PDA

View Full Version : Which items work well / don't work for Boss battle's?


WhYdOyOuThInKtHaThApPeNeD
02-18-2010, 07:42 PM
Does Cloak of Shadows function on the Boss fights (Driller, Spider, Frog, etc.)?
My bet is that it doesn't, too good to be true. (meaning it does nothing, not applicable?)

Does anyone have a complete list of what items function and don't function on the boss battles? If not I encourage people to post their experience with what items worked well and which did nothing. Or if everything works then which items you found to be the most useful for fighting all boss's in general or specific boss battles.

Obviously items that boost units attack / defense function. I assume items that boost resistance types also function.

Also, is there a guide that lists all the different types of attack and what type damage each Boss does?

DGDobrev
02-18-2010, 08:04 PM
There is such a guide, it's in the KBAP info thread in the bosses section.

As for what works, Ive noticed the new sword from gift bad that reduces physical resistance by 50% never drops it below 0%, and I didn't see much difference against bosses.

I wish there was a way to check the boss' stats dinamically while doing the battle. It would make things a lot easier.

WhYdOyOuThInKtHaThApPeNeD
02-18-2010, 09:58 PM
How is it 17 Trolls do around 3000 physical damage to the Driller while 14 Cyclops with the same attack rating, also doing physical type only do around 50-90 damage? It says the base attack is the range attack for the cyclops. Must be something I don't know about the Cyclops. I've never seen such a dramatic damage difference when using the Cyclops.

It's not just the range aspect since Evil Beholders & Arch-Mage do A Lot of damage to the Driller (granted magical type). And it's not just the physical type since the Troll does A Lot of damage. What's special about the Cyclops range attack? I have noticed the Cyclops generally does do more damage when using his regular attack melee as opposed to range, but not that dramatic a difference.

...

Ok it's gotta be the Driller. After the Cyclops consistently doing super low damage for several rounds now it's consistently doing 1500-2000 damage. But now the Demons which usually do around 1500 damage are consistently doing around 50 damage. If this is some special thing about the Driller it might be good to add to the info thread about him. Otherwise maybe it's some game mechanic I don't know about?


http://img683.imageshack.us/img683/8941/cyclopsdoes82todriller.jpg (http://img683.imageshack.us/i/cyclopsdoes82todriller.jpg/)

WhYdOyOuThInKtHaThApPeNeD
02-18-2010, 10:19 PM
hmm, I've read the info thread a lot, guess I missed it, but now I see it.

Yeah, it would be really good to be able to access a complete stat screen for bosses (and Gremlins).

DGDobrev
02-18-2010, 10:42 PM
The answer to your question lies in the total amount of HP the driller had left on this position. Looking at the screenshot, it changes positions, so the cyclopses most like took ot the rest of his health with that 82 damage hit.

WhYdOyOuThInKtHaThApPeNeD
02-19-2010, 12:14 AM
Seriously crazy, please check this out, I really want to know what's the deal!

So I did an attack run on the Driller with 1 Troll stack. Everything exactly the same each time except 1 time with the Cloak of Shadows, the next time without the Cloak of Shadows.

Here is the damage done by the Driller on first run with cloak:

347
878

355
412

912

368
929

384

379

364
941

346
942

Here is the damage done by the driller after reloading and without the cloak

347
878

355
412

912

368
929

384

379

364
941

346
942

So apparently the cloak is useless for the Driller and probably other bosses.

But this really proves something else I've been noticing. Certain things don't seem so random. It's impossible that the damage done by the Driller on both runs would coincidentally be the exact same for each hit for every single hit.

I've seen this in many other cases. I often play the same battle over again several times. Sometimes to get a no loss or in case something really bad happens. But usually I do it just to try different things to learn how things work and because some battles are fun. I've got a whole collection of save points just before good battles.

Sometimes I'll do the exact same first move in a battle. For a Red Dragon once I remember it did the exact same damage every single time no matter how many times I did it. Got the same critical hit every time.

So what is up with that?

DGDobrev
02-19-2010, 12:23 AM
Well, if you do the same thing in all situations, the outcome will be the same, because the AI playes the same way and does the same calculations. This is how you can mold the game to your liking. You change something - like waiting with a unit, digging after a certain unit has/hasn't moved - and there you go, the game suddenly changes.

Your numbers seem accurate as well. When the boss retaliates, it does 2.5 times less damage, which is clearly shown here.

WhYdOyOuThInKtHaThApPeNeD
02-19-2010, 12:32 AM
re: the Cyclops damage. I guess it must be as you say. I've played the battle over now a few times and it's never happened again. It tricked me the first time because it was the first time I went into battle against the Driller with the Cyclops and it was the first and only attack I've seen the Cyclops do vs the Driller and the very next attack by the Cyclops did the exact same thing, about almost exactly the same amount of damage. So it looked like that's what the Cyclops is going to do vs. the Driller, being those were the fist 2 attacks I've seen the Cyclops do vs. Driller. (I think my very first run through the game I used Cyclops, but I forget), so I was like how can that be. But then it just never happened again. weird.

Then as soon as the Cyclops starts doing normal damage, suddenly the Demons are doing what the Cyclops was doing, about the same quantity of damage, 50-80, twice in a row. So it tricked me to think maybe there is some special defense the Driller has sometimes.

On the next few runs I occasionally saw the Demon units do a very low quantity of damage, but that's when the stack was almost wasted away. So that explains that. But the Cyclops never died at all so it must be as you figured for the Cyclops. Anyway what a trick of the mind.

