I recently realized that redesigning the top ten is an immediate fix and doesn't need to wait for the next stats update.

Maybe you guys can consider while the servers come back up: what would be a reasonable criterion for making the server top ten; should we measure by games and kills?

Some figures we tossed around earlier were 20 games / 100 kills.  If we really wanted to do it right, I might say 100 games / 1000 kills to make the top ten.

yeah 100 games 1000 kills sounds reasonable to me.

Aye I think that sounds good as well

I would do 150 games / 1,000 kills. (I'm assuming they have to meet both critera, not either...)

As time goes on and more stats are collected, I don't see why these numbers can't increase a bit more.

I second that.

hmmm...question is do i play enough...YUP i think so lol. practicly every dammmn night hahaha

Hmm; as we speak the only one who meets that criterion in the top ten is Blitzen at #7.  Minimum match will lead to quite a shake up.

For the next update, the way rank is calculated will be significantly different: you have to play ranked players to get ranked at all; and only after the first twenty ranked matches or so will you graduate from “provisional” to ranked.

Unranked vs. unranked, and ranked vs. unranked count as unranked matches.  Question is: what would count as a reasonable number of matches to graduate from provisional to ranked status: 20, 100, 500?

I think 100 would do the trick. 500 is way too much, I think, considering we DO need to have some ranked players on the server for this to work. Similarily, 20 is too low.

I'm not so sure that losing rank when dying is handled as well as it could be. It doesn't seem to take into consideration the fact that Free For All's generally end up with people waiting on the outskirts of a fight to come in and beat all of the opponents when they're 1 hit away from death.

It would be quite a challenge to come up with a mathematical model to describe people waiting on the edge.  When the stat-parser is crunching stats, we only have raw kill data to go from.

As it stands, rank is treated like a fixed resource: in order to gain rank, someone has to lose it.  The killed loses as much rank as the killer gains.

Either we reduce the amount of rank gained or lost in any given transaction, or “massage” the amount lost by kill.  I might opt for the former since it's a little less arbitrary.

Ah, I was unaware. Carry on with the way you were going. I trust you'll be able to work something out with the given resources.

I like 150 games 1,000 kills

Would it be possible (or sensible) to make kills in the one-on-one arena count for more rank gain/loss then FFA kills?

Although that would only encourage those accursed five-deep waiting lines...


I like the ideas being thrown around here. Sounds good so far.

Originally posted by YellowHair:
Would it be possible (or sensible) to make kills in the one-on-one arena count for more rank gain/loss then FFA kills?

That would be technically feasible; but what would be the reasoning: one-on-one is more representative of a person's skill because there are less intervening influences?

You may have a point; if we went that route, however, I'd almost like to divide the rank into several individual categories and one composite category.  The individual categories would consist of, say:

• ffa
• one-one-on
• headball
• skolf

and the composite category would somehow consolidate these into a single synoptic ranking.  That's the way it's done, for instance, at UT; and I hadn't seen a need previously since we didn't have any well-defined games.  It could be an idea whose time has come, however.

If we implemented it, the data would start from the next update; everything hitherto will be purely composite.

Speaking of, there are tons of great gameplay modes. I was chatting with mano about it, and we were thinking that it would be nice to just alter the map with a portal or something that leads to an identical map, but one that has headball, skolf, and medium/high cal. It would be fun to have all of em on the same map, and also a shame to lose any of our current ones. His main concern was the map file size for downloads.
Just wondering your take on that, and if such a project might be workable sometime in the future?

I noted the portals to the Low Caliber Arena and the Low Caliber Cycle from Faison's castle map. I think it would be a great idea to add these sorts of portals to the Low Caliber Network.

That way, we could have 2 arena servers that players could easily switch between, while also having the ability to go over to skolf and headball.

However, if we did something like this, we would have to devise a method of command-permanence.
ie: If someone is muted/nosafed in one server, they should be muted/nosafed in whatever other server they jump to. Otherwise, it would merely be a simple means of escaping such member commands.

Or have it so the person having a tag on them isn't allowed to go through the portal. I would only favour having skolf, headball, skockey, CTF, TDM and our arena. No High or medium calibers because those defeat the purpose of being called Low Caliber.

quote:

---

That would be technically feasible; but what would be the reasoning: one-on-one is more representative of a person's skill because there are less intervening influences?

---

That was the reasoning; a greatly skilled person can hold the floor in one-on-one almost indefinately; while even the most experienced warrior can die by a randomly thrown weapon in FFA. It requires a different sort of skill to stay alive and on your toes in a frenzied melee, but isn't to be compared to the focused and determined battle of two chieftains in an honor duel. Individual categories would help differentiate between the skills.

quote:

---

[...]thinking that it would be nice to just alter the map with a portal or something that leads to an identical map, but one that has headball, skolf, and medium/high cal.

---

I like that idea; would minimize the space cramping and perhaps neutralize the need to remove my beloved boxing arena.

Would there be room for mace-wars?

*begs on his ancient arthritc knees* Please! Separate the 1v1 and Melee rankings!!! lol

But great ideas in here. I like most of 'em. Especially separate rankings and a composite . . . *sigh* Finally my favored and most honoralbe one versus one getting the respect it deserves heheh. 'Tis enough to make this old man fall down and weep.

I'd prefer 100 games or 1000 kills, as opposed to 100 games AND 1000 kills.

PS - I hope to start playing again soon. Automata and graphics programming have proven to be infinite time-sinks, however.

Originally posted by Ministration:
However, if we did something like this, we would have to devise a method of command-permanence.

Good point, Min; we'd just have to determine when the tags get turned off: say 15 minutes by default, but allowing you to specify? Example: “nosafe 7 5” no-safes player 7 for five minutes.

We need a consensus on six games and their division according to arena; for instance:

Arena:

· ffa

· one-on-one

· boxing?

Team Arena:

· skolf / skock

· mace wars

· headball

Some of you guys mentioned team games; the thing with team games: we'd have to go with the whole server team or no team, otherwise we risk complications.  No problem, though: we'll introduce a Team Arena as our second server; that opens up the door for team ffa and maybe capture the torch.

I really could have used Fais's help on the skolf portion, though.