Spire squad size calc - we need your data!

[michmarc]

12 / 27 / 334 / 86 / 1545 / 101 => 1298 / 577 (3/22/2020)
12 / 27 / 347 / 91 / 1665 / 103 => 1401 / 623 (4/ 5/2020)
12 / 27 / 353 / 92 / 1665 / 105 => 1427 / 634 (4/12/2020) (1 extra expansion was premium)
12 / 27 / 356 / 95 / 1665 / 105 => 1437 / 639 (4/19/2020)
12 / 27 / 361 / 100 / 1665 / 105 => 1468 / 652 (4/26/2020)

12 / 27 / 367 / 104 / 1788 / 106 => 1520 / 675 (5/ 3/2020)

 

[DeletedUser5371]

16 / 35 / 597 / 290 / 4626 /127 => 4060 / 1803 [2020/05/03]......differences are 1 AW level, beerfest, teahouse, imp senate (built) , union tower

 

[Deleted User - 3932582]

16 / 35 / 597 / 290 / 4626 /127 => 4060 / 1803 [2020/05/03]......differences are 1 AW level, beerfest, teahouse, imp senate (built) , union tower

So you've got 4 new research techs as listed - correct? Were all of these completely unlocked (i.e. goods paid for, green checkmark) before Sunday? Thanks!

 

[Enevhar Aldarion]

This will be the last one with me still in chapter 15. I added 6 more AW levels, for an increase of 39 for squad size, or 6.5 per AW level, so that is staying right in there with the roughly 6.X per AW level that has been happening.

15/35/556/188/4626/127 - 3340/1483 [5/3/2020]

 

[Deleted User - 3932582]

12 / 27 / 367 / 104 / 1788 / 106 => 1520 / 675 (5/ 3/2020)

Everyone reporting - when adding new expansions, can you please mention which kind is it (i.e. research/map/premium)?

Also, on a going forward basis, please add how many new research techs were unlocked (hopefully it's not based on KPs, that would be a pain). Probably the easiest way to report on that would be to take a screenshot of your research tree (the last part), so it is easy to see what was unlocked vs previous week. This way we would also be covered if research part is based on KP.

Thanks everyone!

 

[Deleted User - 3932582]

This will be the last one with me still in chapter 15. I added 6 more AW levels, for an increase of 39 for squad size, or 6.5 per AW level, so that is staying right in there with the roughly 6.X per AW level that has been happening.

15/35/556/182/4626/127 - 3340/1483 [5/3/2020]

So 188 total AW levels then, correct?

 

[Enevhar Aldarion]

@MinMax Gamer Yep, that is what it says..........now. lol

 

[Deleted User - 3932582]

Now I am comparing mine and hondamike's numbers and he has 108 more AW levels than I do and a spire squad difference of 720. I have been seeing around 6.25 to 6.5 squad increase per AW level, so if you multiply that out by 108, you get between 675 and 702. Subtract that from his squad size and it becomes a difference of roughly 18 to 45 that his being in chapter 16 and five completed researches is adding.

Not sure which numbers you're comparing As of this week, your differences are 102 AW levels, not 108. Mike also completed 7 researches total, not 5.

 

[qaccy]

15 / 35 / 507 / 270 / 3573 / 143 => 4243 / 1884 [Apr 19 2020]
16 / 35 / 507 / 270 / 3573 / 143 => 4261 / 1892 [May 03 2020]

Two changes this week compared to last week - one research was unlocked, and chapter increased by one (because that was the unlocked research). This week I'll be unlocking the optional Union Tower research to be able to better see what effect an individual research that isn't the first of a chapter causes.

 

[Enevhar Aldarion]

Not sure which numbers you're comparing As of this week, your differences are 102 AW levels, not 108. Mike also completed 7 researches total, not 5.

I deleted that before you posted. You might have had the message open for too long.

 

[Deleted User - 4646370]

16/35/444/178/3699/113 --> 2979/1323 (The 4 expansions are 1 research one (that I have had in inventory since I ended chapter 15) and 3 scouting ones. The AW upgrade is Simia to level 11. I also researched Senate and both Embassies.)
8/21/167/31/570/65 --> 469/209 (In this city I researched this week both cultural buildings at the beginning of Orcs, and the Main Hall upgrade.)

