1upt No So Cool

My number one complaint about 1UPT... Drumroll please... When you are defending a city and you have a unit (usually a ranged unit) garrisoned in the city... You CANNOT send another unit through the city to attack an enemy in an adjacent tile.

I just had a war that was waged over a city I built on a narrow isle so that I would be able to send naval units through and have a more direct route through the continent. So anyways, the AI had 2 land tiles on his side of the city. I had 1 land tile on my side with a catapult garrisoned in the city.

Well, he would line up his units on both tiles on his side. I had all of my knights and longswordsman eagerly awaiting to attack his units but they couldn't because I had a catapult garrisoned in my city and no other land route around to the other side!

Soooo frustrating.

Another frustrating thing is I've had a frigate and a cannon both in a city at once, but the unit icons where stacked ontop of each other. When I clicked the icons, only the cannon icon would come up because that was the one that was ontop. Nothing I did could access the frigate and I couldn't just move the cannon because my city was under seige and I need to bombard the enemy units! This REALLY frustrated me as I had no way to access the unit.

IIRC, in every previous civ you could access all of the units garrisoned in there by simply going to the city view screen. But what do I know?

I disliked 1UPT at first, but as I'm getting to know how to play the game it seems to growing on me. I'm in a game now that I have had to figure out how to attack through bottlenecks. So far I have had to get to 3 different enemy areas via bottlenecks, yes it took a while but at least I had to think.

I will have to start taking notes because I think this is happening and may have been said already, but it appears that:

When I send a new unit into an area in my war zone that is a ten turn journey and then subsequently move another unit into that space the journeying unit stops dead in its tracks.

I think that is what happens and it probably has been said in a more eloquent way by someone else or I was just imagining it.

Centerfinger said:

I disliked 1UPT at first, but as I'm getting to know how to play the game it seems to growing on me. I'm in a game now that I have had to figure out how to attack through bottlenecks. So far I have had to get to 3 different enemy areas via bottlenecks, yes it took a while but at least I had to think.

Click to expand...

Spaaaaaaaaaaartans! What is your profession?

It definitely adds another variable to the equation when choosing city placement, even if it is a minor variable.

mmmfloorpie said:

My number one complaint about 1UPT... Drumroll please... When you are defending a city and you have a unit (usually a ranged unit) garrisoned in the city... You CANNOT send another unit through the city to attack an enemy in an adjacent tile.

I just had a war that was waged over a city I built on a narrow isle so that I would be able to send naval units through and have a more direct route through the continent. So anyways, the AI had 2 land tiles on his side of the city. I had 1 land tile on my side with a catapult garrisoned in the city.

Well, he would line up his units on both tiles on his side. I had all of my knights and longswordsman eagerly awaiting to attack his units but they couldn't because I had a catapult garrisoned in my city and no other land route around to the other side!

Soooo frustrating.

Another frustrating thing is I've had a frigate and a cannon both in a city at once, but the unit icons where stacked ontop of each other. When I clicked the icons, only the cannon icon would come up because that was the one that was ontop. Nothing I did could access the frigate and I couldn't just move the cannon because my city was under seige and I need to bombard the enemy units! This REALLY frustrated me as I had no way to access the unit.

IIRC, in every previous civ you could access all of the units garrisoned in there by simply going to the city view screen. But what do I know?

Click to expand...

if you go to the top left of the screen you can select a unit list - that should allow you to select your frigate

Post Monkeh said:

if you go to the top left of the screen you can select a unit list - that should allow you to select your frigate

Click to expand...

I went into the military advisor panel and I tried everything possible to get the camera to centre on that specific unit/activate it but I couldn't get it to work.

mmmfloorpie said:

I went into the military advisor panel and I tried everything possible to get the camera to centre on that specific unit/activate it but I couldn't get it to work.

Click to expand...

when i go into the unit list in the top left and click a unit it immediately selects the unit, guess the bug's stopping that from happening. the only thing i could suggest is if you have another catapult, bring it up behind the city on a road and use it to bombard while you remove the one from the city to get at your frigate, either that or rush a catapult in the city, which should force you to remove the units already there.

mmmfloorpie said:

I went into the military advisor panel and I tried everything possible to get the camera to centre on that specific unit/activate it but I couldn't get it to work.

Click to expand...

Once you've selected a unit from the military drop down list, click on the portrait of the unit at the bottom left of the screen and it will center on the unit. Same way it worked in previous civ games.

mmmfloorpie said:

I went into the military advisor panel and I tried everything possible to get the camera to centre on that specific unit/activate it but I couldn't get it to work.

Click to expand...

Or just press c. That has been since number 1 too...

Jharii said:

<snipped>

There are many more complexities to pathfinding. One additional one is that if the target tile is occupied, clear the existing path and activate the unit. Another is if the path is blocked, and the target tile can no longer be reached, clear the existing path and activate the unit.

Click to expand...

That's a minor annoyance. If you are moving an army to a rally point, then once a unit reaches the rally point, other units will stop in their tracks, which may be across the map. They should continue until they can get no closer.

