 goldorak
|
|
Total Posts: 1091 |
Joined: Nov 2004 |
|
|
I just gave it a second thought. I had forgotten that you do not always get 2s, but sometimes 4s directly.
If half are 4s, you may reduce the sum by 2^14=16384
Guys, looks like you are legit again!
|
If you are not living on the edge you are taking up too much space. |
|
|
 |
|
time not well wasted here
 |
First Commander of the USS Enterprise |
|
 |
|
Good point Goldorak. Now I think we can infer the frequency of 4s from our final scores, just to piss svisstack off :)
The downside is that we cannot prove peerless colleague's score is fake... But if it is real, it is unlikely to be from a human player. It takes ~1h to get to 4096, so it would take about 16h to reach 2^16. The problem is that there is a luck factor because of the random position in which the tiles appear, and the random appearance of 4s, so you would have to try many times in order to reach 2^16. I'd say 100 times maybe (that's a wild guess). Now, who has 1600h to burn? That's, like, an entire work year in France.
It is possible to code an algo to play, and in that case, I think the maximum possible tile is 2^17. This is what the board will look like just before you get it, it's pretty clear you cannot get a larger tile:
 |
Inflatable trader |
|
|
 |
|
No simulations of this yet? Time vs Score? C'mon, guys, you're not wasting enough time on this. |
|
|
 |
|
This is actually a good source of interview questions, I will definitely use it ... |
Inflatable trader |
|
|
 |
 goldorak
|
|
Total Posts: 1091 |
Joined: Nov 2004 |
|
|
@Svisstack
May be that is the difference between students and experienced people: the second group know where their time is well spent.
My skydiving day went pretty well. Thx.
|
If you are not living on the edge you are taking up too much space. |
|
 |
 Cheng
|
|
Total Posts: 2870 |
Joined: Feb 2005 |
|
|
A little follow up:
Depending on the code the relation of 2s and 4s varies but I think my implementation gives about 90% 2s and 10% 4s. Now go and adjust for the probabilities to see that I am an honest broker . |
"Don't try to run there's no way to survive / A vicious surrender I'll eat you alive / No one is safe until I get my fill / All hear my warning I'm king of the kill"
|
|
|
 |
 goldorak
|
|
Total Posts: 1091 |
Joined: Nov 2004 |
|
|
.1?
.1%, I mean.
|
If you are not living on the edge you are taking up too much space. |
|
 |
.gif) benji
|
|
Total Posts: 197 |
Joined: Feb 2005 |
|
|
I must be way behind the cool kids on that one ...
agar.io |
|
|
|
 |
 goldorak
|
|
Total Posts: 1091 |
Joined: Nov 2004 |
|
|
For the ones still interested in 2048.
2048 |
If you are not living on the edge you are taking up too much space. |
|
 |
|
Carousel shaping seems to be the right method when doing research on playing games with ML :-) |
|
|
|
 |
 NIP247
|
|
Total Posts: 560 |
Joined: Feb 2005 |
|
|
So agar.io copycat slither.io is apparently raking in 100k$/day ? |
On your straddle, done on the puts, working the calls... |
|
 |
|
Valorant is one of the best games to play when you need to pass the time and improve your gaming skills. Of course, practice makes perfect and even if you master basic techniques and strategies it's sometimes hard to take that extra step forward. That's where Valorant boosting comes in handy. Boosting can provide an invaluable opportunity to learn from professional Valorant players who have mastered their craft, as they will help lift your gameplay up skills to higher levels, bringing your win-rate up with no effort on your part. This way you can end up enjoying both a leisurely game session and a beneficial boost in skill level at the same time! I was reading an interesting article on this site earlier https://www.zupyak.com/p/3585888/t/valorant-boosting-a-comprehensive-guide-to-choosing-the-right-service |
|
|
|
 |