16/35/444/181/3699/113 --> 3010/1337 (Upgraded timewarp from L9 to L11 and thermal springs to L3, and researched both beer festival and tea house).
8/21/169/33/570/65 --> 482/215 (upgraded Abyss from L9 to L11, and researched barracks upgrade, elder treant, gems manufactory upgrade, city expansion 33, training grounds upgrade and orc warrior).

 

[Deleted User - 4646370]

15/35/598/440/4140/149 ==> 5592/2483 (the +22 is probably due to wonders upgraded Sunday last week between Spire starting and data collection in Monday)
12/25/303/71/2133/100 ==> 1242/552
12/26/295/54/2349/100 ==> 1202/534 (to be confirmed - no wonder had been sold so AW levels shouldn't decrease).

16/35/617/444/4140/150 --> 5791/2572
12/26/303/73/2253/100 --> 1280/569
12/27/303/57/2349/101 --> 1239/551
All these cities also do research, but I don't check detailed data on these.

 

[michmarc]

Added expansion information. Not included are the six starting expansions. For past research, I did take the three optional squad size updates in Ch 1-4 (because I didn't know any better) but have skipped the rest.

Ch / MH / Prov / AW / SS / Exp: Research-Province-Premium
12 / 27 / 334 / 86 / 1545 / 43-47-6 => 1298 / 577 (3/22/2020)
12 / 27 / 347 / 91 / 1665 / 43-48-6 => 1401 / 623 (4/ 5/2020)
12 / 27 / 353 / 92 / 1665 / 43-49-7 => 1427 / 634 (4/12/2020)
12 / 27 / 356 / 95 / 1665 / 43-49-7 => 1437 / 639 (4/19/2020)
12 / 27 / 361 / 100 / 1665 / 43-49-7 => 1468 / 652 (4/26/2020)
12 / 27 / 367 / 104 / 1788 / 43-50-7 => 1520 / 675 (5/ 3/2020) [Research: 19 done in Ch12 -- everything through and including the column with workshops]

 

[Deleted User - 3932582]

I'll be unlocking the optional Union Tower research to be able to better see what effect an individual research that isn't the first of a chapter causes.

That should be a great next datapoint because Union Tower is so cheap (~1/10 of other techs in this chapter). Hopefully, we should be able to see if it is the tech number, or contributed KPs that are important.

 

[Enevhar Aldarion]

15/35/555/182/4626/126 - 3273 [4/19/2020]

15 / 35 / 593 / 281 / 4626 / 126 => 3876 [2020/04/12]

I went back and found two weeks where Mike and I were the same, except for AW levels. He had 99 levels more than me and a Spire squad 603 higher. That is 6.09 troops per AW level. That is definitely less than my small sample size of 6.28 and 6.5, when I added 7 and 6 levels during a couple of weeks with no other changes. And way less than the 6.9 @MinMax Gamer last posted as the expected amount. The only other number different is the completed provinces, but is not supposed to have an effect.

And now today's numbers.

Mine: 15/35/556/188/4626/127 - 3340/1483 [5/3/2020]

Mike: 16 / 35 / 597 / 290 / 4626 /127 => 4060 / 1803 [2020/05/03]

The difference in AW levels grew by only 3, but he is also in chapter 16 and has completed seven researches. His Spire squad is now 720 higher than mine, or an increase of 117. If the three AW levels account for about 19 of that, then the seven researches account for an increase of roughly 98. 7x14=98, so could it be as simple as all chapter 16 techs are worth 14 squad size each?

 

[Deleted User - 3932582]

...so could it be as simple as all chapter 16 techs are worth 14 squad size each?

I wish @qaccy above added 1 chapter 16 tech, and it added 18 points. Another city added 11 and 9 chapter 16 techs in two weeks, and each added about 24-31 per tech... These increases are definitely correlated with base Spire SS itself (meaning cities with higher SS get bigger increments), that's how it works for pretty much all parameters so far. Tiny cities get tiny increments on AW levels, research and expansions etc.

