Review-in-Progress (Open Thread)

[yin26]

Originally posted by Shadowlord
Hmm. I take it the GalCiv II AI doesn't refuse to trade techs for a while if they're important? That was an innovation that I think worked really well in Civ IV.

As far as I can tell, the AI puts its techs out for sell the minute it gets them.

Does it price techs higher if they allow wonders that haven't been built yet?

Hmmm, I can't tell. Interesting question.

[yin26]

Originally posted by drank
That's a pretty funny post, Yin!

LOL. Thanks.

On a smaller map, I find some fun in these various games. The StarbaseShuffle and TradeTango become tedious quickly on a big map, because you just don't have adequate UI tools to coordinate the building of dozens of starbases with hundreds of constructors.

Excellent point. For one, I can't see to toggle a starbase option on the minimap. Is there one?

The DiploDance is potentially a fun way to play. I agree that Diplo is a very powerful bonus, and indeed, the player can do quite well tech trading early on. By the 100th time you're dragging the slider on the Diplo screen to find what the AI will pay, it's lost some of its charm. A better UI (and an AI that better understood its strategic interests in tech trading) would make this genuinely fun in my book.

Agreed. All the ingredients are there...they just need to be mixed a little better and left in the oven on high for about 2 hours.

[Shadowlord]

Hmm. I take it the GalCiv II AI doesn't refuse to trade techs for a while if they're important? That was an innovation that I think worked really well in Civ IV.

Does it price techs higher if they allow wonders that haven't been built yet?

[drank]

That's a pretty funny post, Yin!

On a smaller map, I find some fun in these various games. The StarbaseShuffle and TradeTango become tedious quickly on a big map, because you just don't have adequate UI tools to coordinate the building of dozens of starbases with hundreds of constructors.

The DiploDance is potentially a fun way to play. I agree that Diplo is a very powerful bonus, and indeed, the player can do quite well tech trading early on. By the 100th time you're dragging the slider on the Diplo screen to find what the AI will pay, it's lost some of its charm. A better UI (and an AI that better understood its strategic interests in tech trading) would make this genuinely fun in my book.

[yin26]

First, I'd like to thank everybody for such great posts.

One thing that occured to me this morning (finally got some more time to play): GalCiv2 excels at giving the player a number of somewhat low-level, simple, but engaging mini-games for the types of players who enjoy them. Consider:

1] The "Starbase Shuffle" (tm) -- If you zoom out to the strategic view (which is awesome feature, Brad, thank you), you'll see the AI literally spamming space ("Space Spam" (tm)) with constructors on their way to do something. Zoomed in, you miss this. So, you have an entire realm of starbase possibilities that become their own game within GalCiv2. Granted, for the player who would rather just focus on taking over planets, this mini game is not that big a deal.

2] The "Trade Tango" (tm) -- Here, too, setting up trade routes, particularly ones that aren't going to get wiped out...and then running econ starbases along the routes...is its own mini-game. Again, simply taking over other planet populations cuts to the chase, and trade taken to this extreme can also be given only moderate attention if you want.

3] The "Diplo Dance" (tm) -- I am experimenting this morning with a custom civ using huge diplo bonus (Master?). As Sirian has noted, you can leverage the diplo bonus like crazy. I put ZERO spending to research for a good part of the early game still stayed ahead in the tech race by leveraging my diplomacy. To some extent, this is also an interesting kind of mini game.

4] The "Customizing Cha-Cha" (tm) -- I think most everybody agrees that the ship customizing feature of GalCiv2 is a potentil life sucker all on its own! I've seen some reviews joking start with "You mean, GalCiv2 is more than ship design software?" Again, you could ignore this feature, too, if you didn't find it engaging or relevant or balanced.

With some beer, pretzels, and no real pressure to beat the AI like crazy, hours and hours can be blissfully employed plotting a new starbase, tweaking another trade route, squeezing out another spectacular diplo session, or crafting the coolest looking (and minutely speced out) ships in the galaxy.

So, my theory is that depending on whether or not you see those elements as either a) micro management hell b) fun mini game or c) virtually irrelevant / part of enabling problems in the "larger" picture if it's too easy to take over planets in the first place will have a profound effect on your review. This makes sense with what I know of Sirian and Solver (whom I would put mostly in the "c" category) and myself (I'm some place between "a" and "b" with a long term, but not yet pressing, look toward "c").

So if anybody is reading and trying to decide about GalCiv2, ask yourself if you enjoy those kinds of games within the game and have a high tolerance for micromanagement. If not, would you be willing to deal with those if the AI ultimately were capable of putting real hurt on you in intelligent ways?

