What is "point dumping" bad? My thoughts:

% If someone has only one point and wins an 8-player game against
1000-score opponents, they get 800 points max, because of the
100-points-per-player limitation.

% Once the dumper gets 800 points, winning more games doesn't really
help him that ...

Are random invites not allowed, just annoying, or are most people OK w/ them?

I often create games and have people way below my point level join
them (beneficial to neither of us), so instead thought I'd start
inviting/reserving games for people closer to my point level (using the
leaderboard ...

I'm in

I'm in!

The odds against that are only ~5563 to 1 against, and, given the number of dice rolls on conquerclub, that's not too impressive.

Losing 100 vs 2 is unlikely enough to be suspicious.

$d1 >= $a1, don't you want to do @d[0] >= @a[0] (or something like that)?
(I don't know perl, so I'm guessing a bit)

This is it. I guarantee he's banging his head on the desk as he reads it.

BANG BANG BANG! OK, I corrected my error and got the accepted results.

I now officially suck at both ...

sherkaner wrote:$d1 >= $a1, don't you want to do @d[0] >= @a[0] (or something like that)?
(I don't know perl, so I'm guessing a bit)

ARGH! That might be it. I'm using variables called $a1, and $a[1]... that's just bad programming

I wrote a simple Perl script to compute "intensity cube" (dice) odds,
but my results differ from the generally accepted odds. Could someone
take a quick look at:

http://github.com/barrycarter/bcapps/blob/master/bc-conquerclub-cubes.pl

and see what I'm doing wrong? When I do this:

bc ...

Dako wrote:Yes, but it doesn't show armies numbers. But it shows the progress of the game.

Right. I think the original poster already realized that you can't get exact troop numbers, since the logs don't have enough information to reconstruct that data.

Probability of that happening is 1.18571*10^-26

well then there is a probability

don't forget to take into account that the odds are stacked in the defender's favour.

Well, there's ALWAYS a probability. However, ~10^-26 is way too low a probability to occur by chance (at least not consistently ...

hairy potter wrote:i was playing a moderator and i had 100 and they had 2 and i lost and it was my 1 versus their 2 and i lost the game

Probability of that happening is 1.18571*10^-26

What game number was it?

Using the API, one could create something even closer, but if the current system is good enough, no point.

It looks like the API has a new field, game_state, which briefly broke ccgames.db.94y.info

It's A or F, so I'm guessing that's "Active" or "Finished"?

http://gaygamer.net/images/l2_46730.jpg

I get a 403 (authorization denied) error when accessing this.

The chance of losing 967 to 1 is 5.745487928449193*10^-355, which suggests that the dice may be misprogrammed.

I'm assuming ConquerClub's source code is not open?

I might be missing something, but doesn't http://userscripts.org/scripts/show/83035 already do this if you click on the 'Action' arrows?

OK, I just tried this and it brings up a confirmation box asking if you want to auto-assault. There's no "how many troops" option?

Game Finder now has a 'search by map' option, so has this been resolved?

You could use the API to pull your games and create a page of 'join' links to them? Or am I missing something? [the team part would be harder]

I actually cheated a bit here and used the more extensive data from
the HTML pages, which meant I had to download 495 pages. This only
took a few minutes (I didn't time it exactly).

The Perl script to process the pages and put them in an SQLite db also
took a few minutes.

I agree that this ...

To demonstrate what I plan to do w/ API data, I've now created a
public read-only database of my conquerclub games:

http://conquerclub.db.94y.info/

Reminder: this only includes *my* games, not everyone's.

Questions I can now answer (SQL query appears on listed page):

% How many games ...

Create a really big database to answer many of the questions in the other threads.

Whose played the most games at a time. "What's my best map"? "Which opponent am I best against?" etc etc etc

I'm already doing this for my own games, but there might be some players out there interested in other ...

Given permission, I'd love to simply download the full logs (HTML) from all games, but given that there are over 6 million of them, I'm guessing that's not going to happen

Yup, you're right. OK, I still think the game initialization time is what most people count as the "start" of the game, so, if I had to choose one, that would be it.

Ideally, both, but if I had to choose one: game initialization time.

We could then answer questions like "what's the highest number of games I've had 'open' at one time?", since most people include "initialized-but-no-turns-taken" as "open". Of course, most people joining a game that now becomes ...

I tried downloading XML data for all games, but had to stop when I
overloaded the server.

Could I get a dump of this data?

If not, could I try downloading again, this time inserting a 1 (or 3
or 5) second delay between API requests?