Allegiance 2

[HSharp]

QUOTE (cashto @ Feb 5 2013, 08:21 PM) Not yours. P1's.

Ah k, just seemed like a weird time to bring it up especially without quotes and p1 didn't post between our posts
QUOTE In response to your latest ... I guess I could talk until I'm blue in the face about the futility of trying to ad hoc a ranking system by assigning points to in-game events. I hope you do get the data you're looking for. I think that would be the only way to convince you ... for you to try to find, out of the exponentially large search space of possible permutations, the right combination of factors which people would look at and agree, yeah, that's a more accurate rating of kill than what Trueskill achieves.

(Tho frankly, you could probably get 90% of the way there by looking at kill-death ratio alone, because that's largely what people notice anyways -- dogfighting ability).

Edit: I should mention that there is an objective way to determine the validity of a ranking system beyond the "nebulous opinion of experts" -- use your ranking system to make predictions of game outcomes, and compare with the actual result. By random chance you would get 50% right. SgtBaker used to claim something ridiculously high, like 85%, which I never quite believed (even if Allegskill had zero error, which of course it doesn't, either pretty much every game is just massive stacked, or the weaker team is still going to beat the stronger team, by chance, something more than 15% of the time). But anyways your ranking system should do something better than random chance, and the degree it predicts games better than random chance is the degree it's better than another ranking system.[/quote]

With the current implementation of points it might not be accurate at all, however with more metrics there could be more possibilities, also I have been thinking and AI might not be too far off from the solution actually, another possible idea of mine which links with some AI work im doing but would need a lot more metrics, but more metrics are possible as DataMine proved.

---

Also on the off-chance TE reads this, http://www.freeallegiance.org/forums/index...showtopic=62121

Gimme my CSVs!

[Adept]

Truly successful Allegiance (2) would have to work like World of Tanks. You log on and press play, and it throws you into a (more or less balanced) game. No half an hour wait to sort out who will command, what settings, map... etc. All that would have to be automated.

The commander's role would have to go away, aside from professional (SG style) games. Think of LoL / DOTA2 style games where creeps automatically push out wave after wave. In Allegiance the creeps would be the miners and the constructors. Find alephs to new sectors and cons will go out to secure them. Once control has been establlished, miners will try to mine there. That sort of thing. Bombers unloack when a certain amount of time has passed (10 minutes maybe). Mined He makes the cons go out faster and speeds up tech gain.

Something like that. A simplified game without the commander, that keeps the focus on the pilots.

[cashto]

I could imagine a system in which:

Upon turning on autobalance, all players are automatically flushed.Players indicate which teams they are willing to join. A player may request to join multiple teams simultaneously. Double-clicking NOAT will indicate that a player is willing to join any team. Double-click NOAT again to cancel.Commanders cannot accept/reject players; the server will assign them only.The server will not allow the game to pass a certain imbalance threshold. In general, this will mean that while team sizes are still small, players would only be admitted in pairs.A game automatically starts once a threshhold of non-afk players is reached.Players who are booted or who leave will automatically be replaced if there is another player requesting position.There is no guarantee that the server will assign players in order which they arrive. E.g., if a high-ranked veteran is waiting to play, and a newbie arrives, it's possible that no matter which way the server assigns the players, the imbalance threshhold would be crossed. Therefore, the system would keep them both waiting. If a second newbie then arrives, the system will admit the two newbies, one on each side -- and the high-ranked veteran will remain waiting.

[HSharp]

I had that similar idea in my had years ago, a sort of automatic pick pool which lets people join in waves after game has started.

[pkk]

QUOTE (HSharp @ Feb 6 2013, 12:11 AM) I had that similar idea in my had years ago, a sort of automatic pick pool which lets people join in waves after game has started.

It will remove the fun from game...

[zombywoof]

People make money streaming World of Warcraft.

No, I don't really play anymore. I don't play anymore because it was 30 minutes between each game that launched 5v5 and, on a good day, finished 12 v 12. You better believe that if I could get 4-5 good games (starting at least 10v10) in a day one after another I'd be right here still flying.

