FreeOrion

Forums for the FreeOrion project
It is currently Fri Dec 15, 2017 8:24 am

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  [ 91 posts ]  Go to page 1, 2, 3, 4, 5 ... 7  Next
Author Message
 Post subject: My Feedback
PostPosted: Sat May 21, 2016 11:28 pm 
Offline
Space Floater

Joined: Sat May 21, 2016 9:11 pm
Posts: 33
Let me first say that as a community built freeware game this one is rather remarkable and those that work on it are really doing a good job.

Last time I was coding was in Fortran in the 80's so it goes without saying that any critiques I make come with that basic lack of code understanding as to the difficulties involved. Like a car driver giving feedback to a car builder any criticism must come from a point of certain ignorance. On the other hand I have been playing 4X and strategy games a very long time. I still own the boxed sets of my original SimCity, Civ1, and original CnC games with their 4x floppy discs. Windows was in its infancy when I was playing games like this. So as driver to builder criticism goes I'm not without experience and perspective. My ignorance will stem purely from the difficulty of coding, not games. Pardon the long disclaimer but I didn't want to start a row on my first day here on the forum.

Feel free to respond to any comment of mine, but please state your perspective for answers as either someone who helps build the game or is just a player. Also state whether you just don't like my idea or you are stating it really isn't possible to put code in for it. I have tolerance for those who do the heavy lifting, but very little for those who are just fanboy types who don't like criticism of their favorite toys.

I find the game getting a bit repetitive even after about 20 runs through it. The challenge remains the same because the opposition tends to remain the same. The opposition all colonizes the same way (filling systems completely and linearly as the find compatible systems in the order as they find them through expansion), and they are all uniformly aggressive to expand. True, they have differing degrees of aggressiveness but they all look to fill the space in front of them and want the player to lose if you are in their way. It all boils down to overcoming their ship stacks.

It's like dealing with the same AI player but with minorly different degrees of aggressiveness depending on the species chosen. The player at the outset has option to set the AI aggressiveness and what I think would help is to add some of those aggressiveness traits to the individual species. It would also be nice to see other options than xenophobic aggression. Cooperation would be nice to see too. A desire to mingle would be nice. Sharing space, research, trade, cooperation against monsters, or cooperation against other species would be nice. The game has a lot of species but other than other than favored terrain types and artificial limits on their ability to make ships or colonize they just seem to act the same.

I'm not sure who has played a game called Imperium Galactica 2 but it was one of the last 4X games to have some success before the age of online MM gaming deflated the desktop single player trend. Those who code might be interested in grabbing a copy to see how very unique species interact. All of the races were intent on expansion to some degree but they went about it quite differently. Some with brute force, some with cooperation, some with espionage and trickery, and some were just content to ride your coattails while you forged ahead in front of them.

The IG2 game also was filled with events and specialist opportunities. Some wise and some not. But it created branching paths to cut the linear aspect of the game. The game also had cut scene graphics and while very nice for the time they weren't necessary to create these divergent paths to expansion. It had ship and land battles too, but again not necessary as you had the option just to tell the game to simulate these battles via algorithm results. So while that game had features not suitable to include in this one, some of the underlying functions to break up the linear aspect could definitely be used. That and adding truly different personalities to species type could go a long way to making each game fresh.

While I've played about 20 times through the game I've started so many other games and resign almost immediately, many times in the first turn when looking at the home system and perhaps going a couple turns to see the ones next door. A limiting home system may add challenge but isn't always wanted. Instead of starting 20+ games just to get one decent home system the start screen should include the option for a rich home system. If the seed notation plays a factor in that then maybe just release what the seed code means and let those who want to tweak the seed to their liking be able to do that. Or just force fill a home system with a list of essential extras. A gas giant, an asteroid belt, and several other planets in varying degrees of usefulness. The game already forces a terraform to conform with the species chosen, so forcing more is just more artificial intervention, not something new. If players want a jump start then just give it to them instead of putting them through an hour of restarts to get what they want. Criticize them as non-purists if you like but don't irritate them enough that the prospect of a couple dozen restarts just to begin playing has them choose to play a different game instead that evening.

Ships. All the interesting ship technologies in the tech tree and ultimately I rarely use any of them. Once into robotic ships the linear aspect and need to fight the uniformly aggressive AI means many of the ship types aren't worth the bother to study until you just need research slots to fill. Most are too weak and don't offer enough of a change to bother putting to use. Too few useful equipment slots for most of the hull types and in the end you are just looking for speed, hull strength, and enough slots for what you wish to do. The same also goes for weapons. Jump to lasers first and then go for the death ray. The rest are just research slots to burn through. Maybe there is some minor differences but the goal ends up being brute force anyway, because that's what the AI does. You want a stack of ships bigger than theirs in firepower and hull strength. I say all this without a lot of good alternative ideas to offer. For ships I would suggest considering letting different races have advantages of different hull types or completely locking them out of some hull types. Maybe give Exobots bonuses to robotics hulls or organic being bonuses to organic hull types. You get my drift. All hull types for all playable races seems wrong somehow. Some races should be able to speed through research for one hull type and bog down or be locked out for others. You need far less premade ship types offered on the build trees as they just fill up space to scroll down. Let the players design their ships. As for weapon types I would suggest more, make them more different against certain hull types, limit or restrict them like the hull types to certain races, and come up with more options. Mine layers, sand casters, solid shots, missiles, etc. I would also suggest adding system-only hulls to create more defensive options. Monitor style defense platforms that have faster build hulls so star fleets aren't tied to defense. A star system should be able to completely defend itself and full destroy opposing fleets if enough build points are added to it. A focused laser using lenses to direct a star's sunlight could be an absolutely devastating system weapon.

Monsters. Low monster setting should be lower or add an even lower initial setting. Not every system with a special in it needs a monster. There should be a start setting where monsters are barely just color for the game instead of a bunch of expansion roadblocks locking a player (or the AI players) onto a handful of systems. Spawn rates for that new setting should be lower too. Finally there needs more options to make monsters more than just cannon fodder types. Galactic flippers to speed up tamed monsters. Sharks with lasers ideas would be interesting too. Put something to actually kill nests instead of just taming nests would be great too. If you can create a black hole then surely you can kill a Kraken spawn nest. Monsters are definitely a new place for game expansion ideas instead of just a lot of repetition.

Specials. Why is it that I get only Neutronium makers, multispectral shielding, death ray 1, kill spawners, and the odd dragon tooth ship? And about 2/3 up the tech tree I never get anything more? I mean those are nice gifts and all, but surely there could be something else. Offhand, I'm not sure what the krill spawner does and would like to know. More techs should be awarded. Some of those unavailable specials would be nice to check out (ash and cloud generators, dimensional cleavers, precursor shielding, void eruptors). Only one type of ship seems lame too. And cutting off any more after a certain point is wrong too. Getting nothing should remain as a random option but should never be automatic. Research bumps, uber lighthouses or scanners, special power generators, some of the existing but never found techs, more ship types, etc, would make the race to get more specials a lot more fun. They are a perfect vehicle to create more randomness and yet they have become boring fast. Once I get my early multispectral shielding and a neutronium maker I quit trying so hard. The risk/reward isn't worth it. I know I'll get cut off completely soon from them and I don't get anything else of serious note. a couple dragon tooth's early on are nice but they don't dominate for very long under the stacks of AI ships.

Game lag. Yeah, it makes the late game too big of a bother sometimes. It's got to be all those ship stacks. The AI piles up the ship stacks and player has to match. The monsters only exacerbate the problem because they just keep spamming away. Something to convince the AI to quit piling up the ship stacks may help. Have them recycle old ships more often and design stronger ships instead. Some kind of bonus for both AI and player to recycle old hulls would be nice. Adding a one time bonus to new ship construction would be a great way for encouraging ship scrapping.

Well I'll leave it at that for now. I had a longer metal list but I have already forgotten some of them. I'm sure I will think of them later and perhaps come in later to talk about them.

I'm definitely interested in responding opinions and I would hope those reading don't take offense at my criticisms. If it helps to know, I surely would have paid money for a retail version of this game a decade ago. Remember that I still spool up and play games I bought a decade ago. For community made freeware this is some of the best out there.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 4:58 am 
Offline
Programmer

Joined: Sun Feb 14, 2016 12:08 am
Posts: 359
Hi Zeke, most of your post mirrors my own thoughts (and I suspect a lot of others).

Krill Spawner creates an unowned Krill Spawn when the ship is in a system with an unowned asteroid belt. Of limited use, but I've found it is one of the few decent pairings to a Camouflaged Asteroid Hull (all internal slots).

The only species bond for ship hulls so far is the robotic interface shield. I've toyed around with expanding this quite a bit (for own use), but think there would need to be at least one general/main ship line before others become more specialized. Probably need at least 2-3 more hull lines for balancing this with the 5 metabolisms, though I'm not clear if/how far the dev team would want this to go.

Lots of creative ways to expand on monsters. Per the Krill Spawner and monster taming, I'm sure more is planned, will be exciting to see.

Expansion of ancient ruins gifts might be more of a late stage balance change. In the interim, maybe add a chance to pick from a random unlocked tech?

A couple of things that might help the lag a bit: modify the effect accounting in the options and (if you don't have issues) change the log-level to ERROR (in config.xml or as a command line arg). Personally my only dreaded screen is the research tree with all techs shown, but for those with a slow HDD or anti-virus it could be a game changer.

If you feel up to playing around with new ideas/content changes, the markup definitions are pretty straight forward. The Effects reference has the majority of the options on it, if you start getting more detailed.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 7:59 am 
Offline
Space Floater

Joined: Sat May 21, 2016 9:11 pm
Posts: 33
dbenage-cx wrote:
Hi Zeke, most of your post mirrors my own thoughts (and I suspect a lot of others).

I figured that very little of what I mentioned, if anything, was actually new material. But as long as the same folks don't keep repeating themselves on a regular basis to create irritation then some repetition from a new person has a bit of initiative to it.

Quote:
Krill Spawner creates an unowned Krill Spawn when the ship is in a system with an unowned asteroid belt. Of limited use, but I've found it is one of the few decent pairings to a Camouflaged Asteroid Hull (all internal slots).

Good to know but don't really see any value in it as Krill are just weak monsters. Maybe if you could stealth real well and lure Kraken with it then you could create a minor bit of distraction for an opponent, but that assumes you are very idle to mess around irritating an opponent without really going after them.

Quote:
The only species bond for ship hulls so far is the robotic interface shield. I've toyed around with expanding this quite a bit (for own use), but think there would need to be at least one general/main ship line before others become more specialized. Probably need at least 2-3 more hull lines for balancing this with the 5 metabolisms, though I'm not clear if/how far the dev team would want this to go.

Lots of creative ways to expand on monsters. Per the Krill Spawner and monster taming, I'm sure more is planned, will be exciting to see.

Maybe just make new monster evolutions caused by building additions to the spawning outpost. Say the next stage up from great kraken is the meson beam kraken as the 4th stage size that only happens at a modified outpost nest.

Speaking of monsters, them all being on the same team just seems wrong. Old relic maintenance ships that attack every player and ignore a kraken just makes no sense. I realize one team for all monsters is easier but I don't like it.

Quote:
Expansion of ancient ruins gifts might be more of a late stage balance change. In the interim, maybe add a chance to pick from a random unlocked tech?

Look in unavailable weapons in the game now. Lots more relic types listed that I never see in game. No need to create new ruins gifts. Just activate what is already there and put them in the rotation. Addong a few more ship types would be super easy too. So would opening more existing techs. No real need to add a lot of content for ancient ruin prizes. Just put what they already have to use and stop cutting it off so early. Heck, why cut it off at all? They may not impress you much late game, but a free dragon tooth is still a fairly useful attack ship at any stage of the game to add to your ship stacks.

Quote:
A couple of things that might help the lag a bit: modify the effect accounting in the options and (if you don't have issues) change the log-level to ERROR (in config.xml or as a command line arg). Personally my only dreaded screen is the research tree with all techs shown, but for those with a slow HDD or anti-virus it could be a game changer.

I'll look into those first items to cut lag but when you've got a gaming machine as robust as mine then limited graphics freeware games shouldn't be bothering it. Mine is a liquid cooled i7 stuffed full of ram and high end graphics card. It's not hard to figure out what's slowing it down. It's the ship spam and keeping track of it. The AI needs to quit trying to stack ships so deep and move up to bigger ships. That and letting the monsters just keep multiplying incessantly in fringe areas. Objects clutter taxes the game engine too much.

Quote:
If you feel up to playing around with new ideas/content changes, the markup definitions are pretty straight forward. The Effects reference has the majority of the options on it, if you start getting more detailed.


If I chose to help it would be on documentation. Too much of this game has too weak a description to help the novice player read up without having to just go try it and see what happens. I've never tried a nova bomb because the descriptions make poor sense and just look risky to use in the vicinity of my late game defenses. Documentation should be good enough for the player to understand the original plan for something. Trying something out to test it should be for playing with quirks and exploits, not using for using something as intended. Before anything is added or tweaked in the released game the Pedia should be clear on intent and how to implement it. Coding was in my very ancient history but I still remember the professors harping hard on documentation. It was expected that documentation took as much time or more time than the actual coding. Didn't matter if it fell into an infinite loop or worked absolutely the first time, if you didn't document hardcore you could fail the class project even if it worked exactly as you planned it to. There was to be no intuiting on the part of the next user. Every line of code was fully explained in detail. It was a tedious pain but mistakes were easily solved later and later users knew exactly what was going to happen. The weakness of documentation has sunk many a good retail game over the years. Too many people just shoot right to getting the code results they want and skip the documentation process. It takes leadership to demand it be done and holding those results oriented types responsible ruthlessly and without pity to get the culture change. Coding needs to be treated like a science, not an art. In science the masterpiece is only allowed to shine once everyone understands it.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 1:38 pm 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3291
Side observation (I promise to come back later for a more thorough reply, I agree with some and disagree with others).

A lot of the parts you're seeing in "unavailable" aren't unused relics, they're monstrous parts or similar, the void disruptor, cloud spawner, etc are all things put onto the Nebulous Body to create the stealth causing monsters, they're not designed for nor meant to be for player use (I keep meaning to assign the monster icons to each part, the 'ruins' icon is the default for any part that hasn't got an icon defined).

_________________
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  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 4:08 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4296
Location: Sol III
Something important to keep in mind is that this game is still in alpha stage, which means that many essential game mechanics are missing or incomplete. The AI being a prime example - it's still far from what we want it to be. A lot of the things you brought up will (hopefully) make it into the final release. It will still take us years to get there however...


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 4:46 pm 
Offline
Programmer

Joined: Mon Feb 29, 2016 8:37 pm
Posts: 204
IamZeke welcome.

I think that everyone here appreciates well considered criticism.

Keep in mind that freeorion is software libre, so it is not completed and released, but more of a work in progress. I think that the upcoming version number 0.4.6 implies where freeorion is in terms of progress towards completion.

Your observations of what is happening are clearly stated. I've looked through some parts of the code, so I'll try to shed light on why it is happening.

Aggressiveness
Aggressiveness is a proxy for difficulty. At the lower, levels it restricts the AI from colonizing, invading, researching certain technologies etc. The individual AIs are not personalized in any way, except for a 25% chance that they backed off by one step from the maximum aggression set by the player.

Startup conditions
The specific letters and positions of the seed notation have no significance. The seed only guarantees that starting from the same seed gives you the exact same initial galaxy state on all of the supported OSs.

Players have asked for different and conflicting things in terms of startup options. Any specific request is satisfiable, but the full set is impossible.

I also spend time polling for startup conditions. I'm in the midst of submitting a series of patches to speed up startup times. This will reduce the time spent polling for a desirable starting state and I'll consider a partial solution to the problem.

Monsters

There is a "No Monsters" option in the development version. If you can compile from source, pull the latest version from https://github.com/freeorion/freeorion/ and follow the instructions at http://freeorion.org/index.php/Compile. Or if you run Windows/OSX, then you can download the weekly test builds that Vezzra produces at http://freeorion.org/forum/viewtopic.php?f=2&t=4208&start=330 or http://freeorion.org/forum/viewtopic.php?f=2&t=5587&start=405

The monsters all being on the same team is an embedded assumption of the game that everything not an AI or human player is on the same team. I also find that inconsistent from a game play perspective, but not enough to try and change it.






.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 8:50 pm 
Offline
Space Floater

Joined: Sat May 21, 2016 9:11 pm
Posts: 33
MatGB wrote:
Side observation (I promise to come back later for a more thorough reply, I agree with some and disagree with others).

A lot of the parts you're seeing in "unavailable" aren't unused relics, they're monstrous parts or similar, the void disruptor, cloud spawner, etc are all things put onto the Nebulous Body to create the stealth causing monsters, they're not designed for nor meant to be for player use (I keep meaning to assign the monster icons to each part, the 'ruins' icon is the default for any part that hasn't got an icon defined).

Good to hear. And yes, a monster icon seems like the next move to stop confusion. But there may be some room for a version of those powers in actual ruin gifts.

Vezzra wrote:
Something important to keep in mind is that this game is still in alpha stage, which means that many essential game mechanics are missing or incomplete. The AI being a prime example - it's still far from what we want it to be. A lot of the things you brought up will (hopefully) make it into the final release. It will still take us years to get there however...

No need to apologize for an incomplete product. That's what incomplete means. But you must critique an incomplete thing to get it closer to being complete.

LGM-Doyle wrote:
Aggressiveness
Aggressiveness is a proxy for difficulty. At the lower, levels it restricts the AI from colonizing, invading, researching certain technologies etc. The individual AIs are not personalized in any way, except for a 25% chance that they backed off by one step from the maximum aggression set by the player.

As I said, it still creates a linear aspect to what you face in opposition. The AI opposition wants the space you are in and uses direct attack to get it. It won't go around you or try another direction. It won't try to "live and let live" while taking property you pass up on. It won't negotiate or trade for what it wants. It won't try to subvert your property discreetly, just bashes until it takes it or balks in place if at numbers disadvantage. It just boils down to your ship stacks versus theirs.

I'm fully aware that creating multiple personalities for AI behavior is a HUGE deal. It's not even remotely a rules tweak or bug patch. If this was a corporate setting and the command came from above to do it you would see the coding room collapse in various sighs, growls, and cries that their job just got 300% larger in just one directive. Just getting the one AI mentality right is a lot of work.

But it is this multiple personality type that greatly reduces the repetitive aspect of the game. I always know now how the AI is coming at me. It's coming right for the throat and trying to push me out with brute force. 100% xenophobia.

So while you push to finish the basic aspects of this game it may help to tinker on the side occasionally with other AI type behaviors. As I said, playing a few rounds of IG2 might give you a taste of what I'm getting at, as well as show how events can also disrupt the linear aspect of each play by introducing decision trees.

Quote:
Startup conditions
The specific letters and positions of the seed notation have no significance. The seed only guarantees that starting from the same seed gives you the exact same initial galaxy state on all of the supported OSs.

Players have asked for different and conflicting things in terms of startup options. Any specific request is satisfiable, but the full set is impossible.

I also spend time polling for startup conditions. I'm in the midst of submitting a series of patches to speed up startup times. This will reduce the time spent polling for a desirable starting state and I'll consider a partial solution to the problem.

Good to hear this is a known issue. I'm sure players will ask for black holes or neutron stars or specials at startup, but you can't satisfy everyone. But a "rich home system" switch would be nice. A couple extra planets that cover the basics would go a long way to cutting seed polling. A gas giant for power generation and asteroid belts for ship building, and a few varying normal planets with at least one more of good/adequate type seems easy enough. Leave the specials and star types random as now. Players want to get both an industrial and a research planet going early AND they want to use that one colonization ship to grab a good strategic location outside their home system. With the long build times for colony ships the home system having a second useful planet that can be had for a fast colony base ship is worth a lot of startup restarts. Just jamming a few extra common planet types in the home system might really cut back on the number of restarts.

Quote:
Monsters

There is a "No Monsters" option in the development version. If you can compile from source, pull the latest version from https://github.com/freeorion/freeorion/ and follow the instructions at http://freeorion.org/index.php/Compile. Or if you run Windows/OSX, then you can download the weekly test builds that Vezzra produces at http://freeorion.org/forum/viewtopic.php?f=2&t=4208&start=330 or http://freeorion.org/forum/viewtopic.php?f=2&t=5587&start=405

The monsters all being on the same team is an embedded assumption of the game that everything not an AI or human player is on the same team. I also find that inconsistent from a game play perspective, but not enough to try and change it.

My version is 0.4.5 so I have the "no monster" option already. That's nice to change the dynamic (and to reduce late game lag, btw), but the low monsters version still jumps too far with the idea. Monsters at that lowest setting do not need to be at all or even most specials. More like 25% or less of specials deserve a monster. And the spamming amount of new monster generation needs to be lower at those creation sites. On the low monsters setting the monsters should be a novelty aspect, not a tactical one. At present they still limit expansion progress at that low setting.

I'm not saying to nerf the existing settings, but add a new lower one so they still remain but now no longer represent a notable tactical challenge.

And adding monster effect buildings to outposts on nests is definitely a wide open new avenue for game expansion. There could definitely be a new style of play where for basic attack stacks the player could rely on monster ships heavily or exclusively. Right now the monsters are most just small irritants or big heavy hit point machines with weak attacks. Nothing but meatshields. But it would be interesting to play where your normal shipyards just make scouts, outposts, and colony ships and your monster nests make all your attack and defense ships.

Unlike my suggestion about AI personalities that would require crazy amounts of time to even bring one new AI personality type to the mix the monster tweaks could radically alter game play linearity without a lot of attempts at AI behavioral aspects. A completely new style of play with mostly just a few new building types and monster ship options added. And because you can't build a living being too many different ways from its original state you don't need to add a ton of premade monster ship types.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Sun May 22, 2016 10:46 pm 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3291
It's clear we need to make more of a distinction between monsters and guards for this discussion.

Specials that get guards get them because balance dictates the ought not to be usable within the first few turns as it gives an advantage. I, personally, intensely dislike that guards were turned off for the No Monsters setting, they're not wandering monsters.

For the sake of balance, if one player (say, an AI) gets an unguarded Ancient Ruins and unlcoks Death Ray or Multi Spectral Shields really early, that makes for a very difficult game for other players (ie, you, the human, in single player).

If you want fewer guardians, have fewer specials, they're not monsters and are, deliberately, not controlled by the monsters setting (that's been made clearer in the Test releases but it's probably not as clear as is ideal.

On the other hand, I have myself raised concerns about the number of actual monsters you still get in a Low Monsters galaxy and while we've made some tweaks I'm still not sure the balance is right.

There are two discussions: Whether specials should have guards and if so how many, and whether the Low Monsters setting creates the right number of wandering monsters. They are different.

_________________
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  
 
 Post subject: Re: My Feedback
PostPosted: Mon May 23, 2016 12:42 am 
Offline
Vacuum Dragon

Joined: Wed Aug 26, 2015 6:15 pm
Posts: 507
I don't totally agree with the OP, or even some of the responses. However, a lot depends on the specifics of your game. I like to play ones with a bit of space for development, and I actually find the AIs don't do a good job of colonizing their systems.

In the case of "Maniacal", they tend not to do much at all. In the games I play I found them to be so easy to beat that I quit choosing that setting. They would just research military tech, colonize one or two worlds, then build a fleet and set out for the neighbors. By they time they arrive, I have a good industrial base and can switch to a war footing and pump out some ships. Once their fleet is knocked down, it's a simple matter to back-track and take their territory away.

BTW, if you (IamZeke) aren't mixing up the galaxy sizes and structure, then you're missing one of the ways to freshen up play.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Mon May 23, 2016 5:10 am 
Offline
Space Floater

Joined: Sat May 21, 2016 9:11 pm
Posts: 33
MatGB wrote:
It's clear we need to make more of a distinction between monsters and guards for this discussion.

Specials that get guards get them because balance dictates the ought not to be usable within the first few turns as it gives an advantage. I, personally, intensely dislike that guards were turned off for the No Monsters setting, they're not wandering monsters.

For the sake of balance, if one player (say, an AI) gets an unguarded Ancient Ruins and unlcoks Death Ray or Multi Spectral Shields really early, that makes for a very difficult game for other players (ie, you, the human, in single player).

If you want fewer guardians, have fewer specials, they're not monsters and are, deliberately, not controlled by the monsters setting (that's been made clearer in the Test releases but it's probably not as clear as is ideal.

On the other hand, I have myself raised concerns about the number of actual monsters you still get in a Low Monsters galaxy and while we've made some tweaks I'm still not sure the balance is right.

There are two discussions: Whether specials should have guards and if so how many, and whether the Low Monsters setting creates the right number of wandering monsters. They are different.


Guardians on empty fortress planets? Yes, I agree they should be there regardless of the settings used. But even on low monster settings you see sentries and maintenance ships at every star system that has any kind of bonus in it. A planet with Ki spice doesn't need a monster guardian at the start of a low monsters setting game. In low monster games the actual guardian type ships should only be at empty fortress worlds and at inhabited planets where the indigs have evolved some self defense capability. My last game had a system with only 2 asteroid belts in it and somehow put that lithic mineral special on both. Not a system with bothering with early on, but it was in the way of early expansion and came with two maintenance ships! That basically cut off the map for close to 60-80 turns in that direction unless I was willing to lose ships to break through that logjam. It was just a silly roadblock in a game where I didn't want a lot of monster hassle.

A planetary bonus shouldn't rate a guardian ship monster in low monsters setting. There needs to be a bigger reason for one to be there. Empty fortress planets, indig races that have time to boost their planetary defense, and monster nests should be the only areas that rate a monster or sentry guardian type ship in this low monster setting.

defaultuser wrote:
BTW, if you (IamZeke) aren't mixing up the galaxy sizes and structure, then you're missing one of the ways to freshen up play.


Yes, I do tinker with galaxy structure a bit, yet all I notice is how much it isolates me at the start of the game. The AI opponent might not fill systems if you encounter it early enough, but with enough time they always fill every system they choose to settle into. And if you settle somewhere on the lines between two of their systems then they are going to come after you. The AI never ignores you if you are right in front or somehow get between their areas of operations. 100% xenophobia every time by the AI. They will only wait patiently if your ship stack is stronger than theirs.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Mon May 23, 2016 8:06 am 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4296
Location: Sol III
IamZeke wrote:
But it is this multiple personality type that greatly reduces the repetitive aspect of the game.
Totally agree. I fully expect different AI personalities to be implemented at some point.
Quote:
But a "rich home system" switch would be nice. A couple extra planets that cover the basics would go a long way to cutting seed polling. A gas giant for power generation and asteroid belts for ship building, and a few varying normal planets with at least one more of good/adequate type seems easy enough.
I don't know how apt you are at Python (considering that you've done coding in Fortran you shouldn't have any difficulties picking up Python), but it's actually not too difficult to tweak the universe generation scripts to get what you want here, if you don't mind tampering with those scripts a bit.

Of course that's no replacement for proper settings provided by the game setup, but until we get around to add some more of them, that's a nice alternative for people who want to experiment a bit and are sufficiently comfortable with messing around with Python scripts. ;)


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Mon May 23, 2016 10:27 am 
Offline
Dyson Forest
User avatar

Joined: Wed Jul 16, 2014 6:14 pm
Posts: 212
Location: Black Forest
MatGB wrote:
I, personally, intensely dislike that guards were turned off for the No Monsters setting, they're not wandering monsters.

For the sake of balance, if one player (say, an AI) gets an unguarded Ancient Ruins and unlcoks Death Ray or Multi Spectral Shields really early, that makes for a very difficult game for other players (ie, you, the human, in single player).


I got this in an older version a few times, one time I got Multi-Spectral Shields (IIRC) and that was a I-WIN-BUTTON. In an another example I got the warning that an AI found something very early on and thus, I abandoned game.

I therefore share your opinion. And I esp. like that some non-wandering monsters were created that are more defensive-typed, like the maintenance ship, so the AI won't loose that many ships against it while it still takes reasonable time to take down.

IamZeke wrote:
A planet with Ki spice doesn't need a monster guardian at the start of a low monsters setting game. In low monster games the actual guardian type ships should only be at empty fortress worlds and at inhabited planets where the indigs have evolved some self defense capability. My last game had a system with only 2 asteroid belts in it and somehow put that lithic mineral special on both. Not a system with bothering with early on, but it was in the way of early expansion and came with two maintenance ships! That basically cut off the map for close to 60-80 turns in that direction unless I was willing to lose ships to break through that logjam. It was just a silly roadblock in a game where I didn't want a lot of monster hassle.


A maintenance ship only has a 3-attack - build a single ship sporting a Deflector Grid on whatever hull and this ship can fight infinitely against even multiples of Maintships.
Same goes for a Sentry - one Robohull with 4*MD3 will kill it lossless (ie repair back after fight).
I do find the early game, and the fight against Monsters, to be the most interesting aspaect of the game - because here, the researchpattern can still greatly differ in respect of what the actual map holds available/unavailable.

IamZeke wrote:
Yes, I do tinker with galaxy structure a bit, yet all I notice is how much it isolates me at the start of the game.


From a player standpoint on, this is what I like, because it makes more up for a challenging game. For an AI however that can be crippling.
If a "Rich Homesystem"-option is included someday I'd like to request an extra option "only for AI" as well - or perhaps a "Desolate Homesystem"-setting that can be set for AI/Player (perhaps new players may set it for the AIs....)

Thing is, if I start a game where all the mentioned stuff (GG, Asteroids) are already present I know I'll easily win, and it will force me to take the same researchpath as well (Micrograv etc early). A few early GGGs are enough to get sufficient production running so it'll overcome any existing weak-prod ability & you can go full mongo on research --> easy win.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Mon May 23, 2016 10:50 am 
Offline
Dyson Forest
User avatar

Joined: Wed Jul 16, 2014 6:14 pm
Posts: 212
Location: Black Forest
Re: on different AI behaviour

What could also be done is to restrict several game items from different races, that way they'll have to play things out differently (in lack of alternatives). This could basically be done by restricting the techtree. As an example, Robotic Races should not be able to use Organic-based Hulls. As a compensation, they'd have less costs in researching Robotic-based Hulls.

The advantage, to my eyes, of such a method (over codeing multiple AI rationales) is that this actually has an impact on the player as well.

Ultimately, it could mean that every starting race gets a unique techtree for themselves, with individualized costs for any techs, and perhaps even a few unique techs reserved only for them (or their traits) nd some even lacking/missing. [*]

(in FO an empire consists of multiple races so that might be a contradiction on first sight, but research could be seen as being centralized or coordinated from the starting planet or Megalith planet)

Nevertheless, different AI would be very nice. There should be those that basically invest all their capacities in order to increase their prod/research - such an AI will be the victim of early military rushstorming but if left alone, will become increasingly dangerous from midgame on.
Another AI type should militarize up very soon (at the cost of stiffled expansion) such an AI could successfully defend itself against being rushstormed but it will loose in midgame if it can't overtake some other empire.

And those different strategies should be applied somewhat randomly to the enemies so it's not always clear to the player which strategy is run by a spotted AI.

One thing to realize is that the function of the AIs is to basically thwart the build-up of the player. And they do so better if they (silently) cooperate in that undertaking as if all of them plays to the same strategy which keeps them all mutually low.

edit:
[*] or perhaps, once you captured a new species you gain access to some of their unique techs, as well...


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Tue May 24, 2016 1:10 am 
Offline
Space Floater

Joined: Sat May 21, 2016 9:11 pm
Posts: 33
While I mention multiple AI behaviors I realize it isn't something to be done fast. The primary reason is you need alternative methods to succeed. A merchant minded AI requires a path to victory that allows trading, and therefore you need a trading aspect to the game. An espionage minded AI needs the system to have a spy aspect of the game. A peaceful colonization AI needs a way to victory using just gross population numbers. Even with straight military AI's you need to offer different types of tech tree systems in order to keep the opposition from looking the same. With this game you might even have some kind of "monster tamer" primary path to military victory.

But each AI personality would require a significant addition to game complexity because you need a tech tree and buildings to first enable it and then you have to get an AI to use it reasonably effectively.

I see this line of suggestion by me as something much longer term and not even something worth trying to force before beta stage. Instead I see it as something for those who want a distraction to tinker with on the side and not even all at once. Just getting one new path to victory will be a job, much less several all at once.


As for going to learn Python, Vezzra, my time is too pinched in my professional life to revive a skill that is 30+ years rusty. I realize you guys could use more help than just my opinions and playtesting but that's really all I have time for. I'm a middle aged man with far more hobbies already than I have time for.


Yes, Kassiopeija, after enough time your first robotic ships can start whittling down those big hit point low attack monster ships but it also depends how early you run into them. If you hit them within the first few systems then there is a limit on the amount of production you can put on the problem and still keep building your empire. Attack ships and colony ships compete for the same build resources. And to be fair to the AI opponents they face the same problem. The AI players are reluctant to press through those monster ship blocked areas. There just needs to be a lower grade setting for monsters than the ones offered now. Of all my suggestions so far this is one I do see as an option that could be coded easily enough. The player at startup would have 5 instead of 4 monster settings. None, Almost None, Low, Medium, and High.


OK, I have something new to bring up. I've played a couple games where there was a star system not connected to a star lane. Is this a bug or a feature? If a feature then what does it offer to the game? Can it be exploited or is it just a novelty with no realistic input to the game?

Also something else of use would be a species matrix in graphical form that showed acceptable planet types for those races that can colonize. It might help newer players trying to populate their empire with their subject races. Searching the Pedia is a bit of a pain and I'm only now beginning to memorize that list. Brand new players would appreciate it. I sure would have appreciated about 15 games ago.


Top
 Profile  
 
 Post subject: Re: My Feedback
PostPosted: Tue May 24, 2016 1:21 am 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3291
Feature. It's a victory condition (and we haven't documented them very well yet, we will at some point). Research Omni Scanner to see more info.

_________________
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  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 91 posts ]  Go to page 1, 2, 3, 4, 5 ... 7  Next

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