This game was previously discussed in the thread [Agora] Delicate Dice Pools.

So tonight we played a bit of Agora, which I would say is about 80% complete.  We played through Descent and one round of scenes; it took us a couple hours.

Judson played a Hierarchy Legionnaire, which is about exactly what you would expect.  Imagine Commander Adama minus compassion.  I played a Diamond Alliance Legionnaire, a pirate captain.  Mark played a Uprising Bucolic Cesar Chavez type.

Character Generation was very straightforward and, minus a couple hiccups, quick.  I need a dividing line in the Resources box to separate people from other stuff.

Descent -- a tutorial-like last stage of character creation play through a scene thingy-thing -- went very well.  Each stripped-down simplified scene was relatively rocky, but that was sort of the purpose of the Descent scenes -- to get the players familiar with the mechanics, and this phase served that purpose admirably.

We then played through one round of scenes, so each one of us started one.  Judson's troops attempted to set up a food processing facility but met with a great deal of natural resistance -- animate trees, apple-sized mosquitoes, swampy terrain -- to the point that they ordered an orbital strike to wipe a big chunk of forest off of the face of the planet.  I played his Obstacle, the Black Woods, and rolled high for its stats, and Judson did not play with a Lieutenant helping him out.  Judson won out, but he had to fight really hard to pull it off.

For my scene, instead of mucking about trying to grow things, my merry band of pirates wanted to look around and find some targets for raiding.  Mark played my obstacle and set up a rudimentary militia police force.  I suspect he thought that we would duke it out (most of the scenes thus far had entailed tactical or strategic combats) but instead the pirates decided to subvert the idealistic little police force.  I had Judson play one of my Lieutenants and we made the police dance like a little puppet on strings.  It was very nice seeing the die mechanic work just as easily for social manipulation as out-and-out fighting.

For Mark's scene, he wanted to recruit the yeti-like native inhabitants that had been established in my Descent scene; instead of playing the yeti as obstacles, Judson decided to create a faction of Ardent Orthodoxy missionaries who were also trying to recruit the yeti.  I played one of Mark's lieutenants, Sister Courage, and in the end we set the yeti to rioting against the Ardents.

We called time there and discussed the game in general.  The biggest problem is a dice currency issue (there's a big surprise) where it's relatively easy to get into a situation where you can reroll to avoid a tie, but the only thing that will help you is rolling a one, which means that you will beat the other guy by one die, forcing him to lose a die, while you lose the one that you just rolled -- in the end, you both lose a die, which is exactly the same result as would have happened in a tie if you hadn't rolled the die in the first place.  I'm rethinking a large part of the dice-currency to avoid this issue.

We also discussed the long-term direction of the game, and Mark especially felt like he was flailing a bit without a sense of what the point was -- he had this faction of people who were there on the planet, but there was no signposts saying 'do this' or even a reinforcement cycle to encourage him to do anything in particular.  While this isn't necessarily a bug in everybody's estimation, I do concur that some direction options and 'win conditions' -- even if slight -- would help give the game experience a little more shape.

Even flawed, the game was entertaining to play through, so I'll definitely be developing this one.

Imagine Commander Adama minus compassion.

Actually, you could pretty easily replicate Commander Adama.  While your Faction and Culture ideals are picked off of lists, you create your Background and Descent ideals whole cloth.  An Adama-clone could have ideals stressing the importance of due process, democracy, loyalty, family, what have you.  And actually, those smaller-scale ideals will get sucked into play (and have a massive impact on the entire planet) based on both the player's and his Obstacle's actions.  Both of them have a vested interest in putting as many ideals in play as possible.

The uber-conflicts and long-term game pacing is something that I am coming to agree the game needs.  Presently the 'long-term' development of the game is expressed in a couple ways: (a) your ideals change to reflect the impact and character of Agora, as you pick up keywords from your opposition, (b) the Obstacles pick up keywords from all of the characters, thus the setting is changed by the PCs, (c) Obstacles are exhausted of Complications, and are thus 'tamed'.  So already there's a mechanical impact there, where the world changes you and you change the world.  You leave your imprint, which is good.  What I need is a metric by which you ensure that your imprint lasts, and that I'm not entirely sure how to implement.

I'm pretty sure I don't want an explicit win-condition endgame, though.  I don't want people to 'win' Agora; I want them to articulate their ideals in the face of adversity.  I want everyone at the table to be able to 'win' in that sense at the same time, and I want to make it possible that some people 'win' and others 'lose'.  Perhaps a Dogs-ish "reflection" phase at the end of the game where the players discuss who had the biggest impact on Agora, and what they expect would happen afterwards or something.

One of the suggestions that Mark had was that there be a limited number of Obstacles that could be created in the game.  This would lead to a semi-natural end point, where all or most of the Obstacles are 'tamed', but I fear that this would lead to a sort of dying-energy game where everything just sort of winds down.