|
| Thread: Strategical thoughts / strategical Questions |  This thread is pages long: 1 2 3 4 5 6 · «PREV / NEXT» |
|
zsa
Famous Hero
|
posted January 02, 2005 04:27 PM |
|
|
yes i was very well aware of that but there are so many factors to consider like what spells does the opponent have, what race is (s)he and what units (s)he has.
Against castle, inferno, necro stronghold and fortress putting your titan there is the right move. I never play conflux (or most other ppl for that matter) so that leaves 2 towns with breath attack capabilities out of 8. Majority rules  .
|
|
csarmi
Supreme Hero
gets back
|
|
posted January 02, 2005 05:27 PM |
|
|
|
Another (now bigger!) problem is that it only works if you are the attacker: if you are on the defending side, your titans can be attacked from 3 hexes.
|
|
Zsa
Famous Hero
|
|
posted January 02, 2005 06:03 PM |
|
|
|
lol, yeah you're right, forgot about that. In that case it wouldn't be very good to put your titan there, maybe swich the master gremlin with the titan positions at the end of every turn. :\
|
|
Sir_Stiven
Honorable
Legendary Hero
banned
|
|
posted January 03, 2005 03:59 AM |
|
|
i still think the first question is a toss though since its too easy.
only advantage terek got vs hack is faster units at mainly low level. That leads to eating up retal with lesser worth units. And off course having ogre magis reaching in first round is important as they are solid.
But really.. if tower doesnt have tactics youre only gain is to reach with ogre magis first round.
Since both towns have 11 at fastest unit youre opponent might have a speed art or you have to fight on snow ground or get attacked... meaning opponent will have first move... then the advantage of having an all out attack also will be removed as opponent will have time to arrange troops.
Would you really sacrify hack as a main and a possible huge attack bonus for that?
i wouldnt.
as for the 3rd question... which opposite town?
|
|
csarmi
Supreme Hero
gets back
|
|
posted January 03, 2005 11:00 AM |
|
|
Not really - actually it's even better if tower has no tactics, because you can move your shooters in range too.
The real problem I have with Hack:
Look at a normal line-up. (assume 21 attack, level 20 Hack)
9 ancients ~ 810-1350 + 20% from offense
27 t-birds ~ 300-500 (ranges) + 60% from offense
28 ogre magi ~ 150-300 (ranges) + 60% from offense
90 raiders ~ 2*(270-360) (ranges) + 60% from offense
150 hobgoblins ~ 150-450 (ranges) + 60% from offense
+ shooters
You get the picture - Crag Hack does not go well with ancients at all. Which is sad, because ancients are by far your most damaging troop.
Something to consider though (regarding the Terek strategy) - it's not such a big idea to take retal with wolves.
|
|
Sir_Stiven
Honorable
Legendary Hero
banned
|
|
posted January 03, 2005 06:17 PM |
|
|
21 attack and has an army of week 5 units?
lol
hardly the case 
and the higher attack the more the 60% becomes in dmg.
and in the end... your gain might just be having the ogre magis getting over first turn.
Is that really worth risking all the extra dmg for?
as for the so called shooters idea, how often do you get enough crystal to get clopses early on? and even then.. do you rather spend it at getting those clopses or upg. beths?
id go for the beths. slot 7 i try to fill with wyerns or angels from creature banks.
as for the wolf raiders taking retal i thought about that earlier aswell, and only solution i could come up with was getting gretchin as main hero as then the gobs could move ahead of wolfies.
but then you lose the all out attack possibility and having ogres reaching first round even if opp has tactics. Or you lose all the extra dmg.
Even yet.. if i had terek, hack or gretchin day1 from tavern i know who my pick would be no matter opposition.
tereks only possible gain except mainfight is clearing the map first week when hack havent developed his 2ndary skills yet.
|
|
Xarfax111
Bad-mannered
Supreme Hero
The last hero standing
|
|
posted January 03, 2005 06:48 PM |
|
|
...welll, thats what i expected:
An open discussion, sharing opinions
Thanks!!
...therefore there was this provokative header "Terek vs. Hack".
If u want to compare them u should take a reasonable endfight zenario:
Late week 2 to early week 3 army on a fixed map 100%.
Middle week 4 army on a random 160% big blocks.
About Level 20 is ok in both cases, the army u described csarmi is to big in my opinion.
