Longturn.net

maho

Player

Posts: 161

Re: LT31 will start soon

Maho, your suggestion of letting players just submit orders and not have anything actually happen ingame until TC would turn freeciv into a PBM which indeed eliminates all RTS issues.  It would probably require quite some coding, UI design and server design work to make even a rudimentary implementation of that a reality

Not really. The only change should be: at beginning of every turn (but after automatic/scheduled moves), iterate through units and set remaining movepoints to 0.

so simple.

BTW: what is PBM?

Regarding declaring people who move after 'Turn Done' is pressed being RTS-ers I have a better solution: simply block ALL orders from a player who has pressed 'Turn Done' until TC!

Then nobody will press TC, like now.

ifaesfu

Player

From: Huelva - Spain

Posts: 93

Website

Re: LT31 will start soon

I think we must be sure if we want RTS or not. Any kind of loggers, autoattackers and so on or not. Then take measures.
I'm not for RTS and modified clients to get advantage over the standard freeciv client. The only measure it seems to work and it's easy (I think) to implement is the time limit. At least, it minimizes the effects.

Also, would it be possible to be online 24/7 but only to be able to move units within the time limit?

The only thing I found really obviously unacceptable and which I simply cannot find a place for in the LT games was hacking players accounts and thrashing them.

I think that actively breaking a rule isn't very far from that. One was banned, but many other not.

Modeemirotta

Player

Posts: 14

Re: LT31 will start soon

Also, would it be possible to be online 24/7 but only to be able to move units within the time limit?

This would be a good choice. If it was doable? You could use time planning and chatting and not do actual moves. After all the part that takes the most time is planning?

IllvilJa

Player

Posts: 51

Re: LT31 will start soon

Not really. The only change should be: at beginning of every turn (but after automatic/scheduled moves), iterate through units and set remaining movepoints to 0.

so simple.

...but IMPRESSIVELY effective vs logging/RTSing/TC-gaming given the small change in the server.

I can think of a host of, well, let's call it CHALLENGES that the players have to cope with when this server modification is in effect (e.g getting units to coordinate their behavior, especially when belonging to multiple players, moving into unknown territory with ships or not to mention managing FIGHTERS), but still... it would make a LOT of issues simply GO AWAY!

And all the problems that happens to surface due to this modification will affect all players alike, so the modification would in that sense be fair!

YES, I'm quite enthusiastic about this suggestion!  Definitive one I'll look into implementing on a freeciv server!

BTW: what is PBM?

PBM = "Play By Mail".  Sometimes called PBeM = "Play by Email".

In essence, in these days it's more by "Play by Web" or some call it just turn-based play.  Main concept is that a report or "Current situation" is provided to all players which in practice means that the world is totally static, NOTHING changes in it while the players analyze the situation, contact each other for diplomacy etc and craft up their orders.  Then, at the update (TC that is), the PBM server will take all player orders that has been submitted (usually there is a deadline for submitting the order before the update takes place) and processes them, running an update of the world (stuff moves, is produced, combat takes place etc etc) and during that update the players are COMPLETELY locked out, they cannot interact with the update.  Once finished, the world have progressed to the next turn, and a new set of reports are provided to players.  A long time dream (or rather fantasy) of mine has been to create a freeciv-ish/empire-ish/xconq-ish PBM game with infantry, cities, submarines, nukes and the lot......

With your fix, the movement stuff would be very PBMish, whilst diplomacy, city management, research management is more RTS-ish.

And there will be no need for the exercise of identifying and kicking/banning RTSers, because from a movement+combat perspective, there won't be any around!

Regarding declaring people who move after 'Turn Done' is pressed being RTS-ers I have a better solution: simply block ALL orders from a player who has pressed 'Turn Done' until TC!

Then nobody will press TC, like now.

Marduk's suggestion was to have an informal rule that required players to press TC once they had done their moves or else risk being identified as RTSers and potentially banned.

