Edit:
Ok, so now I was able to check the multiplayer system and it actually works completely different to anything I have seen before and it makes my whole posting pretty redundant.
So what I will do instead is praise the Developers for their implementation and explain it to everyone who wonders:

GalCiv3 works like this:
TURN1
-Player A's Turn
-Player B's Turn
TURN 2
-Player A's Turn
-Player B's Turn

Now within the complete turn, that means including the other players turn you can do almost everything:
-You can set up the research
-You can build buildings
-You can rush buildings
-You can plan moves
-You can change taxes/distribution

The only things which are exclusive to your own turn are:
-You can execute moves
-You can construct Space Stations

Now when playber B finishs his Turn and with that finishs Turn 1, it transitions into Turn 2 and the progress is made. For example the building is now actually built.

This is a pretty great system because it completely circumvents all the issues I mentioned while not really needing to change much gameplay wise in comparison to the classical turn system.

--------------------------------------------------------

My Original Posting:

Hey there.

I want to reiterate on what I mentioned in my first feedback thread and check what other players think about this topic specifically and it would also be cool if a Dev could say something about the practicability.

In the past I played tons of Civ 5, mainly in Multiplayer with one or more friends. We played with and against AI, and the more we played the more it became apparent how broken the game is for multiplayer. Civ 5 is implementing AI Opponents not as competitive individuals but as “stuff existing to be conquered”. This can be seen best by taking a look at the “great art” system which was implemented with the latest expansion. The AI will trade it no matter what. It's purely implemented for the player to beat the AI. The mechanic is not used by the AI at all. And if you interact with the AI it doesn't care about what would be best for it. Now there are a couple of systems like that, together with big balancing issues which don't matter in singleplayer.
These are all things which could be “easily” fixed.
But – there is an underlying system which also breaks a lot of mechanics which can't really be “fixed” in that way, or can it?

How it is implemented

I am talking about the way Civilization 5 implements the turn system. And for those of you unaware of how Civ 5 does it, it's the same way as GalCiv2 and GalCiv3 currently implement it:
You move in real time. You select a ship, click on the destination, the ship moves. You have movement points left, you can move again, and the ship moves. If you did this with all your units you click end turn. Then it's the AI s turn and then it's yours again.

And this works very well for single player games.

But in multiplayer this causes several issues as some examples from Civ 5 Multiplayer show:

-Player A and B want to get to ruins. The player who clicks faster gets them.
-A unit discovers a city with his first movement point but gets in range, if the other player attacks him with his city fast enough, he will hit/kill the unit before the other player can use the second movement point.
-If a unit was badly hurt in one turn, you must be very fast in clicking to flee with it in the beginning of the next turn, if the other player is faster then it will get lost.
-There might be troops facing each other with a small gap in between. But no player wants to start moving because it will start a click-fest. If the opponent is watching the situation the unit might be dead before it can actually do something. A planned out attack might become impossible, more is depedent on who clicks when instead of actual tactics
-If you play with a turn-timer you might want to wait and plan a while and then quickly attack and move at the end of the turn, so that before the opponent can do everything the timer runs out on him

In order to “solve” the issue in Civ 5 Firaxis first erased all movement animations and later implemented a system which switches to true turn based whenever there is war. But this slows down gameplay immensely. So what my friend(s) and me did here is to make house-rules instead. Once there is war, one player always is first with managing troops and the other player moves when the first one said that he is ready. And from that point onwards the first player is no longer allowed to react. It is possible to play that way, but it's far from optimal and of course it only works with friends.

A Solution

But there are other games which showed how it could work.

Master of Orion 2 is one example. Armada 2526 also solved the issue. And if you play tactics games you might know “Frozen Synapse”, which is a different genre altogether, but it displays very well, how the system could work: These games have in common that they split the turns of the game into “Planning” and “Execution”.

What does this mean? It means that You tell your ship that it should fly from A to B. The display tells you “it will need two turns” (very much like it already is displayed in Galciv3) but instead of flying the first half right away, the ship does not move at all.
That is until you press “End turn”. As soon as >all players< pressed end turn, the commands are executed.

This solves all the issues other simultaneous systems have and which I listed before. But it brings some new problems. However these are mainly problems which can either be solved easily or which won't appear in GalCiv 3 in that amount as they do in Civ5.

-what if you want to take a specific route? Normally you use 2 movement points to go to A and then then the other 2 movement points to go to B, so that you don't fly through a certain field for example. Well, you could solve that like you normally control in RTS games. Set waypoints/ queue actions. Click on the ship, click on A, then shift-Click on B and it will take the path you intended.

-in Civ5 this would cause problems with exploration as the field of view can be very narrow. Meaning if you have double movement in forest, you might run into a forest without knowing what is “behind” the forest. So it might be possible that you can't access the tile and sent the unit there when planning. This isn't a big problem in GalCiv. As it is your decision how big the sensors on your ships are. As the space is empty for the most part, the movement and vision are much more free. Also you can move on every single tile if it's not occupied.

