|
| Thread: Different approaches to town/creature tier lists | |
|
Otuken
Famous Hero
|
posted July 01, 2018 07:46 PM |
|
|
Different approaches to town/creature tier lists
Hi all!
I would like to give you different approaches to your town and creature tier lists. I usually see that people tend to compare creatures with their counterparts but this method usually gives us false informations due to different growth rates costs and different dwelling costs. For example comparing Black Dragon with Ghost Dragon gives us no informations because Ghost Dragons are more reachable. Also I think we have to ignore heroes (Except gold andresource bringers) of towns becase every hero may be selectable from a different town.
We have to take into account dwellings and their requiement building costs. For example Dragon Cave’s dwlling cost should be taken as Dragon Cave+Mage Guild level 1+ Mage Guild level 2+ Mage Guild level 3.
Converting every resources into gold is must fort this measurement. I will consider every woods and ores as equal to 500 gold and other resources are 1000 gold. It is optimal consideration. So in sum up only dwllings, horde buildinds and requirement buildings are included the total costs. Unique buildings are unmeasureable. I excluded them. Here is the town rank from the chepest to most expensive town according to their buildings.
1) Conflux 158.600
2) Fortress 178.500
3) Necropolis 189.400
4) Stronghold 197.700
5) Inferno 217.600
6) Tower 235.200
7) Rampart 238.000
8) Castle 241.500
9) Dungeon 267.900
Here is weekly troop costs ( I know they are already exist in variety of website)
1) Stronghold 25.130
2) Fortress 28.660
3) Necropolis 28.920
4) Rampart 30.620
5) Dungeon 31.205
6) Inferno 32.090
7) Conflux 32.500
8) Tower 34.800
9) Castle 35.580
Somethings are missing because We haven’t taken into account the heroes.
1) Castle 1 Gold bringer hero+1 estate skilled hero, 3325 gold per week
2) Rampart 1 Gold bringer hero, 2450 gold per week
3) Necropolis 2 gold bringer hero, 4900 gold per week
4) Tower 1 gold bringer hero and one mercury bringer hero, 2800 gold per week
5) Inferno 1 gold bringer hero and one sulfur bringer hero, 2800 gold per week
6) Dungeon 1 gold bringer hero and1 crystal bringer hero, 2800 gold per week
7) stronghold 1 gem bringer hero, 350 gold per week
8) fortress none
9 conflux 2 gold bringer hero 2 estate skilled hero, 6650 gold per week.
I wish Every town had to play with only their heroes. But we have to assume it for to get more accurate stats. Anyway if we take into account the heroes, the new weekly troop costs will be something like this:
1) Necropolis: 24.020
2) Stronghold: 24.780
3) Conflux 25.850
4) Rampart: 28.170
5) Dungeon: 28.405
6) Fortress: 28.660
7) Inferno: 29.290
8) Tower: 32.000
9) Castle: 32.255
Well these are just economic stats. But Conflux (Also Rampart) has potential to become the chepest town in the game because two of the Conflux Heroes start with estate skill. Here is 1.st level troops stats concidering prices. These lists a just sign of cost effecienciest not measurement of creatures. For example in here Master Gremlin and its cost is constant.
1) M.Gremlin: 40 gold×32 Growth=1280 Gold
Total Fight Value=32×55=1760
2) Sprite: 30 gold× 42.7 Growth= 1280 Gold
Total Fight Value= 42.7×70=2989
3) I.Troglodyte: 65×19.7 Growth=1280 Gold
Total Fight Value=1654.8
4) C.Captain: 90×14.2 Growth= 1280 Gold
Total Fight Value= 14.2×115=1633
5) Skeleton W: 70×18.3=1280 Gold
Total Fight Value= 18.3×85=1555.5
6)Gnoll M:70×18.3=1280 Gold
Total Fight Value18.3×90=1647
7)Halberdier: 75×17.06=1280 Gold
Total Fight Value: 17.06×115=1961.9
8) Familiar: 60×21.3=1280 Gold
Total Fight Value: 21.3×60=1278
9) Hobgoblin: 50×25.6=1280 Gold
Total Fight Value: 25.6×65=1664
1) Sprite: 2989
2) Halberdier: 1961.9
3) Master Gremlin: 1760
4) Hobgoblin: 1664
5) I.Troglodyte: 1654.8
6) Gnoll M: 1647
7) C.Captain: 1633
8) Skeleton W: 1555.5
9) Familiar: 1278
No surprize except the tier of C.Captain
2.level creatures, Zombie is constant
1) Zombie: 125×16=2000 Gold
16×160=2260
2) H.Hag 170×11.77=2000 Gold
11.77×196=2306.92
3)Wolf Raider: 140×14.29=2000
14.29×174=2486.46
4)Lizard Warrior: 140×14.29=2000
14.29×130=1857.7
5) Storm Ele: 275×7.27=2000
Total Fight Value: 7.27×324=2355.48
6) Marksman: 150×13.34=2000
13.34×115=1534.1
7) Battle Dwarf: 150×13.34=2000
13.34×209=2788.06
8) Obsidian Gargoyle
160×12.5=2000
12.5×155=1937.5
9) Magog
175×11.43=2000
11.43×210=2400.3
1) Battle Dwarf 2788.06
2) Wolf Raider 2486.46
3) Magog 2400.3
4) Storm Ele 2355.48
5) H.Hag 2306.92
6) Zombie 2260
7) O.Gargoyle 1937.5
8) Lizard Warrior 1857.7
9) Marksman 1534.1
Omg, this list is full of surprise. Everybody know that 2. Level archers are normally better than slow sturdy units in early stages of games but in the final battles, sturdiness becomes more and more important and it probably makes Battle dwarf the best 2.nd level unit in the game. Wolf Raider is also fragile but it has huge damage potential.
