So from the balance suggestion thread, and several other threads and statements, I've finally put togather something that seems to me like a great idea that might, if not too difficult to code, make the game a much more enjoyable expirience for all sorts of players.
One of the things that always bothered me about the Beta is that the Gods came into play late and erraticaly, but are at the same time most important and a key to beating sucesfull dungeons. God's coming into the beta slowly, I coldn't really complain, but now they're all here, I can be constructive about my observations. My current greivances with god's related to gameplay and story progression as well as the learning curve are the following:
1) There is no tutorial on gods, and the way the story is sturctured there can't be one.
2) They are unlocked by subdungeons, with the effects of running into one often causing you to abandon a run. The most egregious examples are probably EM, and especially Pactmaker, who gets to screw 3 of your runs while denying you a subdungeon. Dracul would also be a probelm if he wasn't bugged. Yes, you can avoid some of this by checking out the subdungeon the last thing you do, but it's counter-intuitive and really, really annoying.
2.1) Some of their subdungeons give out more resources than most other subdungeons, main offenders here being TT and Mysteria - if you choose not to unlock TT every time you run into his subdungeon you get a whole map worth of extra regen. If he was a low power god, it would make more sense NOT to unlock him then.
2.2) If you miss out on a subdungeon, or don't unlock the god the first time you enter it, it takes scumming to find that god again, and with gods that are deffinite "start the run over" types, this is more than annoying.
2.3) Some of the Dungeons, like Binlor's one, are easy to completely missunderstand. Ive herd people who've ran into it posting on the forums asking for how to unlock Binlor.
3) With gods baing random, the only way to get really proficcent (or at all profficent) is to either scum for certain gods repeatedly or unlock only the ones you want to test. Since some gods have harsh penalties and steep boon costs, figuring out how to get mileage out of them can sometimes take ages, the way I only recently "discoverd" GG, or how people are only recently discovering TT's potions.
3.1) With so little reliable practice time with gods, the ones that lash out with bad punishments screw players over in runs that would've otherwise be fine (for example taking Lifesteal from Drac and switching to GG thinking you can still fight undead because you don't have poison). If such things happen often, and with some gods they do, the "difficult" gods start being avoided.
3.2) What also happens is that players can't gauge the difficulty of certain quests or merits of certain strategies because they aren't aware of resources at their disposal, the way TT's potions are unbelievably good but haven't been discovered for ages. The only reason god manipulation is at the top of the learning curve is because of "fake difficulty" caused by the way gods are introduced into the game.
3.3) This is further excerbated by the use of Agnostic Collar to desecrate the altars you don't like in favour of the ones you've managed to learn to use.
3.4) What also happens is that the whole community seems to "discover" boons or deities that have been ridiculous all along "all of a sudden" - with such a small window of opportunity to experiment, powerfull stuff keeps lying around unused, and once some madperson goes out and uses it, everybody can if the guy wrote down the algorithm. Once the game goes live, such stuff will be all over the wiki's, and we'll be in a paradoxical situation that beta testers had a harder time helping balance the game than regular players have exploiting overpowered features they can read up about on a wiki. A good examle would be the way I "found" just how stupid is the amonut of piety you can reap from GG by regen fighting a poisoner - and that was easily fixed in a matter of days. But the method was there for months.
3.5) The previos point is also the reason we, as a beta testing community, can't properly gauge optimal power levels for stuff - someone who got to learn how to use abuse TT and someone who got to learn how to abuse GG aren't playing the same game, and again, once it's all in a wiki we'll all look like a bunch of dopes.
My very timid and actually in no way "high and mighty" or arrogant suggestion.
Remove the "god" subdungeons from the game, and have a series of "idols" or "scriptures" or whatever planted around regular dungeons in all 4 directions of the map. Make them always there unless found. Have one for each god.
You can make quests to get them from the exlorers guild, as the explorers guild has a big part to play in all this.
Then, when the player brings the idol back, say that the priests have managed to construct an Idol of whatever deity that the playe can practice "harnessing it's powers" or whatever from.
And then give the player a god themed puzzle pack for each idol, that would allow newbies to get to know a god before they let him start spawning in their dungeons.
You can also help veterans and dedicated testers reduce their scumming time while testing out complex strats by allowing the final puzzle be a simple "unlocking" - completing it brings a gold or some other reward to goad newbies into bringing the god in, and not completing it keeps the god locked to reduce god scumming. The unlocking ceremony can be some version of the original "god subdungeon".
You can also work in the cusader unlock require you to unlock 4 gods and then go find his subdungeon, and you can make sure everyone unlocks all of them by informing them that it makes the vicious Gaan'Telet a lot easier.
If noone has time or energy to come up with these puzzles, I volunteer, as I think all this would be a great, great thing for the game, and save everyone a lot of time and grief. And I seem to be good with puzzles
Before all this is done gods need balancing and reshuffling of abilities to "fix" them in place as features that puzzles can be made around. But the sooner done the better, as I think all of what I've said here is reasonable, possibly very beneficial to the game in many ways, possibly even necessary. The only thig I'm not certain of is the amount of coding needed, but I offer to help in any possible way to get this idea implemented.
I've been working on it for a while, and gave it a lot of consideration.