Another problem is with the use of roads. Sometimes it makes sense to wait a turn for the road to clear, in other words, to move as a column along the road. There's no easy solution to this without some user input. A more advanced mechanic would have them form a column, then reassemble into battle formation at the rally point. But that's probably beyond what can be reasonably expected.

All-in-all, though, 1upt is one of the best new features of the game.

Zachriel said:

That's a minor annoyance. If you are moving an army to a rally point, then once a unit reaches the rally point, other units will stop in their tracks, which may be across the map. They should continue until they can get no closer.

Click to expand...

Ah, yes! I completely agree with this. They need some type of grouping or follow feature.

Agree with the last 2 posts, a grouping feature and or follow feature would be keen.

I suspect a grouping feature would actually help out the AI as well, because presumably, the group, or army moves at the speed of the slowest unit, so you won't have melee units unsupported moving up.

My biggest complaint with 1upt is the micromanagement of this aspect. I don't want to have to micromanage moving my units across the map. I'd much rather group them and set a gather point the way an RTS would.

My biggest complaint, like someone else pointed out, is the fact there's no option to toggle something that allows me to see planned unit movements. In this game, it would be useful to have an option showing ALL the planned movements of units since it greatly helps with planning.

YoHo said:

Agree with the last 2 posts, a grouping feature and or follow feature would be keen.

I suspect a grouping feature would actually help out the AI as well, because presumably, the group, or army moves at the speed of the slowest unit, so you won't have melee units unsupported moving up.

My biggest complaint with 1upt is the micromanagement of this aspect. I don't want to have to micromanage moving my units across the map. I'd much rather group them and set a gather point the way an RTS would.

Click to expand...

It's difficult to do such a thing. because the map is too big in scale. So mountains, forestes and hills will hinder the formation 'cause the movement restriction the units have... If it was a map on scale like Panzer General, were plains are much bigger and occuppy large portion of the map, the forest and hill tiles are grouped , because they represent one entity, it could be more doable....

you could easily see where your units were heading in civ4 by clicking on them, i haven't figured out any sense in them removing it. i'd also add my vote to the "don't cancel route until it's physically impossible to move any further" camp, it's really annoying to have to re-issue 10 orders every time an enemy worker moves into my target point.

JLoZeppeli said:

It's difficult to do such a thing. because the map is too big in scale. So mountains, forestes and hills will hinder the formation 'cause the movement restriction the units have... If it was a map on scale like Panzer General, were plains are much bigger and occuppy large portion of the map, the forest and hill tiles are grouped , because they represent one entity, it could be more doable....

Click to expand...

I don't see why this is a problem. They don't need to maintain exact formation. Just all get to a location roughly as a group. I'd be fine if it was done single file and then they reformed after arrival.
There's always a compromise, no matter what the scale between convenience and micromanagement. If I let the computer do it, then I can't expect absolute ultimate precision, but I can expect all the units to arrive at the same time.

In this case, you wouldn't group your units and click on the city you wanted to attack. You'd have to choose an appropriate gathering point.

I was quite excited/enthousiastic when I heard about the new Hex+1upt system. After playing the game a bit, I feel the implementation is just...wrong. A tactical warfare system on a strategic level map? To be honest, I find it somewhat ******ed, as I can never really experience the tactical / operational feel of the war.

If I had any talent in modding, I think I would try to give the warfare system a more strategic feel. Something like this perhaps :
- No more 2 tiles movement for land units except for really fast moving ones (cavalry/motorised/tanks, etc);
-Allow stacking at some point, army-group style : One/two infantry + one range or support (+one civilan maybe);
-Get rid of the utterly ******ed 2-3 tiles range attack; thus, range attack only to adjacent tile. Sorry, but I can't get over the fact that my bowmen or canons can shoot over the alps or adriatic sea.

These are some thoughts / impressions I had after playing. I didn't spend a lot of time thinking them through though, since I somewhat stopped playing and caring 3-4 days after release, out of total boredom

YoHo said:

I don't see why this is a problem. They don't need to maintain exact formation. Just all get to a location roughly as a group. I'd be fine if it was done single file and then they reformed after arrival.
There's always a compromise, no matter what the scale between convenience and micromanagement. If I let the computer do it, then I can't expect absolute ultimate precision, but I can expect all the units to arrive at the same time.

In this case, you wouldn't group your units and click on the city you wanted to attack. You'd have to choose an appropriate gathering point.

Click to expand...

Yeah, it wouldn't be too difficult. The hexes are more than likely stored in an array. Simple enough to navigate through. I've done some hex-based tile coding, and it works no differently than square-based with a very simple formula to convert the coordinates.

Jharii said:

I disagree. 1upt is outstanding. And it's a game. By nature, it's "unrealistic" out of the box. Sim games may give you a better feeling of "realism" that you may be looking for. Actual life does it for me, though.

Click to expand...

