Are you, in fact, a pregnant lady who lives in the apartment next door to Superdeath's parents? - Commodore

Create an account  

 
New experimental version

I think the crash is caused by the overlay engine - it's swapping out the parent segment because the game wasn't complied assuming that segment is a parent for where I'm calling. Fortunately, I can put the calls outside where they are proven to be safe.
I suspect this is what prevented calling unit rating from the intercontinental procedure, too, as that is also a child segment of a likely unplanned parent.
Reply

This game I observed so weird damage rolls I feel there might be a bug behind it.

More specifically, two war bears (and two other units that didn't matter) attacked my 5 sprites.
No buffs on the bears or sprites. Bears had 6 shields because the wizard was tactician.
The first bear went down in 6 shots. So 6 shots did 18 damage to it.
The second bear on the other hand...took 5 total damage from 14 shots.

First bear averaged 3 damage taken/attack, second was 0.35 damage/attack.

That is far too much difference to be just bad rolls. (each attack involves 5 figures so the sample size is 100 total rolls...)

Did anyone experience anything similar?

(wouldn't have normally noticed it but this game it was the second time something refused to take anywhere near the expected amount of damage from sprites despite a large number a shots - almost as if that unit had higher defense.)

Edit  : Calculated and the expected damage from the attacks should have been around ~2.5. Also checked through the entire attack procedure code but found nothing wrong.
Reply

I didn't pay close attention, but I had a unit of sprites do about 50% of a naga in one shot, and then the next 3 shots combined couldn't do the same damage to the next naga.

Can you run 10000 sprites vs war bears attacks? This may just be 'clumpy' rolls in the number generator.
Reply

Well, with a 2.5 expected damage per shot, getting 0.35 average from 14 shots definitely feels wrong....but without finding a bug in the code I have no choice but to assume it was bad luck. I wonder if anyone can calculate the probability of that. If it's like, 0.5% then we're still good, if it's 0.0000005 it might be worth looking for a bug longer.
Reply

Probability of one figure doing 0 damage = all missing + 1 hitting and at least 1 blocked plus +2 hitting and at least 2 blocked + 3 hitting and at least 3 blocked + 4 hitting and at least 4 blocked = 12.96% + 34.56%*88.2351% + 34.56%*57.9825% + 15.36%*25.569% + 2.56%*7.047% = 67.6%

So one figure has a 67.6% chance if doing 0 damage.

So one unit has a 14.1167% chance of doing 0 damage.

So 9 units have about 0.00000222637681% chance of doing zero damage.

So even if 1 damage per unit is exactly average (and its not, 1 is still below average) for the first 5 units attacks, 9 units should not realistically do 0 damage.

If you want I can calculate the exact probability of 14 units doing at most 5 damage. But that would take a lot more work, and I think what I've done so far us enough to show its odd. (Since the best you can hope for is that 100% of the time 5 units will do at most 5 damage; so the 0.000002% shown above is the absolute highest the probability could be. In reality, its going to be closer to 0.000000005% but yeah, more work)
Reply

I just had a thought about the new ability to do minimal damage regardless of armor. How much does that step on the toes of sky drakes, storm giants, and death knights? For instance, I would expect that sky drakes should do more damage than great drakes to high armor targets. But the ability is based on raw damage, so don't great drakes end up better?

Numbers: against 40 armor, great drakes will do average minimum of 15-16 damage with its breath. Actual damage will probably end up being 8, which will increase to the minimum. On the other hand a sky Drake breath will do a minimum of 10 or 11; and then actual damage will end up around 9, and also increase to the minimum, making the lightning part completely irrelevant.
Reply

Quote:I just had a thought about the new ability to do minimal damage regardless of armor. How much does that step on the toes of sky drakes, storm giants, and death knights? For instance, I would expect that sky drakes should do more damage than great drakes to high armor targets. But the ability is based on raw damage, so don't great drakes end up better?

   

Sky Drake vs Great Drake - their damage output is close to equal against extremely high armor, and Great Drake was better in the first place at lower armor. Having minimal damage changes nothing here.

Admittedly against 80 armor targets the Sky Drake might do 1-2 damage and the Great Drake can't but...that doesn't matter. You can't exactly ram 25 Sky Drakes into the hero to kill it. If anything, Minimal Damage helps both dragons equally to damage things they couldn't have damaged otherwise.
At 60 armor the Great Drake was better to begin with - it did 2x3 damage while the Sky Drake only did 1x3.6.

