Game balance in 0.4.4 release—feedback needed

Describe your experience with the latest version of FreeOrion to help us improve it.

Moderator: Oberlus

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 Reply
Message
Author
User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Game balance in 0.4.4 release—feedback needed

#61 Post by Kassiopeija »

Dilvish wrote:
godel wrote:Most advances are more powerful then the ones before, but laser is less than Mass driver 4.
It is true for all the weapon tiers that Weapon 4 of tier X will be stronger than Weapon 1 of tier X+1 (and I think generally weapon 2 of tier X+1 also). This is intended to make for more interesting decisions about when to possibly skip immediately to the next tier versus completing the current one (is my recollection about the reasoning). You can review the strength of all these weapons in the ship design window.
I might add that MassDriver modules are more cheap than Laser, so esp. early in the game when practically nobody has shields the MassDriver makes more sense or, is stronger than its successor.

godel
Space Dragon
Posts: 267
Joined: Tue Mar 24, 2015 12:58 am

Re: Game balance in 0.4.4 release—feedback needed

#62 Post by godel »

It almost never makes much sense to go for laser, except as a stepping stone to something better.
So, yes, I agree on the mass driver.

User avatar
Kassiopeija
Dyson Forest
Posts: 212
Joined: Wed Jul 16, 2014 6:14 pm
Location: Black Forest

Re: Game balance in 0.4.4 release—feedback needed

#63 Post by Kassiopeija »

actually all weapons are justified in some context, however it depends on your games setup if something is falling out. for example, in a huge game with over 50 systems per AI you will have enough time to build up sufficient research & production before getting in contact with the AI empires. in such a game, you might as well skip massdriver entirely and go directly to laser.

godel
Space Dragon
Posts: 267
Joined: Tue Mar 24, 2015 12:58 am

Re: Game balance in 0.4.4 release—feedback needed

#64 Post by godel »

Thanks for that perspective.
I have just started a 100 node game.
I found going for m4 first still won. I will pass through laser 1 on the way to the P canon.
I think the goal is many ways to play? Your comment made me realize that the ways are many.

marcOSX
Space Squid
Posts: 55
Joined: Sat May 31, 2014 5:20 pm

Re: Game balance in 0.4.4 release—feedback needed

#65 Post by marcOSX »

OK I had a a few games with FreeOrion_2015-05-04.d742fbf_MacOSX_10.7

Game is running great. I tried a few very small galaxies, and I confirm that asteroid hulls are still overpowered. I usually go research first, raising production and military through technology. In this case I usually skip totally mass drivers, go for laser 1 then up to death ray ->4. I also skip armor as rock armor is quite good and comes with asteroid hulls. Exception is if by any chance an enemy planet is up for grabbing in the first 30 turns. But usually I go for a ship able to deal with sentinels and reduce defenses of some Mu Ursh or Hhoh planet before building any fleet building.

I noticed there were no Victory message when defeating all enemies. I pushed the game towards transcendance and got no message neither... Is it as intended or did I miss something?

I have not seen any experimentors in those games. Playing on maniatical AI as usual.

In my experience there are really ineffective technologies, or some that come into play in a wrong moment.

.1. Fuel tanks. I don't see any interest into developping them as it is far more effective to raise your supply lines than to venture towards out of reach planets. It may reveal somewhat useful in a very poor environment with no planets in many jumps... And truth is that at the beginning internal slots of my warships are filled with a shield and some speed booster. No idea why anyone would even put a ramscoop in a ship... The only exception is if the experimentors live very far away and you have to send a full expedition to get them...

.2. Spatial distortion Generator/ Gateway to the Void ... Well I have yet to find a game where I am not leader in research and I should use that to deal with some enemy. Ususlly when I get this technology my scattered hull biggies are already ruling the galaxy. Did anyone actually used it? Same thing with Nova bomb... Please say if you really use it, at least against the AI.

