AFL Spring Season

[Lucifer]

Predictably, I've changed my mind and setup a vhost with mod_python. That doesn't rule out php, I just prefer python....

Edit: We can put this idea on hold for a few weeks and I'll see if I can't get the app written before school starts, but after my accounting program is finished. I also have to work on the GLXP thing a bit before school starts....

[Concord]

@epsy: In two seasons, to my knowledge, we've so far never needed a single recording to arbitrate a game. I'm willing to lift the requirement that every game have a recording, but only if the screenshot requirement is toughened (we blew it off for a few games last fall because there was a recording), and only if server admins provide chat logs of the game. A recording would still be recommended so that teams can cover their own asses.

I loved the recordings
as a improvement tool
I would
watch and reproduce


ok
that was off center

[DDMJ]

What's needed now are tools to ease administration. Durka had one in development (and it was used last fall) that, if I could get him to send me the code, I could probably extend and make it even more useful.

You have to keep in mind that when I decided to organize the Fall Season, I had to make the site immediately. I think it took me less than a day. This is because we needed to start the season ASAP so we weren't still playing around Christmas/New Years.

Keeping that in mind, I didn't bother using sql or php in terms of the game reports. At least 3-4 times I got different results from different people so I had to ask around to find out which was right. Then there's the part of "if it's automatic and I'm on UCF I'll make it so we win every game lol lol lol".

Recordings are nice to have, but whatever. Screenshots are good though...but, for some reason the majority of us can't find our screenshot button. On over 10 occassions none of the 12 players playing in the match could press that damn button. That's when recordings are nice. That's when being the organizer sucks. I remember once I had to go through 3 recordings to get screenshots. Thankfully ~ X ~ and Kyle helped me so it didn't take forever.

Anyways, you'll here more from me when I figure out what to do to the existing site...

[~*PsYkO*~]

I wouldn't mind tossing a round length limiter in there. It would be a deathzone big enough to cover the whole arena and kill everyone at the same time, or it would have to be a random and growing deathzone. Preferences?

Psyko: it's not specifically management that is a problem, more like it's the amount of work that each game week requires to finish it up. We've offloaded as much of that onto the teams as possible, but there's still more that has to be done. Personally, I would have an easier time committing to X amount of work every week than committing to 2X amount of work every two weeks.

But if we split it up amongst several people, say 3, then everybody has only X/3 work to do every week.

Those of you who have offered help running another season, would you be willing to install a subversion client? A plan is coming to me.... (Depends largely on how quickly I can put up a web form for uploading game results. I've sharpened up my sql skills recently, and php isn't that hard to work with)

tbh one of the bad things about specifically the AFL is that you guys are creating too much work for yourselves. Screenshots are cool w/e recordings I don't see the point cuz i think at least 60% of the pop still use .2.8 and how much of that 40% actually watch the videos. Eliminating those two aspects would make it so much easier. Because when the day is over, the only things people care about are playing (hopefully winning) and see who else won/lost, everything else is added work. Don't get me wrong, it looks nice it looks professional its how almost any other game tournament is constructed with MVPs screenshots and recordings but we do not have the same resources sinc ethis is a small community.

Definitely implement a fast deathzone from the middle, fast enough to kill everyone at once and create a draw game. Having a random one will only increase the "no fair the deathzone came closer to me" as close as some of the AFL matches are nowadays there's no need for added stress

[Lucifer]

Well, the "added work" is there to make sure the evidence is there when a dispute arises. So far, no serious disputes have arisen, but we know that sooner or later, something will happen. The question is what do we need to know to deal with it? Answer: we won't know until we have a dispute, and then no amount of recordings and screenshots may be able to provide the information needed.

Other than that, we've had a few minor disputes that occurred over stuff outside the games themselves. Ironically.

I've been thinking about it some more and I don't think a round timer of any sort will get what epsy's after. I also think that if 2 hours are too long, we should drop back to 2/3 instead of 3/5 for regular season games. Also, a match timer, which the game easily supports, will help. So the winner is the first team to reach 100 points OR the team with the highest score after 10 rounds OR the team with the highest score after 20 minutes (or whatever is deemed a suitable length of time for a match).

Here's the breakdown:

Assume it takes 10 rounds rather than 100 points. What's a reasonable length of time for a round? I started with 5 minutes. That's 50 minutes per round, or 150 minutes for 3 matches. Clearly 5 minutes is too long. 1 minute is too short, we already know that much. How about 3 minutes? That makes a 30 minute match, for a total of 90 minutes for 3 10-round matches. Why not just set a match timer for 30 minutes, then? That guarantees the same maximum time period for the same game, but it gives teams the same options during a round that they'd otherwise have.

On the other hand, it could be argued a match timer will strengthen defense, whereas the effect of a round timer in the form of a death zone will depend largely on the circumstances of the round, possibly giving teams different tactical options and could add to the enjoyability of the game.

I've found that in single play, a match timer adds significantly to the enjoyability of the game. I don't really know how it will translate to a team game. Is anybody willing to host a server for awhile to test these options? A regular fortress server should work, but I can update afl settings and upload them if you guys prefer.

A match timer adds another possibility, of course. Just set the match timer for 45 minutes and let score and rounds increase to infinity. Then every game is 45 minutes, period. I'm actually inclined to go to 2 matches, each 30 minutes long, and add the two scores together for the final score. That makes it more like soccer. (football for you europeans that are short by one sport)

[Concord]

timer definitely would
add new strategies
not sure they are positives

I can foresee a 9th round
with one team ahead 75-70
and a man advantage of 2-1
and just staying put and
letting 30 minutes elapse

[Lackadaisical]

I'll see if I can find some time to write up a little php script for aggregating game results. Mailing lists are no problem for me, other than the minor requirement that you have to email me to ask to be put on it. I could probably make a web form for collecting email addresses, but I'd still have to manually add them. I don't know what Durka did to manage the mailing list, I think he just BCC'd everyone on the list.

I recall that there was only a small portion of the people who had signed up to the mailing list right before the season started, and Durka had to do extra work to make more people sign up. So how about making people give their emailaddress when they sign up?

[~*PsYkO*~]

i definitely dont think 3 matches would be a good idea because so many times there was a comeback when a team was head 2-0 or 2-1

[Lucifer]

So you want to keep the 3/5 game? Is anybody (like wrtlpwned? ) willing to test a match timer vs round timer? I can throw my server back up with afl settings to test the two, but it's really a lot more than two. It's "figure out the optimal match timer and optimal round timer, then decide which is better, if any". The only catch there is that my server lags a lot.

A round timer that's a little short will tend to emphasize core dumps over the fortress, which may be a good thing. I always found in regular fortress play that if you had 5-6 people on a team, having a person or two going after core dumps really multiplied the score (assuming they were successful, of course). A round timer also helps the team with an impenetrable fortress but a terrible offense to keep the scores from drifting too far apart, and possibly winning. I don't like that at all, to be honest, because it emphasizes defense and makes it possible for a team to work only on defense and still win via core dump scoring. A match timer addresses that directly by allowing rounds to go as long as they need to be, but might encourage a team that's in a bad spot to just camp it out and run down the match timer. So we'd still need a winzone or a round timer to stop that, where the round timer here is "big death zone at a point where you really should have won by now", like 6 minutes or so.

Of course, an alternate is to just reduce the score requirement to 80 and 8 rounds.

Lack: it was actually an original requirement that folks give email addresses that we tossed during summer registration when a lot of people didn't want to give them, and we didn't really have a way for them to do so without suffering spam bots. Definitely worth revisiting.