Now for other specific responses:

QUOTE (Bacon @ Feb 5 2013, 11:49 AM) Spunky I'll concede that maybe I had no idea what was going on back in 1999 as I was 13 years old. I just remember seeing ads in PC Gamer and finding the game on their "demo disc," and seeing a large # of people in the AZ-beta lobbies during this time. It had a shot. It really did. Maybe the marketing campaign wasn't the greatest, but with the internet all you really need is word of mouth. Had Alleg been the "diamond in the rough" we all think it is, word of mouth would have spread and it would have been a massive hit (think Minecraft). The thing is, it appealed to a niche. We are that niche, and that's why we like it so much. Niche games are VERY loved by their target audience, but not by many others.

It appealed to that niche in 1999. The internet has increased so much since then with online gaming taking off. Moba, as a genre, wasn't even around until after DotA on Warcraft 3. There's more potential for people now than there was in 1999, and it did well enough in 1999.

QUOTE Anyway, this idea is a loser. I'm sorry. I wish I could be more positive, but when it comes to Allegiance, I have my opinions and I have them for a reason. This game was great for what it was, and I think it had an impact on a lot of our lives moreso than most video games, but trying for a sequel is like trying to recapture lightning in a bottle. Alleg was Alleg because it came out in 1999 and was played by versions of us in 1999. It's 2013 now, and the world has changed. Alleg doesn't really fit anywhere, and as Kage just said, nobody here has a compelling vision on how to make it fit. It'd be amateur-hour trying to build a sequel, and I foresee a lot of people losing a lot of time and money if you actually pursued this.[/quote]
Allegiance was Allegiance because it intermixed the genres of space-sim and RTS.

Also who the $#@! cares if this community doesn't agree on what "Allegiance" is? This wouldn't be the Allegiance community's project, it would be MY project. I just was asking here, since it's the largest collection of people who would know that I'm aware of off the top of my head, how much it would COST.

QUOTE (aptest @ Feb 5 2013, 12:38 PM) Alleg was an outsider when it first came out. now it is main-stream as the MOBA (this is what alleg is) gnare is taking more and more space in gamesphere. I can certainly see alleg working as a project however the question remains:

how is this going to be funded?

Kickstarter (derp)

QUOTE every moba out there to date has some kind of revenue plan behind it. it can be unlockables (LoL), pay2win (Savage2), pay2play (HON) or some other methods for generating cash flow.[/quote]
Pay to play with eventual revenue streams coming from tourney sponsorships and advertising space on tourney streams.

QUOTE Making alleg 2 seriously will require serious cash. It'll need skilled people investing big time into the project and these people will have to be payed for several months of their time. In addition, servers and bandwidth will have to be rented, advertising bought, et cetera.

Where is the money going to come from, not just on a one-time basis, but consistently to keep the project running?[/quote]
First, kickstarter (because that'll generate the capitol), then through a combination of game sales and/or advertisement revenue from streams.

With regards to further comments about game time: the average game, from what I remember, took about 30-35 minutes and could last anywhere from 10 minutes to a couple of hours (with even 2 hour games being rare and probably more of a core design issue than game style issue). League of Legends lasts anywhere from 20 minutes to 1.5 hours (with certain outliers, for example I have experienced games that are anywhere from 12 minutes to 3 hours on Summoner's Rift) with an average of about 30-40 minutes. Even League games that are stomps tend to last around 25 minutes if the surrender vote fails.

For game speed, most games went something like this iirc:

1-10 minutes: People rushing off to scout in their various directions and meeting up for the first furball of the game.

Announcers: could be talking about the choice of faction for this particular map/settings, the direction each side chose to push and what it means, who's flying what and how that matters, predictions for the opening furball, musings on what the miner chain is going to be. During the furball could zoom in and call the action, highlighting famous and/or exceptionally gifted players.

11-20 minutes: Economy wars. There are various buildings placed and various miner rushes/defenses, sometimes more intricate carrier and supremacy play, other times it can turn into a "farmfest" like in league where both sides are just building up money.

Announcers: could talk about endgame techs, speculation on where the miners will go next, discussions of what each team is doing to hide or find enemy miners, discussions of what each team is doing effectively to mine or to stop their opponents from mining, where the next constructor's going to go.

21-end: End game. Teams are pushing cons to get some HTT/Bomb action going or setting up SBs or whatever.

Announcers: talking about what the TP2 scout is doing well, whether he was spotted or not. Talking about the anticipation for whatever the next fight is going to be.