(ifaesfu and Modeemirotta, I saw your two last posts.  Got sorta kinda a reply ready in my head but don't have time to write it... Real Life (TM) demands attention for now...)

elrik

Player

Posts: 108

Re: LT31 will start soon

Heh... I can't even remeber when You tell about such modification for the first time Maho:) Maybe this time it can really happen. I think it is a GREAT idea:) But i am affraid that this simple change can be too big to cope for some players:) I really we will at least try a test game like that. That brings so many tactical possibilities:) The only weak in some ways point is FIFO que on the server. Sometimes setting orders just after TC can decide about everything

IllvilJa

Player

Posts: 51

Re: LT31 will start soon

Heh... I can't even remeber when You tell about such modification for the first time Maho:) Maybe this time it can really happen. I think it is a GREAT idea:) But i am affraid that this simple change can be too big to cope for some players:) I really we will at least try a test game like that. That brings so many tactical possibilities:) The only weak in some ways point is FIFO que on the server. Sometimes setting orders just after TC can decide about everything

Well, the FIFO queue of pending orders can be addressed like this: before the orders are executed, randomly rearrange their sequence.  That way you don't know if it is YOU or your OPPONENT who manage to move a settler to that juicy city spot B first, no matter when or in what sequence you two enter your orders for that move :-).

Of course, in that situation, the player hitting 'B' first will still be the first to build a city regardless of where his settler has ended up (and assuming those two settlers end up in tiles next to each other, building that city effectively prevents the other player to build a city at HIS settler's location), so all RTS-ish artifacts of the game won't go away.  (But hopefully, enough of them will so the advantage gained by RTS-ish and TC-focused play is small enough for safely eliminating all informal rules intended to reduce RTS-ish play and play around TC.)

So, no matter in what order you provide your move+attack commands, you don't know in what order they will be applied, and you don't know how many enemy commands they will be interleaved with.

Ofc, this might prove challenging regarding multiple-action units like howitzers on railroad who have enough movement points for attacking multiple times in a turn or for fighters who need to both attack as well as return to base in the same turn, but this can either a.) be solved by providing a way to specify multiple commands in sequence for a unit or b.) accept the tricky situation as it will be fair in the sense it's equally tricky for all players.  Sure, fighters will be rendered useless (unless theres a rule change allowing them to stay up during 1 TC like bombers) but that's artifacts to be pondered as the game is further developed.

Another interesting thing to ponder: how do we treat the paradrop command?  Will that be a delayed command which happen in a random order or will it be an RTS-ish command?  Not to mention NUKES and their (potentially accidental while en route to target) detonation!

Not to mention how tricky it will be to pursuit an enemy unit who's trying to escape (or land aircraft on a carrier which is moving... you better keep that carrier static ;-) ).

Of course this scheme has places where it may be 'exploited' or 'abused' or would 'not work' but when it comes to those three phrases, just do a search and replace like this: replace 'exploit' and 'abuse' with 'use that hurts me because it is different from my personal preferences and I insist on sticking to my personal preferences' and replace "don't work" with "works in a way I did not bother or manage to anticpate and which hurts me because someone else was innovative enough to figure out that way".  (Actually, I find that substitution scheme useful quite often elsewhere when discussing game designs... ;-) )

Questions, questions, questions.... and hypothesises, hypothesises and hypothesises...

So many things to play around with... and SO many nasty PITFALLS!  Can it get any better?

Time for some serious experimentation!!

(And ooh, how badly would not this change to the game defeat schemes like using en-masse  scripted attacks...? Just asking...)

elrik

Player

Posts: 108

Re: LT31 will start soon

With random mix of order que is one big problem... - imagine attack 4 catas followed with 3 def units to cover that. Now mix it and take a look at results:D

Maybe we can mix only player in que? one turn you have luck and are first, another turn the last.

