Defensive Mana - Troops/Differences/Turn- Based Mana Calculations and it's Discovery

[Zack]

@Faeldin

How do your values compare

Following your list I’ve prepared a similar table now:

	total tiles	mana generation bous	tiles in styx1	tiles in styx2/3	MP in styx1	MP in styx2/3	total MP	required MP
mana modifier			0	100
Styx 1	5	20	5	0	600	0	600	600
Styx 1	4	50	4	0	600	0	600	600
Styx 1	3	100	3	0	600	0	600	600
Styx 2	8	20	5	3	600	660	1260	1200
Styx 2	7	43	5	2	715	486	1201	1200
Styx 2	6	67	4	2	668	534	1202	1200
Styx 2	5	100	3	2	600	600	1200	1200
Styx 3	11	19	6	5	714	1095	1809	1800
Styx 3	10	30	5	5	650	1150	1800	1800
Styx 3	9	50	4	5	600	1250	1850	1800
Styx 3	8	75	4	4	700	1100	1800	1800
Styx 3	7	100	3	4	600	1200	1800	1800
Styx 3	6	150	3	3	750	1050	1800	1800

Those are the same numbers except for

• styx2 goes with 8 tiles at 20%
• styx3 goes with 11 tiles at 19% (with 18% I get 1798 MP)

@Faeldin

has anyone perhaps verified some of these breakpoints already?

I think so far I have only verified that styx1 is charged with 5 tiles at 20%

[G_H_O_S_T]

Yo just got done finalizing a merge so after war I’ll start testing on the tides heroes @Zack

[Groot61]

@Elioty33

defense

My Godfather , my best friend and mom died this year so I don’t understand half of your discussion with others.
Can you send me basics of tactics for dragon s, raid tournaments and I have a problem with leveling food amd metal. It stops at 20 half of them.

Thanks Ghost

[Groot61]

Don’t worry (be happy)for them. They are properly rewarded for this hard job.

[Zack]

Been some time, work is getting crazier than usual - it’s getting closer to the end of the year.

We wanted to check tile mana for 4* to verify that they also get 75 MP for each combo1 tile, so here are two tests with a fast 4*.
As we now know, fast 4* charge with 899 MP on defense.

Test 1: Mist with +16% mana generation bonus from mana troops level 21 (+11%) and Zhabog (+5%) getting mana from 4 regular turns, 2 combo1 tile and 5 turns with -64% for 898 MP.
898 = 400 x 1.16 + 2 x 75 x 1.16 + 500 x (1.16 - 0.64)

Expectation: 4* get 75 mana for each combo1 tile just like the 5* do, so 898 are not enough to charge Mist

=> Mist is not charged, as expected

Test 2: Mist with +16% mana generation bonus from mana troops level 21 (+11%) and Zhabog (+5%) getting mana from 7 regular turns and 1 combo1 tile for 899 MP.
899 = 700 x 1.16 + 75 x 1.16

Expectation: Mist will get 75 MP from a combo1 tile and charge from the 899 MP

=> Mist is charged, as expected
=> 4* get 75 MP from a combo1 tile, as expected

Bonus test 3: Fast 5* Akkarog with 11% mana generation bonus from mana troops level 21 (+11%) getting mana from 2 regular turns, 3 combo1 tiles and 6 turns with -34% for 933.75 MP.
933.75 = 200 x 1.11 + 3 x 75 x 1.11 + 600 x (1.11 - 0.34) = 934?

Expectation: The 0.75 mana from the three tiles is added and then gets rounded up and Akkarog is charged from the 933.75 MP. Otherwise the mana for each tile is treated separately and gets rounded down and he is not charged from the 933.75 MP.

=> Akkarog is not charged
=> The MP for each tile gets treated separately

Bonus test 4: Fast 5* Akkarog with 14% mana generation bonus from mana troops level 11 (+9%) and Zhabog (+5%) getting mana from 7 regular turns, 1 combo1 tile and 1 turn with -64% for 933.5 MP.
933.5 = 700 x 1.14 + 75 x 1.14 + 100 x (1.14 - 0.64) = 934?

