Reply With QuoteIs Tom reviewing this one?
PS I loved IGN going on about the "canny AI".
Started my first release-version game today.
WHY DOES THE INTERFACE STILL FADE OUT AT THE UPPER TWO ZOOM LEVELS?!
Fuck.
P.S.
I really liked the new intro video for my chosen faction.
P.P.S.
And I heard some new tunes, also. Great stuff.
Is Tom reviewing this one?
PS I loved IGN going on about the "canny AI".
I played a bunch today for my review. I still hate combat.
Has the AI been improved since the earlier versions?
I watched a LP a while ago where the dude playing was rambling about the AI spamming colonies where he had to carefully select his sites due to their happyness reduction mechanic. Also, it looked like while there's a limit to individual fleet sizes (a pretty harsh one) there seems to be almost no limit on how many flees you can have in total - is that right?
_____
rezaf
Yep, no limit to fleets in total, just how many can be rounded up for one fighting stack at a time. AI is better than in beta I think, but still quite shit, even enormous amounts of cheating don't help it that much.
The happiness reduction isn't too harsh really. If you go and madly grab planets then yes you will run into it and it will hurt. As long as you scout first, you will usually find a decent terran/ocean/jungle planet within the first few jumps of your home system. Once you have that running it's easy enough to find some decent arid/tundra worlds. The asteroid/gas giant heavy systems are insanely powerful late game, but if you rush for their colonisation techs early without researching the left hand happiness techs you'll just get striking colonies and jack shit happening.
Early-early mid it's essential to keep happiness as high as possible, and either buffer your happiness or dust so you can squeeze in as many turns at 'fervent' approval as you can. You don't really have the capability to bring in dust that will pay for much more than rush building a farm on a new system or hiring heroes/hero upkeep so I generally try and keep population growth happening over pretty much any other concern in the first 50 turns. After that you can look at getting some of the more extreme planets and boosting industry, but again, if you can't reach 'fervent' even temporarily, you're sabotaging population growth and everything else by some quite scary numbers.
I think I played this game out in beta, and I almost finished one retail release game. I can't finish it because ES keeps crashing (Heap allocation failures). I played this last game on a large galaxy with 8 players, one of each race. I played it on the 2nd from the top difficulty with me as the Amoeba race so I could see anything.
Anyway the AI is beyond bad. It is down-right stupid. Even with the insane advantages they had, they just bumbled around making stupid decision after stupid decision.
My main strategy after the initial land grab was to hunker down, research and upgrade my worlds. I didn't build any influence based improvements (this pisses off the AI). I maintained peace with everyone except the cravers who were far away from me. I just traded and teched up. I did make a mistake, I allied with someone, and once that is done you can't un-ally. It is bugged, you can drop the alliance from the diplomacy screen, but it still says you are allied.
Anyway in the turns in the 200s, (I disabled most victory conditions that allowed you to sit in one place and win, such as economic, science, and wonder victories), my FIDS had risen to 2nd place. 2ND PLACE?! How? The AI has a huge advantage from the difficulty, AND has far more worlds than I did.
It is just terrible at managing its own worlds. Eventually I got all the military tech I wanted and went to war (well war was declared on me). I began exploring the idiocy of the AI.
Stupid things I found about the AI:
1. Despite the fact that they had more worlds and a monster FIDS bonus, somehow I became 2nd in research in the late part of the game.
2. I got my FIDS to #1 in the mid 200s while owning the least amount of systems by a significant amount.
3. An AI who was at war with several people and losing, declares war on me, only to have me walk all over them.
4. While they had the tech to upgrade worlds to Terran (I found several systems of all Terran worlds) they owned, they consistently failed to upgrade most of their worlds.
5. They love their dust. I think they like to buy everything since that is the most common planet exploitation.
6. They apply every single system upgrade to each system they own, even if they do not need it. A massive dust cost for no reason. IE stuff that upgrades asteroids and gas giants in systems that have neither. A system that is ecstatic will still get even more moral upgrades.
7. As soon as they get the tech to turn food into production, they apply it everywhere. I found countless systems at like half population cap because all the food was going to industry instead of maxing out the population.
8. They just throw ships at you when attacking without any bigger strategy. They decide they want some system and will go crazy to get it even if it is a losing proposition.
edit to add: 9. I do not think the AI uses system Governor heroes. I think they all end up in fleets which IMHO is the worst use of an academy slot.