The extra damage that Hack gives will be in most cases about 30%, as peoples will try to get offense with terek too, same as peeps try to get Tactics with Hack.
On the other side it would be plain stupid not to get Tactics with tower, to avoid that Stronghold explodes in your face.
To wolfrider soacking retal i say that there is no other oppurtunity, as they go straight after Thunderbirds. U can change the scenario by letting the Thunderbirds attack the NQs and concentrate all power to kill the only damage dialers that tower do have in the infight = Titans and NQs.
I took this example to discuss these things ...the pros and contras using different heroes. Especially to mention that "changing the prehit situation" is a hidden clue in Tereks fight.
To the third question:
Its not necessary to state the opponing town, but as desired i say its a castle town. The Question is what are the main thoughts if u place your Endfight army...what do u think is to object? I often see that high ranked player do master chaining and stuff, but if it comes to endfight it seems that they dont think about their formation. I know in 90% of the games those stuff is not decisive, but sometimes u need this little extra plus.
Are there no opinions about Zsa s formation?
Xarfax1
____________
|
|
csarmi
Supreme Hero
gets back
|
|
posted January 03, 2005 09:19 PM |
|
Edited By: csarmi on 3 Jan 2005
|
Of course the army is too big, but the army size is of no importance - substract one week growth, oh well, and of course normally cyclops and ogre magi lower.
If I play (that does not happen often) I play with no joiners accepted (banks included).
I'd switch cyclopes for some fast troops too, though.
Or - whatever.
Tower and damage dealing.
Let us say:
5 titans - 24 attack, 200-300 damage, 1500 hp
14 naga queens - 16 attack, 420 damage, 1540 hp
21 master genies - 12 attack, 273-336 damage, 840 hp
28 archmagi - 12 attack, 196-236 damage, 840 hp
42 iron golems - 9 attack, 168-210 damage, 1470 hp
88 obs. garg. - 7 attack, 176-264 damage, 1408 hp
200 m. greml. - 4 attack, 200-400 damage, 800 hp
So you are attacking the wrong opponents.
Master Genies do - by far - the most damage per hp.
Should you concentrate on the higher level creatures and you'll find yourself in a big trouble.
Ancients are destined to kill the titans, that's sure, but the other cratures should focus on the genies and - maybe - the naga queens.
Actually, as I keep testing this fight I am more and more sure that Stronghold is bound to lose. Especially if opponent has fire magic (fire magic is very powerful if you get frenzy and sacrifice).
One of the main reasons is having two shooters.
A good tower formation - againts no dragons:
Get ammo cart, and ballista, maybe first aid tent too.
Master Genies
(ammo cart)
Titans
(ballista)
Iron Golems
Master Gremlins
Naga Queens
(-)
Obsidian Gargoyles
(1st aid)
Archmagi
Titan and Archmagi well-defended. Master Genies not easily attacked either. Iron Golems and Naga Queens in the centre are well-placed.
http://www.math.bme.hu/~csarmasz/heroes3/tvss.h3m
(a map for tower vs stronghold testing)
http://www.math.bme.hu/~csarmasz/heroes3/Terek attack 1.GM2
(the save before enfight)
You can't blitz with two shooters and no tactical advantage.
|
|
pacifist
Famous Hero
|
|
posted January 04, 2005 02:48 PM |
|
|
Very interesting thread! Congrats Xarfax1 for sharing.
Personally I don't think Terek is a good choice with mid-level armies (around week 2 or 3). The strenght of striking first becomes more important when there are more creatures just like the creature vs creature fights that have been discussed many times on the boards. Also the tower army just asks for the ennemy to be in range of his shooters, 2 of them having no melee penalty. It could be interesting though to study optimum battle formations in town vs town fights and more important (as Csarmi pointed) the best desirable targets in those fights each round. That doesn't mean raw damage comparison only but taking into account the A/D, buff spells and offence/armorer modificators. Most desirable targets are no retaliation and double atack stacks but only if they deal more damage than other units. Also the fight is very dynamic and desirable targets vary as fight goes on and some stacks are reduced.
The ultimate goal of the fight being to reduce ennemy HP to zero while sustaining the less damage, each opponent unit must be checked to measure his offensive possibilities vs our units and divide that value by their total HP. Some times it can be good to not attack at all when opponent has no more no retaliation troops or no more archers or blocked archers and instead rely on damaging spells and defending only.
