FreeOrion

Forums for the FreeOrion project
It is currently Tue Dec 12, 2017 12:13 pm

All times are UTC


Forum rules


Always mention the exact version of FreeOrion you are testing.

When reporting an issue regarding the AI, if possible provide the relevant AI log file and a save game file that demonstrates the issue.



Post new topic Reply to topic  [ 10 posts ] 
Author Message
PostPosted: Wed Jun 14, 2017 10:27 pm 
Offline
Space Squid

Joined: Thu Dec 11, 2014 3:22 am
Posts: 53
Hi,

No expectation of seeing any of this. But...

* Random target selection in combat should be replaced with an algorithm. RNGs waste time. They produce strange results when the situation being modeled isn't truly random or based on properly assessed probabilities. The game design already prefers to avoid randomness when possible and reasonable. Better to sort ships and apply the algorithm. I have ideas about possible algorithms, but that kind of conversation can be deferred as long as people prefer randomness.

* Fighters
... Should be able to attack planets. And planets should attack fighters if there's nothing better to hit. If fighters cannot attack planets, it becomes necessary to invest in direct fire weaponry, even though it is unnecessary to invest in fighters.
... Should do the same thing on every round of combat. Having fighters not be available for the first round might make sense if each round of combat represents an hour. (Of course, even then it might not: Don't the flux capacitors for the main guns need to be charged? What about waking the thousands of people needed to man the Main Gun? Checking all the fuses that have probably blown out since the last systems check? This clearly takes longer than launching a few unsophisticated light attack craft...) Of course, 3 combat rounds represent an entire turn's worth of time, which is maybe a month? a year? a decade? On that kind of time scale, each turn is not a sortie as much as a major battle, with Turn 1 being Pearl Harbor, Turn 2 being Coral Sea and Turn 3 being Midway. Aesthetic and chrome handwaving aside, it's simply better design. Once balanced, it lets you go from 3 rounds of combat per turn to 2 rounds or 5 rounds and remain balanced. It even lets you include technologies and ship status settings that can change the duration of combat and remain balanced.
... Should be balanced even for non-stealth designs, since it is generally agreed that stealth is kind of broken atm.

* Tech tree
There's continual discussion about balancing the tech tree, but I think it needs an overhaul, not just balancing costs.
... should reign in research snowballing. It's fine if production snowballs based on research, because there is a direct contest among players of production vs production which continues to work no matter how high production output reaches. It is not fine if research snowballs, because the tech tree is quickly exhausted, for the most part, when the inflection point is reached. Dramatically increasing the costs of mid- and late-game tech might fix things, but that comes at the cost of flattening possible tech strategies, since research tech either becomes more important than ever or investment in research becomes pointless. Better to curtail research bonuses. Increasing the depth of the tech tree might also help, and deserves consideration in and of itself, but it will not help much in the face of superlinear research increases.

...should have paths that effectively exclude each other. Right now, there is a Best Ship and Best Weapon, etc, which homogenizes things. (Your first non-basic ship is already certainly Robotic Hull, the ship you hope to produce in megadroves is probably a Solar Hull with Death Rays. I think it would be better to have Organic, Robotic, Energy and Asteroid tech trees, each with particular benefits that stack with each other but not with benefits from other trees. (Or, bonuses never stack and only the best, from deeper in a tree, every apply.) These benefits would include various hulls and weapon systems, but also research, supply and production bonuses, and also various buildings, with a design intention for all to be balanced, so that the most advanced ships of each tree might differ from each other, but be just as good, perhaps in a rock/scissors/paper kind of way. It is probably desirable to offer interesting choices even within a subtree. It is also reasonable to have common techs not part of any tree.

Anyway,

Ken


Top
 Profile  
 
PostPosted: Mon Jun 26, 2017 2:48 am 
Offline
Space Kraken

Joined: Mon Nov 16, 2015 5:57 am
Posts: 105
Yes id re-balance the entire thing in many many ways.

The first problem is research bonus's based on population along one line that gives everything else.
Solar hulls are the most ridiculous ship and one of the ugliest in game and its the best a big sun ship shouldn't easily do 140 speed.

You should have to research specific things to unlock specific techs.
For instance how do you make the leap to death rays well you research other weapons first.
In reality you don't get gun powder from continuously improving bowstrings in bow and arrow research.
You start researching chemistry and stuff.

Same idea should apply you should need a high tech like theory of everything to get deathrays.
For Armour you should need to do construction research. Alternatively better building materials should make for better colonys. ect.

Its not well balanced at all really and it doesn't even make much sense.


Top
 Profile  
 
PostPosted: Mon Jun 26, 2017 12:03 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4288
Location: Sol III
Please note that the tech tree will undergo a major revision, which actually means being basically completely redone from scratch at some point. However, that isn't likely to happen anytime soon, so you'll have to have a bit patience here... ;)


Top
 Profile  
 
PostPosted: Mon Jun 26, 2017 1:05 pm 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3289
Vezzra wrote:
Please note that the tech tree will undergo a major revision, which actually means being basically completely redone from scratch at some point. However, that isn't likely to happen anytime soon, so you'll have to have a bit patience here... ;)

Yup, we agreed in a different thread to do a series of tweaks to get the serious problems sorted, research snowballing being my current priority work, I'm starting by tweaking several of the research techs in both effect and order in the tree, toning them down for the most part but having a better spread of boosts throughout the game.

