Alpha 5.3(.0)

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#41 » 07 Dec 2016, 16:01

pier4r wrote:
Ehm actually the leaderboard did update. But only counting my defeats. In the meanwhile I had 2 draws and 3 victories but those are not reflected in the leaderboard. This would create quite a bad mood for players.


Further proof that this happens. I'm losing against the same guy of league2 5 times in a row. First I get always the same amount of minus points (-11) and this cannot be because his score should grow every time and I should lose less and less. Second which K factor is used if me from league1 is against one from league2? Third, in the leaderboard I still have 10 games played when I played way more.
Last edited by pier4r on 07 Dec 2016, 16:13, edited 1 time in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#42 » 07 Dec 2016, 16:12

Christian wrote:Imagine a fresh player dipping into multiplayer. First match vs second league player using shotgun. The handicap should be the other way around when matching like that.


Yup it is quite frustrating. I'm testing so there is no problem, but the majority just quits. The idea of L-1 and L+1 is not bad but it requires either a lot of player playing, so the punishment is distributed, or protection on some maps when only with poor AI one can lose with more special bots.

So either is a long quest to find balance, or the some matchmaking rules should be dropped.

I would suggest, by feeling (one should measure to get better results):
  • get the 100 "oldest" (ordered by creation time) available matches with opponents in leagues between L-1 and L+1 (L being player's league)
  • variant1: avoid league1 to be included
  • variant2: avoid league1 to be included on certain maps with low number of bots
  • variant3: avoid l-1, l+1 on certain maps with low numbers of bots
  • variant4: avoid l-1, l+1 altogether
  • get the best from these matches using these comparison rules:
    • match with an opponent in the closest league (compared to player's league) is the best
    • in case of equality: match with an opponent using a different IP is the best
    • in case of equality: match with an opponent never met or met the furthest from now is the best (using a 1 hour tolerance)
    • in case of equality: match with an opponent with the closest ELO score (compared to player's ELO score) is the best (using a 100 points tolerance)
    • in case of equality: match with a creation time the furthest from now is the best

If no match was found, create a new one using the existing rules (random map, random left/right team).


On varian4: the fact that match can be spammed and the random maps is good to have less waiting time compared to alpha 5.2. If one does not want to be "used as source of points", then he will just spam 1-2 matches not more. The ones that are confident will spam more games. The alpha 4.3 (yes four) rhythm was nice, moreover the additional matchmaking rules will avoid to be matched always against the same guy as soon as other players will play in the same moment.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

ConsciouS-0nE
Autonomous Entity
Autonomous Entity
Posts: 740

Re: Alpha 5.3 Release Candidate

Post#43 » 07 Dec 2016, 16:52

:( :o :shock: :roll: :!: :?:

Gfx Is the matchmaking final?

You did 5.2 random maps for apparent 'reasonS' from 4.x,
But now the 5.3's enemy matches is not far different from 4.x.


1. For example if i match mgblitz for the 2nd time, the possibility of matching him 8 more times is really high.
Screenshot_2016-12-07-23-37-29.png
8th time, didnt even change in 4hours.
Screenshot_2016-12-07-23-37-29.png (127.2 KiB) Viewed 1368 times


Further, contruction of anti-enemy ai program because you already know whats up.
Its not exploitation but that was the problem before and got revived.
And alot more.


Or this will change when players will start to crowd up?

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#44 » 07 Dec 2016, 17:31

I guess will change when not only 10 players are playing. I like the quick pace, and if one does not want to be exploited does not leave a lot of pending matches. In you case blitz is just winning so he is spamming the matches.

The point is, why does your score change (the more you lose, the less points you lose) and mine not? The score should be a function in the database not depending on android version and stuff. After hours I still have the same 10 games played, while I did at least 20.

edit: I just got my 11th game registered! Yay!

edi2: nothing to do, when I play with this SunwayTaihuLight games are not registered. Lost, -12 points, I still have 11 games on the leaderboard.

edit3: played again, again -12, again only 11 games registered.

edit4: played again, again -12, again only 11 games registered. This time split teams. Maybe games are not registered when done across the leagues?
Last edited by pier4r on 07 Dec 2016, 17:43, edited 3 times in total.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

ConsciouS-0nE
Autonomous Entity
Autonomous Entity
Posts: 740

Re: Alpha 5.3 Release Candidate

Post#45 » 07 Dec 2016, 17:36

Nah i sent damp bots so i can take a peak on league 3,
I stopped spamming matches weeks ago.

