Cheer on the underdog!

The "Load" button?

Well said! 

If you look at it the other way, I think it would be very frustrating if you had the upper hand and wanted to finish off an opponent and the game would constantly help him in some way to stay in the game...

I think this issue comes back to the issue of steamrolling and 'killer stacks.' Many times, the player with the killer stack is able to steamroll any opposition with ease. The player with the best economy is often able to build a killer stack the fastest, and thus underdogs in the economy get absolutely owned by a stack. I think as long as it takes a fairly long amount of time for two empire's borders to meet, due to lots of area between two starting cities, both players should have a decent chance to build up their empire before having to deal with enemies swarming towards them. It will at the very least give a few turns in which to catch up to the enemy in army production. I do not think any handicap for the loser is neccesary or appropriate. If one is losing, the distance buffer between the production cities of one empire and the cities of the other empire should be sufficient enough. Defending your territories is great, but we want small games to end in a few hours at most, not last a week. Also remember, for huge maps with 10+ AI you will want to be able to have a huge war without being arbitrarily handicapped for winning it.

Really I think it depends what you mean by underdog. Frogboy made a great post about victory conditions a while back (sigh, I can't find it...). But to summarize, one player might have a unstoppable military, and it looks like it's only a matter of time before the entire world succumbs to it. But another player might be well on his way to casting a major world-wrecking spell that will completely change the game - the player with the unstoppable military might topple over, indeed most empires would probably suffer huge damage, leaving the caster of the spell in a very good place. And yet another player might be well on his way to completing his Master Quest, snatching victory from both the military superpower and the magical destroyer. And who knows what other victory conditions might be available.

I think that's the best way to handle the 'underdog' problem. Make it very difficult to tell who is actually an underdog; if a powerful empire sends a huge force to roll over a smaller, seemingly innocuous nation, maybe just before the killing blow is struck to the smaller nation's capital city, the defending channeler will utterly destroy the invading army with the culmination of a powerful spell that has been a long time in the making. But at the same time, if you (or any other player, human or AI) are on the losing side of a struggle and have no tricks up your sleeves, the game should definitely not come and save you. Basically - if you are straight up inferior to another nation, it should be quite a challenge to defeat said nation.

Basically, I think that providing many options with which to handle a situation is the best way to make underdog scenarios less clear. You shouldn't necessarily be doomed just because the nearest military superpower has declared war on you - you may not being able to match it militarily, but maybe your magic or allies or something else you've been preparing can save you; and these means that you never were really the underdog - it just means that instead of focusing on military might you focused on different, but equally important, aspects of your empire.

as pigeonx2 says, the only way to fix the underdog situation is to have a different way to win.   You cannot let the underdog respond with the same tactic, because then you're just going to have a longer game, not a better one.    What that also means, is that you may need to leave on other victory conditions besides 'dominate' or whatever its called in this game.   (I know many many people who have the tendancy to turn off all but 1 victory condition)

That's me! I often turn off all but one or two military conditions, but because of me not other players. Sometimes I'm just really in the mood to take over the world, but somehow it almost always turns out to be a lot easier for me to win through an alliance, tech or culture victory. In fact I often find that on my way to a conquest victory I accidentally win by another means, or have to actively stimmy some aspect of my empire to prevent a premature victory through a means I don't want. Although at the same time the other methods of victory can be convenient ways to get your victory without having to do all the boring mop up at the end of world-wide (or galaxy-wide) conquest...

Hopefully Stardock figures out the right balance so that it's rather difficult to... stumble into another type of victory while actively striving for a conquest victory.

Was this what you were remembering, pigeon?

From the bottom of the OP in that thread:

Definitely one of my top five favorite info scraps so far.

I think that MoM went in the right direction with the distance cost multiplier when you Wizard cast spells in combat.  If you expand this into a CIV like corruption effect you can make large empires less efficent.  It has been shown that this is not enough to change the course of a game that has too large of a ballance of power difference but it allows for a small efficent empire to be more then a speed bump.

Sammual

Yes, that is the quote I was thinking of. And yes, that whole post/thread is a goldmine of information. I'm really excited to see how they implement 'action-driven' victory conditions for some of the more abstract ones.

I hated corruption in civilization. It basically penalized you for playing on big maps, it made it essentially impossible to have a nation spanning across multiple continents, blegh. If there is a distance cost multiplier for anything it would definitely have to scale with map size, for one. Personally I like the idea of uniform magic costs within your channeler's domain, with maybe increasing costs outside of it.

I agree, it needs to scale with map size.

