|
| Thread: Stone Skin vs Shield |  This thread is pages long: 1 2 3 4 · «PREV / NEXT» |
|
Commando
Known Hero
Unleasher of the Bloodthirsty
|
posted August 29, 2002 06:39 AM |
|
|
Nothing wrong - try repeating the fights I carried out and see what results you get. Keep us posted 
____________
|
|
Frank
Promising
Known Hero
|
|
posted August 29, 2002 06:42 AM |
|
|
Try my test and I`ll try yours? 
|
|
Wub
Responsible
Famous Hero
|
|
posted August 29, 2002 10:44 PM |
|
|
Quote:
At this point don't you think it would be worth reinstalling Heroes III? I'd guess you've missed out on quite a few interesting maps, which I shall refrain from mentioning here unless explicitely requested...
There will undoubtedly come a moment that I will reinstall Heroes III to my computer. Not in the first few months, but still... So let me request explicitly which maps you had in mind, the more so as I sense someone being modest (again  ). Does this map happen to be called 'unleasing the bloodthirsty' ? I read in a thread about XL maps that it was supposed to be a very good one (but is it playable with RoE?). I also have the feeling that we have similar map tastes since the end battle from my map resembles the one that you mentioned in the 'which spell is best with HUGE forces' thread (or something). So when I decide to reinstall the game I will surely give the map a try and maybe others that you have in mind.
By the way, I am also interested to know the results from both tests mentioned in the post above...
____________
|
|
Commando
Known Hero
Unleasher of the Bloodthirsty
|
|
posted August 30, 2002 12:12 AM |
|
|
Unleashing the Bloodthirsty? Never heard of it
Will let you know how the tests fare - as soon as I get around to them, that is. Let me know if you receive my e-mail.
@Frank: deal - will try, will post
____________
|
|
angelito
Honorable
Undefeatable Hero
proud father of a princess
|
|
posted December 28, 2004 11:13 AM |
|
|
Revived.
Interesting and funny posts i think.
Itīs a real pity Frank and Commando didnīt post the results of their tests though...
____________
Better judged by 12 than carried by 6.
|
|
pacifist
Famous Hero
|
|
posted December 28, 2004 02:35 PM |
|
|
|
I made some tests to try to find a particular case where shield doesn't add to armorer specialist reduction. Tested with lv 59 Mephala with 9999 unicorns vs Sir Mullich with 9999 Naga Queens. Both had 99 in all stats, Mephala of course had armorer expert and Sir Mullich no offence skill. Normal damage NQ did 128349 to WU, but with expert shield 89844 wich is 69,99% thus 30% damage reduction. Same with bless involved : 132629 damage reduced to 92839, same 30% reduction. With Sir Mullich 30 attack vs 99 defense from Mephala : 36671 damage was reduced to 25669, again same 30% reduction. I made same test with reduced armies 10 and 1 NQ respectively to see if there is a rounding factor somewhere, still the same : 70 damage reduced to 49 or 7 damage reduced to 5. I tried different grounds too but was quickly convinced it had no impact on damage reduction. I guess we'll have to wait more details on the battle configuration of Frank to be able to duplicate it.
|
|
Consis
Honorable
Legendary Hero
Of Ruby
|
posted December 28, 2004 05:12 PM |
|
Edited By: Consis on 28 Dec 2004
|
Awesome Thread!
Ok, this threads rocks. Angelito and Pacifist, Commando conceded to Frank. You guys must have missed it. And then I tested it myself with some fixed damage units(Naga/Enchanter/Archangels).
Frank is correct. Wub was a great bystander though, LoL. What a great thread this is. It's getting marked in my book for one of my favorites. I think Commando and Frank deserve a Qp each for the zany antics. Wub was a killer comic sidekick.
____________
Roses Are Red And So Am I
|
|
Q
Hired Hero
Moo,I say.
|
posted December 28, 2004 06:19 PM |
|
|
This is a very good thread IMO but does anyone has the time and effort to test shield and stone skin together?I would like to see the results.
____________
|
|
Nebuka
Promising
Supreme Hero
Save me Jebus!
|
|
posted December 28, 2004 06:55 PM |
|
|
Quote:
Ok, this threads rocks. Angelito and Pacifist, Commando conceded to Frank. You guys must have missed it. And then I tested it myself with some fixed damage units(Naga/Enchanter/Archangels).
