Recorded best scores.

pier4r
Skynet
Skynet
Posts: 3319

Recorded best scores.

Post#1 » 19 Nov 2016, 21:33

Update, recorded best scores:
See post and attachments in the thread for scores up to 2000+

Nullpointer 1866 viewtopic.php?f=7&t=60&start=100#p1976

Nullpointer 1818 viewtopic.php?f=7&t=60&p=1269#p1268

Nullpointer 1800 with screenshot viewtopic.php?f=7&t=60&p=1262#p1261

Nullpointer 1794, 513 games, 401 victories. viewtopic.php?f=7&t=60&p=1260#p1260

Nullpointer 1784, 508 games, 397 victories. viewtopic.php?f=7&t=60&p=1260#p1258

Nullpointer 1777, 507 games, 396 victories. viewtopic.php?f=7&t=60&p=1257#p1257

Nullpointer 1761, 393 wins, 504 games. viewtopic.php?f=7&t=60&start=40#p1256

1740 from nullpointer! (497 games played and 387 victories). viewtopic.php?f=7&t=60&p=1257#p1250

1724 nullpointer viewtopic.php?f=7&t=60&start=20#p1236

1718 pier4r nvidia k1 viewtopic.php?f=7&t=60&start=20#p1196

#################
After some days of domination, someone went ahead of nullpointer score (although for just 3 points).

I was thinking that either nullpointer would have lost the interested in the game (and then being hidden after 7 days) or he would have been at the top until more people would have joined the game.

Currently ntk, lord valcor and ritter runker are ahead of nullpointer.

Another interesting fact is that yesterday the scores between 1400 and 1600 had not so big gaps, instead now the 7th, drinaus, has 1441 while the 6th, johnbob, has 1553. A gap of 100+ points (likely this happens when at the top some players do not play and keep the score).

Side note: some active players (otherwise out of the leaderboard) are really active.

ntk 794 games, ritter runkel 836, maaku 868, etc. I wonder were do you get all those games. When I try to spam more than 15 games I get quickly 5 games in pending and I have to wait enough to focus the interest elsewhere.

Anyway, I have to polish some decision trees for some maps (mind game, meeting point) while for other maps what is my Achilles heel is the bot selection (due to rock, paper, scissor behavior that can be limited with options but i'm lazy), and then I will try to get the 1500 wall as well. So far I get 14xx and then i drop to 12xx due to new tactics or rock/paper/scissors conditions.
Last edited by pier4r on 06 Feb 2017, 12:38, edited 10 times in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

ntk
Algorithm
Algorithm
Posts: 75

Re: After some days of domination, someone went ahead of nullpointer score

Post#2 » 19 Nov 2016, 22:31

When u got up on score is hard

Example me: On win on player with low score got 5 points, on draw lose 4, on lose lost 20 points.

The true problem of nullpointer is he didn't want lose the top of score, he play litle games (low number of games played). He didn't know the new strategies of game.

When he lose the top of rank he wanted get back and play and lose...

Is hard be on top and play ritter got 1660.
If u lose a game u can and lost 30 points.

It can be wrong, but i think it

ntk
Algorithm
Algorithm
Posts: 75

Re: After some days of domination, someone went ahead of nullpointer score

Post#3 » 19 Nov 2016, 22:40

The trick of go up is never lose

I got more problem on circle for many games, now never lose on it

Then mind now is my nightmare but can draft

The other maps win easy, not al games :geek:


For play more games use smartphone. Enter on game play exit do something then enter again.. like facebook

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#4 » 19 Nov 2016, 23:22

These last couple of days are being tough, a lot of awesome AIs. I like how competitive this is becoming.

Valcor, Ritter, NTK and me have been in 1st today :D
I'm back to the 1st position now, but I may ALT+Tab and find out this is not the case anymore lol.

NTK, I have 680 games with my 2 accounts, I'm not avoiding playing the game to keep the 1st place, I'm playing everyday, including against yourself.

I may have to play less though and I'm struggling to find time to design/tweak so many AIs for so many maps, I hope the 2 new maps will replace other 2 instead of just being added.

User avatar
Ritter Runkel
Neural Network
Neural Network
Posts: 498

Re: After some days of domination, someone went ahead of nullpointer score

Post#5 » 19 Nov 2016, 23:38

I like the fast changes on top. One lose and you are down some places. And every time I think that my bots are pretty cool, someone of you has a better idea and toasts me.

Amount of games is a due to often checking, place five games and check again one hour later. Would be cool if the next update brings a possibility to play against someone only for fun, not for points.

Johnbob
Algorithm
Algorithm
Posts: 62

Re: After some days of domination, someone went ahead of nullpointer score

Post#6 » 20 Nov 2016, 00:17

Like ntk, my nightmare was 'mind game' today and i just spent something like 2 hours finding a solution, i'm pretty satisfied with it but it needs to be tested 'on the field' now :-)

Also today i adopted the shotgun technique for 'meeting point': really fun ! And i had to improve a little for circle of death. I will try to go first place but i feel like i'm behind regarding the snipers evading techniques in front of shotguns.

ConsciouS-0nE
Autonomous Entity
Autonomous Entity
Posts: 740

Re: After some days of domination, someone went ahead of nullpointer score

Post#7 » 20 Nov 2016, 03:07

Im just sitting here..
Laughing at my own games ahaha
Im not the best player, but atleast im something.I'm 8
Science is madness! 3:)

MGBlitz81
Automaton
Automaton
Posts: 135

Re: After some days of domination, someone went ahead of nullpointer score

Post#8 » 20 Nov 2016, 03:35

I'm pretty much out until we get an update. Not fun playing and adjusting an AI every day because of the limitations of the bots strengths and rock, paper, scissors gameplay mechanics.

ntk
Algorithm
Algorithm
Posts: 75

Re: After some days of domination, someone went ahead of nullpointer score

Post#9 » 20 Nov 2016, 05:51

MGBlitz81 wrote:I'm pretty much out until we get an update. Not fun playing and adjusting an AI every day because of the limitations of the bots strengths and rock, paper, scissors gameplay mechanics.