Sammual

Well one multiplayer option would be to the option to disable the graphs progress bars.  Within Dominions_3 you have the option to enable or disable graphs which displays army size, research, gold income, gem income, etc., etc., .  Disabling of graphs makes it more difficult to judge the overall strength of opponents.

In regards to the singleplayer sessions the AI opponents should have multiple personalities and this should include personalities which setup alliances with other opponents to attack the strongest enemy.  Far too often I see AI opponents ALWAYS target the weakest player which has two painful side effects.  First the weak human player is overwhelmed by the multiple attacks from multiple opponents.  Second the very strong human player continues expanding/growing while the weaker AI opponents fight amongst each other which can be boring.  It's like watching highschool kids fight while a strong college bully walks up from behind each one snapping their neck.  At least some AI personalities should seek to destroy the strongest opponent on the map via alliances.

For the game as a whole... each players capital and nearby terrain should provide extra defensive and offensive bonuses.  Thus a weak player can hang on longer or possibly recover and not really useful for a strong player.  Each player should be allowed to move their capital with a small cost of resources and time, while also be allowed to rebuild their capital with a slightly greater cost of resources and time.

AoW:SM and Dominions_3 included a magic area known as a domain or dominion where the player would recieve bonuses to the morale of units and/or casting of spells.  This overall provided benefits for those defending their realms.

I advise not changing the course of a games history in your favor because that's a huge unfair advantage.  I won't go into detail and this topic has been previously discussed within another thread so let's not throw this topic off course with responses.

I just wanted to comment on a personal experience where I had the option to quit a game I was losing, yet I chose to keep playing.  It was in Dominions_3 where a new enemy appeared on my borders and was much stronger.  I quickly organized my best armies and supercombatants to face him.  Since my opponent was an AI I felt I had the advantage yet soon learned his greater numbers crushed my strategic battlefield orders and placement of units.  My best armies fell one by one and each turn I was losing more and more land to him.  Every battle he seemed to have larger and larger armies with more mages.  He pushed me all the way back to my capital and my strongest army which was wounded had no choice but to guard the neighboring province below my capital in an effort to protect my remaining provinces providing my last sources of income and gems(my belly).  I felt certain he would march onto my capital and begin laying siege...  I lost so much that winning the game seemed utterly hopeless, but I kept playing.    Surprisingly one turn passed where he didn't attack my capital or my army to the south... it was a breath of fresh air to strengthen my defensives.  Then the second turn passed and he once again did not attack, but his forces were still too large for me to do anything except hold my ground.  Then a few more turns passed and his armies on the border began shrinking... I had no idea why, but every turn I was healing and strengthening.  Then eventually the truth behind the mystery appeared a much more powerful empire was actually swallowing him and took over the territories he stole from me.  I felt like a coyote watching a T-Rex chow down on a lion which had me pinned.  Surprisingly the stronger empire(T-Rex) did not attack me which allowed me to attack and expand into other areas of the map and eventually, much later, take down the mighty T-Rex which unknowningly saved me earlier in the game.    

I would have missed that incredible set of events and my mighty comeback if I would have quit that game or used some type of reload (via 3rd party tool).

Personally I think that not just different ways to win, but different ways to fight are pretty important.  Lets look at Dominions 3.  A huge army that doesn't employ magic well can be defeated quite handily by a small army that does employ magic effectively.  However even then there are multiple ways to fight.  You can outfit heroes with large numbers of magical items that turn them into supercombatants (in Elemental it sounds like imbuing them with essence should also work).

I see the ability to fight with large armies as important, but building up your heroes should have equal (but different) results.  Turning your avatar into a powerful personal combatant by imbuing him with all your essence should also be effective, and just as effective as expending your essence casting a nation (or world) destroying spell.  At the same time some kind of sustainability should come into play.  There should be very real economic costs for building that massive army that result from the feeding of it (while it is on your territory or on barren ground) and the taking of so many able bodied men from the land.  At the same time by focusing on your heroes or your avatar, or saving up your essence for that world destroying spell, your economy will probably suffer because you are less able to spend your essence at building up the land, making it livable.  A balanced playing style (lets say a mix of heroes, economy and army) should also be viable, just as much as going all out in one particular area.

I like these ideas. I would actually like to be able to disable the graphs for singleplayer, too. The AI solutions are great too; although instead of having totally separate AIs behaving drastically differently (some ganging up on the most powerful opponents, some picking on weaker ones, etc), I would prefer an adaptable AI that can make the decision, "will I be better off by conquering that weakling to my east, or by joining the alliance forming to put a stop to Blue Player's unchecked expansion and unmatched might?" It would make games more unpredictable in a very good way, although I imagine it'd be real hard to code.