Frank is correct. Wub was a great bystander though, LoL. What a great thread this is. It's getting marked in my book for one of my favorites. I think Commando and Frank deserve a Qp each for the zany antics. Wub was a killer comic sidekick.
How is Frank correct when just above you there's post that proves otherwise.
Or I'm missing something?
____________
|
|
Consis
Honorable
Legendary Hero
Of Ruby
|
posted December 29, 2004 02:52 AM |
|
Edited By: Consis on 28 Dec 2004
|
Conceded His Point Here:
Quote:
Read carefully - the Hydrae were blessed before attacking. Repeating the fights yielded the same identical results every time. Happy?
First page, 19th post down. This is him admitting he'd been using bless on the attacking units which, in turn, would equate to a higher damage reduced % by the defending unit.
Frank then ignores this post and continues to say that he "doesn't know what Commando is talking about" and that he "only knows his tests are proving correct each time".
So basically, Frank is correct and Commando had been secretly using the Bless spell to alter his own test results to support his own incorrect assumption(i.e. he only wanted to win the argument) that damage reduction can go higher than 30% even if your opponent does not have the Offense skill.
____________
Roses Are RedAnd So Am I
|
|
Commando
Known Hero
Unleasher of the Bloodthirsty
|
|
posted December 29, 2004 09:54 AM |
|
|
Quote:
Quote:
Read carefully - the Hydrae were blessed before attacking. Repeating the fights yielded the same identical results every time. Happy?
First page, 19th post down. This is him admitting he'd been using bless on the attacking units which, in turn, would equate to a higher damage reduced % by the defending unit.
Frank then ignores this post and continues to say that he "doesn't know what Commando is talking about" and that he "only knows his tests are proving correct each time".
So basically, Frank is correct and Commando had been secretly using the Bless spell to alter his own test results to support his own incorrect assumption(i.e. he only wanted to win the argument) that damage reduction can go higher than 30% even if your opponent does not have the Offense skill.
That's not quite what I was saying, mate. I used bless in order to make sure damage was maximized and always the same on every test. I realise I could've used fixed-damage units such as angels or nagas and it would've worked just the same - a mistake that had no effect on the tests' results.
A big hello to Pacifist and Nebuka
|
|
Pacifist
Famous Hero
|
|
posted December 29, 2004 10:30 AM |
|
|
Quote:
Ok, this threads rocks. Angelito and Pacifist, Commando conceded to Frank. You guys must have missed it. And then I tested it myself with some fixed damage units(Naga/Enchanter/Archangels).
Frank is correct. Wub was a great bystander though, LoL. What a great thread this is. It's getting marked in my book for one of my favorites. I think Commando and Frank deserve a Qp each for the zany antics. Wub was a killer comic sidekick.
Can you explain how you tested it? Some more details? Maybe then we can understand what's going on and how different testers give different results . As you see, in my tests only Nagas were involved with and without bless and in all cases the 30% reduction was applied.
____________
http://www.youtube.com/user/alkoriak#g/u
|
|
Commando
Known Hero
Unleasher of the Bloodthirsty
|
|
posted December 29, 2004 11:09 AM |
|
|
Ok, I just did a quick test.
Human: Ufretin all stats 99, expert in all magic schools, 600 naga queens. No other significant skills.
Comp: Orrin all stats 99, insignificant skills, 600 naga queens.
Human never attacks, only moves.
1st attack: comp does 20,700 damage
2nd attack: cast expert shield, comp does 14489 damage
3rd attack: cast stone skin, comp does 11654 damage.
4th attack: defend (+23 defence), comp does 4409 damage.
5th attack: cast expert curse, defend, comp does 4262 damage.
6th attack: cast expert weakness, defend, comp does 3653 damage.
Repeat test but replace Ufretin with Mephala all stats 99, expert armourer at level 88, expert in all magic schools, 600 naga queens.
Comp: same as before.
Human never attacks, only moves.
1st attack: comp does 3932 damage
2nd attack: cast expert shield, comp does 2753 damage
3rd attack: cast stone skin, comp does 2214 damage.
4th attack: defend (+23 defence), comp does 837 damage.
5th attack: cast expert curse, defend, comp does 809 damage.
6th attack: cast expert weakness, comp does 694 damage.
So basically we have gone from 20,700 all the way down to 694, which is slightly more than a 97% damage reduction. It would be interesting to try this giving the human angels, for instance, and seeing how blind penalisation fits into the picture.
