Is it possible... *possible stat manipulation*

Recent

Hottest

Forum Index » General Discussion

Author

Message

Oct 21, 2011 12:49

Subject: Is it possible... *possible stat manipulation*

xxxFISHERxxx

Messages: 4,
Joined: Oct 21, 2011,
Offline

For a +10 walker on regular ground to attack a +5 Plasma tank on "hilly" terrain and only hit for one damage?

This happened yesterday in a rated game. I have been playing a long time and have never seen this happen. If it IS possible I need to apologize. If it is not, I have a screenshot that i will take the time to post and report these opponents.

Thank you for your responses!

Oct 21, 2011 16:26

Subject: Re:Is it possible... *possible stat manipulation*

nfong

Messages: 576,
Joined: Mar 07, 2010,
Location: where it's fun
Offline

Here's the formula for you. A dev posted this a while back.

First, the variables:

A = Attack strength of the attacking unit
Ta = Terrain modifier for the attacking unit
D = Defense strength of the defending unit
Td = Terrain modifer for the defending unit
B = Gang up bonus
H = The attacking unit's health points.

Next, the formula:
p = 0.05 * (((A + Ta) - (D + Td)) + B) + 0.5
if p < 0 then p = 0
if p > 1 then p = 1

This is how to figure out the damage:
1. Pick one unit to be the attacker and the other to be the defender.
2. Use the formula above to obtain p.
3. Take H and multiply by 6. This is the number of random numbers (r) generated between 0 and 1. For every r < p a hit is counted.
4. The total number of hits divided by 6 is the number of damage the defending unit will receive.
5. Switch roles between the units (attacker becomes the defender and vice versa).
6. Use the same formula above.
7. Once both units have attacked the damage points are calculated into their health points.

I will use a quick example:
Underling vs Marauder both on Plains
If you use the formula you see that the Underling has a 30% probability of dealing damage for each of its health points whereas the Marauder has a 75% probability of dealing damage for each of its health points.

If the Underling were on a Mountain though, the numbers change to 40% for the Underling and 55% for the Marauder.

That's it!

When life gives you lemons...

"TAKE YOUR DAMN LEMONS BACK, LIFE. I DON'T WANT YOUR STINKIN' LEMONS."

Oct 21, 2011 19:02

Subject: Re:Is it possible... *possible stat manipulation*

hydrogen

Messages: 19,
Joined: Apr 04, 2011,
Offline

I dont have a pen and paper in front of me. And that math looks a little difficult, does that math give you a range or an average of what damage will be dealt? Has anyone ever seen a full health walker only deal one damage, when both units are on nuetral terrain? I have tested this well over 200 times now, and i normally deal 2 or 3 damage, with an occasional 4 damage being dealt. Can anyone give us insight? Because me and fisher have never seen this, and were not exactly newbies...maybe my copy of uniwar is different then everyone elses, but if the most expensive ground unit in the game is truley capable of only doing one measily damage, then i think that ill probably quit playing this game.

Oct 21, 2011 19:40

Subject: Re:Is it possible... *possible stat manipulation*

Sphink

Messages: 120,
Joined: Dec 22, 2009,
Location: USA, Arkansas
Offline

Well first of all a 5 health is just as strong defensively as a 10 health tank. Attack wise, the 5 health is weaker.

And it's my opinion that if the average attack delt to a plasma tank by a full health walker is 2-3 with a "random number" hit of +1 for four... I dont see why the random number couldn't go the other way on a 2 hit and -1 for a hit of one... Make since?

The average is 2-3 then the random number generator might bump a 3 hit up to for, or a 2 hit down to 1..

I never tested this out and that is purely guesswork. But when I here the average is 2-3 then I have to think random number generator could affect it +or- 1 point... That's my opinion, but I never did a single test.

Good luck, Have fun!

My biased opinions are open-minded.
-Sphink

Oct 21, 2011 19:50

Subject: Re:Is it possible... *possible stat manipulation*

hydrogen

Messages: 19,
Joined: Apr 04, 2011,
Offline

Yes what you are saying makes sense fully and i already thought of that, but my hundreds of test beg to differ. Thats it, im gunna do a thousand test, i want to get to the bottom of this.

Oct 21, 2011 19:51

Subject: Re:Is it possible... *possible stat manipulation*

Sphink

Messages: 120,
Joined: Dec 22, 2009,
Location: USA, Arkansas
Offline

After I did a little offline testing on bog. Building plasmas with one and walkers with the other. I had a 10 health walker hit a 4 health plasma for 1 point... I don't believe you tested this over "200 times" I got my hit after probably 10 rounds... I'd appopoigze.

Good luck, Have fun!

My biased opinions are open-minded.
-Sphink

Oct 21, 2011 20:13

Subject: Re:Is it possible... *possible stat manipulation*

hydrogen

Messages: 19,
Joined: Apr 04, 2011,
Offline

After one thousand tries i finally got one damage to a tank. Sorry to hectopascal for the accusations.

Oct 21, 2011 20:24

Subject: Re:Is it possible... *possible stat manipulation*

nfong

Messages: 576,
Joined: Mar 07, 2010,
Location: where it's fun
Offline

Yeah just figured out the formula, there definitely is some chance of hitting a plasma for 1 damage.
Basically in the equation p = 0.4, and there are 60 tries to get a number > 0.4, with each try ranging between 0-1.
After you find the total number of tries >0.4, you divide that by 6 ,and get the damage done.
So therefore, in your scenario, only 6/60 randomly chosen numbers were >0.4.
Not very likely, but seems possible.

When life gives you lemons...

"TAKE YOUR DAMN LEMONS BACK, LIFE. I DON'T WANT YOUR STINKIN' LEMONS."

Oct 21, 2011 20:37

Subject: Re:Is it possible... *possible stat manipulation*

Sphink

Messages: 120,
Joined: Dec 22, 2009,
Location: USA, Arkansas
Offline

Yes, it is pretty unlikely. But it will happen. I continued testing and haven't got another in about 50+ attacks... But I did get the one which means it is completely possible. Now if it happened like 3+ times in a game... Then you MIGHT have something. But even then the odds of winning the lottery are pretty unlikely too. Lol someone is gonna win.

Good luck, Have fun!

My biased opinions are open-minded.
-Sphink

Oct 21, 2011 21:11

Subject: Re:Is it possible... *possible stat manipulation*

hectopascal

Messages: 4,
Joined: May 06, 2011,
Offline

Thank you, and apology accepted!
But I still insist its due to our *brilliant* strategy rather than your bad luck that we are winning

PS. also thanks to nfong and Sphink for testing it out.

Oct 23, 2011 21:40

Subject: Re:Is it possible... *possible stat manipulation*

waxoid

Messages: 442,
Joined: Aug 07, 2010,
Location: Seattle, WA
Offline

for the math nerds... pretty sure p = 0.3.

Take the difference between attack and defense (4, here) and each point is 5% moving away from the default of 50% (0.5) chance of hitting. A p = 0.3 with 10 chances to hit (hit points) gives a most likely hit of 3 (by a modest amount... fifty-some percent) and the next most likely is 2 because it rounds down. (The rounding down also accounts for the fact that a 1 hit point unit almost never does damage unless it is p = 1 i.e. attack - defense >= 10, usually meaning an attack against a healer.) The formula multiplies then divides by 6 to reduce variance, but it doesn't reduce so much that a 1 or a 4 aren't possible.

Forum Index » General Discussion