There was a thread in this forum some time ago where a fight of a Dungeon vs a Rampart army was played round by round. It involved high level spells but the same can be done with level 1-3 spells only and other towns vs towns fights. Some very interesting comments of Zhuge of the Inner Pentacle concerning best selection of targets for grand elves (evil eyes or medusas).
What is the average army size on online games in random maps? Do Tower players go for giants first week (with citadel only) if lv2 dwelling is built or castle with nagas and gremlins upgraded?
Same for Stronghold : Build cyclops week two or never?
That could be a good base for a starting setup.
____________
http://www.youtube.com/user/alkoriak#g/u
|
|
zsa
Famous Hero
|
|
posted January 04, 2005 03:45 PM |
|
|
what you build on a random is a lot dependent on resources.
For tower my goal is to usually get nagas if i can week 1. If the map is rich i might go for giants. But giants also means no castle, so i might sometimes prefer going for castle nagas, get upgrade , and clean the map with the nagas.
2. For stronghold if the map doesn't have enough crystals, you might not put cyclops at all. The 20 crystals used for that could be better used to get the upgrade for the behes. Last time i played stronghold i did put behe and cyclops week 1. But that is mainly cause i managed to gather enough crystals and because i had a cyclops dwelling around. The map wasn't rich in money tho, gold chests or banks, crypts, so i had serious money problems second week.
|
|
angelito
Honorable
Undefeatable Hero
proud father of a princess
|
|
posted January 04, 2005 04:07 PM |
|
|
Average army size of randoms in online games surely depend on the template u will use.
And about the amount of the specific numbers of your army, it surely depends on your building order. If u have castle for example, you probably have NO archers, NO griffins and NO horses built in first week, coz u go for angels and castle. In Tower u will probably have all level 1-6 and castle built. So u see how it differs from town to town. (dwellings could change amounts also...)
On average to rich templates, u will normaly have your endfight in middle of week 3 (L size, no under)
On poor to average templates, u should give it a week more.
For XL also give 1 week more.
Just my experiences though...
____________
Better judged by 12 than carried by 6.
|
|
pacifist
Famous Hero
|
|
posted January 04, 2005 04:23 PM |
|
|
Ok, thanks. Now I have a good idea of the likely configurations. I suppose you always buy no matter what the starting native hero that comes each week in tavern? (or both for inferno and necropolis towns or for fodder in fights).
There is still a luck factor since sometimes 2 heroes of the same type can appear and that can decide the final battle outcome. I will assume only one hero per week then.
____________
http://www.youtube.com/user/alkoriak#g/u
|
|
angelito
Honorable
Undefeatable Hero
proud father of a princess
|
|
posted January 04, 2005 04:51 PM |
|
|
I don´t know if u r referring to a specific tournament (Native warriors), where u r allowed to buy native heroes only (2 in week 1, every next week 1 more..), or if u r talking in general.
In online random games, u try to get as many heroes as u can afford on day 1, usually 6-8, to get all the things going u have to do in your area.
And yes...u could be very lucky on day 1 when u r playing dungeon or stronghold, coz additional to your starting hero, u could find SHAKTI (about 100 troggs) and GRETCHIN (about 70 goblins) in your tavern, which would give you a great fighting army for the first week.
On the other hand, u could play with Castle, starting hero is SANYA with only 16 pikemen, archer dwelling is not prebuilt, and in tavern u find SYLVIA with 12 pikemen and STRAKER with 9 walking undead.
So your army on day 1 would be 42 pikemen and 9 walking undeads....compared to about 150 troggies (goblins) and 100 goblins (troggies)....see how unfair this game can be...
____________
Better judged by 12 than carried by 6.
|
|
csarmi
Supreme Hero
gets back
|
|
posted January 04, 2005 04:54 PM |
|
Edited By: csarmi on 4 Jan 2005
|
Well that angels+castle strategy requires very specific conditions, I guess...
I'd love to watch you fight anything just half-serious week 1. :-)
You summarized well what I hate about heroes 3.
I'd never play with so unfair conditions.
In my opinion level 2's should be always prebuilt and starting army differences should be much smaller.