I think I am done with the game. Each game plays out the same more or less and it is not any more fun to play against a moron who has infinity symbols for his resource counters.
Last edited by DeepT; 07-09-2012 at 06:55 AM.
Yikes!
It's a pity the promise of this game wasn't fully realized. Oh well, let's see if they patch it enough.
DeepT: I agree that the AI is rat-retarded.
Play multiplayer! It's super-easy to drop in and out and is really fun.
I have to wonder how much of an economic advantage they gained by pushing this out early. It's a shame. This could've been great, but now it will be forgotten quickly unless something terrific happens.
How does that work? You get people who are willing to sit around for 3 or 4 hours to play a small game? Are turns simultaneous? It seems like it could be really slow especially if you just want some research to finish and wait for 5 turns to go by.Originally Posted by Giaddon
Basically, when dudes set up a game they can decide to make the game "open" (you can do this too!) When the game is open, it shows up on the "join game" list you can access from the main menu. You can hit someone's game to take over from an AI, and when you quit it will go back to the AI. Turns are simultaneous and it goes pretty quick, I've found.
If you want to play a game start-to-finish it's probably best to organize with players you know, (though I'm sure you could complete a small game if you had a few hours to spare) but I've found it fun to drop into a game, get my bearings, and play as best I can from there. If I have to leave, I quit guilt-free knowing it goes back to the AI.
I think it's a really good system.
To quote myself from a few weeks ago:
I knew it would come to this and sadly I was right. Another 4X game I can add to my list of "potential is there but everything else lacking".
When will game devs understand that we don't need another 4X game with crappy AI and even less features than games of this kind could provide 10 years ago?
Even the nice graphics and art are wasted with the current combat system which is neither on a tactical nor strategic level fun and gets boring pretty fast due to its repetitve strcuture and total lack of balacing.
I really wonder why the 4X genre is one of the worst when it comes to polish and balance.
This game has a battle system that's quick and fast for multiplayer, it makes it awesome.
Try to play a game of SOTS 2 multiplayer and you will NEVER get anywhere, its just too slow paced.
Combat is about ship setup, heroes and cards, it's strategy as you can certainly win by playing better.
I think its a really good 4x game, the AI is not the best, but if people stop playing on 'easy' and start playing on some of the harder levels, you WILL loose, and I personally would like to rip HEADS off if I see another guy complain about cheating AI.
Total War franchise had an AI that didn't understand the game, it was WAY WAY worse.
Don't conflate drop-in / random multiplayer with all multiplayer. I play long TBS games like SOTS/CIV/EU3 etc with my friends over the course of a few weeks. While the niche is undoubtedly small, there's still room for slow-paced multiplayer games.
I don't mind the cheating so much as that it cheats in a way that is no fun to play against. I mean - take the effect of those cheats on combat - what purpose is served by producing and then suiciding ships at a ratio of 73:1 other than boring you to death? It's good at levelling your ships up, which makes the next fight with the next suicidal AI even easier and even more boring.
What do they expect? "That's top cheating guys, well done" ?
won my first post-release game against the AI:
Hissho
Normal difficulty
Large map (6-arm spiral)
Six races (Hissho [me], Hissho, Sophons, Cravers, Pilgrims, Sowers)
Everything else standard
As long as you just keep killing people, the Hissho are pretty easy to play against the AI. Their low research can be overcome fairly easily by doing some modest research into... research. Then just build some labs and you're pretty much on-par with everyone other than Sophons.
And you need that research power: one of the strengths of the game is the inter-dependance of the tech-tree. YOu can't just go all-combat because you have to research up the tech to find the minerals to build your weapons. Also, you need the finance techs to refit your ships. Plus, you need the warp tech to get to your enemies. Etc. Etc.
I eventually won by conquering 75% of the galaxy. I think it was around turn 185. And thank goodness: after I had wiped out two of the six races (Hissho and Sowers) the writing was on the wall, and it would have taken another fifty turns of incredibly tedious fleet-management to kill the two that were still alive when the game threw in the towel. My people despised me, and I guess that's OK because very, very few of them were actually my race... 90% were subject-races and I gave up on keeping them happy. They seemed to build my replacement fleets just fine while on "strike".
