SCIENCE! Peer Review at its finest! or Nothing to see here, move along, move along

Compact Subject
SCIENCE! Peer Review at its finest! or Nothing to see here, move along, move along
OR
Why are you clicking
OR
Scram
OR
Vamoose

OLD Subject
Live! Naked! SCIENCE! more messy, more dirty than your secondary science teacher told you! not clean and clinical like the science textbooks want you to think!
OR
A very important lesson in science and peer review
OR
We can do this with a little help from our friends
OR
Using 2* 1.11 micro trainer heroes or Comparing Recruit Training 11 & Color conversion vs. Recruit Training 8- Dark training [Analysis, leveling, Skill Roll]- Updated 2018-Jun-07

OLD Subject
Using 2* 1.11 micro trainer heroes or Comparing Recruit Training 11 & Color conversion vs. Recruit Training 8- Dark training [Analysis, leveling, Skill Roll]- Updated 2018-Jun-06

Edit
==Simple Advice==

Edit: While an interesting experiment, the results ended up being pretty boring.

-If you have the ascension items for 4* 4.70 or 5* 4.80, just do rainbow feeding, Zero has a great tutorial.

-If you have the 4* Heroes and 5* Heroes from summons, just feed them the same color 1* and 2* heroes because they have twice the skill roll of off color 1* and 2* heroes.

Edit:
-If you use Dark training, or any recruit training 5- 9, there is no huge advantage, but there is also no huge disadvantage over creating same color trainer heroes.
This seems to be a function of Dark training usually being bad, but no worse than rainbow feeding or training heroes from loot drops.

==Complex Analysis==

==2* 1.11 micro trainer hero or Color conversion of extra low cost training ( Recruit Training 11 )==

Use 1* 1.1 to level 4* or 5* hero of same color
4x 1* different color + 4x 2* different color to level one 2* 1.1 to 2* 1.11 micro trainer hero,
Use 2* 1.11 micro trainer hero to level 4* or 5* hero of same color.

Edit:
2* 1.11 micro trainer hero has a lower cost than turning a 1* 1.1 hero into a micro trainer hero because the same XP ascends a 1* 1.1 hero past 1* 2.1 .

Example
Use purple Nightshade 1* 1.1 to level purple 4* Sabina
Use 4x 1* different color + 4x 2* different color to level one purple Layla 2* 1.1 to 2* 1.11 micro trainer hero,
Use purple 2* 1.11 micro trainer Layla to level purple 4* Sabina

==Comparing 2* 1.11 micro trainer heroes/ color conversion ( RT 11 ) to Dark training ( RT 8 )==

==Update: Corrected math==
Dark training ( + 9.25% / 1k XP ) is better for skill roll than color conversion ( +6.849% skill roll/ 1k XP ). But more importantly Dark training is better XP for one recruit ( +64.8 XP / recruit ) than color conversion ( +43.8 XP / recruit )

While Dark training is not as good for XP per 1k food ( +38.11 XP/ 1k food ) than color conversion ( +44 XP / 1k food ) that is not worth creating micro training heros. Especially since recruits are much harder to acquire than food.

Color conversion also offers the advantage of working for any color, where Dark training only has the advantage when leveling a Dark hero. However, finished heroes can be left in extra low cost training camps indefinitely ( link ) so a better approach is to simply save finished heroes in camp, until you have one hero of each color you want to level or have all the ascension materials to level a single hero to maximum tier, maximum level.

==End Update==

==DATA==
==Creating a 2* 1.11 micro trainer hero ( RT 11 )==

Example
purple Layla 2* 1.1
+4%, +468 XP used to level same color hero

2* 1.1= +4x 1* 1.1 different color ( +600 XP )=
4.8k food
= 1.4= +4x 2* 1.1 different color ( +1560 XP)=
5.088k food

= purple 2* 1.11 Layla micro trainer
+4%, +696 XP used to level a same color hero

Food
1k Layla
+4k four 1* 1.1
+4.8k leveling Layla with four 1* 1.1
+4k four 2* 1.1
+5.088k leveling Layla with four 2* 1.1
= 18.888k food

Recruits
2 Layla
+8 four 1* 1.1
+8 four 2* 1.1
= 18 recruits ( edit: original mistake said 10 )

