.com.unity Forums
  The Official e-Store of Shrapnel Games

This Month's Specials

Raging Tiger- Save $9.00
The Star and the Crescent- Save $9.00

   







Go Back   .com.unity Forums > Illwinter Game Design > Dominions 3: The Awakening

View Poll Results: Do you think that MA Ulm is underpowered?
Yes 52 85.25%
No 9 14.75%
Voters: 61. You may not vote on this poll

Reply
 
Thread Tools Display Modes
  #1  
Old September 22nd, 2007, 10:19 PM
Tuidjy's Avatar

Tuidjy Tuidjy is offline
Major
 
Join Date: Apr 2004
Location: La La Land (California, USA)
Posts: 1,244
Thanks: 0
Thanked 30 Times in 11 Posts
Tuidjy is on a distinguished road
Default Petar\'s \"Range Weapons\" flamebait

> Arbelests simply do the most damage per combat round, as I have shown, so it's
> not up for debate.

CUnknown, you are an arrogant prick, but that's not the problem - after all, so
am I. The problem is that you are wrong, and by touting your misconceptions, and
generating as much noise as everyone else together, you may deceive people into
thinking that the community is more divided on this subject than it actually is.

You have not shown [censored]. First of all, your numerical analysis is totally
worthless for every single reason under the sun. I cannot even accuse you of
being deliberately obtuse, because your bad assumptions swing both ways - some
lend power to your argument, some fail to address the arbalest's strengths.
No, you are simply ignorant. And you analysis sucks. I am repeating myself,
and typing slowly, because I know you do not get things right away.

First your analysis reduces everything to a worthless DPR value. According to
you the damage per round of a short bow firing at a protection 10 unit is 0.
When you cannot deal negative damage, and your average is 0, all your values are
zero. And all this time I have been building short bow archers...

Second your analysis shows that you have no understanding of how damage is
computed. You are grossly underestimating crossbows and arbalests. I'll
enlighten you, no worries. Read on.

Third, disregarding precision, range, rate of fire, presence of shields, and
army orders is lame. The assumption that your opponent's as bad as you are
is also unjustified.

And last, production capacity and marching speed does matter. The high resource
cost and low strategy movement devalues arbalests. Real players fight real
battles, and in real battles, you use only what you manage to bring to the field.

OK, I'm almost done with insulting CUnknown...

CUnknown, I once crushed your Ulm with the race you believed was weakest,
in 15 turns, without blessing or tramplers. I am ready to do it again,
with any Dominions II land race. If you intend to answer this post, please
do it by proving my math wrong, or by picking up my challenge. I'm tired
of non-substantiated nonsense.

Now, I'm done. The rest will be worth reading, I promise!

Lets remember how damage is computed. Once a hit has occured, both the damage
dealt and the armour value get two open ended dice added to them. The average
value for such a roll is 8.5 (I will not back up my math here. Anyone who doubts
it will have to bathe in my vitriol in a different post)

Thus when firing at a protection 10 unit, damage 10 short bows will fail to
do damage about half the time. 18.5 vs 18.5 - it's a wash. Now consider a
damage 10 crossbow. The average case is 18.5 armour-piercing (AP) damage versus
13.5 armour. Five points of damage are dealt when the rolls match.

Next I'll throw some numbers at you. I will consider five armour levels, and
examine how four different weapons affect them. I will be tracking kills,
no damage hits, and the average damage for all remaining hits.

Armour levels: 0 (none), 10 (light), 15 (heavy), 17 (elite), 20 (black plate)
Weapons: short bow, longbow, crossbow, arbalest
The targeted enemies are assumed to have 10 hit points.

First case: Unarmoured targets. Remember, we are tracking only hits

short bow (10) ---- Kills: 46.0% -- No damage: 6.7% --- Hits: 47.3%(6.79)
longbow (13) ------ Kills: 68.4% -- No damage: 3.2% --- Hits: 28.4%(7.09)
crossbow (10AP) - Kills: 46.0% -- No damage: 6.7% --- Hits: 47.3%(6.77)
arbalest (14AP) --- Kills: 74.4% -- No damage: 2.5% --- Hits: 23.1%(7.12)

Once we adjust for rate of fire, it becomes clear that against unarmoured targets
arbalests are simply abysmal - about three times worse kill rate than longbows,
and nearly twice as bad as short bows.

