Round table — Online Tournaments

In reply to Oelm

Statistics

Registered: 4/6/2008 3:16 PM UTC

Posts: 474

8/22/2012 2:48 PM UTC #3092

Auto-translated

Oelm

An armor specialist is more effective than an attack specialist.

Once, Vic wrote a detailed post, and I'll try to reproduce it from memory, as I couldn't find the original.

I can't find Forrest's post right now. In it, Forrest mathematically, using formulas, proved that a defense expert is better than an attack expert.

Discord 5884

Statistics

Registered: 4/29/2007 10:30 AM UTC

Posts: 11,704

8/22/2012 4:34 PM UTC #3093

Auto-translated

Tazar is better than Kreg because in the game it’s more important not to lose than to kill faster.

Statistics

Registered: 9/30/2011 7:06 PM UTC

Posts: 1,260

8/30/2012 1:45 PM UTC #3095

Hack is clearly better than Taz for severel resons.

1. You take less loses with Hack against neutrals. The way we fight neutrals is that we get first strike to avoid loses. After Hacks first strike there is no or little retaliation. Battles where the neutrals get first strike we avoid or we bring powerstacks and fodder. The ai rarely hits the powerstacks because its affraid of the retaliation. Also a balista is much more usefull on Hack than on Taz.

2. Hacks chance of getting armor is higher than Taz chance of getting offence. 6 vs 5 isnt that big a diffrence in theory, but in practise Taz is left with no offence way to often.

3. Hack gets better magic skills than Taz. You allmost allways get earth and air with Hack, the same cant be said about Taz.

4. In a main battle where you have one of these 2 chances are that you will have the upper hand in might. So in case you are up aginst heavy magic Hack has a better chance of deeling with it by his high attack and mass haste.

Statistics

Registered: 2/23/2010 12:22 AM UTC

Posts: 2,567

8/30/2012 2:58 PM UTC #3096

Auto-translated

Maretti

1. You take fewer losses with Hack against neutrals.

That's what I tried to prove some time ago – in certain types of battles, there's less retaliation with a heavily boosted attack than defense, but these are considered easy fights and therefore don't really matter. Try to beat, for example, 24 Vyv monarchs with Crack level 7 (1 hydra, 29 lizards, 40 gnolls, 3 gorgons, 25 upgraded flies). Tazar did it (ProtosArchon-Chemist).

2. Hack has a higher chance of getting armor than Tazar has of getting offense.

According to the manual, it's the opposite – 30% barbarians, 25% beasts. Maybe certain heroes differ...

3. Hack gets better magic skills than Tazar. You almost always get earth and air with Hack, but the same can't be said about Tazar.

+1

4. Yes, but in all previous battles, Tazar will likely have *much* fewer losses.

Added after 30 minutes

According to the manual, it's the opposite – 30% barbarians, 25% beasts. Maybe certain heroes differ...

Tested – 10 tries, up to level 8 – Crack had a 50% chance of getting Armorer and a 70% chance of *not* getting crap as the 2nd upgrade; Tazar – 70% (offense), 80% (no garbage creature).

In reply to zoog

Statistics

Registered: 4/1/2009 3:57 PM UTC

Posts: 756

8/30/2012 6:13 PM UTC #3097

Auto-translated

Tootsie Pie

Tested - 10 tries, up to 8lvl - Crac had 50% of getting armorer and 70% of not getting crap @ 2nd up; Taz - 70% (off), 80% (ngc).

There are only 250+ trees for each hero, so 10 is too few to draw conclusions.
The probability of defense 6 for Barbarians and attack 5 for Swamp creatures is a well-known fact. To judge whether this probability applies specifically to Crag and Tazar, you need to check a much larger number of trees than 10.

Награда: Лучшему игроку сезона Осень 2009 за замок Конфлюкс

Statistics

Registered: 2/23/2010 12:22 AM UTC

Posts: 2,567

8/30/2012 8:14 PM UTC #3098

Auto-translated

Medium_HP,

Tootsie Pie

The point is that the _total_ number of trees is an astronomical quantity (roughly speaking, level 2 has 10 options, level 3 has 100, and so on almost the same way up to level 24). It is evident that 99+% are high-level branches, and only 0.0000000000000001 of the total is allotted to the first 8 (important to us) branches! Therefore, even considering that the authors changed this proportion — relatively few trees should still be assigned to the first options, roughly about 20 by eye, while for various Tazar types it's 5..7 for the first 3 levels, which allows Adelaide to get Diplomacy with a 90% probability.

In reply to zoog

Statistics

Registered: 4/1/2009 3:57 PM UTC

Posts: 756

8/30/2012 8:42 PM UTC #3099

Auto-translated

I didn't immediately grasp what was being discussed, as I'm used to calling something else a skill tree — namely, the full set of possible progression paths for a specific generation. There are 255 such trees.
If we're evaluating the number of specific progression options up to, say, level 3, then by eye it seems to me there aren't 5, or 7, or 20, but significantly more, even accounting for choosing only "good" skills. I could be wrong — the eye is not a precise instrument.

Награда: Лучшему игроку сезона Осень 2009 за замок Конфлюкс

Statistics

Registered: 9/26/2008 2:20 PM UTC

Posts: 2,571

8/31/2012 9:04 AM UTC #3100

It is used to think that Monere much better than Krag.

Земля нуля.
Все в ней нулево и кукольно.

Statistics

Registered: 2/23/2010 12:22 AM UTC

Posts: 2,567

9/4/2012 4:32 AM UTC #3101

Auto-translated

Medium_HP, what is the disagreement about? I was talking about these trees. Maybe there are more, but not by much. By the way, you yourself are confused: a tree is first a collection, and then there are as many as 255 of these trees.

Statistics

Registered: 4/29/2007 10:30 AM UTC

Posts: 11,704

9/4/2012 7:51 AM UTC #3102

Auto-translated

A skill tree is a combination of skills that are offered when a hero levels up, until all 8 slots are filled. There are a great many branches in the skill tree. However, there are only 255 such trees for each hero.

Statistics

Registered: 12/4/2010 3:26 PM UTC

Posts: 1,791

9/4/2012 9:20 AM UTC #3103

Auto-translated

I only know that in the damage calculation formula for a unit, 15% armor + bonuses are calculated last, which gives Taz'ar an advantage over Craig, where the attack, as far as I remember, is calculated somewhere at the beginning of the formula.

ICQ 335-665-133
Discord Sevensan#3379