Expectation: The .5 MP for the tile gets rounded up and Akkarog gets charged from the 933.5 MP

=> Akkarog is not charged
=> The 0.5 MP gets rounded down for tiles

@CrazyChemist3891 Sorry, I didn’t find the time yet to continue with the overspill mana tests, but I also want to first find a suitable overview table for these tests that allows to check on each mana overspill for a systematic approach

[Elioty33]

@Zack

=> The MP for each tile gets treated separately

Which is what I expected too And it need to be remembered when working on overspill of mana charges as only one tile does the overspill. The tiles before have the current (before getting hit by tiles of this combo rank) mana modifier (if any), the tile that does the overspill as the same mana modifier and then the other tiles of the same combo gets the mana modifier of the next charge applied (if any, no modifier for ninja for example). At least, that’s how I expect it to work.

[G_H_O_S_T]

I won’t be in a place where I can check on the tides heroes until the 1st. Been really wrapped up lately

[Zack]

@Elioty33

@Zack

=> The MP for each tile gets treated separately

Which is what I expected too And it need to be remembered when working on overspill of mana charges as only one tile does the overspill. The tiles before have the current (before getting hit by tiles of this combo rank) mana modifier (if any), the tile that does the overspill as the same mana modifier and then the other tiles of the same combo gets the mana modifier of the next charge applied (if any, no modifier for ninja for example).

Great conclusion!
That result actually came as a side effect for the test

@G_H_O_S_T

I won’t be in a place where I can check on the tides heroes until the 1st. Been really wrapped up lately

No worries, but I may go for some attempts against a Phorcys in the meantime. My tests with combo2 tiles went phenomenally well, so I feel lucky that tests against Phorcys may also go somewhat smooth.

[Zack]

New tests for combo2 tiles because:

@Elioty33

0.75 tiles worth of mana is for each tile that hits a defense hero. And for each combo, tiles should give 0.1 tiles less worth of mana for the hit defense heroes (or minus 10% of the base 0.75, not totally sure on this one). Example: on the first move (without a combo straight from the start actually), each tiles should give 0.75 worth to the hit defense heroes, on the second combo, 0.65, on the third 0.55 and so on with a minimum of 0.10 tiles worth of mana (if my first assumption is correct, would be different if it’s minus 10% for the base 0,75 for each combo, in this case, the minimum would be 10% of 0.75 = 0.075 tiles worth of mana).

So how does the progression for tile combos go?
We verified that combo1 tiles give 75 MP.
Do combo2 tiles give 10 MP less, i.e. 65 MP, or 10% less, i.e. 67.5 MP, which then is either 67 MP or 68 MP?

Test 1: 5* Anne with +2% mana generation bonus from her family bonus with Sargasso getting mana from 7 regular turns, 1 combo2 tile and 4 turns with -64% for 934 MP or 932 MP.
65 x 1.02 = 66.3 MP
67 x 1.02 = 68.34 MP
68 x 1.02 = 69.36 MP

934 MP = 700 x 1.02 + 67 x 1.02 + 400 x (1.02 - 0.64) = charged
933 MP = 700 x 1.02 + 65 x 1.02 + 400 x (1.02 - 0.64) = not charged

Expectation: -10% progression and therefore Anne will charge.

=> Anne is charged
=> combo2 tile gives at least 67 MP
=> combo2 tile gives 10% less, not -10 MP

Test 2: 5* Lianna with +9% mana generation bonus from mana troops level 11 (+9%) getting mana from 5 regular turns, 1 combo2 tile and 7 turns with -64%.
67 x 1.09 = 73.03 = 73
68 x 1.09 = 74.12 = 74

934 = 500 x 1.09 + 68 x 1.09 + 700 x (1.09 - 0.64)
933 = 500 x 1.09 + 67 x 1.09 + 700 x (1.09 - 0.64)

Expectation:

=> Lianna is charged
=> combo2 tile gives 68 MP (the 75 x 0.9 = 67.5 gets rounded up)

I have no idea yet on how to set up a reasonable test for a combo3 tile

edit: Is there a raid video somewhere with a long cascade with combo 10 or more where we can see, if the mana bar is increasing at all from combo 9 onwards?

I might very well remember incorrectly, but I think that from approx. combo 9 onwards the raid target gets basically no more mana from tiles or close to no more mana.