There are some other things I agree with in the OP, but for Fighters we strongly agreed to have them not fight planets nor have planets fight them, this is partially due to the way the planetary defences currently work, I'd love to rework them from scratch at some point and have planets launch fighter squadrons, etc and have multi-shot defence weapons, but until that happens the current split is the best approach. Also it gives some very interesting strategic choices, especially for stealth, that I find fun to play with.

_________________
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.


Top
 Profile  
 
PostPosted: Mon Jul 31, 2017 9:48 pm 
Offline
Space Krill

Joined: Mon Jul 31, 2017 9:26 pm
Posts: 1
Change the planet habitability chart. Going counterclockwise from Terran could be first Tundra, second Desert, and Third Barren. This would better reflect the difference in water available for life from abundant on Terran, to frozen on Tundra, to minimal on Desert, and absent on Barren.

4.7 on Mac Sierra


Top
 Profile  
 
PostPosted: Thu Sep 14, 2017 5:11 pm 
Offline
Space Krill

Joined: Mon Sep 11, 2017 10:25 pm
Posts: 6
Quote:
Random target selection in combat should be replaced with an algorithm.

How about an option at the start of the game? Random target selction false/true

Quote:
* Fighters
... Should be able to attack planets.

I respectfully disagree!

Quote:
I think it would be better to have Organic, Robotic, Energy and Asteroid tech trees, each with particular benefits that stack with each other but not with benefits from other trees.

I agree. Maybe wihtout rock paper scissors since that might give a luck based advantage/disadvantage. And you cant re-research you ship line from scratch in mid game.


---My suggestions

o) Retreating
There should be the possibillity to retreat during combat. If the ship is set to passiv stance, it should do a check every round. If the fleeing vessels are faster, they have chance to end combat depending on the difference.

o) Anti Snowballing
Ressearch: For example, technologies get more expensive by 1% every X spent RP.
X is customizeable at the beginning of the game - basically how difficult ressearch is. X dobules every 1% change. So for example 1% increase after 100RP, 200RP, 400RP, 800RP (...)

Production: The more "present production value" you have, the more production you lose (due to maintanance). PPV would be increased by built ships and buildings. This would slow down the snowball effect.

o) Radar Range
The radar range starts out decent I think. The active radar is a great increase. It gets ridiculous with the first sensor technology. Ranges get out of hand and huge.
-) I think, you should not see enemy radar ranges. That they have radar - ok, but for sure not their range nor detection value.

o) Maintainance
That is something for a poll: Buildings have a maintainance toll which reduced effective population. For example:
Planet with 11,4 Population. It has a building (like shipyard) which reduces effective population by 0,5. So when calculating the planetary focus, the planet is considered to have 10,9 population.

In addition to that, I suggest new buildings, like Planetary Shield and Planetary Naval Batterys - which all reduce effective population but further harden a planetary frontline.
Difference to tech planetary defensive network: They are an active defence. They cost production points, the have maintainance costs and you build them only, where you need them.

yay or nay? What do you think?


Top
 Profile  
 
PostPosted: Thu Sep 14, 2017 10:27 pm 
Offline
Dyson Forest
User avatar

Joined: Mon Apr 10, 2017 4:25 pm
Posts: 241
Captain Tofu wrote:
How about an option at the start of the game? Random target selction false/true
That's a plausible option. However, it could take out some of the fun because of the total predictability of the outcome of a battle. Randomness always brings fun (ask anyone in a Casino). Therefore, I would go at most for an stochastic (semi-random) option. Something like flak cannons chance to aim at a fighter is increased over expected uniform distribution of shots, but I don't know how this actually translates into the game (maybe some statistician around...). And example may explain what I mean:
For 99 fighters and 1 capital ship, uniform distribution will give you 99% chance to hit a fighter and 1% chance to hit the capital ship. To increase by a 50% percent the chances to hit a fighter would mean... nonsense. You can maybe raise that 99% to a 99.5% and lower the 1% to 0.5%. So, what would be the general formula for any possible distribution of ships?
So I'm fine right now with flak cannons being almost useless against non-fighter fleets and spinal cannons being mostly useless except against only-capital-ship fleets. I just adapt, an get a lot of fighters on every one of my fleets :)

Agree with fighters not attacking planets. I picture it as if every planet, regardless of tech level, has plenty of antiaerial cannons that could reduce to ashes any fighter fleet that gets into atmospherical flight but that are useless against orbiting ships.

Captain Tofu wrote:
Maybe wihtout rock paper scissors since that might give a luck based advantage/disadvantage. And you cant re-research you ship line from scratch in mid game.
Rock/paper/scissors makes best strategy games. If there is no strategic cycle, if there is one best strategy that beats all the rest, the games reduces to get production/research advantage over your enemy, get strong and finish him over. It's true that there might be different ways to get that advantage (more PPs or more RPs?) and that brings replayability to the game. On the other hand, with rock/paper/scissor techs you have another dimension of strategy and the possibility to beat an otherwise unbeatable enemy. I'm thinking of Star Craft, for example. With no R/P/S cycle you will always go for photon turtle base and mass protoss carriers, boring as hell after 100 games. Now I've played more than 10000 games and I'm still excited when I play it (SC2 actually).