Second case: Lightly armoured targets (armour 10)

short bow (10) ---- Kills: _5.2% -- No damage: 54.0% --- Hits:40.8%(4.07)
longbow (13) ------ Kills: 10.6% -- No damage: 31.6% --- Hits:57.8%(4.67)
crossbow (10AP) - Kills: 16.8% -- No damage: 20.7% --- Hits:62.5%(5.31)
arbalest (14AP) --- Kills: 38.5% -- No damage: _8.4% --- Hits:53.1%(6.59)

After adjusting for rate of fire, the arbalest kill rate is slightly better
than that of the longbow. But if we combine the kills and hits that did damage,
the arbalest is twice as bad. And the latter is what determines whether most
enemies will break. Furthermore, when you have three times the hits, and the
average hit is 4-5 points of damage, the kills add up. Thus, longbows and
regular crossbows soundly beat the arbalest in this case.

Third case: Heavily armoured targets (armour 15)

short bow (10) ---- Kills: _1.5% Misses: 83.2% Hits: 15.2%(3.80)
longbow (13) ------ Kills: _3.2% Misses: 68.4% Hits: 28.4%(3.93)
crossbow (10AP) - Kills: 10.7% Misses: 31.6% Hits: 57.6%(4.17)
arbalest (14AP) --- Kills: 25.6% Misses: 13.4% Hits: 61.0%(5.52)

Finally, the crossbows start to shine. Even adjusted for rate of fire, the
bows cannot compare. But the regular crossbow still has a much better combined
total for kills and hits than the arbalest.

Fourth case: Knights, elite infantry (armour 17)

short bow (10) ---- Kills: _0.9% No damage: 89.3% Hits: _9.8%(3.73)
longbow (13) ------ Kills: _1.9% No damage: 79.2% Hits: 18.9%(3.85)
crossbow (10AP) - Kills: _8.4% No damage: 38.5% Hits: 53.0%(3.92)
arbalest (14AP) --- Kills: 20.7% No damage: 16.8% Hits: 62.6%(5.18)

After adjusting for rate of fire, the crossbow's combined total is still higher
than the arbalest's. The arbalest may be better in this case, as even when the
crossbow does wound, the damage is lowish, and even multiple hits will not result
in many kills. But it's close. 4.2% kills vs 6.9 and 26.5% wounds vs 20.9%.

Fifth case: Black plate of Ulm (armour 20)

short bow (10) ---- Kills: _0.4% Misses: 94.8% Hits: _4.8%(3.67)
longbow (13) ------ Kills: _0.9% Misses: 89.3% Hits: _9.8%(3.73)
crossbow (10AP) - Kills: _5.2% Misses: 54.0% Hits: 40.8%(4.08)
arbalest (14AP) --- Kills: 13.4% Misses: 25.6% Hits: 61.0%(4.98)

No argument here. The arbalest has no equal for shooting (in the back) those who
wear Umlish armour. The crossbows may inflict a few wounds, but only the
arbalest will get rid of those pesky Black Plate infantry.


So far we have established that even when we oversimplify the analysis, the
arbalest may be better than the crossbow only against armour 17 and higher.

But everything else plays against the arbalest.

Its higher range results in a badly aimed first salvo, and by the time they have
reloaded, the enemy is either in melee with other Ulmish troops, or engaging
the crossbowmen themselves.

The high resource costs mean that one cannot produce many arbalests in the first
few turns, and that gold gathers unused until more castles can be built. Once
those go up, the low strategic move prevents the crossbowmen from being where
they are needed.

When "Flaming Arrows" comes around, arbalests benefit the least from it, due to
their abysmal rate of fire.

And of course, if your opponent is worth anything, he will draw the enemy fire
with low resource troops with shields. Pythium and Ermor's Velites, Tien Chi's
footmen, Machaka's warriors, etc, etc, etc... all of these are cheap and are
best dispersed with a higher rate of fire which the arbalests lacks. The
arbalest's high damage is perfectly unnecessary here. Once again, its only
purpose is to kill friendlies.