I'd say at this point in GalCiv2's development that the AI can and likely will improve greatly as Brad hears from fans on what's working and not working. In the meantime, there are plenty of micro-management mini games to keep you very entertained IF you like those sorts of details. Thus, I believe a lot of the very positive reviews rightly focussed on those elements of the game that were more accesible and easier to manipulate while probably not reflecting the kinds of laser-surgery being performed by the likes of some in this thread who cut to the AI heart of the matter.

[VetLegion]

Hey Solver, post some screenshots too to liven it up a bit!

[Solver]

A few very short points - I will have part 1 of today's extensive comments in several hours.

I believe that the reason for professional reviews being somewhat different in their verdict is that, indeed, the bar is set lower - but in a slightly different area than you would expect. I think GalCiv2 is excellent on the surface and with some truly awesome concepts. However, I am setting the bar pretty high for strategic depth - that is, the lack of situations with one right choice, which seems to be a weaker point of GalCiv2. And the second part is the AI - yes, I would also agree that the GC2 AI is one of the best in the genre overall, but it's still not good enough. During the development of Civ4, I was paying a lot of attention to the AI and performing analysis of how it works - with that in mind, I am not so excited about the GC2 AI.

I'll play this night, I think, then maybe put it down until 1.1. I do feel like I owe Brad and everyone else to be fair.

As said, more extensive stuff in several hours, complete with more (possibly useful) gameplay observations and maybe even the time will be ripe to post my fix ideas.

[bonscott]

Originally posted by drank
To use your example with the Drengin invasion, one of the things I've learned in GC2 so far is that if the human player just fights, he can beat a seemingly tough AI force. I'll bet, for example, that the Drengin ships had movement 2 or 3 - and your ships were much faster. The attacker gets a big combat advantage, so you probably have enough speed to hit the Drengin, run to a planet to heal, and come back and hit them again, long before they actually prosecute their invasion. You'll also do a better job of levelling your units, so that they soon will have far more hitpoints than their Drengin counterparts.

In short, when the AI attacks, it's a paper tiger. It has a large quantity of ships, but unless it's got an overwhelming tech advantage, you'll probably beat it with quality ships and good tactics.

In general I would normally agree with you but not this time. I had 2 fronts going on. The 3 Drengin worlds at the very back corner of my territory. So most of my forces were down there taking care of business. I only had 1 fleet on my main border. But the Drengin were bringing it. I calculated it and I wouldn't have gotten all my main fleets back to the other front in time to avoid losing at least 3 core worlds (including homeworld) and a couple very key starbases. So getting peace was the better idea and I didn't have to give to much for it since I had already inflicted some damage to the Drengin.

If I wouldn't have had that 2nd front then yea, I probably could have stood up to the Drengin and destroyed their incoming fleets. Now that I've eliminated that front I can get all my forces in position to avoid this again.

My biggest gripe with the AI is this: They declare war but don't have the ships in position to take it to you (usually). If they are far away then you've got a good amount of time to muster up your fleets in position and perhaps even meet them halfway. This is where the Civ 4 AI is sometimes better. The Civ 4 AI tpically won't declare on you until their SOD is right on your border or darn close (however the good player can still deflect this tactic easily by bribing another AI to attack the one that declared on you and now the pressure is off you after the first wave.) This is certainly a spot the GC2 AI could be improved. However I haven't played above Tough so I can't say if it's any better on smarter levels about this but sounds like probably not.

[Gufnork]

I also think that many who praise the AI does so because of the messages they get from other civs. They tell them they see you building up, so they decided to declare war before you do. They declare because you're hurting their trade and so on. In addition to that, they often do it in a very funny way. This causes people to think the AI sees right through you and makes you want to think the AI is smarter than it is. I must praise Brad for that, because how good the AI appears to be is really what matters.

For me it all breaks down when a minor civs declares on me because I was building up, when I was in fact transporting my troops to the front line. It's declaration of war caused me to halt my troops, invade his poorly defended world and move on. Another declared war on me because I was hurting his trade, the very turn after he paid me a kings ransom to get out of the war.

The AI also has quite a few weaknesses that people who are good at reading the AI can exploit (and of course I'll exploit it, that's how you win competetive games. You exploit your opponents weakness while trying to minimize your own weaknesses). It's predictable once you learn how it works. I've kept track of the official forums and I've seen the gradual change in people's opinion of the AI. People learn and adapt. The AI is very predictable so eventually people will learn and be able to beat it. I don't think the AI as it is can last that long, sooner or later a majority will have bested it.

The way I see it improvements to the AI is necessary to keep people from figuring the AI out. Even people who don't analyze their opponents very closely needs this, there's just no hurry in that case. But of course, I may be wrong in this. I'm a very analytical person that has played computer and board games for a very long time. I always analyze my opponents even without thinking, so I may be overestimating the transparency of the AI.