Meaning in your comparison with hondamike, Mike would be getting bigger incremental increases on AW levels on average, and in limited data that I've seen so far the same applies to research impact.

6.9 parameter was used to fit the most cities with chapter 15 data. As I already mentioned, constant increases per AW level is definitely not how the actual calculation works, but it was a good approximation for specific set of end-game cities. Difference between 6.1 and 6.9 over 100 AW levels of difference is only 80 SS points, which was well within the accuracy of the end-game model. This model does not work for cities other than the ones at the end of chapter 15.

 

[Deleted User - 1178646]

I have the feeling that there are x parameters and those get multiplied with eachother.
This would suggest that if 1 value is high an increment somewhere else would also create a bigger increment.

Could this be the case?

 

[Deleted User - 3932582]

I have the feeling that there are x parameters and those get multiplied with eachother.
This would suggest that if 1 value is high an increment somewhere else would also create a bigger increment.

Could this be the case?

Yes, that's my current hypothesis. Though what is getting multiplied is already some other function of parameters (as e.g. zero AW levels does not mean zero Spire SS). It can also be a sum of some products etc. This also makes analyzing multiple variable changes difficult as impact of combinations is non-linear, so you basically need to guess the general function for dependencies, then fit parameters and see if it works out. Because of this non-linearity datapoints with only one variable changing are so much more valuable.

 

[Enevhar Aldarion]

I wish @qaccy above added 1 chapter 16 tech, and it added 18 points. Another city added 11 and 9 chapter 16 techs in two weeks, and each added about 24-31 per tech... These increases are definitely correlated with base Spire SS itself (meaning cities with higher SS get bigger increments), that's how it works for pretty much all parameters so far. Tiny cities get tiny increments on AW levels, research and expansions etc.

Meaning in your comparison with hondamike, Mike would be getting bigger incremental increases on AW levels on average, and in limited data that I've seen so far the same applies to research impact.

6.9 parameter was used to fit the most cities with chapter 15 data. As I already mentioned, constant increases per AW level is definitely not how the actual calculation works, but it was a good approximation for specific set of end-game cities. Difference between 6.1 and 6.9 over 100 AW levels of difference is only 80 SS points, which was well within the accuracy of the end-game model. This model does not work for cities other than the ones at the end of chapter 15.

All that makes it sounds like there should be a much bigger difference between mine and Mike's numbers, especially if he got the same 24-31 per tech those other players got. Even if he only got 18 per tech, like qaccy, that would be 126 more from the first comparison to today for the seven techs, plus about 20 or so for the AW levels, so 145-150. But the difference was only an extra 117.

Well, the only thing changing in my numbers this week is completing the Senate research and officially being in chapter 16. So we will see if I get the same 18 increase that qaccy got for doing the same thing or something different.

Also, what were the normal squad sizes for the players who got those 24-31 increases? Didn't you say once before that maybe the other variables increase in smaller amounts for the players with higher squad size? Maybe Mike did get 14 per because of his squad size, qaccy got 18 because his is smaller and the ones who got 24-31 had even smaller squad sizes?

 

[Deleted User - 3932582]

Also, what were the normal squad sizes for the players who got those 24-31 increases? Didn't you say once before that maybe the other variables increase in smaller amounts for the players with higher squad size? Maybe Mike did get 14 per because of his squad size, qaccy got 18 because his is smaller and the ones who got 24-31 had even smaller squad sizes?

No, regular squad sizes are unlikely to be the driver of differences here. I tried that, there is no obvious correlation there. The city with high increments has almost the same regular SS as @qaccy. I have practically the smallest possible regular SS, and my chapter 16 research increments seem to be about 13 so far.

I don't have much data on research yet, certainly not uncontaminated with other changes. Limited data from the last week looked like it lines up very well, but the data from this week does not fit nearly as well so far. Perhaps there is some subtle impact to moving into a new chapter after all. But I'll need to see more data to see if any pattern emerges. Also need to see how it works in the lower chapters...