Morale ticking only at month's end is really really stupid.

[TehJumpingJawa]

- Win a battle on the first day of the month, and you have a good chance of being able to chase the retreating army to their destination before they have a chance to regain any morale, thus delivering a wipe.
- Win a battle on the last day of the month, and you have zero chance of chasing them down before they recover enough morale to resist a wipe.
- Beyond a certain army size (at any given tech level) it becomes impossible to inflict enough morale damage early in a battle to wipe out an army that has received 1 ticks worth of morale.
- It's almost impossible to predict the day of the month a battle will end, thus the successfulness of pursuits is demoted to random chance.

As army wipes from successful pursuits are often the crucial turning points of a war, I find their randomness compromises the integrity of the game's combat mechanics.

The solution is simple enough; tick morale every day.

 

[wei]

Agreed, should tick daily. Could set in additional parameter like no tick in first 7 days of retreat etc.

 

[Seelensturm]

Agree.

 

[downing]

I think this might just be to much work for the computer. Otherwise I'm sure Paradox would already have included this...

 

[Mann42]

While I totally agree with this, I'm going to assume that morale updates once a month for performance issues. If the game was calculating the morale of every single army every single day of the game, it would most likely slow down quite noticeably.

There are probably some ways they could optimize, or a compromise (calculate once every 10 days instead of a month start), but I wouldn't expect any day calculations in the near future, especially after they optimized the game in 1.2 to pretty much never calculate anything on a daily basis.

 

[unmerged(784568)]

I am pretty sure this is intentionally as it adds quite a lot operations to the game and therefor slows your cpu down. The same could be said about Your income or tradesteering. You have to draw the line somewhere.

Just picture the moraljump as payday.

 

[Vanillamarine]

Absolutely agreed.

 

[unmerged(804580)]

Probably too much work for the CPU. The game lags every end of the month on my laptop...

 

[TehJumpingJawa]

If calculating the morale increase itself is a complex operation, sure, do it monthly, but the morale really needs to be incremented every day.

//every day
foreach(Battalion b : battalions) {
b.morale+=b.dailyMoraleGain;
}

//every month
foreach(Battalion b : battalions) {
b.dailyMoraleGain=someComplexCalculation()/daysInCurrentMonth;
}

A few thousand extra floating point additions in the day update code isn't going to have the slightest impact upon performance.

 

[nOxr]

You could let the monthly morale tick vary slightly so that it doesn't always happen on the 1:st, but anytime betwen day 25 and day 5.

 

[rbl0010]

Probably too much work for the CPU. The game lags every end of the month on my laptop...

The devs confirmed (somewhere, please excuse lack of reference) indeed that computing things on a daily basis would be too much and likely to provoke lags on pretty much all but high end machines

 

[Xara]

You could make it so that when an army starts battle, immediately beforehand it gains the elapsed proportion of that month's potential morale gain

i.e. if an army is attacked 3 weeks into september, immediately before the battle begins both armies gain 75% of the morale they would gain on October 1st
would have the same effect as a daily update yet only be calculated when needed

-exchange "since the 1st of the month" for "since last battle", so if they fought 1 week into september then get attacked 3 weeks in, they gain 50% of the Oct 1st amount, not 75

 

[DarkThug]

Just picture the moraljump as payday.

OK I can live with this reasoning :rofl:

 

[GAGA Extrem]

I think daily morale is a bit over the top.
Weekly would certainly suffice.

 

[Vanillamarine]

The smallest measureable amount of time for EU4 is one day, with something as crucial as complete army wipes one actually has to be quite thorough IMO. Calculate the amount of morale on a monthly basis but distribute it with daily increments.

 

[knul]

A solution could be to give a victorious army an immediate morale raise equal to a month's increase. At least it is computationally much, much more efficient than a daily morale update of every army on the planet.

 

[Xara]

A solution could be to give a victorious army an immediate morale raise equal to a month's increase. At least it is computationally much, much more efficient than a daily morale update of every army on the planet.

Why would you want to buff the winning army? the issue is a routed army being caught on feb 28th vs march 1st

I hate winning battles late in the month

 

[Redron]

You could make it so that when an army starts battle, immediately beforehand it gains the elapsed proportion of that month's potential morale gain

i.e. if an army is attacked 3 weeks into september, immediately before the battle begins both armies gain 75% of the morale they would gain on October 1st
would have the same effect as a daily update yet only be calculated when needed

-exchange "since the 1st of the month" for "since last battle", so if they fought 1 week into september then get attacked 3 weeks in, they gain 50% of the Oct 1st amount, not 75

This sounds good to me.

 

[knul]

Why would you want to buff the winning army? the issue is a routed army being caught on feb 28th vs march 1st

I hate winning battles late in the month

Oops, I meant the losing army. Just shows that it's quite counterintuitive, but it would at least remove the problem of easy wipeout when you win a battle at the beginning of the month.

*EDIT* Never mind, Xara's suggestion is much better. See what happens when you don't read thoroughly?

 

[zodium]

If calculating the morale increase itself is a complex operation, sure, do it monthly, but the morale really needs to be incremented every day.



A few thousand extra floating point additions in the day update code isn't going to have the slightest impact upon performance.

It's my understanding that efficient use of CPU cycles is a priority for Paradox precisely because EU is already very computationally taxing on lower end computers. The AI in particular is a hog, so what might not make a big difference in a production app can be a critical difference for a game like EU.