Here are my lessons-learned from the game:
- The game is stacked against bigger ships. You should research up to dreadnought... but only to get the command-rating boost that it gives you. Building anything larger than a destroyer hull (100 hull-storage points) is not cost-effective.
- Almost all defensive improvements are not worth the space. That includes kinetic armor, missile-defense, shields, HP-increasing armor, etc. I'd also throw in bigger engines, larger sensors and self-repair gizmos. All these things just take up space and ton-for-ton another gun is a better use of your allocation than a countermeasure... especially since you'll be only building small vessels that can be replaced quickly. I suppose that if your empire is GARGANTUAN, an engine upgrade could be worthwhile, but there seems to be plenty of tech that increases ship speed without actually buying better engines.
- When you conquer a system, it seems like you get all their buildings intact. That was nice of them.
- I you're going full-offense, you might as well not concentrate on learning how to colonize, say, a methane world because if your victims have that tech you can just colonize their planets and not worry about it. Later on, when your research value is much higher, you can go back and grab the ones you still need when they only take a turn to research.
- The tactical AI is terrible. There were many. MANY cases where they would play the "Chameleon" long-range tactics card, which increases their anti-missile effectiveness by 40% and decreases their kinetics effectiveness by 20%. That was generally stupid because (a) I never mounted a single missile, and (b) their primary armament was almost always kinetics. I saw that type of stuff all the time. Disappointing.
- Almost all my space battles were resolved in the "long range" round, and although some of my casualties might come from beam weapons, generally I only suffered ship-loss from postmortem missile hits.
- My long-range card varied depending on what my odds were and what the enemy weapons consisted of, but my medium and short-range cards were always the nano-repair option. The reason for this is that although the battles were by-and-large decided in the first phase, the computer still goes through your last two phases anyway... and the amount of damage you can repair using that card is about what it would take six turns to repair orbiting one of your planets. I actually had one occasion where one of my fleets came out of a battle significantly healthier than it was when it went in.
- Fleets should travel in packs. Not only is is easier to stop opposing fleets in numbers, but you can consolidate your ships into the maximum size after each engagement... and it's ALWAYS better to have seven ships in a fleet than it is to have seven one-ship fleets. But it's also tough for a single fleet to burn-down the defenses of a system in a reasonable amount of time, and it would be better to take one system quickly (and start reaping the industrial output of same) than to burn down two or three slowly.
- Don't bother upgrading obsolete ships. Oh, I did it often enough, and I'll say that in the first 75 turns or so it might be useful since you won't have gone into full-blown "wave" strategy mode yet, but as you conquer, your eternally-angry space-slaves suck up more money than they produce, so you won't have much cash available for upgrading your ships, particularly in the late-mid game where each destroyer can cost 500 gold to incrementally upgrade. At that points it's actually cheaper (in the long run) to consolidate the older ships into suicide fleets that you send to their deaths or just to scrap them.
- If you're on offense, don't bother with missiles. They only fire once per phase and only against one ship, which results in a huge amount of overkill and inefficient shooting. It also means that the battles will go past the long-range round and deprive you of the "nano-repair trick". I found that beam weapons work pretty well.
So I'll start playing on the harder difficulty levels, but I think some of the issues with combat probably need to be addressed at a fundamental level. The hull-space model needs to increase non-linearly so that there is a reason to buy larger-hulled vessels. The ratio between offense and defense needs to skew more towards the latter than it does now.
My buddies and I have been playing multiplayer and it is a great deal of fun. The drop in/out model is ripe for exploitation (probably) but since we're not out to cheat each other this really doesn't matter to us. I think it was a good decision.
I am very impressed with the tech tree. There are several different paths to success and although the races aren't really all that distinct, the unique tech and their pre-game research makes the research meta-game nice and variable game-to-game.
I am similarly impressed with the display. There are multiple ways to get the screen you need, and the iconography and hover-notes REALLY give you a huge amount of data once you get past the first big hill in the learning curve.
The late-game is super tedious, but that's standard for a 4X game,a nd I don't see much that they could do to make it better. But I do have some suggestions (that I will also submit to them):
- I'd like to see an option to suppress building options that I will never use for a give star system. If all I have are "large" and "huge" worlds, showing me that I can buy the improvement that makes "tiny" worlds better is just annoying. Likewise, if I have a system where the planets have no moons, I really don't want to have to scroll through all the moon-improvement options. Being able to gray them out or shunt them to another tab or something would be nice.