.3. Moving planets stuff. It is definitely fun and sci fi. I love the idea, and did it for fun in a few games. But truth is the investment is heavy for little real use. If you need a shipyard at a black hole, it is much easier to colapse some red star. It can be fun in heavy big multiplayer tho maybe to remove planets from some lanes so that enemies have no supplies... Never used it recently.

.4. Mind control. Again it comes late into play, and at that time I almost never get my planets bombed by enemies, let alone organic species... I would love that to be effective. It comes too late and is too limitative to be the gameplay alternative it should be.

.5. Sealth. I tried a few games raising sealth early, anad have yet to find any real use (against AI, it could be fun against players). As an enemy annoyance it is a bit effective if you neglect to raise your detection. It does not protect you from space monsters. Can anyone comment how you really use it into play to get something you won't get easier with well armed warships?

Go on with the good games. Do you plan any translation to French?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Game balance in 0.4.4 release—feedback needed

#66 Post by Geoff the Medio »

marcOSX wrote:Do you plan any translation to French?
https://github.com/freeorion/freeorion/ ... les/fr.txt

godel
Space Dragon
Posts: 267
Joined: Tue Mar 24, 2015 12:58 am

Re: Game balance in 0.4.4 release—feedback needed

#67 Post by godel »

Krill Swarm:
I have not seen any experimentors in those games. Playing on maniatical AI as usual.

Replying:
4)I got mind controlled once. Took my victory fleet totally away from me. I was wary and researched it. I never took a fleet from the ai, and it seems that only worlds can do so, not fleets. Which limits its usefullnesss, it is defensive only. likely better so.
1) I put fuel tanks in the spare place at first. I have never researched a better fuel tank, mostly that slot is either a faster engine or more often a better shield. Fuel tanks are expensive even if only 10 research points.
I had a problem with some ships being faster because of a faster engine -- the got there first and my force was defeated in detail.
5) Stealth is generally too expensive for non strealthy races. It gives a huge point disavantage if you are against such: you have to catch up. Which is why the Etty have a very hard time against such.
2)I found myself in a game where I tried to use a nova bomb. It took out my homeworld. OOPs! back off and try again. Spacial distortion generator looks like it could kill my forces better then it could kill the enemy.
3)Never used that moving planets stuff.
0) the victory conditions have not generated a win congrats banner ever yet, the game is young. I asked too.
I have played on all conditions of AI. I have never found maniacal. Nor have I ever found the AI to be over aggressive. It is almost passive. I am using both the experimental Linux and the Win 7 versions. The win7 lacks much new, but does not foul up and not let me create troops. The linux version does not allow ships to be created if they use certain technologies. I doubt I could build asteroid anything in that game and have been winning using the robot track--without the robot shield. I might not be familiar with experimenters. There is a monster generator that tends to limit games to 250 moves. I have stopped playing with that.
There are *so* many different games that can be made with the game set.

User avatar
MatGB
Creative Contributor
Posts: 3310
Joined: Fri Jun 28, 2013 11:45 pm

Re: Game balance in 0.4.4 release—feedback needed

#68 Post by MatGB »

marcOSX wrote: Game is running great. I tried a few very small galaxies, and I confirm that asteroid hulls are still overpowered. I usually go research first, raising production and military through technology. In this case I usually skip totally mass drivers, go for laser 1 then up to death ray ->4. I also skip armor as rock armor is quite good and comes with asteroid hulls. Exception is if by any chance an enemy planet is up for grabbing in the first 30 turns. But usually I go for a ship able to deal with sentinels and reduce defenses of some Mu Ursh or Hhoh planet before building any fleet building.
One thing I'm loving reading these is the vast variety of different approaches people have to researching the weapons techs, and no one has mentioned using my approach, which proably means we've got the balance of choices about right for the current stage of the game and the different weapon techs available.