Edit: this test was done in a rush, so let me know if you notice any flaws.
|
|
Nebuka
Promising
Supreme Hero
Save me Jebus!
|
|
posted December 29, 2004 05:05 PM |
|
Edited By: Nebuka on 29 Dec 2004
|
Frank said:
But what you have to know is a lvl 20 armorer specialist gains absolutely no extra defense from casting shield expert. Somehow, % of damage taken off is limited to a specific number. So once you reach 30%, forget about shield.
Now, unless I completly misunderstood that above, Frank is wrong as Pacifist and Commando showed. I tried as well, and you get extra defense from shield (30%).
So, Consis, have no idea what you're talking about...bless in Commando's case was there to give fixed damage, not for cheating lol.
Anyway, hola Commando, long time no see.
Making maps?
EDIT: Oh well, haven't misunderstood Frank...but looks like he did his tests on 1.0...
Anyhow I redid the test correctly and still, expert armorer specialist will take the same damage from an angel or naga attack with expert shield or without. Come with any other results and we should be looking at if you have 1.4 or some other weird version.
And about that...I have no idea. Could install it and check.
____________
|
|
Commando
Known Hero
Unleasher of the Bloodthirsty
|
|
posted January 01, 2005 12:34 PM |
|
|
Quote:
Anyway, hola Commando, long time no see.
Making maps?
Heya Nebs
Got a few odds and ends lying about - who knows? 
Cheers mate.
|
|
DonIgnacio
Tavern Dweller
newly hooked strategist
|
posted April 05, 2006 11:41 AM |
|
|
The mathematical truths behind Commando's test results
The NQ have constant damage of 30.
600 * 30 = 18000.
Against their equals they have +3 A/D with +15%.
18000 * 1.15 = 20700.
Expert shield IS a 0.7 factor in later patches.
20700 * 0.7 = 14490 (-1 for rounding?)
Expert stone skin lowers A/D by 6 to -3. Negative A/D goes in 2.5% increments. So we are at 92.5%.
18000 * 0.925 * 0.7 = 11655 (-1 again)
Now the huge effect of the defensive stance surprises me. But if it says +23 defence in the message, that is what we need to use here. So A/D goes to -26, just short of the -28 cutoff mark.
18000 * 0.35 * 0.7 = 4410 (-1 again)
The expert curse on the Nagas does little to their constant damage, just a single point. So attack number five calculates as
(600 * 29) * 0.35 * 0.7 = 4263 (-1 again)
The final weakness brings -6 to attack but the effect is just -2 for minimum A/D of -28 and a factor of 0.3
17400 * 0.3 * 0.7 = 3654 (-1 again)
As for the expert armorer skill, this is obviously another multiplier. All numbers from the first test are reduced to 19%.
Armorer special skill at this level would be 15% * (1 + 0.05 * 88) = 81%. I had this wrong before editing this post. Thanks to Ecoris for setting things straight, also with regard to the curse effect.
The resulting reduction comes from the multiplication of a rather small curse effect on NQs (29/30), minimum A/D (0.30), expert shield (0.7), and maximum armorer special (0.19). This is indeed below 4% of base damage.
This thread rocks!
|
|
night_on_earth
Hired Hero
|
|
posted September 15, 2006 05:14 PM |
|
|
Commando
Quote:
Human: Ufretin all stats 99, expert in all magic schools, 600 naga queens. No other significant skills.
Comp: Orrin all stats 99, insignificant skills, 600 naga queens.
Human never attacks, only moves.
1st attack: comp does 20,700 damage
2nd attack: cast expert shield, comp does 14489 damage
3rd attack: cast stone skin, comp does 11654 damage.
4th attack: defend (+23 defence), comp does 4409 damage.
5th attack: cast expert curse, defend, comp does 4262 damage.
6th attack: cast expert weakness, defend, comp does 3653 damage.
[...]
Edit: this test was done in a rush, so let me know if you notice any flaws.
I was getting curious when I read the change due to the curse spell on expert level. The manual is quite vague here:
"((80% of minimum damage) -1)"
- observe the brackets!