- I'd love to see an auto-forward option for new ships that sends them to a rally-point somewhere instead of building up in the planetary hangers. This isn't such a terrible thing since the UI has several ways of showing you where your ships are sitting, but it would be nice. Maybe there could be a "naval depot" building that allows you to be able to send ships to a planet?
Overall, this is a worthy successor to MOO2.
The point of the game when you start it, is to win, can you do that? Or are you just complaining that its too hard?
Also KevinC, if you got the time and friends willing to do MP in those games, its awesome, no doubt about it, however I think Endless can provide just as awesome experience, one where you can actually declare a winning instead of speculating aswell.
I don't really care if you think the Total War series had worse AI. That doesn't excuse what's going on here.
This AI is awful and it does cheat. Rip my head off if you want. A "cheating" AI isn't all bad. In some games, it's actually a good thing to have the computer cheat a bit. Where a cheating AI falls down is if if cheats in a manner that makes it plainly obvious what's going on (seeing the wizard behind the curtain so to speak) or cheats in a way that makes winning impossible.
As some of us have pointed out, this game cheats badly. It does stupid things and still cheats so you can see the cheating happen. Cranking up the difficulty doesn't make it better. It just makes it harder. Some would say that losing to a stupid cheater on a harder difficulty is less interesting than playing against a stupid cheater on easy.
Yea, I think we all can expect AI to cheat because it just won't compete with a human on an even playing field. I think what really turns me off on games is when an AI cheats in a way that takes a lot of strategy off the table. For example, if an AI in a game can conjure armies/fleets out of nothing, then there's no point in trying to knock out their industrial worlds, no economic warfare, no blockades, etc. I hate when strategy games do that, and it sounds like that's the current state of Endless Space.
Maybe I'm wrong, but I'm definitely going to hold off until the AI is patched to the point where it doesn't use it's massive cheats to headbutt you with stupid until you submit. That doesn't sound like fun to me, at all.
Picked this up Sunday night as part of the Green Man sale, and finally started my first game. The tutorial is a little dense, but I'm liking it so far - although I'm not nearly far enough into to complain about the AI yet.
Parts of it just feel... unfinished, though. The big thing that I haven't figured out yet is how to know how long it will take to travel to a system. Did I just miss something, or is that information not there?
It's not great, but once you give a fleet a destination, you can zoom in and the slightly larger dots along it's course indicate places where it will stop at each turn. I believe I've read the devs are planning on adding something a little better at some point.
Oh another one to join SOTS 2 on the pile...
(Well, no, to be fair I pulled SOTS 2 out of a bargain bin for 1.99 and don't yet have endless space lol)
Endless Space may not be finished, but its beta form was better then SOTS2 is even today. Its not a fair comparison by any standard.
I don't really view balance and a decent AI as optional ><
I didn't say it was optional. But I picked up both SOTS2 and Endless Space at the same time a few months ago and put far more time in ES because its still by far in the better shape of the two games. All I'm saying is that ES is in better shape then SOTS2, not that it is where it should be for a release.
The decision to release it now was disappointing given the promise of their development model. And if selling the beta couldn't keep them going then I'm worried about what releasing it will mean for the company.
That all said, I usually have more fun in these types of games building my empire up through colonization and expansion then I do when I get to the point where the borders are defined and its time to see who the more powerful is. AI failings are usually more obvious in the latter part.
My favorite 4X of all time, MOM, was horrible at both.
Yes, well, there are more interesting kickstarters to fund and other stuff to play before I buy this then :P
It sounds like I'd end up doing my own balance pass on the game before I found it enjoyable (if that's possible), and that's time I'd rather spend playing.
Too bad about the AI troubles. I picked up this game when it was in beta figuring that it looked too promising not to snatch up.
It isn't the dire AI that bothers me so much as the bland sterility of the game and tech advances. +2% here, +2% there, shuffle the huge cascade of fleets around until the enemy is crushed. There's no feel of grand strategic decisions, only of incremental micromanagement.
It seems like one of the better MOO-style games, but I'm not really excited about it. No espionage? Simplistic planet-building? Rather boring techs?
MOO2 was 1996 - that's 16 years. How hard can it be to take the best parts of that game and improve upon the formula?
Posting Permissions