A couple of questions...

Erasmoth

First, why not use two bards? One is a skald and the other is a vanilla Bard. I made a test game and realized that their songs do indeed stack, and having both songs up at the same time would turn any party member into a full fledged killing machine for most of the game.

Second, question is about shapeshifters. As i have found on the forums, many people consider this be be OP, then UP as the game progresses. I like the idea of a druid that can turn into a killing machine, but would like it to be viable. Are there any mods out there that can balance out the shapeshifter? I couldn't find any like the Rogue Re-balancing ones

CaptRory

As for the bard question,

There are only six characters in a party.
There are a lot of classes you can take.
A Skald and a Bard would team up very well but that's two guys you have mostly standing around singing during fights. You could make that more viable with a strong focus on summoning which bard music also buffs and which are also buffed by group magic.

Erasmoth

What are the summons you would consider buffing? I have tried a couple with my skald group but they just fall over like everyone else haha

CaptRory

Well, whatever is on hand really. You'll get better and better summons as the game goes on. The idea is to spend resources on summoning to make up for the people not actively fighting. If they fall over real easy they still soaked up stuff for your party.

Shikao

This summoning analysis is for HoF, but should still give you idea which summons spells are better to use.

Lord_Tansheron

Simplified comparison:

f1[1-6]=(6-x)*(x*a)

where x = number of buffers and a = average damage they add to all party members

f2[1-6]=y*b

where y = number of damage dealing characters and b = average damage they deal per hit

Obviously you can see that f2 is entirely linear, i.e. more damage-dealing characters increase your damage output in direct proportion.

f1 is different, as it's a peaked curve (with peak at x=3, obviously) that ends at 0 (obvious, 6 buffers 0 damage dealers = 0 damage).

Now, if you made an overlay of both curves, you'd see certain intersection points where more buffers become better / worse, depending on the values of a and b in your respective party.

[Ok technically these are all discrete values and not continuous ones so it's not an ACTUAL curve, but let's not go into the details TOO much :P]

Anyhow, the comparison between the two will yield optimal numbers, contingent on the actual values of the buff-gains resp. the damage output. This is a simplified version, of course, and intended for illustration of the PRINCIPLE, not for actual plotting of values. The reason being, of course, that neither buff values nor damage values are uniform (at least in most cases), so you'd actually have to do a summation of the individual respective values rather than averaging them all out - and then you'd have to look at what class brings how much and go through all the permutations etc. etc. An exercise for those who want truly accurate data representation, but probably a bit too much work for most :P

I find that such reductions to simple principles help me understand the mechanics of an issue better, and maybe that'll help you, too. This is a question I've seen come up on these forums constantly, so it seems there is definitely some need for clarification.

TL;DR: Don't overdo it with the buffers, try to find a good balance. Exact formula depends on your particular party choices.



[Disclaimer: I suck at math, big time. I'm a language major, it comes with the job. If you find something blatantly wrong, please let me know. Gently.]

EDIT: Thanks @Shikao‌ for pointing out a rather glaringly obvious mistake :P

Shikao

Lord_Tansheron said:

f2 is different, as it's a peaked curve (with peak at x=3, obviously) that ends at 0 (obvious, 6 buffers 0 damage dealers = 0 damage).

It's F1 that's different =]

Lord_Tansheron

Fixed! Thanks!