuse code you either use subVIs or a loop…)
This is how my Snake game looks like:
It uses an event structure to catch key down events, like cursor keys…
@alhebu wrote:
3. I don´t understand... could you explain why not ocurr at the same probability?
Because you are rounding to the nearest index, so e.g. 0 occurs for values between 0 and 0.5, while 1 occurs for values between 0.5 and 1.5, a range twice as wide! Correct is to multiply by 11 and round down towards the next lowest integer. Now 0 occurs for values 0..1, 1 occurs for values 1..2, etc. All ranges have the same width! Note that the random number is never exactly 0 or 1, so you don't need to worry about rare edge effects.
Here's a quick demo. Note that since we are in 2D, edges get half and corners only a quarter with your code. You'll never get a job working for a Vegas Casino if your random numbers are biased. 😄
Gerd already answered most points. Here's some additional info:
- You don't need to use an event structure to read keys. You can also just read the keyboard. 😄
- An event structure is actually problematic because events break the timeout wait and if you use external timing instead, events can accumulate in the event queue unless you take precautions.
- Of course you should not call it a snake game, because it is not a snake, just a flea that tries to avoid corners and the red square. 😄 (A real snake game is one of the very early video games and forms a snake that grows in length)
- I answered the probability issue in detail above.
