BGForums

BattleGoat Studios
It is currently Dec 11 2017

All times are UTC-05:00




Post new topic  Reply to topic  [ 7 posts ] 
Author Message
PostPosted: Nov 14 2017 
Offline
Sergeant

Joined: Nov 14 2017
Posts: 10
Human: Yes
Good afternoon, ladies and gents! I'd like to foremost preface this post by extending my commendations towards Battlegoat Studios for creating one of the most intriguing grand strategy titles I've experienced to date! With over 1500 hours of gameplay; I've thoroughly enjoyed the title and it's scenarios. Hailing from the proud hometown of Waterloo; and as a fellow Canadian, gentlemen, I applaud your efforts both past and present to provide both great service, and a great product!

The insight I'd like to offer today is based on empirical game play observations over the 1,500 hours I've spent playing Supreme Ruler Ultimate across various scenarios as multiple nations, with many variations of settings.

Thread introductions aside!

Today, I'd like to introduce a topic of analysis regarding the impact of military efficiency and DEFCON war elevation, and how this interaction impacts the outcome and flow of warfare in Supreme Ruler Ultimate. In section one; I'll be providing my interpretation of DEFCON military efficiency impacts and mechanics in war. The analysis will be performed in a theoretical fashion. The goal of the following theoretical situation is to take a closer look at the interaction between two land units in war. The analysis will shed insight on what I believe is a positive feedback loop that exists as a core underlying factor leading to undesirable conditions and outcomes when two AI players "go to war".

Section One : DEFCON Military Efficiency Bonus: Stronger units becomes stronger; The Postitive Feedback Loop

In a theoretical situation of war in Supreme Ruler Ultimate where both nation have a relative base efficiency parity and DEFCON efficiency bonus of 20% each - this can be interpreted as moot or irrelevant at first, as both nations are "20% more effective". This interpretation however couldn't be any closer to incorrect.

In a given theoretical scenario, Unit A, of a technologically superior nation whose average stat value of 30; gains the multiplier of 1.2. Unit B, of a technologically inferior nation, whose average stat value is 20, gains the same multiplier of 1.2. The initial difference in overall stat value is 10, as the average values of Unit A are 30 and Unit B are 20 respectively. When the multiplier is applied, Unit A; 30 * 1.2 = 36; Unit B; 20 * 1.2 = 24. After the multiplier is considered, the statistical difference grows in disparity to 12 stat points; in theory.

Why is this a problem?

It's true a technologically superior unit should be stronger. However, under the pretext of war, this gap widens even further than the initial statistics suggest, as after efficiencies are applies, there is a stat difference of 12 (36 to 24) versus the initial stat difference of 10 (30 to 20).

Unit B in the above scenario, in a one on one confrontation scenario, is meeting the damage threshold to "route/retreat or turn away for repair" before Unit A, as it would in any scenario, war or not, however, at a greater rate under DEFCON elevation. Unit B is also inflicting less damage due to the greater disparity of stats when efficiency is applied. When this occurs in gameplay, the probability that unit B will successfully route or retreat is greatly diminished. When the first stack falls, and the AI player who is technologically inferior suffers it's first down-tick of MAR, is when the positive feedback loop begins it's accelerated pace towards "military collapse".

Section Two : Singularity Point MAR; "Military Collapse"

The small change in relative stats has now caused major repercussions in game-play. The inferior nation (Unit B) will typically suffer military losses until a threshold is reached, this threshold being defined by the MAR percentage in which said front line unit has zero probability to successfully route or retreat. At this point, the full effects of the feedback loop are realized. This is a point, which quite literally is a singularity point, henceforth dubbed "Singularity Point MAR".

When this mathematical point is met is a matter of time, it's how war works in Supreme Ruler 2020. The speed at which this happens however is extremely rapid at any setting. The pace at which the deterioration of a military force occurs is a function of the relative base stats of the units and the relative MAR of contending forces. As such, this can be easily controlled and mitigated by adding a bottom cap on MAR. Whether this value should be 30% or 50% or otherwise would require testing, but based on empirical game play experience, it should be strongly considered as a means of controlling this described feedback loop which undermines AI versus AI warfare in Supreme Ruler Ultimate.

Section Three : Do Land Unit Turn Times decide the Fate of a Nation?

In the described scenario, early, Unit B was described as "Turning away to route for repair." In this scenario; when a unit does get destroyed, it's typically while turning away, or while moving away from combat. A major part of the feedback loop as described above is indeed tied to global physics. The importance of unit turn times is a physical game-play mechanic that must be given some attention and consideration in regards to its overall impact on warfare in Supreme Ruler Ultimate. Turn times have to be reduced given the current behavior of "loss tolerance"; and units must be given the ability to successfully route or retreat for AI players to wage longer wars. Considerations to reduce repair times significantly would also compliment the decreased rate of front-line attrition.