= purple 1* 1.1 Nightshade
+2%, +180 XP used to level same color hero
1k food
2 recruits

Total
+6% ( +6.849% skill roll/ 1k XP )
+876 XP ( +44 XP / 1k food, +73 XP / recruit )
19.888k food
20 recruits ( edit: original mistake said 12 )

Extra low cost
2 recruits, 1k food
1* & 2*

==Dark Hero ( RT 8 )==
5 recruits, 8.5k food
1* & 2*

purple Layla 2* 1.1
+4%, +468 XP used to level same color hero
= purple 1* 1.1 Nightshade
+2%, +180 XP used to level same color hero

Total
+6% ( + 9.25% / 1k XP )
+648 XP ( + 38.11 XP/ 1k food, +64.8 XP / recruit )
17k food
10 recruits

==Edit: Original mistake==

Click arrow only if you are good at keeping two similar conclusions separate

Dark training ( + 9.25% / 1k XP ) is better for skill roll than color conversion ( +6.849% skill roll/ 1k XP ).
But color conversion is actually cheaper XP for food and recruits ( +44 XP / 1k food, +73 XP / recruit ), this includes the cost of creating the micro trainer hero, than Dark training ( + 38.11 XP/ 1k food, +64.8 XP / recruit ), offering +15% XP per 1k food and +12.6% XP per recruit.

Color conversion also offers the advantage of working for any color, where Dark training only has the advantage when leveling a Dark hero. When Dark training is used to level a different color hero, the skill roll drops to +4.625 / 1k XP which is lower than the +6.849% skill roll / 1k XP of color conversion.

==End original==

Read this…tried. Reminded me of story problems in calc. 35 years on and still hate them.

So, let’s try and explain this to dense old uncle bud. You are saying that levelling a dark 4/5* can be done cheaper but longer process…right?

Dark only and 2* is the key player.

Exactly how many 1* do I feed Layla? Looks like four and no dark. Can that be done while waiting for the (four) non-dark 2*?

Big question is how far along the process of levelling the legendary should these faux trainers be put to use? It has been written that low levels in each tier are fed 1s with 2 feeders further along with 3* and trainers at highest levels to save on ham; so where in that mix are super-Laylas best used?

Trust your work based on previous experience, so I know I will be using this…when I understand it.

Thanks

I used to be good at calc and believed my English is above average, however you lost me after say Subject. It seems it’s time to grab a spreadsheet…

1 Like
Edit:This is based on incorrect math, do not follow this advice, see original post for corrected advice

There are several things to take away:

-Dark training is bad. Any of the recruit training 5- 9 are bad. All the numbers back it up.

-If you have the ascension items for 4* 4.70 or 5* 4.80, just do rainbow feeding, Zero has a great tutorial.

-If you have the 4* Heroes and 5* Heroes from summons, just feed them the same color 1* and 2* heroes because they have twice the skill roll of off color 1* and 2* heroes.

-Rainbow feeding for 4* 3.60 and 5* 2.60 ( farmable ascension items only ) is bad because off color skill roll chance is half. It is better to use the same color 1* 1.1 from extra low cost training ( RT 11 ) and feed the off color 1* 1.1 & 2* 1.1 to the same color 2* 1.1 ( making it same color 2* 1.11 ) until you are happy with the special skill of your 4* hero or 5* hero because 4* 3.60 is only guaranteed 3/8 special skill and 5* 2.60 is only guaranteed 2/8 special skill.

-Rainbow feeding gets expensive when you get to 5* 3.63

-Same color 1* feeding gets expensive when you get to 5* 3.63

1 Like

I get the basic ideas and agree completely.

  1. TC5-9 are a waste of resources.
  2. High level heroes, especially 5* heroes are very expensive to feed per hero, so using higher value heroes to feed them works much better.
  3. The odds of special increase appear to be improved by your microtrainer color conversion technique, compared to off-color feeding, 'tho not as good as on-color 1* or raw 2* feeders. (Haven’t checked your math, but it seems believable)

You kinda lost me halfway there with the 2* 1.11 microtrainer color conversion. It sounds fascinating, really clever but confusing to me…there a bunch of math without full explanation, or all basic assumptions. I got a few questions:

  1. Is taking the 2* to 1.11 chosen solely because over a large number of randomly generated feeders, you would expect each group of ten to include a 1* of each color and a 2* of each color, and using those 8 off-color heroes will get your microtrainer to level 1.11?
  2. Have you worked out the food/recruit costs vs XP value for microtrainers of higher or lower levels than that?