Quite frankly, Allegiance games were action-packed and interesting enough that when I was in CDT-2 some of my friends would legit just watch the games with me. There's a lot of downtime when you're a pilot (get podded, flying opening scouts/ints), but how much downtime is there in the game at large? When it's 15 v 15, how much time is there in a game where you can honestly say that a) no one is doing anything and b) there isn't a bunch of stuff that needs to be done?

And yeah, placing probes isn't the sort of glory-seeking amazing play you think of when you think about the NFL or NBA, but it's still something that has to be done, can be talked about, and there is STILL a whole game going on and that's just one aspect to look at occasionally.

The biggest challenge I can see, from having played League, Allegiance, and Starcraft, is figuring out how to create unified game lobbies that don't invite people to abuse whatever system's in place. An engine can be pulled off the shelf, there are so many graphic artists these days you can't spit in a crowd without pissing one off, balance isn't the hardest thing in the world as long as you have a coheisive idea of how the game should play.

And again, stop acting like this has to be a sequel or clone of Allegiance. "Allegiance 2" is just an easy way to describe it. If you prefer, I want to make a "MOBA that combines space-sim and shooter elements with RTS elements that puts a high emphasis on player skill and a higher emphasis on player awareness that's specifically designed and geared with competitive (for money and prizes) play in mind."

Also I know this is the Allegiance community since more than half the posts in this thread are about stacking never change guys!

[Spunkmeyer]

QUOTE (Adept @ Feb 5 2013, 04:22 PM) Truly successful Allegiance (2) would have to work like World of Tanks. You log on and press play, and it throws you into a (more or less balanced) game. No half an hour wait to sort out who will command, what settings, map... etc. All that would have to be automated.

...or cycled (as in maps appropriate to team sizes) or made into quick-votes as the game starts. Both have been done elsewhere.

QUOTE (cashto @ Feb 5 2013, 04:41 PM) I could imagine a system in which:
[snipped]
There is no guarantee that the server will assign players in order which they arrive. E.g., if a high-ranked veteran is waiting to play, and a newbie arrives, it's possible that no matter which way the server assigns the players, the imbalance threshhold would be crossed. Therefore, the system would keep them both waiting. If a second newbie then arrives, the system will admit the two newbies, one on each side -- and the high-ranked veteran will remain waiting.

More or less, except the part I quoted above which would probably be detrimental to gameplay. This would be one spot where it'd be better to make concessions.

The value of having the server assign players is it wouldn't be possible to 'game' TrueSkill by picking your teams appropriately. Next, full rank change should be allocated players who quit early. Finally, the server should attempt to match commanders of equal skill. Combined, these three changes will improve the accuracy of ranks significantly.

[Spunkmeyer]

QUOTE (phoenix1 @ Feb 5 2013, 05:43 PM) Kickstarter (derp)

Kickstarter does not fund ideas (unless they are $#@!ing brilliant but I don't think A2 quite qualifies) At the very least, you need an active, working prototype of what you are doing that also has sufficient eye-candy. That means several months of development and up-front costs for something which may end up not being funded. Certainly doable as long as you are aware of the risk.

[HSharp]

Indiegogo doesn't require prototype, but general consensus is if you want the cash you need to be on kickstarter.

That being said way before that you need an actual lead developer/designer and an actual business plan with numberwang.

[cashto]

QUOTE (Spunkmeyer @ Feb 5 2013, 03:45 PM) More or less, except the part I quoted above which would probably be detrimental to gameplay. This would be one spot where it'd be better to make concessions.

I suspect (but can't prove) there is a sort of Arrow's impossibility theorem at play here, where no matter what system you come up with, you always have the dilemma of "player choice, fairness, unbounded wait times: pick any two".

QUOTE The value of having the server assign players is it wouldn't be possible to 'game' TrueSkill by picking your teams appropriately. Next, full rank change should be allocated players who quit early. Finally, the server should attempt to match commanders of equal skill. Combined, these three changes will improve the accuracy of ranks significantly.[/quote]

I'm not too concerned about the impact of ragequitters on the ranking system. I think people quitting in order to lessen the impact of a certain loss is a rare or possibly non-existent phenomenon.

Trying to match commanders implies that there's more than two at any time who want to command.