If that would be correct, the following progression may apply:

combo	mana points	factor
1	75	1
2	68	0,9
3	54	0,8
4	38	0,7
5	23	0,6
6	12	0,5
7	5	0,4
8	2	0,3
9	0	0,2

[Zack]

Well, well, I found a youtube video with a combo13 cascade here from Yukimura Sanada. The cascade starts at 15:34. Combo 13 happens at 15:46, combo 14 at 15:48.

It’s a 4* bloody battle tournament from 3 years ago, so there are no mana troops involved. At 4-70+18 Chao may have the +2% mana generation bonus, but that should be too small to have an impact on this rough analysis.

The comparison of the two mana bars from Chao shows a slight increase from the 5 combo13 tiles
=> so the defenders still get a tiny amount of mana even after combo9
=> looks like there is indeed a minimum base amount of mana that the defenders get per tile

I took screenshots and tried to analyse the pixels. The full mana bar is something like 58 pixels. The difference between the two mana bars is approx. 2-3 pixels from the 5 tiles. As a fast 4* Chao’s mana bar is 899 MP.

58 pixels = 899 MP
5 tiles = 2 pixels = 2 x 899 / 58 = 31 MP => 6.2 MP per tile
5 tiles = 3 pixels = 3 x 899 / 58 = 46.5 MP => 9.3 MP per tile
=> So the combo13 tiles give something in the range of 6.2 - 9.3 MP

The only way I found to get a similar value at 9 or 10 tiles and onwards is to apply the 10% progression as follows:

combo 1 = 75 MP = 75 MP x 100% (and rounded) (verified)
combo 2 = 68 MP = 75 MP x 90% (and rounded) (verified)
combo 3 = 60 MP = 75 MP x 80% (and rounded)
combo 4 = 53 MP = 75 MP x 70% (and rounded)
combo 5 = 45 MP = 75 MP x 60% (and rounded)
combo 6 = 38 MP = 75 MP x 50% (and rounded)
combo 7 = 30 MP = 75 MP x 40% (and rounded)
combo 8 = 23 MP = 75 MP x 30% (and rounded)
combo 9 = 15 MP = 75 MP x 20% (and rounded)
combo 10 = 8 MP = 75 MP x 10% (and rounded)
combo 11 and higher = 8 MP = 75 MP x 10% (and rounded) (approximately correct)

=> The 8 MP for a combo13 tile falls well into the range that I roughly identified from the mana bars in the video

@Damirius and @G_H_O_S_T please have a look at these pixels as well

[Zack]

Got the data on the tide heroes. Phorcys was leveled to 3-70, which allowed him to survive the tests.

Test 1: 5* Phorcys with +9% mana generation bonus from mana troops level 11 (+9%) getting 933 MP from 7 regular turns and 2 turns with -24% for fast tide speed.
933 = 700 x 1.09 + 200 x (1.09 - 0.24)

Expectation: His fast tide speed won’t be charged with 933 MP

=> Phorcys’ fast tide speed is not charged with 933 MP, as expected

Test 2: 5* Phorcys with +10% mana generation bonus from mana troops level 3 (+5%) and Zhabog (+5%) getting 934 MP from 6 regular turns, 3 turns with -34% and 1 turn with -64% for fast tide speed.
934 = 600 x 1.10 + 300 x (1.10 - 0.34) + 100 x (1.10 - 0.64)

Expectation: Phorcys’ fast tide speed will charge with 934 MP

=> Phorcys’ fast tide speed is charged with 934 MP, as expected

Test 3: 5* Phorcys with +10% mana generation bonus from mana troops level 3 (+5%) and Zhabog (+5%) getting 933 MP from 9 regular turns and 8 turns with -64% for average tide speed.
933 = 900 x (1.10 - 0.25) + 800 x (1.10 - 0.25 - 0.64)

Expectation: He won’t be charged with 933 MP for his average tide speed

=> Phorcys’ average tide speed is not charged with 933 MP, as expected

Test 4: 5* Phorcys with +10% mana generation bonus from mana troops level 3 (+5%) and Zhabog (+5%) getting 934 MP from 10 regular turns and 4 turns with -64% for average tide speed.
934 = 1000 x (1.10 - 0.25) + 400 x (1.10 - 0.25 - 0.64)

Expectation: Phorcys’ average tide speed will be charged

=> Phorcys’ average tide speed is charged with 934 MP, as expected

=> With this we have screenshot verified all mana speeds on defense

[Zack]

With our new understanding of the mana bar of the defense team it is unclear, if the amount of mana you steal with heroes like Onatel, R&N, Ithar, etc is based on the larger mana bar of the defense or if this percentage is converted to the base mana bar.