So you'd just as happily play a game about apple picking if it had equivalent gameplay to Civ5? I wouldn't. In real life I can't live for thousands of years or direct the actions of millions of people. Rather than trying to impose my own narcissistic or sometimes destructive tendencies on the rest of the world, I find it more constructive to play games.

Jharii said:

And you cannot blame an inferior AI on a game mechanic. AI's are supposed to be designed to take advantage of the game mechanics, not the other way around. Your argument goes in the wrong direction.

Click to expand...

Game mechanics can make or break an AI. 1UPT is akin to insisting that an aircraft that is designed to set a speed record also be capable of carrying 100 passengers. Theoretically it may be possible, but it makes the engineering challenge much more difficult. Your assertion is incorrect in that either approach (AI first or game mechanic first) are valid. But given that AI is typically so much more difficult to design than a game mechanic, most designs that do both well keep the AI in mind when designing game mechanics. Some mechanics are simply easier for the CPU to handle than others.

Jolly Rogerer said:

Game mechanics can make or break an AI. 1UPT is akin to insisting that an aircraft that is designed to set a speed record also be capable of carrying 100 passengers. Theoretically it may be possible, but it makes the engineering challenge much more difficult. Your assertion is incorrect in that either approach (AI first or game mechanic first) are valid. But given that AI is typically so much more difficult to design than a game mechanic, most designs that do both well keep the AI in mind when designing game mechanics. Some mechanics are simply easier for the CPU to handle than others.

Click to expand...

Sorry, I respectfully disagree.

Game mechanics can break a game, not the AI. Games exist without AI. People play them all the time in many forms. If Monopoly had broken mechanics, my intelligence would not be broken. I simply could not play a broken game.

The same concept holds true for computer games and AI. When I am working on computer games, I work on the game mechanics first. I document the core mechanics and rules. When it's coded, I make sure that I can play the game myself and that the mechanics are sound and everything behaves as it should. Then I program the opponent.

Do I keep the AI in mind when designing the mechanics? Hmmmmmmm. I have passing thoughts and concepts on how I would handle the mechanic. Maybe jot a note down or two. But most of those are changed during design time because I typically find a better way to do it. But no, it's never in the forefront because the mechanics are a creative and design process typically done well before any coding is done. AI is the furthest thing from my mind when putting pen to paper.

Here. Check out this thread. There are some discussions of a rather complex AI engine with a very simple premise going on, in general terms. You may or may not find it interesting. But it will give you an understanding of how a core AI can be developed, almost as an independant engine, and then plugged into any system.

Jharii said:

Sorry, I respectfully disagree.

Game mechanics can break a game, not the AI. Games exist without AI. People play them all the time in many forms. If Monopoly had broken mechanics, my intelligence would not be broken. I simply could not play a broken game.

Click to expand...

I said game mechanics can break an AI, by which I meant that if I tried to make a game mechanic that is too complex for current computers / software technology the AI would not be able to provide a challenge for the player.

Jharii said:

The same concept holds true for computer games and AI. When I am working on computer games, I work on the game mechanics first. I document the core mechanics and rules. When it's coded, I make sure that I can play the game myself and that the mechanics are sound and everything behaves as it should. Then I program the opponent.

Do I keep the AI in mind when designing the mechanics? Hmmmmmmm. I have passing thoughts and concepts on how I would handle the mechanic. Maybe jot a note down or two. But most of those are changed during design time because I typically find a better way to do it. But no, it's never in the forefront because the mechanics are a creative and design process typically done well before any coding is done. AI is the furthest thing from my mind when putting pen to paper.

Click to expand...

I've never designed a computer game, though I have designed a lot of PnP and board games. I thought of mechanics first too, because all I had to worry about was whether intelligent humans could grasp and would enjoy the mechanics. There are typically a lot of mechanics available for any particular subsystem in the game, it's not like everything or even most things I did were novel to gaming. Were I to develop a computer game you can be sure that the capabilities and limitations of the computer would have a massive impact on which mechanics I employed. I would be more inclined to include more complicated mathematical expressions for instance, as computers have a lot more patience for them than humans do. And I would try and optimize for the AI to the extent possible while still making a game that humans find interesting. Too many games currently fail here, in part because AI is an afterthought rather than an important part of the initial design process.

Jharii said:

Here. Check out this thread. There are some discussions of a rather complex AI engine with a very simple premise going on, in general terms. You may or may not find it interesting. But it will give you an understanding of how a core AI can be developed, almost as an independent engine, and then plugged into any system.

Click to expand...

I have read that thread. Are you talking about memetic AI? It's an interesting concept, basically a learning AI that has some more interesting social applications as well. With a sufficient database a lot of number crunching could be eliminated in favor of these memes, which seem to be more or less dynamically derived heuristics. Developing these memes would require a capacity to initially number crunch a problem, or a lot of analysis of player behaviors that could be mimicked, or a lot of brute force experimentation by the AI. Or all of the above. It would be simpler of course when the game systems are themselves simpler.

Anyway, thanks for the interesting discussion.