That on blitz 1game for 5 mins,five in total.
After 4 hours deployed 3 more games.
Im not the best player, but atleast im something.I'm 8
Science is madness! 3:)

ntk
Algorithm
Algorithm
Posts: 75

Re: Alpha 5.3 Release Candidate

Post#46 » 07 Dec 2016, 20:31

Gost gost gost game :mrgreen:

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#47 » 07 Dec 2016, 20:34

For the bug that I reported, that my score oes not update, I'm 90% positive that is due games done between two different leagues.

Is anyone playing against people in different leagues? Could this person check if his/her score and number of games get updated accordingly?
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

mcompany
Autonomous Entity
Autonomous Entity
Posts: 872

Re: Alpha 5.3 Release Candidate

Post#48 » 08 Dec 2016, 04:24

Idk, but I feel like shotguns are now mostly useless because you can't swarm the enemy with them, and they can't effectively do a resource rush on any map except kingmaker.

Also, I'm finally back, and now I have snipers, so clunkymonkey can't just steal all of my score. Although I will say, even though it is a forum released version, it still is ridiculous when I play against the same person 8+ times in a row

User avatar
Chunkymonkey
Automaton
Automaton
Posts: 139

Re: Alpha 5.3 Release Candidate

Post#49 » 08 Dec 2016, 07:30

mcompany wrote:Idk, but I feel like shotguns are now mostly useless because you can't swarm the enemy with them, and they can't effectively do a resource rush on any map except kingmaker.

Also, I'm finally back, and now I have snipers, so clunkymonkey can't just steal all of my score. Although I will say, even though it is a forum released version, it still is ridiculous when I play against the same person 8+ times in a row


:oops: Yeah... Thanks for the points... I setup 10+ games to try and even up my karma.

I have to agree with the shotgun comment. Lone shotgun feels like a sacrificial piece. (Ntk has found a use for it though... Against me at any rate.)

Christian
Algorithm
Algorithm
Posts: 56

Re: Alpha 5.3 Release Candidate

Post#50 » 08 Dec 2016, 08:56

Haven´t tried it yet, but I would use a single shotgun supposedly as kind of a tank. Soaking up some damage and using it´s speed to get out of danger and regenerate. Just need to think of ways how to punish enemies who chose to ignore the shotgun. Recognizing that is going to be the hard part given the testable conditions.

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#51 » 08 Dec 2016, 09:13

Yes. The guy in second league Sunwaysomething is using it as damage absorber. A shotgun that absorb damage is quite helpful, for example in barred spiral.

Did anyone of you went to league3 and played against league4? How was the experience? With a sniper less did you felt handicapped?

For instance, a shotgun that absorb damage from assaults bots is totally a game changer, either the assaults ignore it, but then they get additional damage for free, or they try to damage it but the shotgun is simply immortal. In some test I need a 3 vs 1 to take down a shotgun that retreats after getting a bit of damage.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 5.3 Release Candidate

Post#52 » 08 Dec 2016, 09:17

mcompany wrote: Although I will say, even though it is a forum released version, it still is ridiculous when I play against the same person 8+ times in a row


No it is not. If you are just an handful of players, there is no way could be different. For example with a limit of 1h one would wait (between 2 players) 8 hours to do 8 matches, I would prefer way more how it is now. Because if there are a lot of active players the pairing is random, otherwise it is mostly 1 vs 1 according to the timezone. One can carefully play matches and modify the AI accordingly without just spamming around (of course, who wins will just spam around because, ehy, points).

In this case nullpointer and the others in timezones with less players (EU, asia) could play a bit faster.

Side note, from the 5.2 one can see the new scores (so the score are merged). For example bs314 is very high now. 1473 points, 5th.
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

MGBlitz81
Automaton
Automaton
Posts: 135

Re: Alpha 5.3 Release Candidate

Post#53 » 08 Dec 2016, 09:29

So far, I like it! But we need to seriously get some players here. I have probably played about 50 or 60 games so far and there have been only 5 or 6 different players I've encountered. Mcompany whooped me pretty good this last round though.

I'll probably give a more detailed analysis later but I'm getting sleepy.

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 5.3(.0) Release Candidate

Post#54 » 08 Dec 2016, 12:03

5.3.1 is available >
Locking this thread.

Return to “Release candidates”

Who is online

Users browsing this forum: No registered users and 0 guests