I've reduced my reliance on armour research recently, but I still find you need it for planetary assaults, if a system has 3+ planets and each planet can kill a ship in one shot, you're in trouble. I actually think most of the armour parts (especially rock armour) are too cheap...
I noticed there were no Victory message when defeating all enemies. I pushed the game towards transcendance and got no message neither... Is it as intended or did I miss something?
There is currently no message for a conquest victory, I tried to work on one awhileback but I got stalled, I think Geoff's done some more work with some counting/statistics things that I could shoehorn in. There is, and should be, a message if you research Transcendance, there definitely was last time I completed it, I'll check to see if it's working in a game.
I have not seen any experimentors in those games. Playing on maniatical AI as usual.
Experimentors, in order to spawn, need a location at least 5 jumps away from an imperial capital, so if you're playing a smallish map with lots of AIs and/or medium/high starlanes they're less able to start. There is a plan to rewrite the galaxy generation scripts to include them directly if they're selected for use but such coding is beyond my ability. If you want to see the, play on Low starlanes or have more systems per AI.
.1. Fuel tanks. I don't see any interest into developping them as it is far more effective to raise your supply lines than to venture towards out of reach planets. It may reveal somewhat useful in a very poor environment with no planets in many jumps... And truth is that at the beginning internal slots of my warships are filled with a shield and some speed booster. No idea why anyone would even put a ramscoop in a ship... The only exception is if the experimentors live very far away and you have to send a full expedition to get them...
I am now of the opinion that it's too easy to improve supply connections, especially in the late game. However, on a Low Planets/Low Starlanes setting they can be essential, even with the ease of increasing supply, some of the techs will be of less use in some settings (for example, Artificial Planet is probably too powerful/useful on a Low Planets galaxy, but a waste of time/resources in a High Planet galaxy, I'm fine with that).
.2. Spatial distortion Generator/ Gateway to the Void ... Well I have yet to find a game where I am not leader in research and I should use that to deal with some enemy. Ususlly when I get this technology my scattered hull biggies are already ruling the galaxy. Did anyone actually used it? Same thing with Nova bomb... Please say if you really use it, at least against the AI.
Agree with this, I do want to do some work with them at some point as they're definitely fun ideas but they just, well, don't mean much to me. I have seen reports of others using the Gateway, but they're mostly by players playing strategic defence, I tend to want to conquer everywhere.

I used a Nova Bomb once. To see what it did. I decided it wasn't worth it and I wanted to capture planets not destroy them. Then again, there's currently too much strategic emphasis on conquering people and not as much on destroying people, even for Xenophobic species, maybe at some point it'll be more useful.
.3. Moving planets stuff. It is definitely fun and sci fi. I love the idea, and did it for fun in a few games. But truth is the investment is heavy for little real use. If you need a shipyard at a black hole, it is much easier to colapse some red star. It can be fun in heavy big multiplayer tho maybe to remove planets from some lanes so that enemies have no supplies... Never used it recently.
Mostly agree here. I have in the past moved an entire galaxy worth of planets into one system (really bad idea butfun to do), I've moved adjacent Mu Ursh/Acirema worlds on top of a Blue star or a Black Hole a few times, and did one experiment with a mobile shipyard, but it was too silly. But it's definitely fun.
.4. Mind control. Again it comes late into play, and at that time I almost never get my planets bombed by enemies, let alone organic species... I would love that to be effective. It comes too late and is too limitative to be the gameplay alternative it should be.
I'm of the opinion the current implementation is a Proof Of Concept feature that we need to expand on, in a lot of different ways, at some point, but it's not a priority for immediate work for me. If someone wants to play around with the ideas, especially implementing the ideas hinted at in the Translinguistics tech, I'd be very happy to playtest it.
.5. Sealth. I tried a few games raising sealth early, anad have yet to find any real use (against AI, it could be fun against players). As an enemy annoyance it is a bit effective if you neglect to raise your detection. It does not protect you from space monsters. Can anyone comment how you really use it into play to get something you won't get easier with well armed warships?
At the moment, stealth as a strategy isn't up to much, it is something I've been working on and want to do more on, but we're discussing changing the basic stealth mechanics, in part in order to make it more viable as a strategy, so I sidelined that work. It definitely needs more done.
Go on with the good games. Do you plan any translation to French?
Options/UI

Change the stringtable file from the default en.txt into fr.txt, it's not complete but Ouaz has been doing sterling work trying to both complete the translation and translate all the new features we keep throwing at him, I think it's probably the most complete translation but given I only speak English myself I don't know.