Also it tricked me because vs. regular units even if there is only like 10hp of life left in the stack, an attack may do 10,000 points of damage to end the stack. Not used to the boss only taking exactly how much life it has left. I've only played the whole game through a few times and don't often replay the boss battles.

But that double set of exact same damage done by the Driller in a row isn't a trick of the mind, that has some explanation in game mechanics I pretty sure.

WhYdOyOuThInKtHaThApPeNeD
02-19-2010, 12:43 AM
because the AI playes the same way and does the same calculations.

It may play the same making the same choices in response, but isn't the damage based on a certain range of possibility? For the Royal Griffin it's doing a possibility of 20-30 damage, so it may do 27 points of damage or it may do 22, like rolling a dice. Each time a battle is played over the damages should come out random. This is how I thought. But it's seeming to be otherwise.

Occasionally it does have a different figure. That Red Dragon exact same damage figure and always a crit, after many replays it did eventually not do a critical hit on the move where it always did the same before. So it's not a fixed thing either.

It's as if the possibility of something happening is narrowed down to an almost guaranteed response based on everything that's happened in the game up till that point or something. It's like once the battle starts so many factors are already decided with very little room for other possibilities. Like how almost everything in the whole game is decided right at the start after you create the hero. It's like once the battle starts it calculates, ok if this unit attacks this unit then this will be the result, done stored in memory, if that move takes place, played back from memory what was already decided.

Also, the figures for damage in that set look like totally random figures. They aren't rounded or consistent multiples of 5 or anything. When he does his full damage attack it's a very different number in the 8-900 range each time, same for the retaliations. So it seems like it's rolling a dice somewhere.

Also the possibility of a critical hit for a unit may be 11%. But when the battle is replayed over and over the first blow is always a critical hit, so how can it always make that 11% every time. Almost every time. Often enough that it's not just an 11% chance that's factoring it.

WhYdOyOuThInKtHaThApPeNeD
02-19-2010, 01:38 AM
I've played it through a bunch of times. It always gives the exact same calculations if I make the same moves each time.

78 Guard Droids summoned, always 78

Guard Droids do 172 damage

Troll does a critical blow for 6545 damage killing 65 droids

Guard droids do a critical blow for 61 damage

Troll does 4194 damage to droids killing 13

It always gives the exact same figures for everything, every single time, no matter how many times I play it. Always making those critical blows each time.

--------

For the damage done by the Driller, I've run through the first 3 strikes many many times. It always does:
347
878
355

This is the interesting part:

If after the 878 strike. I move the Troll and hit the Driller from a different angle. When the Driller strikes back it does 346 damage instead of the usual 355.

This is also consistent. If I always move the Troll to that spot and hit from that angle instead, the Driller will always do 346 damage on the 3rd strike instead of 355. Always.

MaroonMaurader
02-19-2010, 02:21 AM
That is true for basically all issues which require a call to the random number generator in this game (damage rolls, critical and evasion rolls, AI decision-making, creature-spawning item and scroll results...). Load the same game twice, do the exact same thing, and you get the exact same result. Load the same game twice, do the exact same thing except for something that seems like it *ought* to be irrelevant, and you might get different results.

I don't know how they actually programmed it, but it looks to me like the save-game file probably saves your random seed in it as well. In that case, you ought to be able to also change the damage and results by e.g. swapping an army to reserve, using a griffin egg then dismissing the griffin, and swapping that army back from reserve just before the battle.

dashcunning
02-19-2010, 03:45 PM
The game is just a deterministic computer program when you get down to it. Given the same input (troll moving to a certain square and attacking a certain thing), the output will be the same. Random number generators are just deterministic programs too and aren't truly random. The randomness they are supposed to produce isn't random, just close enough (or not as the case may be here and in many other games). The fact that the numbers are the same from reload to reload just shows that the RNG isn't based on anything that changes from reload to reload, such as a real time clock.

Zechnophobe
02-19-2010, 05:11 PM
The game is just a deterministic computer program when you get down to it. Given the same input (troll moving to a certain square and attacking a certain thing), the output will be the same. Random number generators are just deterministic programs too and aren't truly random. The randomness they are supposed to produce isn't random, just close enough (or not as the case may be here and in many other games). The fact that the numbers are the same from reload to reload just shows that the RNG isn't based on anything that changes from reload to reload, such as a real time clock.

Well, but it is obviously done on purpose. This is very common with strategy games. For instance, Civ4 uses a seeded RNG where the seed for the same 'effects' use the same progression. For instance, you could get a RN for an 'attack' and then grab a random hut (like a chest) and then reload and do them in opposite order, and get the same thing.

In King's Bounty a lot of it isn't even randomized during play, it is simply fixed at start. The only things that are pseudo random are AI tactics, board make-up and 'egg' items.

dashcunning
02-19-2010, 07:20 PM
Gah, the one time I don't Ctrl-C my post, the board loses it.

Anyhow, I'm not sure where I stand on the "randomized at start and predetermined thereafter" vs "randomized during the game". On the one hand, in game randomization allows for and encourages, to some extent, reloading over and over to get the best result. On the other hand, predetermined means if you ever reload, you know exactly whats going to happen. That encourages testing everything first and then restarting for an optimal playthrough.

Didn't one of the Civs (or many of them?) have an option to switch from the default predetermined nature to the constantly randomized way? I recall finding that much more palatable. Probably due to too many infuriating tank losing to spearman type situations. Reloading easily fixes a flaw (IMO obviously) of basic game mechanics.

pavned
02-19-2010, 07:25 PM
For Civ, I think they really put a lot of efforts in the Civ4 to temper the Tank vs Spearman win by spearman situation.

But I do remember that in Civ 2, everything "randomized" wasn't