Influx of non-team players?

[splitterpoint 10,986]

Of course it would be great if players were completely separated by their skill and put into separate games, but Tanki doesn't have enough active players to do that effectively.

Why does the amount of players decrease the effectiveness of the algorithm?

[Maf 32,326]

Why does the amount of players decrease the effectiveness of the algorithm?

Because with fewer players it means that at any given moment you have less people queuing for battles, so the system has less people to choose from when putting them into battles. If there were 3300 people in the queue at any given time, with 100 in every rank, then MM would be very effective, creating 6-7 battles per rank and separating players of each rank into 6 different groups based on their skill, equipment, effectiveness and any other relevant values. That way all battles would be fair and balanced.

 

But if there's only 200 people in the queue, then the system has to put multiple ranks in the same battle together and not separate them by skill and equipment at all, since if it did that, the waiting times would be much, much longer.

[Hakayosha 10]

Because with fewer players it means that at any given moment you have less people queuing for battles, so the system has less people to choose from when putting them into battles. If there were 3300 people in the queue at any given time, with 100 in every rank, then MM would be very effective, creating 6-7 battles per rank and separating players of each rank into 6 different groups based on their skill, equipment, effectiveness and any other relevant values. That way all battles would be fair and balanced.

 

But if there's only 200 people in the queue, then the system has to put multiple ranks in the same battle together and not separate them by skill and equipment at all, since if it did that, the waiting times would be much, much longer.

So, since fewer and fewer players will be playing Tanki, matchmaking will be getting progressively worse, not better...great news.  :(

[Maf 32,326]

So, since fewer and fewer players will be playing Tanki

Don't be so sure about that. With the release of mobile version there's a good chance that new players will be coming in.

[Person_Random 4,991]

Some people give up when they are losing by a big number, but they could do more than hurt their D/L.

[bozo88 450]

a group of 3

 

 

 

and

 

 

 

MM is not to blame, gutless players who can't properly and maturely express themselves and deal with change the cause.

[wolverine848 13,097]

Because with fewer players it means that at any given moment you have less people queuing for battles, so the system has less people to choose from when putting them into battles. If there were 3300 people in the queue at any given time, with 100 in every rank, then MM would be very effective, creating 6-7 battles per rank and separating players of each rank into 6 different groups based on their skill, equipment, effectiveness and any other relevant values. That way all battles would be fair and balanced.

 

But if there's only 200 people in the queue, then the system has to put multiple ranks in the same battle together and not separate them by skill and equipment at all, since if it did that, the waiting times would be much, much longer.

Would be better to have longer waiting times, or fewer players in each battle versus widening the rank brackets.

 

Nothing wrong with starting 5 vs 5 if the players are of similar "power".

 

a group of 3

 

 

 

and

 

 

 

MM is not to blame, gutless players who can't properly and maturely express themselves and deal with change the cause.

Group of 3 being the 2 sergeants + the one who brought them into the battle?

 

MM has all sorts of issues - "groups" are just one of those issues.

[Maf 32,326]

Would be better to have longer waiting times, or fewer players in each battle versus widening the rank brackets.

 

Nothing wrong with starting 5 vs 5 if the players are of similar "power".

I agree, but perhaps devs determined that the majority of players prefers shorter waiting times to balanced battles. Who knows...

[bozo88 450]

I agree, but perhaps devs determined that the majority of players prefers shorter waiting times to balanced battles. Who knows...

 

Balanced = bozo on top :)

Group of 3 being the 2 sergeants + the one who brought them into the battle?

MM has all sorts of issues - "groups" are just one of those issues.

Yes, the 2 star in a clan brought them in and abandoned the battle a few minutes after I reported him/her.

[LadyMagnum 73]

These players should be condemned to DM, until they earn 10k xp.

[bozo88 450]

Our boy y2 is back. What are the chances I would run into the same player again on my team?

[Benefactor 6,889]

I don't understand those players. Maybe they're really small children, whose parents let them play on their account for a while. Maybe they're using a really bad PC or have terrible lag and can't shoot anything. Maybe they have an actual physical or mental disability that prevents them from playing on the same level as others, but they still want to play the game and you can't blame them for it.

And they made Colonel rank with that kind of play?

[brohomie 45]

 

It's basically botting (using a bot to play the game while you are offline) and this should most likely be punished by a game Block

 

I thought about botting, but couldn't understand why I see this behavior only post matchmaking. Bots still make sense though.

 

And they made Colonel rank with that kind of play?

The Yura fellow made over 9K exp, I think they play until they are bored, and (if they are indeed botting) set the bot to run for 5 o6 hours each day. I have begun to collect data on exp/time played o these accounts, so we can guess at how long they melting each week.

[bozo88 450]

 

I thought about botting, but couldn't understand why I see this behavior only post matchmaking. Bots still make sense though.

 

The Yura fellow made over 9K exp this past week, I think they play until they are bored, and (if they are indeed botting) set the bot to run for 5 o6 hours each day. I have begun to collect data on exp/time played o these accounts, so we can guess at how long they melting each week.

 

I have few reasons for why I don't think bots. But first real quick, a reason we may be seeing an "influx of bots" is that the multi are all mostly gone, so the only thing that sticks out like a sore thumb is bots. 95% of the matches I play have no mults or lazy tankers, pre MM it was closer to 10% having mults.

 The way Y2 interacted me seemed like a fairly advanced AI (GTAV/LSPDFR traffic level) for what I assume it a basic cheap bot-algorithm. (asterisk: I have never made a bot program, nor programmed any AI capable of reacting to game environment. I helped a friend make a clicker for CoinBox Hero (not a multiplayer so no EULA to break, I may be a bozo but I ain't a criminal) and that seemed relatively unsophisticated. I am surprised that someone can make a bot where the tank interacts in a rationale manner with players and the environment in each mode, but can't have a simple turn-turret-toward-nearest-player/enemy-shoot-repeat. At risk of helping the enemy, I'll give a few examples of why I am convinced there is a human controlling Y2 directly (pull down spoiler):

 

 

-when I gave him a shove against a wall, he gave up quickly. and waited for me to stop shoving before trying to push me back. I'd expect a bot to have cruise control set, or else behave like a roomba, reversing and turning away from objects.

 

-steered towards falling drugs to use them

-randomly deployed overdrive once I could see, giving it to one other player

-steered around some mines, but not all

-interacted with ramps and bumpy terrain pretty well

-didn't immediately take off after responding.

-would self destruct after at least 15 seconds of being flipped. I was busy carrying my team so I wasn't paying strict attention to how long this jackass took to do simple tasks.

-stayed till end of match, instead of fleeing 30 second prior so as not to show up in PBR screens as doing nothing.

^^ The latter is a favorite tactic of mults, go inactive for most of the battle and then score a few kills (I've seen capable multing colonels get over 100XP in 1 minute) in the last 30 seconds, so if they are ever called to account for their behavior, they can spin some tale about sitting at base guarding the cap. Also, pre MMS, when people would rush in as the battle ending, often times they would end up higher on the team score chart than some good-faith players who didn't have time to score, making them look as if they are multing. 

 

 

 Unlike in Wargaming where simple participation gets you some reward (there was a Mingles with Jingles recently where he explained botting in WOWS) and even if you are AFK you will spot whoever kills you, thus giving you some XP, hacking this way in TO gets you nothing except a lower KDR and possible ban. (WG is lazy about banning players, esp. those who buy)

 If it is bots, then TO administration really needs to look into this, as it is worse than multing. If a few people out there are responsible for a few hundred bot accounts, TO could have a successful civil suit on it hands if it wanted to make an example of them.

edit: added spoiler due to length

Edited June 21, 2018 by bozo88

[bozo88 450]

Right now lots of clans form groups to come in and cheat everybody out of games and points. Group is a real bad idea that encourages cheating.

Not so much encourage as facilitates. I have seen plenty of groups where their was no cheating, i.e. groups of 3 colonels together.

 

Every team I get on is at least 25% mults. Tanki is that what you wanted your updates to do?

 

Is that helping your money goals?

 

I don't see 2% of the players I started with and ranked all the way up together with. Was that your goal tanki?

I don't think any game expect most people to stick around for 6 years or more. Also, people not making it to Legend 14 is their problem, not TOs.

 

Do you have any concern for the players who are telling you what is wrong with the game?

No.

 

The big buyers and cheaters rule the game.

In what manner? Prove it. The cheaters I experience once every 10 games are really no different from noob teams or gold loving maniacs, and the buyers die just like the rest of us. If anything, the clowns who flee the match seconds before they die to preserve their absurdly high KDR ruin the game.

[u812ic 1,629]

I saw an Asida, a few days ago, with very strange behavior. It never left base, but continuously went from teammate to teammate giving each one a cursory zap, even those who had just respawned. Those who needed healing, didn't get completely healed, just the cursory zap of about a second or less. It just went down the line, giving everyone a zap, went back to the beginning of the line and did it again, completely healing no one. When our base was invaded, the Asida's pattern never changed. When it respawned, it went back to the tank it was zapping and continued on down the line. My thought was that it was a bot, but I don't remember seeing bots in Tanki (maybe I did but thought they were mults). Now, from what I read here, maybe they do exist in Tanki. 

[brohomie 45]

Perhaps these players are deliberately trying to lower whatever MMS uses to detect how good you are, so they will be put in battles will unskilled players, then they can easily dominate. If I was consistently put in battles with WO3-5s who had m0s and didn't turn their turrets, it would Happy Times all over again. After all, good KDR and high efficiency provide no tangible or material benefit to you.

Edited June 23, 2018 by brohomie

[Guest]

Haven't seen any players like that...

[splitterpoint 10,986]

I saw an Asida, a few days ago, with very strange behavior. It never left base, but continuously went from teammate to teammate giving each one a cursory zap, even those who had just respawned. Those who needed healing, didn't get completely healed, just the cursory zap of about a second or less. It just went down the line, giving everyone a zap, went back to the beginning of the line and did it again, completely healing no one. When our base was invaded, the Asida's pattern never changed. When it respawned, it went back to the tank it was zapping and continued on down the line. My thought was that it was a bot, but I don't remember seeing bots in Tanki (maybe I did but thought they were mults). Now, from what I read here, maybe they do exist in Tanki.

 

That's pretty intriguing, sounds like a key-map sequence with some other intelligent software to track tanks or something. I've seen it happen on other online games aswell. Those guys had a program that ran the game itself though so the system could easily detect if you were playing the game trough a dedicated client or an unknown program.

 

I'll be on the lookout armed with my screen recorder.

[SmellThis 117]

Cheat will cheat, i expect they had friends in the other team ... yes its possible even in matchmaking