Here is 3.rd level tier, I used the same method again;
1) Iron Golem: 6921.6
2) R.Griffin: 5090
3) Cerberus: 4139.52
4) Orc C: 4074
5) Wraith: 3681.72
6) Dragonfly: 3500
7) Evil Eye: 2940
8) Grand Elf: 2913.3
9) Ice Ele: 2822.4
In once again: These tiers are may be valid only in the final battle. Everybody know that Grand Elf is better than Iron Golem in the early stages of games. Also is Wraith really 5.th best level 3 creature?Its fighting value is 252 i have no idea does Wraith really deserve this tier. How accurate fight values?
Level 4. Tier;
1) H.Demon: 5688
2) Ogre Mage: 5376
3) S.Pegasus: 4865.52
4) Crusader: 4704
5) G.Basilisk: 4488
6) Vampire Lord: 4172.8
7) Medusa Queen: 4103.1
8) Energy Ele: 3880
9) Arch Mage: 3320.37
The Worst level 4 unit is seen the most cost effecient unit in this tier. I am getting suspicious about the fight values. Probably I was supposed to use AI values.
5.th level units;
1) Dendroid S: 4674.6
2) Mighty Gorgon: 4626
3) Minotaur King: 4183
4) Pit Lord: 4134.06
5) Power Lich: 4000.5
6) M.Genie: 3366
7) Thunderbird: 3354.34
8) Zealot: 3000
9) Magma Ele: 2646
6.th level creatures
1) Magic Ele: 10344
2) War Unicorn: 10271.8
3) Naga Queen: 8520
4) Champion: 7200
5) Efreet Sultan: 6858.72
6) Wyvern M: 6633.66
7) Dread Knight: 6492.8
8) Scorpicore: 5703.36
9) Cyclops King: 4850.7
7.th levels;
1) Phoenix: 26320.86
2) A.Behemoth: 21588
3) C.Hydra: 18768.32
4) Black D: 18314.1
5) Gold D: 16794
6) Arch Devil: 12950.21
7) Ghost D: 12912
8) Archangel: 12066
9) Titan: 11450
Overall;
1) Rampart=27
2) Stronghold=30
3) Inferno=31
4) Conflux=33
5) Fortress=36
6) Castle=37
7) Tower=38
8) Dungeon=39
9) Necropolis=44
Statistically Necropolis produces the least cost effecient troops but with dangerios abilities per week with easily reachable 7th level creature. Necropolis’s specialities which is necromancy, undead natures, life drain ability, animated dead etc are incalculable.So better to exclude Necro from the list. Personally I do not believe the list. This list only shows that Fight Values are not reliable.
BTW is cost efficiency originally include the Fight Values?
____________
|
|
avatar
Promising
Supreme Hero
|
|
posted July 01, 2018 08:28 PM |
|
|
Off topic: This is my favourite creature chart:
http://home.pinknet.cz/~coyot/homm3/creatures.php
____________
|
|
OhforfSake
Promising
Legendary Hero
Initiate
|
|
posted September 02, 2018 01:29 PM |
|
|
Personally I like these kind of analysis a lot, but as you conclude yourself those values don't have much to do with the game.
I think cost efficiency of units is very interesting and it's a shame that it's not that prevalent in the game since most maps are rich enough to get level 7 units in the first or second week, so unit growth becomes more relevant. Unit cost only comes into it after the stronger units have been bought in my experience.
The in game values doesn't really say that much I think.. it's probably more interesting to look at how the typical (yet a little boring) game is played.
The game can be divided into two stages. Stage 1 is creeping which means playing against the map, here the objective is to collect as strong an army as possible as fast as possible before meeting the other human player. In this stage the better you are at taking down large number of computer controlled units with much fewer units yourself the better you'll do. The second stage is the final battle, where each player show what they have managed in the creeping stage and have one battle to decide the map, roughly speaking.
In the creeping stage, I prefer fast hard hitting units, and a focus on the attack primary stat, so heroes and units with these attributes are generally to be preferred, hence in the final battle you also tend to have a higher attack than defense stat. Also it's easier to keep casualties low with high level units as they are tougher.
For the final battle, apart from Angels, Wyverns, and in some cases Demons or Skeletons, the higher level units you have the better, since while the hp and damage may not be that different across the tier ranks due to growth rate, the attack and defense stats makes a huge difference. That is also why the ability to bypass defense in various ways can become extremely important. Funnily this means that for e.g. a Stronghold army, defense stat may be a better stat than attack stat, because once you have two equal high level heroes facing off, the Ancient Behemoth will relatively quickly reach a maximum damage output of 160 in average per AB, meaning more points in Attack is wasted when considering their most important unit. Also there can be cases where low amount of hp is an advantage, because army strength increases much faster than spell power (unless hit & run is allowed to diminish enemy armies via scouts), so rare spells like sacrifice can much easier replenish an entire level 7 stack of e.g. ArchDevils, while overall they are just as good a unit as other level 7's (in my opinion better than most) due to their high attack and defense stat.
So I think a better list would take into account the stages of the game, and probably focus mainly on offensive and fast units and use growth rate in stead of unit cost.
____________
Living time backwards
|
| |
|
|