Thanks for the feedback, very much welcome.
Mat Bowles

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

marcOSX
Space Squid
Posts: 55
Joined: Sat May 31, 2014 5:20 pm

Re: Game balance in 0.4.4 release—feedback needed

#69 Post by marcOSX »

Hello, tanks a lot for your answers.

I just started a huge game with 2,500 systems, low everything, maniatical aggressive 16 AI. Very good news is that even on my old Core 2 duo Mac game was playable. (system generation lasted minutes, each turn took about 15-20 seconds to calculate, but it was still OK). I thought I might face a new challenge, maybe even use some tank parts... But I have been surprised to see how easy it is, even in those conditions, to develop a good supply system. The drawback of the huge and low setting is that I can play unchallenged for 100 turns. I just discovered that I am not alone on this universe recently, and still no clue who are those light blue guys.... A few remarks:

-Don't you think there is a need for some extra challenge beyond maniatical? Look the research and production graphs. The IA is so far beyond that there is no way I can't win this game. And this has been the rule in all my last games with different settings. Small and dense gives for more violent game, huge and depopulated for more "how big is the universe" feeling, but both are too esay to win, with an AI far too predictable. Maybe it is just that the AI is not yet using the new colonization system correctly, but still... I was hoping on such a large map to get some challenge from a far away civilization... Maybe we can use a setting where some IA get in fact Advanced on the player(s). Let imagine they start with 3-10 systems and 3-10 technologies. Wouldn't it be a real challenge to deal with those high tech aliens? That could also give some early use for diplomacy: Facing high tech aliens, some of their neighbours might be willing to ally against them. I would definitely like to try something like that.

-Is there a way to turn the supply system more challenging? When you get to know well the research tree and the (few in this case but still) alien supply potential, you almost never get a system out of your supply line.

-I was thinking about maybe having some "pirates" encounters that can take your ships, pillage your planets, cut your supply, or turn your outpost into rogue ones. Maybe those guys can rebel from your own colonies if dissatisfied... which could bring some usefulness to the happiness meter beyond delaying the planet use. The happiness meter can drop when bombed by enemies, or if no upgrade is done to the system in some turns (just to avoid a happiness building).

-In this game I did not chose any hull as I did not built any warship yet. But is other similar games I opted for the robotic/titan path and I can confirm that it is almost as easy and sure winner as the asteroid one. Globally I think you found the way to balance the different hulls. I would maybe like some early sealthy hulls, but as you said sealth is to be rehauled.

-So my conclusion is lots of good work (cheers up !!), but I would really like a game where I am not sure to win at turn 100. Do you think implementing some advanced civs in the world creation would be hard to code? I have limited programming skills but I can try if well coached.
Attachments
Snapshot at turn 100... I have a winning hand don't I?
Snapshot at turn 100... I have a winning hand don't I?
Turn100.jpg (291.38 KiB) Viewed 854 times

godel
Space Dragon
Posts: 267
Joined: Tue Mar 24, 2015 12:58 am

Re: Game balance in 0.4.4 release—feedback needed

#70 Post by godel »

I have wondered if you would allow research trades as Civ does.
To have research steals when you retake a star would destablize the game: the research heavy type would be messed over.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Game balance in 0.4.4 release—feedback needed