Mechanically speaking, mathematically speaking, Unit B in our scenario will always spend more time than Unit A in a state of turning away, or routing/pathing away towards repair. In the loop, there is no possible way Unit B of Nation B will ever be able to win a war, as the singularity point MAR undermines the potential quantity advantage possessed by one side. If Nation B sends more Unit B, Unit B will suffer a magnification of losses, further speeding up the positive feedback loop of MAR loss in the current model. In this fashion, numbers actually work against the inferior nation as they're a liability to MAR. Each unit Nation B possesses is quite literally a magnification of it's susceptibility to the loop, creating an environment where the greater the force of Nation B, the less effective it is.

If these systems were to be reconciled and balanced; it would create an environment where debt and economy also become a factor in warfare, as nations would be able to successfully retreat units into repair throughout the course of a campaign/war. Nation B, spending more time routing and repairing Unit B, may be able to win the war economically under the correct circumstances if not militarily by force if the proper economic conditions exist.

Section Four : If not ME percentage, what works for DEFCON?

In the place of military efficiency, I'd recommend a significant supply/infrastructure bonus during war. This would represent the concentration of stance expenditure into "total war production and military logistics", the later of which appears to be very under represented in the current scope of warfare in Supreme Ruler Ultimate. (No modern force; realistically would get stranded anywhere... they'd of course have planned logistical supply lines, represented by infrastructure in-game) This would also serve to mitigate pathing and supply issues that remain prevalent in desolate regions (Northern Brazil, Siberia, parts of Africa) through providing a relevant positive modifier through infrastructure for the AI player. Additionally, this option may be a "lower resource" option as it will not require the revision or creation of new code to correct fundamental deficiencies in AI interaction with the supply model. Just buff the @#$# out of supply during war, gents.

Section Five : The Disclaimer

All of the above insight is theoretical in nature; and open to debate.I seek only to offer insight that I hope proves to be valuable in the future success of your product and title! Supreme Ruler Ultimate is THE best grand strategy out there, and it's future success and improvement isn't only important for my well being (hahah!) however, would sincerely make me proud, gentlemen!

Thanks for your time!
LordTyrantFTW

Edit 11/15/17 :: After making observations in a game while using limited military approval settings, this phenomenon continues to persist despite MAR fluctuations playing a minor factor. Inconsistencies in unit turn times and behavior exist. Sometimes a unit will drive "backwards", sometimes a unit will decide to turn away and path. Sometime a unit will engage an enemy "sideways". Although the "point in which a unit has zero probability to route" is never reached with limited MAR effects, the behavior of land unit turn times themselves still creates a model in which Unit B in our scenario is at a much greater disadvantage than intended.


Top
   
PostPosted: Nov 15 2017 
Offline
Lieutenant

Joined: Aug 10 2017
Posts: 96
Human: Yes
Thank you for this in-depth explanation.

I have to add a few points to your Section two:

The rapidly lowering MAR - wich i usually leads to mass deserts of units to the enemy - sometimes seemingly just by contact without firing a shot -
can also triggered by a very high local supremacy to a technological superior but totally outgunned force.
If that the hightech-region units would come back to help on that new front, the MAR would be already at the bottom, and those elite veteran units would just put their hands in the air and change sides. (just disbanded would bad enough or help in this case - but they are added to the enemy army!)

This happened to me vs an AI - did not matter that i had almost everything 10times more and better - the low MAR and mass desertation of units killed me.
Unfortunately the MAR hit does not account for the scope of the losses - if an empire from the german-french border up to bejing, is killed by loosing a few cities near that french border and a few dozen fresh produced green units, then something is not "grand" in this strategy.
Maybe a very easy calculation could help with it: reduce MAR only by the percentage of the units destroyed compared to units total - or even better - use the casualties compared to total army staff including reserves. Numbers that are already tracked for every region.

Since then i use only the "limit approval effects" options and watch my back. I wish it was rather a slider that could be set to any value..like 0.

_________________
my mods
viewtopic.php?f=79&t=25932 (even techs and units for everyone - AI will own you too)
viewtopic.php?f=87&t=26151 (MARSX1)


Top
   
PostPosted: Nov 15 2017 
Offline
Sergeant

Joined: Nov 14 2017
Posts: 10
Human: Yes
Thank you for the further insight, my friend! I too, for the past several days, have now been playing with limited military approval effects to observe differences in warfare without MAR factoring into the equation.

While reducing MAR effects does help to reduce instances of "military collapse", land unit turn times and inconsistencies in their turning actions, combined with AI loss tolerance behavior, coupled with the military efficiency bonus that makes "stronger high value units stronger relative to weaker low value units" still plays a large factor in AI players losing entire armies in a short period of time. It doesn't help that AI behavior is "send all units into the meat grinder until none remain". Resolution? Remove the meat grinder.