I played and go up with a buging AI in meeting point (see need to adjuest every game) after i lose on it 4 games vs the top 7 players i understoud where is the strange bug in it.
But the strange or not is that they use the same AI or more like. I see new ideas only from.player of the middle or low rank.

First rules: don't lose
It's hard, in some map i play with 3 AI, changing and changing create bugs in AI and u can lose VS easy AI.

MGBlitz81
Automaton
Automaton
Posts: 135

Re: After some days of domination, someone went ahead of nullpointer score

Post#10 » 20 Nov 2016, 09:59

it's not a matter of giving up, hehe. I just don't want to fight for top ranks with ever changing AI matches at close and long range. I'll leave that to you guys. I'm patient, I'll wait for the next update before I adjust my AI anymore.

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#11 » 20 Nov 2016, 10:50

ntk wrote:It's hard, in some map i play with 3 AI, changing and changing create bugs in AI and u can lose VS easy AI.


Totally agree, using a lot of AIs in the same map increases the chances of "bugs", you have to lose a few matches before you spot all of them.

ntk
Algorithm
Algorithm
Posts: 75

Re: After some days of domination, someone went ahead of nullpointer score

Post#12 » 20 Nov 2016, 11:15

For this a said the game don't need complex rules o more maps, now it's complex manage AI for more maps. I think need to help player in 'debug' of new AI: i hate after testing an AI vs other AI restore the sequence of AI for all the bots.
On Pc i dunno, on smarthone 5 inchs is not easy with small caracters.

The game need a debug mode and release mode. It is useless show all type of AI name when u play a map: u rarely use AI for a map on another.
And need a list groupping of AI for select easy before a match

It's not a must but think on it

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#13 » 20 Nov 2016, 11:18

MGBlitz81 wrote:I'll wait for the next update before I adjust my AI anymore.


The problem is that there's always gonna be another update. Gameplay and balancing takes a good portion of the development time and I can predict there will be changes on that in every single release.

What would be interesting, for me at least, is reducing the Ranked games to 4 maps only, otherwise it's too much work for us to build or rebuild so many AIs. We have 6 maps at the moment with 2 more coming :(

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#14 » 20 Nov 2016, 11:33

Tough games

Image

pier4r
Skynet
Skynet
Posts: 3319

Re: After some days of domination, someone went ahead of nullpointer score

Post#15 » 20 Nov 2016, 12:14

ntk wrote:Example me: On win on player with low score got 5 points, on draw lose 4, on lose lost 20 points.


That's how elo works and I love it. For example in many games players are obsessed by the win ratio, but if one plays only (even if just y chance) weaker opponents, that's easy.

The point is: either one always wins against weaker opponents, or the first weaker opponent will take away a lot of points. For this comes in handy the K factor, that GFX reduced for the 4th league. The K factor (in the elo formula) says more or less "how quickly the score should adapt to the new situation. Actually makes not so much sense to variate it a lot when one played a lot, so it reduced to allow slowed changes.

Surely the elo is not a predicting index 100% of the time, but statistics shows that works well 75% of the times and that's quite enough. You can see that people apply it successfully also in places with more than one players like football teams.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3319

Re: After some days of domination, someone went ahead of nullpointer score

Post#16 » 20 Nov 2016, 12:18

MGBlitz81 wrote:it's not a matter of giving up, hehe. I just don't want to fight for top ranks with ever changing AI matches at close and long range. I'll leave that to you guys. I'm patient, I'll wait for the next update before I adjust my AI anymore.


Yes that's also a point, especially if the next update will bring more assaults bot. It means the AI will need to be reworked, so one can just play for the moment.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3319

Re: After some days of domination, someone went ahead of nullpointer score

Post#17 » 20 Nov 2016, 12:20

NullPointer wrote:Tough games

Image


wait a second, how could you play 2 matches against the same player? as GFX pointed out this should not be the case (you should wait roughly one hour).
Last edited by pier4r on 20 Nov 2016, 20:19, edited 1 time in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#18 » 20 Nov 2016, 13:24

Dunno, has the matchmaking algo ever been explained?

Do we want that at all? I can play at most 5 matches per hour already.

pier4r
Skynet
Skynet
Posts: 3319

Re: After some days of domination, someone went ahead of nullpointer score

Post#19 » 20 Nov 2016, 13:58

NullPointer wrote:Dunno, has the matchmaking algo ever been explained?

Do we want that at all? I can play at most 5 matches per hour already.

https://www.reddit.com/r/Gladiabots/wiki/index

in particular: viewtopic.php?p=82#p82

I like the idea (it avoids to be 'picked' by someone that was able to break your tactic) but therefore I wonder how people can accumulate 800 games. I can barely play 20-30 per day and then I have to wait.
Last edited by pier4r on 20 Nov 2016, 14:06, edited 1 time in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

User avatar
NullPointer
Autonomous Entity
Autonomous Entity
Posts: 539

Re: After some days of domination, someone went ahead of nullpointer score

Post#20 » 20 Nov 2016, 14:02

Cool, thanks for that.

This list has games from today and yesterday, so I guess the 1-hour rule still applies.

Return to “General Discussions”

Who is online

Users browsing this forum: No registered users and 1 guest