Related question: I’ve noted down feed values for unleveled heroes and trainer heroes (special case leveled ones), and noted that leveled heroes give more XP…but haven’t ever tried to figure out that calculation. Would you share what you’ve worked out on that?

I think I still need to understand how you are creating the micro-trainers.

Then, an explanation of where in the levelling process this creation belongs.

The word dark is confusing us, possibly. I never use those tc levels. Just thought your micro-trainer is somehow a purple only hero cum trainer.

Pls elucidate on your process…words.

This was a fun concept, so I started trying to sort through it in excel myself…and I think your math on the cost of the TC11 2* 1.11 micro trainer is off…

I see the inputs for Layla:
9X output from RT11 = 18 recruits + 9k food
+ 8x1.2k food to level up micro trainer
= 18 recruits + 18.6k food
I’m not sure why I get 18.6k food and you get 18.888k but it hardly matters. The big difference is you lost 8 recruits…

So the total for that plus Nightshade
= 20 recruits + 19.6k food

I’m not building micro-trainers, so I’ll use your feed XP value, and I get:
+876 XP ( +45 XP / 1k food, +44 XP / recruit )

And that’s a mixed comparison vs. RT8; 18% less food, 32% more recruits (and no swords or backpacks, and 5X the time running your TC.)

Probably…still better, but perhaps not enough to justify the extra effort. Now it depends on the limiting factor(s): food/swords+backpacks vs recruits/time.

1 Like

Subject
Well that could have gone better or Peer review, it really works

Click arrow to discover how I REALLY feel

#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@#%!?&@#%*!?&@

Argh. 2+8+8 does in fact equal 18.

This my fine friends is why you need to publish your science.

When I get access to a computer I will rewrite the post.

Edit:
@BarryWuzHere

First
Thank you
thank you
thank you

For taking the time, and effort, to review my math. My entire purpose to posting to the forums is to double, triple and quadruple check my assumptions, math, and conclusions. Science is messy, but the result vastly improves our lives.

If I could give you the forum version of Reddit gold, I would.

Second
Food, and iron, can be gotten in multiple ways ( including IAP directly at the hero ascension menu and the builder upgrade menu ) but recruits can only be indirectly purchased through Fast lane packs ( World Energy flask+ loot ticket= instant recruits ).

So in my view, recruits are the crucial ingredient in training camps. So a savings of 50% on recruits makes a huge difference ( 10 recruits versus 20 recruits )

Third

I agree with your conclusion.

I double checked my screen shots, and my food is accurate for my experiment. I am not willing to waste heroes trying to replicate it.

Maybe someone smarter than me can explain it.

But I think it is a waste of both our times to try and figure it out in this context. We can revisit it if people find a related bug.

2 Likes

You are most welcome.

I hope you haven’t leveled up any 5* heroes with 1.11 micro trainers yet :rofl: That would hurt me, and you seem even more efficiency obsessed than I am!

1 Like

Nah.

My camp “Alpha” has 615 finished heroes sitting in extra low cost training ( recruit training 11 ) waiting for a decent rainbow 5* hero team ( my Summoning Fu is weak ). So I decided to spend some loot drop heroes exploring micro trainer heroes.

My original error made for a fascinating discovery, my corrected post is pretty boring. Not sure I would have even posted it.

All good; glad it was just an experiment :slight_smile:

But methinks you got just a tiny bit carried away re-titling this post? :rofl:

1 Like

I always wondered “How long can a post title be?”

Now I know :grinning:

SCIENCE!!

Edit:
Okay, that is getting annoying

Click to see maximum title length

Live! Naked! SCIENCE! more messy, more dirty than your secondary science teacher told you! not clean and clinical like the science textbooks want you to think! Or Using 2* 1.11 micro trainer heroes or Comparing RT 11 & Color conversion vs. RT 8- Dark trai

1 Like