So I did some tests. Two heroes with zero attack and zero defense. Mass curse on all troops. All troops have attack = defense value except peasants vs. archangels (def-att=29 -> min damage 30% rule):
100 peasants vs. 100 peasants = 100 HP damage
1000 peasants vs. 1000 peasants = 1000 HP damage
1000 peasants vs. 2 archangels = 300 HP damage (30% rule for minimum damage)
100 familiars vs. 100 familiars = 100 HP damage
10 battle dwarves vs. 10 battle dwarves = 10 HP damage (damage 2-4)
10 griffins vs. 10 griffins = 20 HP damage (damage 3-6)
10 medusae vs. 10 medusae = 50 HP damage (damage 6-8)
10 crusaders vs. 10 crusaders = 60 HP damage (damage 7-10)
10 champions vs. 10 champions = 190 HP damage (damage 20-25)
10 archangels vs. 10 archangels = 490 HP damage (damage 50)
I tested this for the "shadow of death version". However at the end I did a quick cross check for version "restauration of Erathia" with the same results.
My conclusions:
1. The 80% are just in the manual and never made it to the game released, which were a big surprise for me.
2. Minimum damage is 1 per creature.
3. The curse spell affects just the (number of creatures x damage) part of the damage calculation. The damage is MAX(1, minimum damage-1).
ATT-DEF, armorer, offense, luck bonus and so on are applied on the result of (number of creatures x damage) as already described in another thread.
____________
night on earth
|
|
Ecoris
Promising
Supreme Hero
|
|
posted September 17, 2006 03:24 PM |
|
Edited by Ecoris at 15:29, 17 Sep 2006.
|
Quote:
18000 * 0.3 * 0.7 = 3780 (+128 over test result)
Do I just let this stand? Any difference beyond +/- 1 cannot easily be ignored. The factor of 0.7 is well proven and applied after the A/D stuff has been calculated. So what is the other factor. Well it is exactly 0.29. The cutoff is lower than we've all been believing so far!!
This is wrong. It is sad that a thread like this includes so many different versions and it is sad that the testers don't work together but actually accuse each other for lying.
And Consis wrote that he thought Frank should have a QP!
And as a warning I might add that other threads contain posts by Frank or Wub with incorrect conclusions or descriptions. Nothing wrong in that , everyone can be mistaken, but unfortunately their claims are not always corrected.
Back to the qoute. DonIgnacio did not consider that the Nagas were also cursed in the test he refers to (#6 with Ufretin). Commando did not mention this, but DonIgnacio should have thought of it before hastily posting "The cutoff is lower than we've all been believing so far!!".
The correct calculations should have been:
base damage: 18000
expert curse: -600
Damage: 17400 * 0.7 (shield) * 0.30 (maximum damage reduction caused by A/D) = 3654
Just a last note: Be skeptical when reading threads like these. If you're lucky you're just confused. If not you may end up believing things like Quote:
But what you have to know is a lvl 20 armorer specialist gains absolutely no extra defense from casting shield expert. Somehow, % of damage taken off is limited to a specific number. So once you reach 30%, forget about shield.
If it's a matter of versions, I'll have to apologize. I also did Frank's Unicorn test mentioned at the bottom of page 1:
100 Unicorns (Mephala level 20, expert earth + expert armorer).
vs
100 Naga Queens (Orrin level 1).
A/D = 18 / 24 = -6 = 15% damage reduction
Nagas attack Unicorns (not in defence mode).
Damage = 1784.
Expected damage: 3000 * 0.85 (A/D reduction) * 0.70 (armorer 30%) = 1785
Mephala casts expert shield. Naga Queens attack Unicorns.
Damage = 1249
Expected: 3000 * 0.85 * 0.70 * 0.70 (exp. shield) = 1249.5
I hope things have been cleared up a bit.
Perhaps we need a quality stamp that indicates whether or not a descrption of "how XXX works" is considered true. I often end up testing things like these myself because I haven't got any other guarantee than the poster's usual credibility.
|
|
DonIgnacio
Tavern Dweller
newly hooked strategist
|
|
posted January 30, 2007 04:04 PM |
|
|
Calculation revised
I didn't accuse anybody of lying. I was just trying to figure out the rationale behind the test results by Commando, which have been prepared with sufficient diligence.
I did indeed overlook the effect of the curse. I've corrected my post and still hope that it gives some insight into the inner workings of HOMM III.
Don Ignacio
|
|
alcibiades
Honorable
Undefeatable Hero
of Gold Dragons
|
|
posted January 30, 2007 04:37 PM |
|
|
Excellent work DonIg! With the corrections pointed out by Ecoris, it does indeed seem to match perfectly.
____________
What will happen now?
|
| |
|
|