Captain Tofu wrote:
Retreating
There is some kind of retreating currently. After each three turns of combat (if there are survivors), you can relocate the fleet if not blockaded. You can configure the game so that there is only 1 or 2 turns of combat for each battle (although that would mean slower pace).

Captain Tofu wrote:
Anti Snowballing
Ressearch: For example, technologies get more expensive by 1% every X spent RP.
X is customizeable at the beginning of the game - basically how difficult ressearch is. X dobules every 1% change. So for example 1% increase after 100RP, 200RP, 400RP, 800RP (...)
I don't know. Numbers should be crunched but with that suggestion I suspect the order in which you take techs would become even more critical than currently. So that the player that starts researching the key techs to increase research would snowball even faster than currently when compared to the players that focused first on production techs.

Captain Tofu wrote:
Production: The more "present production value" you have, the more production you lose (due to maintanance). PPV would be increased by built ships and buildings. This would slow down the snowball effect.
That would be, in rough outlines, another form of the current upkeep system (the more ships you have, the more expensive new ones will be).

Captain Tofu wrote:
Radar Range
The radar range starts out decent I think. The active radar is a great increase. It gets ridiculous with the first sensor technology. Ranges get out of hand and huge.
That depends on the size of the galaxy you play. For 1000 system games, it's not that great. I guess actual range values can be changed as for your personal preference in the FOCS files at your machine.

Captain Tofu wrote:
-) I think, you should not see enemy radar ranges. That they have radar - ok, but for sure not their range nor detection value.
Why not?
Anyway, maybe only when your own radars detect theirs? Not sure if that is exactly what is done now.

Captain Tofu wrote:
Maintainance
That is something for a poll: Buildings have a maintainance toll which reduced effective population. For example:
Planet with 11,4 Population. It has a building (like shipyard) which reduces effective population by 0,5. So when calculating the planetary focus, the planet is considered to have 10,9 population.
Why? What for? If it is to dissuade players from spamming certain buildings, they are already disuaded from that because it requires boring micro-management and because it takes out PPs that you could be devoting to more ships to stomp your enemy (or defend from him).

Captain Tofu wrote:
In addition to that, I suggest new buildings, like Planetary Shield and Planetary Naval Batterys - which all reduce effective population but further harden a planetary frontline.
Difference to tech planetary defensive network: They are an active defence. They cost production points, the have maintainance costs and you build them only, where you need them.
You would need them in every planet, it's a no-brainer. The planet that does not get it becomes the easier target to conquer. Thus you end up building it all over your empire, requiring boring micro-management. That's why in FO you only have defensive techs that once investigated are automatically applied to every planet (although they only get full strength after some time of build-up).


Top
 Profile  
 
PostPosted: Thu Sep 14, 2017 11:09 pm 
Offline
AI Contributor

Joined: Tue Feb 17, 2015 11:54 am
Posts: 224
Oberlus wrote:
So, what would be the general formula for any possible distribution of ships?


You would use standard weighting: For N fighters and M capital ships and your 50% increased chance to hit a fighter, you would get
Code:
p = 1.5*N / (1.5*N + M)
.

_________________
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0


Top
 Profile  
 
PostPosted: Fri Sep 15, 2017 9:23 am 
Offline
Dyson Forest
User avatar

Joined: Mon Apr 10, 2017 4:25 pm
Posts: 241
Morlic wrote:
You would use standard weighting: For N fighters and M capital ships and your 50% increased chance to hit a fighter, you would get
Code:
p = 1.5*N / (1.5*N + M)
Ah, thank you, Morlic. So the general formula would be:

Pij = Cij*Nj / (Sum_j Cij*Nj)

Where Pij is the probability of weapon type i shooting at enemy ship type j, Cij is the "chance factor" (the 1.5, 0.66, etc.) of weapon type i aiming at ship type j, and Nj is the number of ships of type j in the enemy fleet.

Types of ships could be more or less disaggregated, but I would go at first for three types:
- 1) fighters, 2) capital ships and 3) rest of ships

Same goes for types of weapons (more or less categories). I would go for:
- 1) Flaks, 2) MD and laser, 3) plasma and DR.
Also one type for each tier of weapon could be reasonable, and also just two types, flaks vs rest of weapons.


Top
 Profile  
 
PostPosted: Sun Oct 01, 2017 12:45 pm 
Offline
Space Krill

Joined: Sun Oct 01, 2017 11:45 am
Posts: 1
I disagree with the fighters attacking planets, they are fine the way they are. The fighters already have their own advantages, they are practically saving armor on your ships and you get back any lost fighters after each turn if within supply lines and you do not have to upgrade ships when better fighters being researched, and to counteract the fighters your opponents have to install flak cannons weakening the ships


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 10 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


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 Group