Ok, perhaps I'm the odd one around, but I found GC2, and then DA, easy, even at the suicidal level (and even when I went in and modified the AIs from 170% to 200%, and set ALL at max cpu).
The problem was, the AIs were too predictable. I could tell they had the potential, but the way the tech tree was set up, the AI would go for colonization techs, while swarming out colonizers. The techs were cheap enough, within a ten or twenty turns, even on a gigantic map, the AI had hit all the normal planets, and most of the special worlds (due to getting the needed tech while en route to the planets). So, even with special types of worlds, you still got the Torian block, and the Drengin block, etc.
Then the AI would go for weapons. It would trade around, and if not all civs, at least all on a given side of a map, would lock into the same weapons tech and ignore the rest.
Then came the massive buildup of armies, with the inevitable wars between everyone. In between, you'd see a few trade lines, but they would break down quickly, as either the two civs went to war, or other wars made the line too unprofitable to continue.
Essentially, within the first hundred turns, even on gigantic, the entire map was full, each civ's line was pretty much established, and wars were going on continually.
On the research side, after chosing a weapons tech, the AI would go with the corresponding defense tech. It would take each to level 2 or 3, then depending on the AI, would move on to sensors, diplomacy, or habitation techs.
Influence was relatively uncontrollable, except using starbases or cultural centers. I spent last week playing out a 9 year (game time) gigantic suicidal map. I kept it going, just to see if the other civs would ever try to counter my influence bases, or build theirs up more than a single module or two. None of the majors ever even added a single module. The minors, surprisingly, did build theirs up a bit, with the Dark Yor going for three additional modules on an influence base. But no bases of any type from the AIs ever really got going with tens of modules. Mostly the AI used bases just to extend influence, then forgot about them. So, as a player, my ability to use maxed out influence bases without the AI having some way to counter it seemed unbalenced.
Also on balence, Neutral > Evil > Good. Sometimes the colonization events would actually have the Neutral be worse than the good, with evil better than both. However, the alignment benefits just outweighed any of it - and I could always take the colonization benefits and then just pay my way back to Neutral. Plus, on a gigantic map, there's an advantage to waiting to hit Xeno Ethics. If you wait to hit that one until you colonize all the quality 1, 2, 3 planets, you end up with more civ benefits - and again, you can always buy your way back to Neutral.
So what did I want? I wanted the AI to really have to work to colonize a given planet type, and to have to make choices, rather than just hitting all the colonization types and moving on. I wanted the AI to have to choose whether to go for trade, or colonization, or influence, or military. I wanted the AI to have to realize that there were three types of weapons, not just one. (I also found it incredably odd how the AI truely valued Aquatic, trading research treaties for it, but wouldn't even give 1 bc for Toxic).
By the way, on trade, it's the one bit I've not yet fully tested. I don't know if GalCiv2 can handle the max number of trade routes I've implemented. But wha I saw was, trade gets worthless quickly. I prefer gigantic, abundant planets, suicidal, with 9 majors and 8 minors. With a 16 route max, I can't even hit each race. With each route being worth maybe 30-50, on a gigantic map with 800-900 colonizable planets, that trade route quickly isn't even worth 1/10th of a percent of weekly taxes. Basically, trade happened between wars - it's income loss wasn't worth considering when considering going to war (or being paid to do it).
This tech tree isn't for the person who hates gigantic trees, or having to pick and choose. If you like having all the techs quickly, don't go for it. But if you like the idea of, say, the Drengin having a huge empire and a huge army, only to get cut in half by the Arceans when they swoop in and colonize all the Barren/Toxic/etc worlds that the Drengin couldn't (because they were too busy researching war techs)... If you like the idea of the Terrans trying to buy the colonization techs - or at least, just get so influential that they can try to flip them... Or the super spys (who you'd think would be at a disadvantage, but who seem to almost always go for the trade and influence paths)... etc. Then try this, and test it. Gal Civ 2 DA already was getting boring. Now I'm finding myself having to choose which tech paths I want to follow. I'm fighting not just military wars, but influence ones, economic ones, etc. I'm seeing incredably complex galaxies, and having to watch the timeline and the techs available to trade (or found out via spying), just to try and figure out who is doing what, and why that one economy or population or military or etc just spiked/plummeted. Plus, I like the idea of techs having consequences - maybe I want more morale, but is it worth taking a massive hit to population growth? Maybe I want to be a better spy/diplomat/etc - but can I afford the time to research that path?
Oh, I ought to mention, the idea here is to force the AI and the player to have to choose, and to invest time if they choose that path. Almost all of these are no trade/no steal. Some are alignment specific. Only thing I wish is that GalCiv2 supported:
* either/or techs - you can only choose one of tech group
* multiple pre-requisites (if you actually research all of the techs in, for example, colonization - which I now have as 25 rather expensive techs - you get XYZ production bonus)
* I like being able to set no trade/no steal. I'd be nice, though, to also have "no gift", "no demand", and "no trade, no gift, but can be demanded (ie, 'poison pill')" techs.
Most of you can pretty much figure out what the techs are from the titles - I've not yet bothered to add details or descriptions to most of them.
Anyhow, I don't have a website to host the file, but email me and I'll send you a copy of the techtree.xml file for you to try out.

Brian
brian.brianschweitzer@gmail.com