Test 1: Fast 5* Obakan of offense with +16% mana generation bonus getting mana from 6 tiles and R&N’s 50% mana steal from a charged fast 5* (Phorcys with fast tide speed)
934 x 50% / 5 = 93.4 = 93 MP for each hero or
800 x 50% / 5 = 80 MP for each hero

For my Obakan that would either mean
803 = 600 x 1.16 + 93 x 1.16 = charged
788 = 600 x 1.16 + 80 x 1.16 = not charged

Expectation: You get what you take, so it should be 93 MP for each hero and Obakan will be charged

=> Obakan is charged
=> you steal mana based on the larger mana bar of the defense, as expected

Test 2: Fast 5* Obakan of offense with +16% mana generation bonus getting mana from 6 tiles and R&N’s 50% mana steal from a charged fast 4* (Woolerton)
899 x 50% / 5 = 89.9 = 90 MP for each hero or
899 x 50% / 5 = 89.9 = 89 MP for each hero

For my Obakan that would either mean
800 = 600 x 1.16 + 90 x 1.16 = charged
799 = 600 x 1.16 + 89 x 1.16 = not charged

Expectation: Probably gets rounded up and Obakan will be charged

=> Obakan is not charged
=> The 899 x 50% / 5 = 89.9 gets rounded down

Mana stealing
I love R&N’s 50% mana steal (and the rest of her skills) and I’m currently leveling Ithar, who has a 35% mana steal.

So how much mana do Ithar and R&N steal then? And which mana generation bonus is required to charge a fast hero with 6 tiles when R&N or Ithar steal mana from a fully charged fast hero?

Ithar stealing from a 5*:
With 21% mana generation bonus a fast hero will charge with 6 tiles and Ithar stealing from a charged fast 5*.

	Mana points	35% steal	MP for each of 5 heroes
Charge 1	572	200	40
Magic 1	642	224	44
Styx 1	700	245	49
very fast	759	265	53
fast	934	326	65
Styx 2	1051	367	73
Charge 2	1144	400	80
average	1168	408	81
slayer	1284	449	89
slow/styx3	1401	490	98
Magic 2	1436	502	100
very slow	1576	551	110
Charge 3	1716	600	120

Ithar stealing from a 4*
With 21% mana generation bonus a fast hero will charge with 6 tiles and Ithar stealing from a charged fast 4*.

	Mana points	35% steal	MP for each of 5 heroes
Charge 1	550	192	38
Magic 1	618	216	43
Styx 1	674	235	47
very fast	730	255	51
fast	899	314	62
Styx 2	1011	353	70
Charge 2	1101	385	77
average	1124	393	78
slayer	1236	432	86
slow/styx3	1348	471	94
Magic 2	1382	483	96
very slow	1517	530	106
Charge 3	1652	578	115

Ithar stealing from a 3*:
With 22% mana generation bonus a fast hero will charge with 6 tiles and Ithar stealing from a charged fast 3*.

	Mana points	35% steal	MP for each of 5 heroes
Charge 1	529	185	37
Magic 1	594	207	41
Styx 1	648	226	45
very fast	702	245	49
Fast / tide	864	302	60
Styx 2	972	340	68
Charge 2	1059	370	74
average	1081	378	75
slayer	1189	416	83
slow/styx3	1297	453	90
Magic 2	1329	465	93
very slow	1459	510	102
Charge 3	1589	556	111

R&N stealing from a 5*:
With 15% 16% mana generation bonus a fast hero will charge with 6 tiles and R&N stealing from a charged fast 5*.