I one chose Bron with fortress, I had a necro and Gerwulf with 13 gnolls (no ballista) in tavern on 160%. No lizards prebuilt. Bron comes with 15 gnolls and NO basilisks (yes that IS possible).
|
|
LordLazy
Promising
Famous Hero
Wood cleaner
|
|
posted January 06, 2005 12:26 PM |
|
|
What I meant to say there were that even without the moos special, they are still sturdy units. But, as stated, high level units will always be the most important units to all players, except necro, but in that case death stare wont work anyhow
____________
Tagged officially as Noobegian two years ago. This typographic material is strictly copyrighted. All situations containing abuse will be brought to court.
|
|
csarmi
Supreme Hero
gets back
|
|
posted January 06, 2005 12:34 PM |
|
Edited By: csarmi on 6 Jan 2005
|
Funny that we share the same opinion and still ---
Tried to quantify Moo's special but it's not easy.
(12-16)+(OHP/10) damage is a wrong estimate, for example. Funny it gets more exact against lower levels.
What I meant to say is that against lower level units Moo's damage/hp is poor.
Well, a good estimate of Gorgon's damage (funny it works similar to that of Behemoth's - ignores enemy defense; higher defense means higher HP and it balances).
Against level 7's: varies, between 25-40, hard to give it more precisely.
Against other levels: 19 damage/gorgon
(20/19/18/18/19/20 in order)
|
|
Zsa
Famous Hero
|
|
posted January 06, 2005 10:14 PM |
|
|
Why?
Why would you try and make a table like this. It is totally a waste of time. Sure I like statistics, something like the probability of a skill to come to a hero or other stuff like that.
But this is never going to be useful, why?
1. Because unless you factor only dmg/hitpoints, there's always gonna be the subjective decision on the special abilities, speed, you name it.
2. So unless you make functions with n paramaters f( a1, a2, a3, .... ) and take into account all possible major situations in the most objective way possible, those statistics will never tell you who is best to attack.
I hope i made my point clearer now. So yah like Sir Stivie said, play the game, it's much more fun than useless number crunching.
|
|
Xarfax111
Bad-mannered
Supreme Hero
The last hero standing
|
|
posted January 07, 2005 12:13 AM |
|
|
Hi csarmi and all the others, thanks for your posts.
Quote:
Xarfax, I have tried your strategy in a test game against the above-mentioned tower setup (made sure the Ogre Magi can hit the Naga Queens and it was a massacre, but the Stronghold side lost. I have managed to kill 1/3rd of master genies, all but 1 titan and 1/2 of the Naga Queens with the first big blow. Opponent had expert tactics though.
I think I don’t understand. You say it was a “massacre”, but still stronghold did lose. It was either a massacre or it lost…both is not possible. Of course this terek-mass-haste only helpfs for first round, you surely need some more “jokers” for the rounds……. Maybe concentrating full power on the nagas and titan will give better results…. Still the hole fight was just an example to evaluate all the clues that terek gives.
Quote:
Of course the army is too big, but the army size is of no importance - substract one week growth, oh well, and of course normally cyclops and ogre magi lower.
“No importance”. –No sorry ive to disagree. The comparison between your army and the army your opponent may have is one of the biggest thing u must watch for if u play against a human player …its definetaly one of the ”keys to victory”. U must know when your army is normally stronger then the one of your opponent, then u make a plan when and how to attack. Lets see this in an example as it answers one of pacifists questions:
Build Cyclops or not??
Normal Comparison (approx.):
Day 1-10 Tower advantage (Mastergremlins + Nagas against slow fragile Behemots)
Day 11 – 14 Stronghold advantage ( Ancient Behemots; +1 Lvl7; No higher magic)
Day 15 – 21 Tower advantage (Titans, high shooting power; higher magic)
Day 22 ---- Stronghold advanteg (More stable Hitpoints, shooters and high magic getting less important)
Of course this is only seen on average, of course it don’t have to be like this always. On Random it looks a bit else, but about similar.
So if u can assume the impact about day 14 skip the cyclop building, to get the ancients asap and make sure u get the right skills for this strategy. If u assume that u cant reach your opponent before middle to late week 3, built the Cyclops as u will have enough time and now do balance the shooting advantage of Tower.
Quote:
Master Genies do - by far - the most damage per hp.
Should you concentrate on the higher level creatures and you'll find yourself in a big trouble.
