Eu2Vic: Design Discussion

[Sute]{h]

better.

think of prussia and holland. aristocracy vs. plutocracy, both advanced and should have similar literacy levels.

I can live with it.

As to Prussia and Holland I would say that they both have a high trade and infra plus a high innovative slider. Prussia however doesn't have a high freedom compared to Holland. So according to my latest version, they will properly get a lower literacy than Holland anyways.

 

[Brownbeard]

now only the religion question remains. personally i dont find it as relevant, but if others insist it might be a good idea to solve it like the aristocracy/plutocracy issue.

effect of religion should be proportional to narrowminded. at innovative = 10 religion makes no difference.

 

[Sute]{h]

Well the beauty of the current solution is that a modifer for religion isn't really needed. Your literacy depends on your trade and infra level, and since tech is much more expensive in EU2, if you are not in latin tech group, religion has already been accounted for.

 

[montyP]

Ok, I ran a test with the last version of Sute]{h's literacy convertion algorithm:

((1,5 x (trade/(1+(aristocracy x serfdom/100)))) + (2,5 x (infra/(1+(serfdom/10))))) x (1+(innovative/10))

The values for all european nations generally seem a bit to low (there are a few exceptions though). Here is the list. I used the 1795 scenario when testing this.

Country		lit	conv.lit
-----------------------------------
ENG		0.55	0.47
FRA		0.32	0.65
AUS		0.25	0.36
HOL		0.28	0.54
POR		0.13	0.23
PRU		0.80	0.40
RUS		0.25	0.20
SPA		0.13	0.23
SWE		0.80	0.39
TUR		0.04	0.22
VEN		0.30	0.25
POL		0.25	0.19
ADE		0.30	0.17
ALD		0.40	0.17
HEJ		0.01	0.21
BAD		0.80	0.20
BAY		0.80	0.21
ION		0.30	0.18
DAN		0.80	0.33
SWI		0.70	0.38
HED		0.80	0.20
HYD		0.02	0.17
MOR		0.03	0.19
MOL		0.10	0.19
MYS		0.02	0.14
OMA		0.03	0.20
PAP		0.35	0.22
MOD		0.45	0.22
PER		0.03	0.17
SAX		0.80	0.20
SAR		0.45	0.27
COB		0.80	0.20
TUS		0.45	0.22
TRI		0.03	0.17
TUN		0.03	0.17
WAL		0.10	0.19
WUR		0.80	0.20
CHI		0.04	0.15
NIP		0.40	0.12
MUG		0.04	0.14
USA		0.70	0.58
BER		0.01	0.17
SIC		0.35	0.22
GEO		0.10	0.16
LIV		0.15	0.20
KHI		0.01	0.01
KOK		0.01	0.01
AFG		0.20	0.11
CHE		0.01	0.12
HAI		0.08	0.27
ETH		0.02	0
SOK		0	0.05
ZAN		0	0.12
TIB		0.10	0.15
PAN		0.01	0.14
SIA		0.02	0.16
DAI		0.02	0.16
ANN		0.02	0.16
VIE		0.02	0.14
LUA		0.02	0.16
BUR		0.02	0.16
CAM		0.02	0.16
ATJ		0.03	0.21
ZUL		0.01	0.13
GWA		0.01	0.21
KUT		0.01	0.21
JOD		0.02	0.21
MAL		0.02	0.25
BAL		0.03	0.25
BRU		0.03	0.25
EGY		0.02	0.06

 

[Brownbeard]

perhaps it sounds silly, but maybe the eu2 1795 setup does not represent nations as well as it should in terms of slider positions. i wonder what fully played GC conversion would yield.

what i have noticed is that most major european nations have somewhat higher literacy or equal, small german and italian states, along with prussia are way off. my guess is that their sliders are not historically well represented. and i do notice that all pagans are better off than in victoria GC.

it might sound very stupid, but it's not the formula, but the input. :wacko:

sticking to closely to getting results created by converting 1795 scenario to victoria GC setup might make a much greater gap in converting a GC scenario... it's a very difficult question. aiming to get prussias literacy to 80... what if there's no prussia at all after conversion? what if poland annexed them?

could you post the input in columns next to it so it would be easier to trace patterns?

 

[Ilkhold]

What I miss from the formula is the average tax income. Slider positions only tell you about a moment, not about centuries. If a region is very rich I expect it to have a higher literacy than a poor region.
Of course you have to be careful with province sizes. In EU2 european provinces are generally much smaller than others.

 

[montyP]

perhaps it sounds silly, but maybe the eu2 1795 setup does not represent nations as well as it should in terms of slider positions. i wonder what fully played GC conversion would yield.

what i have noticed is that most major european nations have somewhat higher literacy or equal, small german and italian states, along with prussia are way off. my guess is that their sliders are not historically well represented. and i do notice that all pagans are better off than in victoria GC.

it might sound very stupid, but it's not the formula, but the input. :wacko:

sticking to closely to getting results created by converting 1795 scenario to victoria GC setup might make a much greater gap in converting a GC scenario... it's a very difficult question. aiming to get prussias literacy to 80... what if there's no prussia at all after conversion? what if poland annexed them?

I agree that the 1795 scenario is faulty in many, many, many ways. But come on, if nations like ATJ, GWA, KUT, JOD, MAL, BAL and BRU have higher or almost equal literacy to that of countries like BAD, BAY, HED, PAP, MOD, SAX, COB, TUS, WUR or even RUS and POL. Which all have higher tech levels over all, then something has to be wrong!!! Even with this faulty scenario file. Don’t you agree?

could you post the input in columns next to it so it would be easier to trace patterns?