And now, if any retard comes and tells me that the arbalests are the best ranged
weapons and that it's not subject to discussion...
__________________
No good deed goes unpunished...
Reply With Quote
  #2  
Old September 22nd, 2007, 11:04 PM
KissBlade's Avatar

KissBlade KissBlade is offline
Major
 
Join Date: Jan 2006
Posts: 1,055
Thanks: 4
Thanked 29 Times in 13 Posts
KissBlade is on a distinguished road
Default Re: Petar\'s \"Range Weapons\" flamebait

I should probably backup why I view MA Argatha and MA TC as "weaker" than MA Ulm. Assuming all equal player skills, MA Ulm doesn't share several notable problems with MA Argatha/MA TC. MA Ulm can effectively research only evocation to help with its military. You can branch off into construction for diplomacy but for the most part, they have a very straight forward approach when it comes to preparing for war. Its early game, while not spectacular due to the lack of archers, is decent since most of its troops match well against indies in MA. (Xbows/Cavs/Knights are rare in MA) You get a decent amount of free points from Drain to spend on your pretender and you don't need to make him a thug right off the bat since thugs are often used to help with fast exp/harrassment. Also smiths (if I recall) can take an arrow better than most mages. This lets you shuffle them closer to the front lines which is important to help aiming spells.

MA Argatha does have strengths in umbrals and golems but as you've already mentioned they're in seperate skill trees. Worse Umbrals require death gems and you don't start with any of them in generation not to mention your only death gem searcher has d1 AND costs 400g's. Very hefty and deftly not something that lets you flow in death gems early game. Statues are great but slow to acquire. Thus your initial few turns generally will feel very clunky especially due to the subpar attack/precision on your troops. This means in most games, you feel the heat pretty early on as the first few turns are often the land grab turns and you'll be left behind if you don't invest in an awakened pretender. Which is even less points for you to have especially since you are mostly likely going with magic 3 to get your summons up asap. I'm not arguing against what results you may have encountered, I'm simply stating my view of the situation. I have seen MA Argatha do decent and I have seen them flop. But the same I could say for MA Ulm.

MA TC is already realized by many to be problematic so I won't go too indepth with it. While they may look magically diverse, they are actually not very. Their path synergies to almost nothing useful at all especially since their best capital only mages are easily astral dueled away AND often the main things you're hoping for is the A3 thunderstrikers which are pretty damn rare. Summon wise, they are also pretty poor since they lack death magic and only possess minimum nature magic. They can craft nice items with their skill tree, which is decent since you're going to need a SC chassis with them. Sadly enough, MA TC's main strat is consort spamming and that simply doesn't suffice for the mid => late game scenarios where call of the wind is cheap and easy.
Reply With Quote
  #3  
Old September 23rd, 2007, 12:25 AM

CUnknown CUnknown is offline
Captain
 
Join Date: May 2004
Posts: 947
Thanks: 1
Thanked 0 Times in 0 Posts
CUnknown is on a distinguished road
Default Re: Petar\'s \"Range Weapons\" flamebait

I'm sorry, Tuidjy, you're overcomplicating things and for all that not even calculating damage. All "hits" are not equal. It's better to keep it simple and leave out the random numbers, imo. Against 14 protection and over, the arbelest is the clear winner. Even against 12 protection, it is better than any other as has been shown.

Although for sure, the community isn't divided on the Ulm issue, I know and will admit that. It's just that the common knowledge is a little off in this case. People seem to think that Ulm is some outlier, as if it's horribly underpowered. The truth is that it's very similar in power to any other faction in the bottom tier or two (that is, like almost half the factions out there).
Reply With Quote
  #4  
Old September 23rd, 2007, 12:44 AM

Sombre Sombre is offline
BANNED USER
 
Join Date: Feb 2007
Posts: 5,463
Thanks: 165
Thanked 324 Times in 190 Posts
Sombre is on a distinguished road
Default Re: Petar\'s \"Range Weapons\" flamebait

Yeah how silly of him to overcomplicate things by using the same factors the game does to calculate damage. Let's just forget the game, it makes the math too complex.

Your argument that arbalests are hands down the best mundane missile weapon now appears to be "Arbalests do the most damage per hit". Great. I think we can safely say that this is no longer up for debate, because you're the only person arguing that Arbalests are the best and your argument doesn't deserve a sensible reply (you'd ignore or dismiss it anyway, judging by your response to Tuidjy).