	Mana points	50% steal	MP for each of 5 heroes
Charge 1	572	286	57
Magic 1	642	321	64
Styx 1	700	350	70
very fast	759	379	75
fast	934	467	93
Styx 2	1051	525	105
Charge 2	1144	572	114
average	1168	584	116
slayer	1284	642	128
slow/styx3	1401	700	140
Magic 2	1436	718	143
very slow	1576	788	157
Charge 3	1716	858	171

R&N stealing from a 4*
With 16% 17% mana generation bonus a fast hero will charge with 6 tiles and R&N stealing from a charged fast 4*.

	Mana points	50% steal	MP for each of 5 heroes
Charge 1	550	275	55
Magic 1	618	309	61
Styx 1	674	337	67
very fast	730	365	73
fast	899	449	89
Styx 2	1011	505	101
Charge 2	1101	550	110
average	1124	562	112
slayer	1236	618	123
slow/styx3	1348	674	134
Magic 2	1382	691	138
very slow	1517	758	151
Charge 3	1652	826	165

R&N stealing from a 3*:
With 16% 17% mana generation bonus a fast hero will charge with 6 tiles and R&N stealing from a charged fast 3*.

	Mana points	50% steal	MP for each of 5 heroes
Charge 1	529	264	52
Magic 1	594	297	59
Styx 1	648	324	64
very fast	702	351	70
fast	864	432	86
Styx 2	972	486	97
Charge 2	1059	529	105
average	1081	540	108
slayer	1189	594	118
slow/styx3	1297	648	129
Magic 2	1329	664	132
very slow	1459	729	145
Charge 3	1589	794	158

edit: corrected the required mana generation bonus for R&N

[Ufeel]

The last section is very nice table and great info on how “practically” make R&N and Ithar help charge other heroes on 6 tiles … @PlayForFun do you think this info is appropriate to pin on the top (maybe just a link) of R&N and Ithar threads ?

–
Just think for fun that it may be possible Ithar help R&N to charge on 6 tiles, and R&N steal further 50% … So for some average (esp. Yellow/red) speed heroes with suitable bonus can also charge in 6 tiles with 35%+50% mana stealt from both heroes ? – Ithar also has elemental link +10% mana gen, and with R&N, it will applies to all in 2024
----**

Thanks again for your invaluable hard work

[Ufeel]

@Zack Following your way of calculation, it seems possible that this scenario is possible:

We match red 6 tiles, then match yellow 6 tiles. Then,

Ithar steal full-charged 5*-fast , to make R&N full-mana
(with 16% mana-gen-bonus on R&N, like Obakan above)

R&N steal full-charged 5*-average, and make C2 LiXiu full-mana
with 27%-mana-gen-bonus on C2 LiXiu plus Ithar’s +10% EL

But I am a new student just to learn your formula , not sure if I calculate things correctly ?

[Zack]

@Ufeel

We match red 6 tiles, then match yellow 6 tiles. Then,

Ithar steal full-charged 5*-fast , to make R&N full-mana
(with 16% mana-gen-bonus on R&N, like Obakan above)

R&N steal full-charged 5*-average, and make C2 LiXiu full-mana
with 27%-mana-gen-bonus on C2 LiXiu plus Ithar’s +10% EL

But I am a new student just to learn your formula , not sure if I calculate things correctly ?

Yes, that is correct

[CrazyChemist3891]

@Zack every mana speed? Did I miss the higher charges of ninja, magic and styx?

[Zack]

@CrazyChemist3891

every mana speed? Did I miss the higher charges of ninja, magic and styx?

Yes, those were screenshot confirmed in this thread with

• Post 185: styx2 and styx3
• Post 195: magic2
• Post 200: charge2 and charge3

[CrazyChemist3891]

Thank you very much for directing me to those.
Awesome work!

[CrazyChemist3891]

In order to get the overview for myself:

1. Confirmed mana gain modifiers for heroes with multiple speeds/charges.

2. Confirmed star-dependent modifier on defense for the mp amount necessary.

3. Confirmed decreased mana gain per hitting tile on defense.

4. Missing understanding of overspill of ghosted tiles on offense for heroes with multiple charges.

@Zack @Elioty33 correct so far or did I miss some results of your research again?

[Zack]

That is correct.

In addition, we also verified:
5. mana gain at end of turn = 100 MP
6. only whole numbers are valid for MP values
7. each tile gets treated separately, and tile mana gets rounded down
8. On offense: mana steal takes from the larger mana bar of the defense