The info is in the 1795 scenario file.

 

[Brownbeard]

perhaps some sort of curve based on something will have to be implemented as well.

do you have some GC games that have been played fully to compare results?

 

[montyP]

perhaps some sort of curve based on something will have to be implemented as well.

do you have some GC games that have been played fully to compare results?

Yea, we need something more when calculating the literacy value.

I have some fully played games, but I don't have time to test this at the moment. Maybe you could have a swing?

 

[Sute]{h]

My guess at the sinner is the innovative slider... since a high innovative is needed if you want a good literacy.

Also I had a feeling that problems would arise in germany and scandinavia. Perhaps reducing the importance of slider positions and increasing trade and infra importance will solve the problem... Also increasing the potential literacy to 100% might solve some of it...

Btw. How do you run a fast test on this? I could try it on my Inca game, which has been played from 1419 to the very end.

 

[Sute]{h]

((2 x ((trade*trade/10)/(1+(aristocracy x serfdom/100)))) + (3 x ((infra*infra/10)/(1+(serfdom/10))))) x (1+(innovative/10))

While it may still be early to discard my last literacy convertion algorithm, this one might be an alternative. It will increase the importance of a very high trade and infra, and will possibly but very unlikely allow 100% literacy. This might solve some of the problem. Note however that Prussia starts the 1795 scenario with trade and infra 8, and this algorithm will only make it worse, if they don't reach trade and infra 10 by the end of the game. Also this algorithm will severely reduce literacy outside the latin tech group, as trade or infra above 5 is very unlikely.

 

[Sute]{h]

Btw. the high literacy in France is properly because they start the 1795 scenario with trade and infra 9. Which perhaps mean, that the above posted algorithm will only make it worse.

 

[montyP]

Btw. How do you run a fast test on this? I could try it on my Inca game, which has been played from 1419 to the very end.

I put your algorithm in a test version of the converter and printed the "test_literacy" value next to the real value.

I can test your other version later on.

 

[Golden_Deliciou]

Ok, I ran a test with the last version of Sute]{h's literacy convertion algorithm:

((1,5 x (trade/(1+(aristocracy x serfdom/100)))) + (2,5 x (infra/(1+(serfdom/10))))) x (1+(innovative/10))

The values for all european nations generally seem a bit to low (there are a few exceptions though). Here is the list. I used the 1795 scenario when testing this.

Great! Let's take a look at this;

Since this game is converted from 15 years before the end of the EU2 GC, we should expect that the numbers would be a little lower than normal. So we'll look for numbers which are extremely low or somewhat high.

Central Europe (Switzerland, Germany, Scandinavia) seems to be all that's suffering from low literacy.

The vast majority of states which have 1-3% literacy in Victoria seem to have 12-25% literacy in EU2. Also France and Holland have quite high literacy.

Really, we need to get more data. What I would recommend is that the same scenario be fed into the converter using different version of the literacy algorithm, then we could select which version produces the best results, or else come up with another system.

Sute]{h's new formula would be the first one to try. It has more brackets than any of the others so it must be good

 

[unmerged(31960)]

perhaps some sort of curve based on something will have to be implemented as well.

do you have some GC games that have been played fully to compare results?

I think i have 10 or 15 completed games in the computer, so, just tell me what to do, and i'll test (Today if it's posible)

 

[unmerged(31960)]

Yesterday i finished a game as Byzantine Empire (BTW my first finished game in EU2 in first place ) almost a month before ending game and saving to converted, i released Hellas as vassal, as far as i know, Hellas and Byzantine Empire both become Greece in Vicky, right? so, when i runned convertor, it used Hellas as base to create Greece, the results, i got all my byzantine territory as Greece, But i lost all my army, navy, relations and factories, the reason? i think because the convertor used Hellas to create Greece first and then when it found Byzantine Empire, it only gave the territories to Greece but trew away relations, navy and the rest.
so What are your toughts about this?
I say that when the convertor merges countries, it should make a combination of relations and badboy.
Like for example BYZ relations with England = -130 Hellas relations with England = 0
New Greece relations with England = -65
For the army and navy, put both country's armies and navies, and same for factories.

 

[Brownbeard]

I think i have 10 or 15 completed games in the computer, so, just tell me what to do, and i'll test (Today if it's posible)

if you could compile a list like the one montyP made it would be useful to compare results, also, if you could add the relevant sliders and tech levels.

you don't have to bother with the complete list, too much work. focus just on a few representative countries. vic majors, a minor german, and random uncivilized or two and see what you get.

 

[Golden_Deliciou]

I say that when the convertor merges countries, it should make a combination of relations and badboy.
Like for example BYZ relations with England = -130 Hellas relations with England = 0
New Greece relations with England = -65

This would be OK if the figures were scaled to the size of the two states. So for example an 20 province Byzantium with +100 relations and a 2 province Hellas with -100 relations would lead to a Greece with +90 relations.

Otherwise this is something of an exploit- create a vassal and then absorb it to halve your BB points!

 

[Sute]{h]

Perhaps when merging countries the result country should get ½ BB-point per territory they've gained by the merging or something like that? This would discourage players from calculating on a merger... while still not giving a merged country an obsene amount of BB-points.

 

[montyP]

Perhaps when merging countries the result country should get ½ BB-point per territory they've gained by the merging or something like that? This would discourage players from calculating on a merger... while still not giving a merged country an obsene amount of BB-points.

That sounds like an exellent idea! I will take a look at how it can be done.