I was going to run some ingame tests with crossbows and arbalests, behind a wall of Ulmish heavy infantry, vs various enemies to see which one was generally more useful. But now I can see there would be no point, because Arbalests do the most damage per hit. Heh.
Reply With Quote
  #5  
Old September 23rd, 2007, 01:08 AM
Tuidjy's Avatar

Tuidjy Tuidjy is offline
Major
 
Join Date: Apr 2004
Location: La La Land (California, USA)
Posts: 1,244
Thanks: 0
Thanked 30 Times in 11 Posts
Tuidjy is on a distinguished road
Default Re: Petar\'s \"Range Weapons\" flamebait

> All "hits" are not equal.

I know that, which is why I have divided them into kills and damaging hits, and
have displayed the average damage.

> It's better to keep it simple

We have computers so we can crunch the numbers. We have brains so we can
analyze results. This is simple.

> and leave out the random numbers, imo.

No, we should not, because the spread is very significant. I know that my
intuition misleads me when I think about open ended dice. I am pretty sure
yours is no better.

> Against 14 protection and over, the arbelest is the clear winner.

Did you bother reading what I wrote? How the Hell can you say that it is a
clear winner when pure numbers are ambivalent, everything else favors the
crossbow, and protection 14 units are not the what the arbalests will most likely
be shooting at, unless you count your own troops?!

> Even against 12 protection, it is better than any other as has been shown.

What has been shown? Against an enemy of infinite health and protection 12 the
freaking longbow has three times as many landing hits, and 54% chance of dealing
an average of 5.49 points of damage. The arbalest hits have a 89.3% chance to
deal an average of 9.39 points of damage. The hard numbers slightly (6%) favor
the longbow, and every single assumption, including the one about infinite
health, favors the arbalest! Math and English, do you read them?!

> It's just that the common knowledge is a little off in this case.

And you know this because the Pantocrator has spoken to you? Players that are
much better than you think otherwise, Ulm loses in duels no matter how well
you stack the deck in their favor, has no answer to heavy blesses or tramplers,
the numbers speak against them, no one has any suggestions for any late game...

But YOU know the common knowledge is off? I do not have a beef with
people who point out that MA C'tis and Agartha need help. But you keep talking
and talking, and have not advanced any arguments but bad math.
__________________
No good deed goes unpunished...
Reply With Quote
  #6  
Old September 23rd, 2007, 03:21 AM
Tuidjy's Avatar

Tuidjy Tuidjy is offline
Major
 
Join Date: Apr 2004
Location: La La Land (California, USA)
Posts: 1,244
Thanks: 0
Thanked 30 Times in 11 Posts
Tuidjy is on a distinguished road
Default Some more tests.

I decided to test all this crap in game. Here's what I did.

80 Marignon crossbowmen,
80 Ulm crossbowmen,
80 Man longbowmen

vs

80 Tien Chi footmen (shields, low armour)
80 Tien Chi inperial footmen (armour 14)
20 Tien Chi footmen, 60 Tien Chi imperial footmen (footmen draw fire, IF flank)

I positioned the infantry as far back as possible, except for the last
combination, where I positioned them in a half way decent manner. Not in my
favorite arrangement for drawing ranged fire, mind you. I ran every combination
at least twice. The attachment is a setup which you can use to fight Man, Ulm,
Marignon and Tien Chi armies. I am too sick and sleepy to actually bother typing
it all, but basically, the longbowmen win all matchups against the infantry.
They lose to the arbalest guys, but that's because of armour, not shooting skill.

The regular crossbowmen consistently lost to the mixed squad, the guys with the
arbalests lost two and won two. They did very poorly (worse than crossbows)
while shooting, but managed to steal a victory in hand to hand. If Ulm
crossbowmen have a saving grace, it's that they are OK infantry.

Afterwards I ran a few fights with Ulm against tower guards... the tower guards
wipe the floor with any combination of the same gold cost of Ulm melee/ranged
infantry. If Ulm has arbalests, they break earlier - fewer fighters, and the
arbalests kill friends and foes indiscriminately.

Imperial guards beat Ulm even easier, and tie with tower guards - win some, lose
some. Marignon's men at arms slaugter Ulm as well... basically Ulm's 10 gold,
resource 30+ infantry loses to any gold 12-14, resource 24+ infantry, even if you
match them in cost, as opposed to numbers. By the way, heavy infantry with
shields destroys any unsupported ranged troops, but that's only normal.

