Free Allegiance Forums

BackTrak wrote:QUOTE (BackTrak @ Jan 17 2008, 07:44 PM) I don't know about that Finger, it seems that it would get people to quit hanging around on NOAT for a game to start so that they can quickly stack a side. The one minute timeout that Baker suggested would be fine if the user is displayed a timer until they can join. (Some user feedback would be required). This would be much better than just sitting flashing, waiting for the comm to accept. I haven't kept logs, but I'm sure I'm waiting longer than 60 seconds to join on some launched games.

If it's selectable as a game mode, I think it's an interesting idea.
Baker, thanks for the info, very clear.

A 1 minute timeout , could, i repeat, could be useful. But its not really going to deter anyone, just going to make them wait. an i never mentioned anything bad about that.


My issue AND THE ONLY THING I MENTIONED, was with the idea of not allowing someone to play on a team unless there was another person wanting to join the other team. the 'strict mode' "IDEA"

Baker posed an idea he was having and "I" think its a bad one, now baker, my not liking your idea doesn't mean i don't like you or your ideas, just that ONE.

Out of everything ive read so far, i didn't like one thing that looks like might have been an afterthought.

picking one game randomly to 'prove' your point doesn't cut it. or is every game the same. My point on this is later on in the game, when there arent always going to be 2 ppl wanting to join. What about if a team drops a person, how will strict mode handle that( see end of my post)

Heres an idea, only allow comms to allow players in the order that they try to join, or instead of a 1 minutes timeout, make it random, say between 10 and 45 seconds, including your idea (strict mode) all 3 have flaws.


end of post

Or did i miss the point of strict mode? ( not sarcastic)

is strict mode only going to be implemented for the first 1 minute? that might work better...

Will the new after game start AB include a routine where it tries to balance high skilled players?

EG:

team 1: 20-4 -7-4-8
team 2: 17-17-2-3-6

now an 18 wants to join, wouldn't it be a nice touch to only allow him to join the 1st team because the 2nd team (although it has the same team-ranksum) is less in need of a vet than is the 1st team. If the 18 doesn't want to play for team 1 then he'll have to wait for a vet to join team 1 before he can join team 2.
The same thing could be done with newbs to prevent newb stacking.

OK

I want everyone to listen to this

Opinions on the validiity of the ranking system require something other then your gums flappingwithout intending insult [ if that is possible ] think actual number crunching etc etc. My open advice to Baker is to start ignoring these types of posts

HOW it get's implemented is open to discussion and exactly what Baker is trying to accomplish so for everyone sake consider these simple points:

1. A points based systems involving in game actions are in alll examples I can find are simply a behavior modification device and a time sink meant to keep you playing. Until you can show an example [ using icky things like math ] that contradicts what I like to call WoW technique then STFU about it. Honest to God just STFU.

2. Please stop trying to find a gotch ya in the presentation that renders an incredible amount of work totally worthless.... many people smarter then our little Allegiance group have spent many years on it and NO ONE is going to pull some little thread and unravel the TrueSkills sweater.

3. Also stop, really stop, getting so $#@!ing dramatic about the implimentation bits. To quote the immortal Bard [ our Bard not the other one /laugh.gif" style="vertical-align:middle" emoid=":lol:" border="0" alt="laugh.gif" /> ], " Relax this isn't going to be a donkey punch." I mean wtf would Baker be asking these things if he had already decided everything????


Thanks for listening and RL is calling again
MrChaos

You posted an 80 something % accurancy when viewing the old database of games.... can you post the accurancy of predicting Squad Matches?

Ahaneon, NO, Allegskill isn't going to choose a squad for you. Put your big boy britches on and decide for yourself. /laugh.gif" style="vertical-align:middle" emoid=":lol:" border="0" alt="laugh.gif" />

I think its good work Baker.

mcwarren4 wrote:QUOTE (mcwarren4 @ Jan 19 2008, 12:31 AM) Ahaneon, NO, Allegskill isn't going to choose a squad for you. Put your big boy britches on and decide for yourself. /laugh.gif" style="vertical-align:middle" emoid=":lol:" border="0" alt="laugh.gif" />

Hey, give ahan a little credit, I'm sure he's not so conceited as to only join a squad to be on the winning team /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" />

MrChaos wrote:QUOTE (MrChaos @ Jan 19 2008, 05:39 AM) 1. A points based systems involving in game actions are in alll examples I can find are simply a behavior modification device and a time sink meant to keep you playing. Until you can show an example [ using icky things like math ] that contradicts what I like to call WoW technique then STFU about it. Honest to God just STFU.