#71 Post by Geoff the Medio »

marcOSX wrote:-Don't you think there is a need for some extra challenge beyond maniatical? Look the research and production graphs. The IA is so far beyond that there is no way I can't win this game.
I don't think "maniacal" is the same as "most difficult" AI. Particularly for a builder's paradise setup like this, a less aggressive AI behaviour might be better.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: Game balance in 0.4.4 release—feedback needed

#72 Post by Vezzra »

marcOSX wrote:Maybe we can use a setting where some IA get in fact Advanced on the player(s). Let imagine they start with 3-10 systems and 3-10 technologies. Wouldn't it be a real challenge to deal with those high tech aliens? That could also give some early use for diplomacy: Facing high tech aliens, some of their neighbours might be willing to ally against them. I would definitely like to try something like that. ... Do you think implementing some advanced civs in the world creation would be hard to code? I have limited programming skills but I can try if well coached.
Universe generation is handled by Python scripts, and it shouldn't be too hard to do what you have in mind - that is, if you just want to modify the scripts so some AIs get additional planets/techs at game start. Having that as standard options which you can select in the game options dialog box is a different matter.

But if it's sufficient for you if you can do what you want by modifiying some scripts in your local installation, that's already possible. Go to the "default" folder that contains all the content scripts, the universe generation scripts are located in the "python/universe_generation" subfolder. I don't know if you know Python, if not, you'd need to learn it (shouldn't be too hard Python is quite easy to learn), then start by looking at the universe generation scripts and see how far you get. There are comments in there that hopefully help a bit.

If you're stuck, ask. I'm the one who implemented most of that thing, I guess I should be able to give you some pointers ;)

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: Game balance in 0.4.4 release—feedback needed

#73 Post by Geoff the Medio »

marcOSX wrote:Maybe we can use a setting where some IA get in fact Advanced on the player(s). Let imagine they start with 3-10 systems and 3-10 technologies.
Start a new game normally, press turn 50 times before doing anything.

marcOSX
Space Squid
Posts: 55
Joined: Sat May 31, 2014 5:20 pm

Re: Game balance in 0.4.4 release—feedback needed

#74 Post by marcOSX »

Geoff the Medio wrote:Start a new game normally, press turn 50 times before doing anything.
Thanks a lot I will try that. I did not try really a builder's paradise as I put all sliders to low. But galaxy might be too big.

I will try a mix of both Geoff's advices before getting to hack python (I am more used to perl and php but always with emacs so python is near :wink: ).

Part of my problem is that I get the same kind of curves even with a tight galaxy and lots of fights not only "builder's paradise". What I do not build I steal to closeby enemies, and around turn 100 I always have a winning hand. I would like the result to be more uncertain. As an all-time fan of the Civ series, I appreciate that in Civilization you have (at deity) some quite challenging AIs to deal with, some of them so close to victory that you have to cross the world to take their capital city or rocket launching place. It makes all stages of the game exciting.

godel
Space Dragon
Posts: 267
Joined: Tue Mar 24, 2015 12:58 am

Re: Game balance in 0.4.4 release—feedback needed

#75 Post by godel »

marcOSX wrote:
Geoff the Medio wrote:Start a new game normally, press turn 50 times before doing anything.
Thanks a lot I will try that. I did not try really a builder's paradise as I put all sliders to low. But galaxy might be too big.

I will try a mix of both Geoff's advices before getting to hack python (I am more used to perl and php but always with emacs so python is near :wink: ).

Part of my problem is that I get the same kind of curves even with a tight galaxy and lots of fights not only "builder's paradise". What I do not build I steal to closeby enemies, and around turn 100 I always have a winning hand. I would like the result to be more uncertain. As an all-time fan of the Civ series, I appreciate that in Civilization you have (at deity) some quite challenging AIs to deal with, some of them so close to victory that you have to cross the world to take their capital city or rocket launching place. It makes all stages of the game exciting.

If you want challenge use etty, if you don't use Gys. No?

Post Reply