Hell, try it for yourself. The mighty men of Ulm cannot beat anyone in hand to
hand combat. Their only excuse for infantry are the arbalest guys, whose only
saving grace is their heavy armour, that allows them to fight OK in hand to hand.
If they had a buckler and a crossbow instead of an arbalest, they would be quite
the soldiers... but obviously they are too dumb to understand what gear works.
Attached Files
File Type: zip 551425-UlmManMariTienChi.zip (61.1 KB, 114 views)
__________________
No good deed goes unpunished...
Reply With Quote
  #7  
Old September 23rd, 2007, 10:45 AM
Burnsaber's Avatar

Burnsaber Burnsaber is offline
Colonel
 
Join Date: Jun 2007
Location: Finland
Posts: 1,617
Thanks: 179
Thanked 304 Times in 123 Posts
Burnsaber is on a distinguished road
Default Re: Some more tests.

I took some time today to make some experiments regarding Ulmish Infantry. I wanted to test how they fare against the current heavy infatry favourite (Pricipe) mano'a'mano.

I bidded 20 Ulmish Black Steel troops with different weapons against 20 Pythium Pricipes. The Pricipes lost most of the fights (expect against shielded infantry), but when the Pricipes lost they inflicted heavy (7-9) casualties, and all Principes that fled, survived (expect a couple of cripples). Losses were usually 5-7 Blacksteels to 10-14 Priciples.

Then I made a bit more "realistic" test of 30 principes against 20 Blacksteel (due to resource cost you can build 1,5 principes for each blacksteel guy, gold cost is *much* less limitng factor when building troops than gold). It was brutal, not even Guardians had anything against Priciples in this case. The losses were usually 13-17 blacksteels to 8-12 Priciples.

So what does this mean? Ulmish Infantry doesn't *suck*, exactly (they have some results against *the heavy infatry*, but only when magic and resource complications are stripped) , but they're nothing too amazing either. Personally (I have nearly zero experience with Pythium/MA Ulm previously) from watching the fights, I'd take Principles over Blacksteels anyday -> They don't run away so easily, you can make more of them, they move faster, have javelins..

And I think that this is wrong. Ulmish infantry should be awesomely good and badass, since it's pretty much all they've got outside capitol. Not medicore, especially with their magic weakness. While I look at their stats I see base attack & defense 10, they're just your regular infantry in extremely heavy armor. I don't think that it would be at all unbalancing nor unthematic to give them some training bonuses. a mere +1 attack, +1 defense, -1 encumberance, +1-2 action points and +1-2 morale would proabably go a long way to make Ulmish infantry something to actually mention when talking about Ulm's strenghts.

And about Master Smths. CUunknown has a point. They don't excatly *suck* as mages. They're just not enough to give anykind of late-game power (and late-game power is the power you need to have to have chances at winning), taking a heavy-magic pretender can only take you so far. There have been many suggestion regarding twiddling with their randoms and I wholly support them.

Hmm, I just had another idea. How about giving them a slighly tweaked LA Ulm's fortune teller (without the blood pick, perhaps)? Ulmish people are a superstitious folk, afterall. It would allow a lot easier diversification.
__________________
I have now officially moved to the Dom3mods forums and do not actively use this account any more. You can stll contact me by PM's, since my account gives e-mail notifications on such occasions.

If you need to ask something about modding, you can contact me here.

See this thread for the latest info concerning my mods.
Reply With Quote
  #8  
Old September 23rd, 2007, 12:24 PM

CUnknown CUnknown is offline
Captain
 
Join Date: May 2004
Posts: 947
Thanks: 1
Thanked 0 Times in 0 Posts
CUnknown is on a distinguished road
Default Re: Some more tests.

Tuidjy, it's as if you've done your best to be misleading with the analysis you've chosen..

Quote:

I know that, which is why I have divided them into kills and damaging hits, and
have displayed the average damage.
But why even bother dividing it up in this way, unless you're trying to obscure the fact that the arbelest does the most damage against a range of common protection values, even adjusted for rate of fire?
Reply With Quote
Reply

Bookmarks

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On

Forum Jump


All times are GMT -4. The time now is 05:01 PM.


Powered by vBulletin® Version 3.8.1
Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
Copyright ©1999 - 2025, Shrapnel Games, Inc. - All Rights Reserved.