This may lead u statistically to the wrong conclusions – their ratio isn’t that high because of their high damage, merely it is because they have so low hitpoints. Master Genies do have the advantages that they are relatively fast, can fly and do have a nice casting special …therefore the programmers gave then sooo less hitpoints …. They shouldve a shield on there back with the note “please don’t litter – fragile – they even die on retaliation “..with other words: They s*ck in infight. Titans and NQs has a lesser ratio just because they have both: Damage and lots of Hitpoints.
But I get the point behind this ratio … u try to destroy as much damage as u can before the damage could be made. Unfortunately practise is else: Uve to try to destroy those “stable” hitters (Titans, NQs) before its to late ..during the ongoing battle both side will lose fighting power in average ..so if those hard facts stay in fight u wont be strong enough to destroy them anymore …so if u want to succeed my advice is to concentrate on the fighting force.
Btw Moos are only average fighting force so normally u wouldn’t need to care about them. But as their special allows them to eliminate your (!) your hard hitters, it is of course the favourite target when attacking fortress (concentrate on them!), if u r not necro.
3. What is the best Tower Endfight formation (without tactics)? And Why?
Ok here is how I would place them:
Arch Mages
Master Genies
Obsidian Gargoyles
Master Genie (single)
Naga Queen
Iron Golems
Titans
One of the biggest disadvantage of Tower is that opponent goes first. Although doing the “big damage” spell at the beginning of the fight is unusuall nowadays, its still a valid thread for Tower. As Armageddon is banned Meteor and Chainlightning are the one still to expect. As for this reason it is never a good idea to place all the important troups in the middle as Zsa did. Now with Meteor it’s the only chance to either hit Titans, NQs, MGs or Archmages. If Chainlightning on Titans, there will be Iron Golems with only getting (75% of 50%) 12,5% off the initial damage, then first aid tent, and then again the NQs with getting only 12,5% of the initial damage.
They key to win with Tower is to make the Titans shoots as long as u can …they are your damage dialer. Archmages are good too, but u should concentrate on your Titans. With good total amount of hps and their magic damage reducing the Iron Golems are just plain perfect to block them free. Additional to this the gargs are near enough to jump into free spaces around the Titans if there. NQs stay there to protect the block from infighters…the shooting machine is placed.
The Mastergremlins aren’t even in the Endfight; their damage will be so limited in an Endfight that they wont even kill a first aid tent if they are 250-300, so don’t take them with u.
Instead of this take the MGs with you. As same speed as the Titans, place them “above” your Titans (important), so that they are able to make a nice cast on your Titans (even bless or precision, would enable the Titans to do more damage then 250 Mastergrems). Now uve 3 spells more! Additional to this (as same speed) the MG stacks can make prehits for the Titans (when they made their casts), which is also more important then the damage of the Mastergrems.
4. What is the only town for which it is a good idea to place the good troups into the middle? Reasons?
Ok thanks for the discussion.
Xarfax1
PS: I would appreciate if u may take all the statistic stuff into an “statistic thread”. Lets talk about strategy not statistic (although statistics are interesting too).
____________
|
|
Svarog
Honorable
Supreme Hero
statue-loving necrophiliac
|
|
posted January 07, 2005 03:12 AM |
|
|
Guys, I think you're neglecting a key thing about those stats.
I mentioned why those stats have little relevance (speed, specialties, attack/defense difference), but also maybe the most important is total stack HPs. If all enemy stacks had equal HPs, then only the mentioned specifics would disbalance the math. However, in an endfight, all stacks hardly have the same total HPs (due to unequal growth, individual unit HPs, units lost previously in game). Therefore, it is logical that a stack with the biggest number of HPs is almost always the scariest, and should be dealt with sooner than the others.
That is why Sprites are not a priority, even though they have a tricky specialty, and we prefer to kill level 7-s, which dont always have the biggest HP to damage ratio. If a stack of sprites had the equal total number of HPs to a stack of Magic Elementals, they would be far more dengerous, but they never do.
____________
The meek shall inherit the earth, but NOT its mineral rights.
|
|
zsa
Famous Hero
|
|
posted January 07, 2005 06:17 AM |
|
|
actually going for sprites is not such a bad idea. Your opponent gathers a lot of them, they have good speed and no retal.
Dealing with the sprites is a good idea because you can deal with them fast usually. Whenever i played vs conflux, sprites were pretty high on my to kill list
|
| |
|
|