The 7 unit per hex limit should be a HARD limit. We can't have stacks of 100 units insta-gibbing singular AI units moving into defensive positions. This results in a serious steam roller mechanic, where one side that establishes dominance begins inflicting irrecoverable losses to the side that must route first by amassing unmatched firepower on one hex. The AI doesn't create a secondary defensive position when one is lost. They simply continue pathing units to the lost zone.

Changing AI behavior is a serious change, one i don't propose. I think this behavior can work if the unit hex limit was enforced. The hex system in combat is underused currently; and as described, combat is literally 100s of units facing off in one hex versus another.

The best fix in my opinion, again, is not to change AI, but to enforce the hex unit limit. Make that a HARD limit of 7 and let the hex system provide interesting strategic lines of battle.


Top
   
PostPosted: Nov 15 2017 
Offline
Sergeant

Joined: Nov 14 2017
Posts: 10
Human: Yes
Double post apologies aside; i'd just like to briefly post a summary of the above propositions.

1. ME bonus in war is a mechanic that is doubly punishing the technologically inferior nation. Inferiority is already built into the tech and unit design system, and a modifier to the already existing disparity is unnecessary and contributes to faster unit losses in combat, contributing to "short warfare" in SRU. (Evident with MAR limited setting, impacts those playing with limited MAR effects)

2. MAR as a mechanic itself creates a positive feedback loop when combined with the above point that results in extremely rapid military campaigns in SRU. MAR must be given a cap to which it can fall if it's to be viable as a system mathematically. This value should be 50% or more based on game-play experience. (For those that choose to play with MAR approval effects at full)

3. Global unit turn times need a positive multiplier. Give them 150% turn time. The individual unit turning/pathing behavior appears to be inconsistent, however, when units DO turn and path in an engagement, it results in a massive disadvantage in said engagement, and this disadvantage should be mitigated as much as possible to prolong engagements and reduce questionable unit losses.

4. Enforcing the seven unit hex limit would reduce the tendency of AI behavior of stacking their military in "One hex versus one hex" conflicts, where the nation that can stack the most firepower wins. This behavior contributes massively to unit losses in game, as singular AI units continue to path into the conflict, and are instantly destroyed or desert due to the strength of force on one hex.


Top
   
PostPosted: Nov 16 2017 
Offline
Colonel

Joined: Jun 27 2015
Posts: 466
Human: Yes
LordTyrantFTW wrote:
4. Enforcing the seven unit hex limit would reduce the tendency of AI behavior of stacking their military in "One hex versus one hex" conflicts, where the nation that can stack the most firepower wins. This behavior contributes massively to unit losses in game, as singular AI units continue to path into the conflict, and are instantly destroyed or desert due to the strength of force on one hex.

I dont really agree with your other obeservations, or better said only play with limited MAR, but this one reminds me of SR2020. It is not exactly 7. Like 7 stay in 1 hex and when another stack of 7 cross by, only 3 at a time cross. A hard limit of 10 so to say. I think this is very player friendly. Might be more interesting and help AI. Not sure.

I only faced the AI like that a cpl of times. Where France put 600 units in 1 city against Netherlands. I was overrun because it was city vs city, loaded my save game before war, and as soon you face an overwhelming force, make sure your fighting in a city and not them, then they will try other options, if your fighting city vs city you just have to retreat. That's why i think AI entrenching would be very interesting when it facing a bigger army.


Top
   
PostPosted: Nov 16 2017 
Offline
Sergeant

Joined: Nov 14 2017
Posts: 10
Human: Yes
Only witnessed a stacked hex a couple of times? Lucky guy! Thank you for the further insight, my friend!

The AI's behavior in war is to path everything to one hex. Observe a major conflict on any setting. There's no 10 unit limit on a hex, unfortunately. The human or AI player can literally place an unlimited number of units on any hex. The limit seems to fail or become moot when the unit has a target to shoot at, or if you entrench the unit. In this fashion, 10s ranging to 100s of units can and often do occupy a single hex.

I believe 7 units per hex is still too much, but it's a start, and is of utmost importance for balancing warfare in SRU. (Basically, you're still undermining inferior nations by a multiplier of 7 times the stat advantage, relative to the base difference in stats when you have a 7 unit limit. This might still be far too much.)

(edit; grammar, note to self, don't post via smartphone, hah!)


Top
   
PostPosted: Nov 18 2017 
Offline
Colonel

Joined: Jun 27 2015
Posts: 466
Human: Yes
It's very easy to lure the AI to somewhere else. Ai will send troops to hexes which get captured. Not a single hex. If you are passive and only fight in a single hex, yeah then they will send their entire army there.

It would be nice if we can mod the hard limit, but i dont think we can.


Top
   
Display posts from previous:  Sort by  
Post new topic  Reply to topic  [ 7 posts ] 

All times are UTC-05:00


Who is online

Users browsing this forum: No registered users and 2 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Limited