[EternalSpark]

I actually think that how smart the AI is, isn't really very important, in terms of giving a challenge to good players.

++. it's a mistake, I feel, Brad is making.

See, you've got the people on forums like this, saying the AI is dumb because of X, or Y, or Z.

People on forums, like us? We're better, typically. We're not the casual gamer. Our demands for stricter AI hurts those who just want to play and have a little challenge (without it being so dumb that it can't do anything.)

[DaviddesJ]

Originally posted by Sirian

This reviewer wrote: "The computer AI is one of the best I�ve seen in a game, 4X or otherwise."

You know, I really wouldn't disagree with that as stated. Again, this is a comparative thing. However, it is not something I give two whits about. I'm interested in whether the AI provides me with a good experience.

To expand on this, I actually think that how smart the AI is, isn't really very important, in terms of giving a challenge to good players. What is much more important is how the game is designed. If the game is designed to strongly reward things that human players can do and the AI is not likely to be able to do, then the game is not likely to be challenging or satisfying for people who play it very carefully. While, if the game is designed to avoid playing into the traditional weaknesses of AIs, then it may not be all that hard to make an AI that puts up quite a good fight.

[Sirian]

Originally posted by yin26
I'd be interested to hear particularly from Sirian, Solver and Gufnork on how such a seasoned reviewer, who alledgedly took a while to put the reviewer together, draws such a different experience from you guys?

Reviewers play a lot of games for a short amount of time. I play a few games for a long amount of time. The reviewers have a broader experience and probably have to play and review a lot of medicre games and sucky games just as part of their job. I don't. So my thoughts do not reflect any sort of grading curve, where the best of what's out there gets praise just on that basis.

This reviewer wrote: "The computer AI is one of the best I’ve seen in a game, 4X or otherwise."

You know, I really wouldn't disagree with that as stated. Again, this is a comparative thing. However, it is not something I give two whits about. I'm interested in whether the AI provides me with a good experience. If it fails that test, I'll say so, and I may explain why.

Is the bar set lower? I'm not sure I'd characterize it in that way. Reviewers can aim for a different target audience, one not filled with top gaming talent.

Most GC2 customers and players simply won't discover the warts and flaws quickly, unless they come online and read from a few who have, chiefly because these same issues existed in other games, and are readily or even immediately identifiable to people who have extensive experience.

I got six months out of GC1, playing it several times per week. That is right up there in the top fifteen games of all time in terms of value, for me. GC2 is an even better value, I think, but that is in terms of standalone value. As a successor to GC1, it's not as different as I would have wanted.


- Sirian

[DaviddesJ]

Originally posted by yin26

I'd be interested to hear particularly from Sirian, Solver and Gufnork on how such a seasoned reviewer, who alledgedly took a while to put the reviewer together, draws such a different experience from you guys?

I don't think this review is particularly inconsistent with what Sirian or Solver have said about the game. The reviewer points to several good things about the game, which I think most people here would agree with. The difference is only one of emphasis: he doesn't mention (perhaps doesn't notice) the issues that we have. That's not surprising: you have to remember that most people who play these games really aren't very astute about it. As Sirian has said, there may well be things about the game that cause you to enjoy it less if you discover (or are told about) them. Conversely, if you don't look too deeply, those things won't bother you.

[Chronus9]

I was planning to attack a good Thalian planet. I intentionally put two fleets outside of it well in advance of the arrival of Transports. No reaction.

In GalCiv I, the AI would spot your buildup of ships in their sectors and warn you to basically "get out". Does this not happen in GalCiv II?

[drank]

But he gave me a problem because he started outfitting ships on the former Altarian planets with beam defense against my lasers. Smart AI (this is something I've seen in every single game from the AI, very good weapon/defense modification vs. what I'm doing).

This seems to be one of the AI's strong points. It does indeed do a good job of countering the human's ship designs. I think the problems with the AI lie in other areas. If it can't make a sound decision on when to go to war, or how to deploy its ships to fight an effective one, then having the right defense on its ships isn't really going to make it competitive with the human player.

To use your example with the Drengin invasion, one of the things I've learned in GC2 so far is that if the human player just fights, he can beat a seemingly tough AI force. I'll bet, for example, that the Drengin ships had movement 2 or 3 - and your ships were much faster. The attacker gets a big combat advantage, so you probably have enough speed to hit the Drengin, run to a planet to heal, and come back and hit them again, long before they actually prosecute their invasion. You'll also do a better job of levelling your units, so that they soon will have far more hitpoints than their Drengin counterparts.

In short, when the AI attacks, it's a paper tiger. It has a large quantity of ships, but unless it's got an overwhelming tech advantage, you'll probably beat it with quality ships and good tactics.