Great Wyrm - does more than the highest minimal damage on everything, even the most armored enemies. (ok, if the Great Drake also uses the breath then the Great Drake does more - 2*8 instead of 1x9.6. Hard to compete with something that can attack twice. But on counterattacks which is the more frequent case, the Wyrm wins.)

Storm Giant - unit with closest starts and no Armor Piercing is the Great Lizard.
However, since the lizard does not do minimal damage, comparing the is meaningless. In fact, no uncommon or rare summon has the ability and very rares are meant to outclass it anyway, armor piercing or not.

Death Knight - Ineffective against extreme armor due to high figure low attack. On medium armor, they do much more damage than the minimal damage of similar creatures. Note that numbers are per figure as both abilities are applied per figure.

Math :
If a creature has X damage output, the enemy need X damage reduction to disable it.
If it has Armor Piercing, they need 2X. However, 2X will also reduce the damage from a creature with 2X attack strength to zero.

So an Armor Piercing creature is equal to a non-armor piercing creature with double attack power, but in fact inferior to it - they are equal at dealing no damage, but the advantage goes down the less the damage is reduced - if the damage reduction is only X then the AP creature will do 0.5X damage after reduction but the non-AP will do 1.5X.
Meaning the AP creature needs to have significantly more attack power than half of the nonAP creature to be better.

This isn't true for Sky Drakes, Storm Giants or Death Knights - their per figure attack power is low compared to most other very rare creatures.

It's only true for Great Wyrm.

Indeed, the Great Wyrm now has less advantage over Great Drakes regarding damage dealt per attack - if a 50+ armor target does come up, the Drake will now do more damage to it than it was able to while the Wyrm is unchanged. But the Wyrm gets to teleport there and hit twice on turn 1, the Drake does not and that's what matters vs a 50 armor hero army.

So I think it's fine overall.

Edit : For most targets with realistic amounts of armor, both AP and nonAP creatures did more damage than the minimal and thus are unaffected. For heroes specifically, I think the question to ask is not "Is it ok that non-armor piercing creatures are now able to damage them almost as well as armor piercing creatures?" but "Is it ok that heroes don't take any damage at all from non-armor piercing creatures?" And for that the answer is obviously no. Most realms and wizards don't have the option to use armor piercing creatures and the human player will keep the heroes away from them even if an AI does have any. I don't like this solution very much but I don't know better.
Reply

Exp 14 is now up for download! I started one game but I can only hope everything works as intended, as the changes are quite major. So far the AI seemed to not missing any good attack chances due to the changes.
Quote:EXP14
-Fixed : Smithy shows unlocking Warship on races that cannot build it.
-Fixed : When no unit is selected, the Spell button might offer random unit abilities to cast.
-Herb Mastery now works on Death and Undead units on the overland map the same way as in combat.
-Stream of Life now works on Death units - but it still won't heal undead.
-New AI overland intercontinental attack procedure - now uses real unit ratings instead of simplified and might even process faster.
-AI intercontinental attacks performed by 4 or fewer units require twice as much army power as normally sized stacks, 2 or fewer units require 4 times the power
-AI land attacks performed by 4 or fewer units require twice as much army power as normally sized stacks, 2 or fewer units require 4 times the power
-The PWR cheat will now give players all Arcane spells again, except the Spell of Mastery.
-Removed spell : Nature's Wrath
-New spell : Fairy Ring. Whenever another player casts an expensive spell, the owner of Fairy Ring gets a free summoned creature of any type they can summon.
-Clairvoyance has been renamed to Seismic Mastery and now has an additional effect : Whenever you cast a spell that costs 400 or more, you can target a city with an Earthquake effect for free. Spell casting cost has been adjusted. Power production of nodes reduced to 15.
-The AI will not cast Earthquake manually if Seismic Mastery is in effect.
-Adjusted AI overland casting priorities to correctly react to Fairy Ring instead of assuming it's still Nature's Wrath.
-The AI has about halved priority to cast Disenchant Area but uses twice as much power for each.
-Assigned AI garrison priority to Water Elementals
-Fixed Exp9 bug : AI priority for casting Stasis as anti-doomstack measure is addressing memory wrong, causing data corruption.

I would like the next update to be a non-experimental 5.0 version. So please, post as much information about how you perceive each difficulty level, or new AI feature as possible. I obviously can't play enough games all by myself but I don't want to delay the release another 2-3 months just for testing. Ideally, I'd like 5.0 to be released in 2 weeks top.
Reply

I'm playing Exp13 and just noticed that healing Torin doesn't work. I tried it 3 times in combat and he stayed on 7 health each time.
Reply

Were you trying to heal irrecoverable damage? (Wave of Despair, Crack's Call, etc)
Reply



Forum Jump: