Finally tried 0.4.8...

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.
Message
Author
User avatar
EricF
Space Dragon
Posts: 357
Joined: Fri Jul 29, 2016 10:12 am

Re: Finally tried 0.4.8...

#16 Post by EricF »

Oberlus wrote: Sun Nov 04, 2018 10:49 am
Jaumito wrote: Sat Nov 03, 2018 3:55 pmYes, they're way too good. I changed that and made Nanorobitic Maintenance a prereq for SG hulls, and swapped their research cost and time. Now in midgame I have to choose between Nanorobotic and Large Asteroid hulls. At least there's a choice.
That is just great. I shall change my installation the same way.
I was just thinking that SG's are way too powerful for how fast you can get them
and was wondering what I could do about it. This idea seems very good.
Who knows, I might just build a Nano Robotic ship for a change. Never have before. :)

I still think it is a no brainer to go SG route over the Heavy Asteroid route, though.
It has nearly twice the HPs, already starts faster AND has a Core slot to make it MUCH faster later.
Plus it is not twice as expensive as a Heavy Asteroid hull. And I can build SGs anywhere.
Don't have to have an asteroid field. No real choice.
All the Hulls need rethinking to give us any real choice, but this seems like a good start.

Jaumito
Space Kraken
Posts: 189
Joined: Tue May 16, 2017 3:42 am
Location: Catalonia, France, Europe, Earth, Sol, Orion Arm, Milky Way, Virgo Cluster

Re: Finally tried 0.4.8...

#17 Post by Jaumito »

EricF wrote: Sun Nov 04, 2018 1:08 pm
Oberlus wrote: Sun Nov 04, 2018 10:49 am
Jaumito wrote: Sat Nov 03, 2018 3:55 pmYes, they're way too good. I changed that and made Nanorobitic Maintenance a prereq for SG hulls, and swapped their research cost and time. Now in midgame I have to choose between Nanorobotic and Large Asteroid hulls. At least there's a choice.
That is just great. I shall change my installation the same way.
I was just thinking that SG's are way too powerful for how fast you can get them
and was wondering what I could do about it. This idea seems very good.
Who knows, I might just build a Nano Robotic ship for a change. Never have before. :)

I still think it is a no brainer to go SG route over the Heavy Asteroid route, though.
It has nearly twice the HPs, already starts faster AND has a Core slot to make it MUCH faster later.
Plus it is not twice as expensive as a Heavy Asteroid hull. And I can build SGs anywhere.
With my suggested changes, it's not a no-brainer anymore, because of Asteroid Reformation and the dirt cheap efficient armor it gives you. Assuming you have asteroid fields and have researched some basics (Adaptative Automation, Military Robotic Control and Microgravity Industry), you can either go -

Asteroid Hulls > Asteroid Reformation > Heavy Asteroid Hulls
for 410 RP and 14 turns of research

- or -

Spatial Flux Drive > Nanorobotic Maintenance > Contra Gravitational Maintenance
Asymptotic Materials > Monofilaments ........ >
for 1520 RP and 17-24 turns of research, depending on your research potential (the 17 is theoretical, even with Scyllor I never managed to get there in less than 20 turns that early.)

The latter is only one small step away from Reinforced Hull which is nice but no substitute for rock armor plating, so you'll have to research better armor on top of the rest. While if you go the asteroid route early, you can skip the armor research tree entirely, since you'll aim for crystallized armor instead. So at least for research-challenged species, going asteroid seems a very viable alternative (again, assuming my suggested changes.)

User avatar
EricF
Space Dragon
Posts: 357
Joined: Fri Jul 29, 2016 10:12 am

Re: Finally tried 0.4.8...

#18 Post by EricF »

Jaumito wrote: Sun Nov 04, 2018 4:33 pm So at least for research-challenged species, going asteroid seems a very viable alternative (again, assuming my suggested changes.)
Ah yes, I had not taken the tech changes into consideration.
I still think the cost of the SG hull should be doubled to make it an even more difficult choice.

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#19 Post by ovarwa »

Hi,
Oberlus wrote: Sun Nov 04, 2018 10:49 am
ovarwa wrote: Sun Nov 04, 2018 6:31 am*chuckle* I think I'm defining mid-game as the point where RH with bombers and flak no longer cut it for planetary assault, and I want to build SG with something better than flak.
Woa, you mean you are attacking planets with flaks and not even a single MD/laser per hull? If that's what you mean, you are bold.
It actually works, though it takes more ships. 1flak+bomber+2regular armor. These can be built with little rp investment, are cheap to build and work well in groups of 2 or 3 (but not 1!) for staking out territory, especially once laser fighters are reached, which immediately upgrades existing ships. At some point, with good production, it's easy enough to peel off 8 or 9 and start picking off outlying enemy systems. For some enemy systems, 3-4 are quite sufficient. Each ship damages a planet for 9/turn. No rp is wasted on Laser2-4 (or MD2-4). The research can better go toward Zortrium or toward Reinforced Hulls (or both).

I'm still experimenting with which one is better. If you're Gysache, you don't want to use Flak. But if you're George or otherwise light on research...
Also, if at your mid game you only have laser and not plasma, that's a different strategy than mine. Except in some special starting situations, I will get first Nascent AI, then Laser and Robotic Hull with the rest invested on the Adaptive Automation path, and some growth techs if required for non-invading expansion. By turn 50-70 I must have AA, laser weapons, laser fighters and the second tier armour (never remember its name), which is more than enough to take down planetary defenses of systems with several planets. By turn 120 I generally will have Quantum Computation, plasma (weapon and fighter) and self.-grav, and maybe also death ray techs if I wasn't unlucky with the species. I think I also call mid game when my robos with laser or plasma can't stand the big shots of planetary defenses, but by then I must have better hulls.
It depends on what I play. It may be wrong to keep high-production species focused on production and research, but I do that. Either way, a typical research queue, leaving combat tech aside for the moment, looks like:

Physical Brain (only if focused on pp for fast +5rp building); ignore for a long time if focused on rp and not psionic)
Ecology (the +1 cannot hurt, it's cheap, even if there are sufficient nearby Good planets)
Algo
Nascent AI
Subterranean (only if I am mostly setting Adequate planets, otherwise I try to postpone)
Robotic Production
Fusion Production (only if focused on pp)
...AA

And in a perfect galaxy, such as 30 systems per player or High planets, I might not need combat tech yet, especially if I'm high research.

Otherwise, and I've changed galaxy settings so that it *is* otherwise, right after Nascent AI I toss in:

RH
Fighters
Laser1 (either last or first)

Laser Fighters can wait; Laser2-4 and Zortrium might never happen. If I don't go for an early fighter I go for an early shield, which a research focus needs anyway for more research tech.

I try never to build plasma guns. If my research curve is fast, by the time I'm willing to build plasma, I'll have Death Rays not long after. If my research curve is slow, I probably have huge production, and can last a very long time with SG+fighters+laser4s and maybe armor. I'll research plasma, but only to reach death rays. laser4 is better and cheaper than non-upgraded plasma.

For a research focus species (and I include Gysache here since for them I switch starting focus), I can achieve the basic techs plus all the combat techs around Turn40 or so.

If I don't pursue fighters, my early game ship is RH+2Zortrium+2Lasers+DefenseGrid, and a group of these easily withstand big planetary weapons (47-52 structure); I won't lose any unless there's an enemy fleet present, there are multiple older enemy planets or I have bad luck.

I notice that I *can* research SG around 120 in most games, but if I can avoid it I prefer to wait longer, prioritizing industrial and research tech to make my pp and rp ludicrous. Around the time I start building SG in such games, I notice that I have by far the highest pp, rp and strength... and I stop.

It's partially my fault too: I don't really feel the notion of "easy" races and "hard" races, but that choosing a race is every bit as much a part of the game as choosing whether to settle a Good or Hostile planet. (Hey, use your initial Colony Ship to settle a Hostile planet for a more challenging experience!) So if a race is broken, I might not play it, but I think they should be balanced, with different play styles. Including Human, which I simply won't play because they are utterly bland.

Anyway,

Ken

User avatar
EricF
Space Dragon
Posts: 357
Joined: Fri Jul 29, 2016 10:12 am

Re: Finally tried 0.4.8...

#20 Post by EricF »

I always play on Planet Density:Low so getting Subterranean and other Colonization techs are a high priority.

I would suggest playing a low planet density game and see if your strategy still works.

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#21 Post by ovarwa »

Hi,

I could play on Low Density. But I suspect the AI will enjoy it even less than I!

Supply will obviously matter more.

I think I mentioned that I research Subterranean early if I need to settle mostly Adequate planets. It doesn't help early on for Poor planets and isn't a priority if I can get a few Good ones.

If you suggest that I should prioritize it even more, I'm not sure I agree. Ecology is sufficient to let me plunk a colony on an Adequate world, and Nascent AI will give that world +2rp. If I'm focused on Research, Algo will help me get there too, and both will help me get Subterranean. And if I'm focused on production NAI provides some much needed rp. Of course, some situations can change this.

Anyway,

Ken

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#22 Post by ovarwa »

Hi,

SG: In my more recent game, I researched SG (as George) around turn 105, and by Turn 120 had pretty much wiped out my first, nearest opponent, Eassaw (sp). My fleet was mostly RH+2Laser4+1Bomber+1stdArmor, but I had a few SG+3Laser4+2Fighter+1stdArmor, built at a captured Ea world (their good pilots are even better when they're mine :) ). I don't know if I'd have been better off with more but cheaper
RH+1Bomber+1Flak+2Zortrium. I suspect not, but only because of the way ship costs increase based on number rather than value.

I still don't *like* the way fighters have been implemented, though I recognize their utility, even necessity, against an opponent who knows how to use them. I suppose I have a small choice regarding my RH loadout: My first 0-4 ships for normal pilots will be RH+1Bomber+1Flak+2stdArmor, depending on how quickly I need to something out there, and then I can either upgrade the armor, use lasers or even just keep building the crappy start ships, which work fine in groups and are cheap. But my SG choice is even smaller: 2Fighters and whatever armor and guns I have available, with 2Fighters+3Flak actually not being too terrible for a while, perhaps even useful against lots of fighters. Though obviously I want to use DRs and a Core engine as soon as I get them. And at that point, do I even really need another hull? I suppose if Titanics had another internal slot, but they don't, so they are a luxury, and there are probably better uses for 1.5k rp.

So fighters change what the best choices are, but I don't know if they really expand choices. They also do not at all behave the way I'd expect them to behave. No one crashes fighters into missiles or artillery to shield capital ships. Fighters strike first, rather than last. Rebuilding and replacing carrier-based fighters does not require more supply than repairing and reprovisioning a battleship or destroyer. And I think the current implementation makes a mess of the original design abstraction.

I'd consider the design good if I found myself designing ship loadouts as part of strategy or even tactics, reacting to my opponents beyond "what is the best ship I can build given my current state of technological development?" In fairness, I do not mean to say that fighters have made things worse, but I don't see myself thinking "I will build either X or Y, depending on what I think my opponent will have." (I leave stealth aside, because it seems the designers generally consider it broken.)

Flak: Though I use them, because so much stuff is unshielded, I'm not sure how great they are, even against fighters. They do not help against them on Turn 1 because no fighters have been deployed. They do not help on Turn 3 because enemy fighters will get to attack anyway. In a sense, then, Flak exists to kill fighters on Turn 2 so that your other weaponry can attack real ships on Turn 3 and so the enemy fighters cannot attack on Turn 3. But the more flak you have, the less other weaponry you have. So you forgo a better Turn 1 in favor of a better Turn 3. Of course, if you destroy a few ships on Turn 1, they will have less flak to use on Turn 2. So probably the best defense against fighters is fighters, limited by internal slots, and perhaps good armor instead of flak, once flak is not your primary gun. I don't know if there's an actual choice to be made here, or whether there's a best design strategy.

Anyway,

Ken

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#23 Post by ovarwa »

Hi,

My favorite thing about the AI: I like (well, hate: grrr! :) ) the way in the early game it ruthlessly probes my outer worlds, sending an armed ship followed immediately by a very few troops but just enough to conquer its target, if I cannot dispatch a good force there in time. This forces me to be a lot more cautious.

Anyway,

Ken

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#24 Post by ovarwa »

Hi,

I suppose I should talk a bit about what I mean above by 'choice', and I'll do it with a quick, simple combat system design example.

Start with the current FO ship combat system, except:

* No fighters, no flak. Just two kinds of weapon at every tech, first strike and sustained strike.

* First strike weapons only fire during Round 1 of combat. Perhaps they represent weapons that require long prep time to work, or some element of surprise, or long repairs between strikes, or long-range sorties by subsidiary craft or whatever. These weapons do not fire during subsequent rounds of combat.

* Sustained strike weapons fire during all rounds of combat.... *except* for Round 1. These weapons never fire on the first round of combat.

* Planetary weapons are sustained strike weapons.

* There are always at least 3 rounds of combat. After the 3rd and every subsequent round, there is a flat 10% chance that one more round of combat occurs if combat is possible.

* Non-combatant ships are ships that have no weapon systems. These can only be targeted if no other targets are available. (This saves on lots of clicking and annoying mm. It also gives the AI an opportunity to create fleets in which some ships are actually protected, with a mission "protect these guys" or similar.)

* Ships that only have first strike weapons are considered non-combatants, since attacking these can have no effect on the current combat.

Ok. We might want to run a few simulations and balance weapon cost and perhaps damage accordingly. Perhaps even adjust the chances of having extra rounds of combat. At this point, we potentially have an interesting set of choices to make. How much do we like the idea of a first strike that might wipe out an enemy fleet before they have a chance to fire? How much do we want our fleet to have endurance for a longer combat? How badly do we want to protect non-combatant assets in a system?

That's theory, anyway. Maybe there's an optimal fleet, in which case this exercise is yet another case of not offering interesting choices.

For example, we might find out that it is best to combine a fleet of lightly-armored first strike ships sometimes leavened with a few ships that have one sustained strike weapon and lots of armor... in which case this is a failing design.

Anyway,

Ken

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#25 Post by ovarwa »

Hi,

More on stability: My problems are due to linux not being happy with my Intel integrated graphics, which keeps posting unhappy dmesges while running FO, and either stalling or ctd. It seems to be a known issue, but none of the kernel command line recommendations help. I might have to upgrade my distro. Ugh.

Anyway,

Ken

Jaumito
Space Kraken
Posts: 189
Joined: Tue May 16, 2017 3:42 am
Location: Catalonia, France, Europe, Earth, Sol, Orion Arm, Milky Way, Virgo Cluster

Re: Finally tried 0.4.8...

#26 Post by Jaumito »

ovarwa wrote: Mon Nov 05, 2018 8:36 pm Flak: Though I use them, because so much stuff is unshielded, I'm not sure how great they are, even against fighters. They do not help against them on Turn 1 because no fighters have been deployed. They do not help on Turn 3 because enemy fighters will get to attack anyway. In a sense, then, Flak exists to kill fighters on Turn 2 so that your other weaponry can attack real ships on Turn 3 and so the enemy fighters cannot attack on Turn 3.
Well spotted. That's why I always play with 'Combat Rounds' set to 5. You'd argue it'll give fighters more punch, but more importantly it also give more time for them to be thinned down, which makes flak a lot more worthwhile. It also makes interceptors slightly better as a defense against bomber/fighters.

User avatar
Oberlus
Cosmic Dragon
Posts: 5714
Joined: Mon Apr 10, 2017 4:25 pm

Re: Finally tried 0.4.8...

#27 Post by Oberlus »

Regarding the turns in which each weapon is effective, I would prefer this (based on many previous suggestions from others regarding several aspects of the game, such as targetting, better differentiation of weapon roles, and more design and strategy choices):

- Long range (LR) weapons (currently just fighters, could also be missiles) can target enemy capital ships from round 1 (i.e. no launching round, you can consider that launching time is very short or is performed right before battle begins).
- Short range (SR) weapons (MD, Laser, Plasma, DR) can target enemy capital ships from round N, probably 2 or 3 (could be modified by features like leaders, species traits, fleet orders, ship parts, stealth, environment).
- Anti-LR weapons (currently flak and fighters) can "target" enemy LR weapons from round 1. This would couple well with flaks and interceptors not targetting capital ships at all and processing them in a separate fashion than anti-capital weapons, so that only fighters that survive the flak/interceptor fire on a given round can hit their targets in that round). This implies no "fighters will hit for sure on round 2 regardless of them facing an incredible wall of flak fire", and interceptors and flaks get a very definite behaviour. Also, bombers could be made to only target capital ships and fighters keep its double role with more obvious disadvantage against interceptors and bombers than currently.
- The combat duration, M, probably 4 or 5, could be modified by the same features than the first round of short-range combat, N. I certainly would like to have an extra path of research/strategy to confront with weapon types, shields and armour: tactical speed (different from FTL speed for inter-system travel), so that fleets specialised in SR (or LR) can close to SR sooner (or keep at LR for longer).

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#28 Post by ovarwa »

Hi,
Jaumito wrote: Tue Nov 06, 2018 10:16 am
ovarwa wrote: Mon Nov 05, 2018 8:36 pm Flak: Though I use them, because so much stuff is unshielded, I'm not sure how great they are, even against fighters. They do not help against them on Turn 1 because no fighters have been deployed. They do not help on Turn 3 because enemy fighters will get to attack anyway. In a sense, then, Flak exists to kill fighters on Turn 2 so that your other weaponry can attack real ships on Turn 3 and so the enemy fighters cannot attack on Turn 3.
Well spotted. That's why I always play with 'Combat Rounds' set to 5. You'd argue it'll give fighters more punch, but more importantly it also give more time for them to be thinned down, which makes flak a lot more worthwhile. It also makes interceptors slightly better as a defense against bomber/fighters.
I wouldn't argue that at all! It gives them less punch, because they get frittered away. But guns and shields work at full strength, even if the number of combat rounds is set to infinity.

(I suppose I *could* argue that have more punch because instead of being around for 2/3 rounds, they are around for 4/5. But they are ablated, so are not around 4/5.)

Anyway,

Ken

Jaumito
Space Kraken
Posts: 189
Joined: Tue May 16, 2017 3:42 am
Location: Catalonia, France, Europe, Earth, Sol, Orion Arm, Milky Way, Virgo Cluster

Re: Finally tried 0.4.8...

#29 Post by Jaumito »

ovarwa wrote: Tue Nov 06, 2018 6:47 pm
Jaumito wrote: Tue Nov 06, 2018 10:16 am You'd argue ...
I wouldn't argue that at all!
Sorry. I really should have written, "one could argue..."

ovarwa
Space Kraken
Posts: 122
Joined: Thu Dec 11, 2014 3:22 am

Re: Finally tried 0.4.8...

#30 Post by ovarwa »

Hi,

Some more thoughts...

I enjoy exploration, finding cool stuff and then acquiring it. (See lots of small goodies above.) So species that pretty much ignore the goodie game for much of their career are species that I enjoy less, eg Trith, Laenfa and Sly... and I suppose Eassaw.

Phototropic: Still mostly an overall penalty, I think. Except in a young galaxy or with specials set to low.

Self-contained: Powerful, but only if you can get to sufficient numbers of Poor+Good planets early. Otherwise, it is nothing to write home about.

Broad: Hardly a bonus at all. This works out for species that don't really need another bonus to shine, such as George. Otherwise, a great big yawn. It would be far more interesting if it gave 2 Good, Poor and Adequate environments and 3 Hostile. Or 1 Good, 4 Adequate, 2 Poor and 2 Hostile.

Sly: Still figuring them out. I suppose their strongest feature is that the AI has difficulty handling them. And their best feature is that they offer a completely different kind of play experience. But their stockpile feature stops keeping up around the time you'd want Automation. I suppose one could get stockpile tech instead, but then you're not keeping up with Industry... unless you forgo Automation to focus on population boosters until you're ready to create a real supply network using other species, at which point your stockpile benefits and investment no longer matter, and even your stealth benefit is greatly compromised? Until I played Sly for the first time, I didn't notice that 2 planets in the same system are not automatically in the same supply zone! Right now, Sly are a species I'd rather find than start as. I might very much be missing something though.

Trith: I play them as intended, which involves killing off anything that isn't Trith or Exobot. I'm not sure if this is optimal though.

AI: I don't know which settings make life easier for the AI *relative to a human*. Set X might show the AI in its best light, but if a human player can exploit those settings even more effectively, that's less challenging rather than more.

AI and missions: The AI has difficulty understanding how to abort a mission to exploit another opportunity. I was Laenfa, with a long line of doubly-guarded systems on one flank. I figured I had lots of time before I needed to worry about an AI breakthrough, and the stars were dim or worse, so they were not a high priority for me to attack. Except that the species on the other side was Eassaw, which I discovered when I had 2 ships incoming! Uh, oh. But no: They were happy to use their main force of 6 ships and camp out on their primary target to kill the guardians there. Meanwhile, I mutualled with their expeditionary force (which only happened because they were Good Pilots), marshaled a few ships in time to destroy a second expeditionary force with no casualties of my own, and had sufficient ships to defend my system by the time they liberated their original target... at which point they abandoned the system and fled. I did not pursue, because my focus was on the other side of my empire. (My ships: RH+2Zortrium+1Bomber+1Laser4). A human player would probably have dumped the big fleet on my world and made my life suck.

Starting conditions: Possibly the best starting condition is being close to the edge of the galaxy. Unless planets in the area utterly and totally suck, the ability to deal with only 1 or 2 enemies in the early game (whether I'm attacking or defending) is vastly important. If I have to quickly defend against 5 enemies at once, winning involves more luck, depending on who gets ganged up on all at once.

Empires and species: I know it's part of the design of the game, but I continue to dislike the choice to have starting species mean nothing, and starting empire mean nothing. If it's best to have only one planet of some species (say, Eassaw), and murder all the others in favor of some more useful species, that works out just fine, even if you started as Eassaw. No one seems to care that you're killing them off, including other Eassaw. Anyone can integrate any species into their empire with equal ease, and they might as well have been part of your empire all along. If anything, my aversion to this design choice has increased. I'll use species I conquer, of course, but I do turn natives off. I prefer faction bonuses to matter all the way through a game. (MOO2's faction penalties and benefits were a mix of empire/species bonuses, back in the day, and I preferred to choose benefits that I could not go out and conquer, unless they were so good that I wanted to start with them.) Even if bonuses were solely attached to species, I'd be much happier if there were always effects of your starting species. For example, you might not be allowed to use concentration camps on your starting species, you might not get full benefits for other species stuff, especially if you don't have the same bonus (or worse, have a penalty to the same thing), but could perhaps learn tech to improve things; Xenophobic could make it impossible for you to colonize worlds with other species (other than Exobots, of course), and even normal species have to pay more to colonize a world with some other species, representing various costs of letting a subject race expand. And perhaps subject species might not defend their systems as well... So if you are a Gysache, you might really want to acquire a Great Pilot species, but you'll never get their full bonus because your society doesn't value that as much as producing rp and pp. Conversely, you might want to enslave the Gysache to gain their bonuses, but you won't get the full bonus, and those Bad Defensive Troops will be Virtually Non-existent Defensive Troops, because subject species never defend well against factions that haven't committed genocide against them and because Gysache suck at troops to begin with.

That said, I'd much rather see work go into improving the AI, balancing the tech tree, improving the UI, in short, making existing features work well over adding even more features that need work. There's always room for 2.0. (Alas, I'm not volunteering to do this; I'm behind on the coding I get paid for. :/)

Anyway,

Ken

Post Reply