why isnt there a killer tornie like 10 people on each table then the final is 3 people it would be cool

Wrong part of the forum mate

Should be in Game Queries.

It's been mentioned plenty of times before though, so it doesn't look like they will be introduced.

though they would be awesome

I agree, it would be a fresh new aspect to the game, but the tournament schedule might be a little, cramped?

Well there is 80-40 people to each tourney would only be 8-4 games

I mean with the amount of games, there's already 4 game types within the tournament schedule (8US/UK/9Ball/Straight.)

So adding another one might be too many, thus there would have to be less of each of the other ones to accomodate the killer tournaments.

Set max players to 100 (first come first served).

First round you run as many parallel games as it takes, 10 max. Would need a simple algorithm to avoid 81 players = 8x10 players + 1 bye (rather run 9 slightly smaller games)

Second round is simple the winners from round 1 like already suggested, 3-10 players depending on the number of games were played in round 1.

It's not really that hard to come up with several possible schemes for a killer tourney. The problem is that it's going to be nothing alike to the existing ones - meaning an entirely new system must be coded. That takes time and effort, and Nick has been very busy with other enhancements lately. I hope we will see a killer tourney some day, when Nick gets the time to design and implement it.

I like how that set up sounds actually, Jan.

janmb: What if 82 joined?

8x10 players, 2 byes?

yea i discussed tht wid nick ages ago

As I said, you want to avoid byes - they are not necessary.

The simple rule could be to run number of games = (players-1)/10 + 1, then distribute players evenly over those games.

In short, run as many games as you need (in your example 9 games, 8 of them having 9 players, 1 having 10).

yr

If the aim is to have as many players in the final as possible, and bearing in mind that 3 is the minimum number of players per game, we have:

Number of players in tournament - assortment of players in round 1
3-10 - no first round; all players directly into final
11 - 2 games of 4, 1 game of 3
12 - 4 games of 3
13 - 3 of 4, 1 of 3
14 - 2 of 4, 2 of 3
15 - 5 of 3
16 - 4 of 3, 1 of 4
17 - 3 of 3, 2 of 4
18 - 6 of 3
19 - 5 of 3, 1 of 4
20 - 4 of 3, 2 of 4
21 - 7 of 3
22 - 6 of 3, 1 of 4
23 - 5 of 3, 2 of 4
24 - 8 of 3
25 - 7 of 3, 1 of 4
26 - 6 of 3, 2 of 4
27 - 9 of 3
28 - 8 of 3, 1 of 4
29 - 7 of 3, 2 of 4

30 - 10 of 3
31 - 9 of 3, 1 of 4
32 - 8 of 3, 2 of 4
...
39 - 1 of 3, 9 of 4
40 - 10 of 4
41 - 9 of 4, 1 of 5
42 - 8 of 4, 2 of 5
etc

98 - 2 games of 9, 8 games of 10
99 - 1 game of 9, 9 games of 10
100 - 10 games of 10

After 100, it gets interesting...

Now, the other thing is this. Take 90 players for instance. Which is better, 10 games of 9 in round 1 and the 10 winners in the final, OR 30 games of 3 in round 1, the 30 winners going into 10 games of 3 in round 2, and then the 10 players going into the final? In other words, a big final is nice, and probably most desirable for most people, but...

...the rounds leading up to it could (and maybe should) be as small as possible. This can be possible if one can reduce the number of players from whatever it is to a number between 11 and 30 (if I've thought that out right), and then applying one of the above solutions to is. e.g.

Round 1: 46 players: 15 games (14 games of 3, 1 game of 4) to give 15 winners
Round 2: 15 players: 5 games of 3
Final: 5 players

Alternatively...
Round 1: 46 players: 3 games of 3 to produce 3 winners AND 37 BYES
Round 2: 40 players: 10 games of 4
Final: 10 players
(produces largest final possible)

Alternatively...
Round 1: 46 players: 8 games of 3 to produce 8 winners AND 22 BYES
Round 2: 30 players: 30 games of 3
Final: 10 players
(produces largest final possible and smaller, shorter Round 2 games)

There are various compositions possible: the last one I mentioned would involve the most players all the time with the shortest-handed games.

Opinions, folks?

Good point - rather than maximizing players in 1st round matches, you should maximize the number of games, up to 10 - leaving as large game as possible for the final.

As for the game setup for first round, you could easily "formulize" that. Or a simple table like you are outlining will work too... Really not hard to implement at all.

I don't remember much java, but I'll happily draft the algorithm for this if wanted.

What about as small a final as possible? I.e. concetrate each round into a power of 3?

Example: 50 players
Round 1: 50 players: 10 games of 3, 1 game of 4, 16 byes
Round 2: 27 players, 9 x 3
Semi-finals: 9 players, 3 x 3
FInal: 3 players

That;s 3 to the power of something clooney

To the power of 3 would be: 1, 8, 27, 64 etc

Think you want 3, 9, 27, 81

Ah - hung on a comma, as they say... or not!!!

1, 8, 27, 64 are all numbers to the power of 3, agreed. But the definition of a power of 3 is, confusingly, 3 to the power of something, i.e. 3, 9, 21, 81, etc like you said.

A demonstration of my explanation is provided on the Wikipedia page for powers: "In the base ten (decimal) number system, integer powers of 10 are written as the digit 1 followed or preceded by a number of zeroes determined by the sign and magnitude of the exponent. For example, 10^3 = 1000 and 10^−4 = 0.0001." (i.e. 1, 10, 100, 1000, 0.1, 0.01, etc) Otherwise, it would say 1, 1024, 59049, 1048580, 9765625, 60466176, etc.

http://en.wikipedia.org/wiki/Exponentiation#Powers_of_ten

As big final is better imo, but as small as possible is easy too... Just a matter of preference really.

In the interest of keep people waiting as little time as possible though, smaller first round games and a larger final is better I think.

(and warney is right, "the power of N" has no meaning, you have to supply both the base value and the exponent)

And you can easily avoid byes in this format (and then you should imo)

Edited at 10:08 Sat 18/04/09 (BST)