2. Please stop trying to find a gotch ya in the presentation that renders an incredible amount of work totally worthless.... many people smarter then our little Allegiance group have spent many years on it and NO ONE is going to pull some little thread and unravel the TrueSkills sweater.

Trying to get this moved to the rants section eh Chaos? /laugh.gif" style="vertical-align:middle" emoid=":lol:" border="0" alt="laugh.gif" /> Take a chill pill kid /wink.gif" style="vertical-align:middle" emoid=";)" border="0" alt="wink.gif" />

ad 1 : eh trueskill like ... eh ... uses in game actions too to assign relative win positions (8 player FFA, rank1-8 is assigned on kills). To implement this in > 2 team Alleg would mean finding in game stats that allow you to distinguish between 2nd and 3rd place, etc. This would mean using just the point based system you hate so much /wink.gif" style="vertical-align:middle" emoid=";)" border="0" alt="wink.gif" /> Need the math too or do you get the picture ? /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" />

ad 2 : Since Trueskill was never ever tested in a large team environment that is a pretty bold statement to make isn't it? /rolleyes.gif" style="vertical-align:middle" emoid=":roll:" border="0" alt="rolleyes.gif" /> The devs of Trueskill themselves agree that it will take a lot of games in a large 2 team environment before a persons rank even approaches it's real mean. Also Baker tests his data on old ASGS data from a time where peeps played under old rules. If you change the rules peeps might behave differently and your tests based on the old behaviour might be invalid. (eg. 5 minute rule, only AB games count, no imbalanced comms anymore, etc.) . Don't tell me Trueskill is the end-all Holy Grail of ranking systems because there is no such thing /rolleyes.gif" style="vertical-align:middle" emoid=":roll:" border="0" alt="rolleyes.gif" />

Grim_Reaper_4u wrote:QUOTE (Grim_Reaper_4u @ Jan 19 2008, 09:42 AM) ad 1 : eh trueskill like ... eh ... uses in game actions too to assign relative win positions (8 player FFA, rank1-8 is assigned on kills). To implement this in > 2 team Alleg would mean finding in game stats that allow you to distinguish between 2nd and 3rd place, etc. This would mean using just the point based system you hate so much /wink.gif" style="vertical-align:middle" emoid=";)" border="0" alt="wink.gif" /> Need the math too or do you get the picture ? /tongue.gif" style="vertical-align:middle" emoid=":P" border="0" alt="tongue.gif" />

By comparing the maximum playtime length for a team it should be quite possible to accurately say which team was destroyed first. That's your ranking /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />

I don't know if that's the kind of calculation Sarge does tho.

Badp wrote:QUOTE (Badp @ Jan 19 2008, 10:01 AM) By comparing the maximum playtime length for a team it should be quite possible to accurately say which team was destroyed first. That's your ranking /smile.gif" style="vertical-align:middle" emoid=":)" border="0" alt="smile.gif" />

Yeah but we could also use total kills/base kills/max conquest% or whatever to prevent peeps turtling with rix dests in their home /wink.gif" style="vertical-align:middle" emoid=";)" border="0" alt="wink.gif" />

Personally (although this community is probably too small for it) I would prefer splitting up the player base in at least 2 groups : advanced vs novice players for example. In a ranking system that goes from 1 -20 ranks 1-10 would fly in novice and the rest in advanced or something like that. I think this would give us a better Trueskill mean with a smaller standard deviation a lot faster than if we were to toss everyone in 1 big pool indefinately. (yes i realize that the vet pool would be very small in the beginning but you can wait with the vet pool until enough players reach rank 11 or higher and then start the skill separation). So essentially you would play on the newb server until your rank is high enough to let you play on the vet server. This way peeps play games among similarly ranked players in which gross stacks are less likely to occur plus it is easier for the system to find those peeps that need to gain a higher rank and promote them.

Grim_Reaper_4u wrote:QUOTE (Grim_Reaper_4u @ Jan 19 2008, 11:40 AM) So essentially you would play on the newb server until your rank is high enough to let you play on the vet server.

What an utterly horrible idea. Are you trying to sabotage Allegiance or what?