-what if you click on a field in the fog of war and end up at a tile with another player's unit? Well the game could either have fixed behaviour like “if not enemy, stop before tile. If enemy, attack”. Or it could offer you options which could be basic like “Never attack” or “always attack” or more sophisticated like “Only attack if target is weaker”.

-the most important question will be “what happens if you click on a destination intending to attack and the defender orders the unit to flee during the same turn? This isn't much of an issue if you only have “fix” spots like in Master of Orion, but it can be a big issue in Games with a free map like Civ or GalCiv. The big difference is that in MOO for example you can only order ships to retreat to a different planet. And then they are in transit for a couple of turns till they arrive. But in Civ or GalCiv you could just move by one tile every turn so that the opponent can't attack you.
There are a couple of ways to tackle it.
The system could “roll randomly” which unit moves first. That way both units have chances to do what they intend to do. Alternatively some value (maybe speed, maybe sensors) could give the ship some kind of “initiative”, so that a scout will move before a battleship does. And constructors and similar will probably move last.
The third way would be to also make attacks possible if the target is in a tile adjacent to the tile you initially targeted. A player could combine that with the waypoint system mentioned before.
And if any of these options is implemented it might still be possible to jump to flee a little bit, but it won't be a good strategy anymore to gain an advantage in actual war.

-what happens if two units move on the same anomaly? Well, if Unit A reaches it with it's second move and unit B reaches it with the third move, it will go to A. If both would reach with at the same time, either give it to both, none, make it random, or use the “initiative” mentioned before. Same goes for the case where two colonizers want to colonize a planet at the same time or two invaders want to invade the planet at the same time. There is always a factor which you can take first (e.g. population, unit strength and so on) and if that isn't enough, the final answer is to randomize it but this will rarely happen in any game anyway.

-and of course the AI needs to be adjusted to be handle the system which might not be a lightweight

Additional Advantages:

Not enough to have a combat with more fair and fluent combat? A system which doesn't need to become “true” turn based once to combat starts?

Well, there are other advantages: A true turn based system would completely kill the game if you have a couple of players. So for these games you would need to stick to the “broken” simultaneous system for combat. Independent on how many players you have a system with planning and execution phases will always work better and faster.

In addition it would be a big advantage if a PitBoss mode is implemented.

What is PitBoss? Again if you played Frozen Synapse you already know it. Or maybe you played some old RPGs/Stragey Games by snail mail in the 80s and 90s where one guy did all the calculations and then sent a mail back to all the participants with the next situation.

PitBoss Mode means that there is a server which calculates what happens. So the Server says that it's turn 1. And now everyone can log in and do their turn. Once your turn is sent to the server it will save it. When all players entered their moves the result is calculated. The Server will announce the next turn (and might sent notification mails) and every player can log in again.

This system won't work well on the “normal” simultaneous turns. This might already fail if the network structure isn't working well (in simultaneous turns every command needs to be sent instead of the final package) and can also fail gameplay-wise as with simultaneous turns the player who moves last profits, as he can react on what the other players do.

This system works well on “True turn-based mode” where everyone makes his turn one after another. But it can become extremely slow. Not just because they can't all send their input during the same 30 Minutes but because if you Play with 10 Players one round (all players made their turn) can take weeks. Player 1 might not have time on Monday to send his turn and sends it Tuesday Evening. Player 2 doesn't have time in the evening and will need to send it at Wednesday Morning, Player 3 doesn't have time on Wednesday and sends his turn on Thursday and so on.

With a Plan and Execute system this example would be like: All players send in their turn on Monday except Player 1. He sends in his turn Tuesday evening and was the last player. Next Round starts. The slowest player dictates the pacing. The slowness of the players would not add up.

Final Words:

I know that this system is only needed for multiplayer and is not needed at all for singleplayer.

I'd like to know what other players think about it. I know that there are lots of players who would love to play these games in multiplayer if the system would actually really “support” it.

But I also know that Stardock has some own experiences with multiplayer showing that hardly anyone used their multiplayer features (and that's why it was cut from Fallen Enchantress). But if you asked me that was partially a self fulfilling prophecy. If you don't implement multiplayer at release and / or don't implement it in a great way, it will probably fail. Multiplayer works partially exponenstially. If there are hardly players online, it will take ages to find players, which will lead to less players playing multiplayer, leading to even longer waiting times until nobody plays multiplayer but friends in private games. But SC2 also shows that if you implement multiplayer in the right way, it can carry a game. True, it's an RTS. But it has it's reason why SC1 was so popular in Korea. And that it's partially because it focused on providing a great multiplayer experience instead of limiting itself to singleplayer. In case of 4X Games however we don't have such a game yet. There is not a single game pushing multiplayer. Paradox acknowledged the potential of Multiplayer when doing Europa Universalis 3 from what I know. But EU3 isn't exactly a 4X. And it's also completely real time.

So what do other players think about it? Would you give MP a chance if it had a good implementation? You got any experiences to share about different implementations? And ideas about how to implement it?

And what do the devs think? Would it be feasible to offer a “plan and execute” system for multiplayer? Or would it take too many resources? Do you have anything else in mind which could solve the issues occurring in Civ 5 for example? What is your private experience with multiplayer? Every tried to play a game like Civ competitively?