IllvilJa

Player

Posts: 51

Re: LT31 will start soon

With random mix of order que is one big problem... - imagine attack 4 catas followed with 3 def units to cover that. Now mix it and take a look at results:D

Everyone suffers from this problem, that defending units and attacking units won't be guaranteed to act in a coordinated way.  Only way to cope reliably with this is to outnumber your opponent so chances increases that the right kind of units that you have gets their commands executed.  That is, in the example above, double the catas and def units: that would increase the chance that at least SOME catas and def units reach the destination before enemies executes attacks against it.

But yes, it would be HARD to control what happens in detail.  Only way to cope is to get the economy to produce units that outnumber the enemy and tilt the odds to your advantage in the long run.

Maybe we can mix only player in que? one turn you have luck and are first, another turn the last.

Possible way to do it, only catch is that the player that get's the ability to move first with ALL his commands will get an almost insane advantage.   Maybe a better way to do it is to keep the sequence of every players orders intact (so your 4 catas really GETS moved before the 3 def units that cover it) but randomize the way different player commands gets interleaved with each other (some enemy moves might happen before even the first cata moves as well as between each of the remaining 6 units of yours).  Of course, moves between allies probably have to be randomly interleaved as well, it would be impossible I think to ensure that a bunch of cooperating players gets their orders entered in a certain sequence and also, it goes against the anti-RTS idea.

Personally, I'm inclined to implement a fully randomized (not even maintain the relative sequence of your own orders) variant of this, if for nothing else it is 1.) easier to implement and 2.) easier to grasp for the players.

monamipierrot

Player

Posts: 171

Re: LT31 will start soon

But yes, it would be HARD to control what happens in detail.  Only way to cope is to get the economy to produce units that outnumber the enemy and tilt the odds to your advantage in the long run.

This would be the death of military tactics, if not of military strategy: just focus on economy and let the brownian motion of units decide a whole war!
Forget about those surgical navy operations you always dreamed of. Just prepare for the D-Day and convert a whole continent in a giant bloody Omaha Beach.
You will leave your thousands of rambo boys alone in the action and become a bureaucrat without much to think about.
Maybe not bad in last parts of the game, where moving units is a hell. But not definitely my game style.

Possible way to do it, only catch is that the player that get's the ability to move first with ALL his commands will get an almost insane advantage.   Maybe a better way to do it is to keep the sequence of every players orders intact (so your 4 catas really GETS moved before the 3 def units that cover it) but randomize the way different player commands gets interleaved with each other

How to randomize? Player A moved 100 units, Player B just 2, how do you manage this? A rational conclusion would say: 1st B unit moves during, say, the 1st half of units of A, and/or most probably around #25. 2nd unit would move during the second half of A units, and/or most probably around #75. But how do you know player A didn't move dozens of useless units at the end of his turn just to push down the first units in the priority list to be sure to beat B in time? You can't.
Then, there should be a simple rigid scheme: subturns. 1st subturn: movent for ALL 1st units of ALL players, in a randomize sequence. 2nd subturn: ALL the 2nd units, in a newly randomize sequence, then the 3rd ones and so on. Biggest players with many units to move (or attacking players) would be heavily disadvantaged by this, and forced to a static or slow warfare (not even mentioning that we are speaking about 1-move-per-unit vanilla-Freeciv-style, because it would be too frustrating and useless to move units in the unknown).
I don't see any other option (except the complete randomization, i.e. brownian motion in Freeciv)

In any case, it would change completely the game philosophy, or at least the warfare philosophy of LT. I don't know if it would be good or bad. Simply too different from what we know.

And then there's the technical problems. I'm not a programmer and althou I respect all of you who have this skill, I simply don't believe it can be easy to implement such a huge change. The main problem could be the UI. I can't even imagine how would it be. If you're sure it can be done... well I may try believe you.
But if all of this is just to get rid of RTS... please think some other thing (e.g. reduce MP multiplier).

Did anybody think about a "freezing" button? You can push it to ask the server to "freeze" one or more selected player(s). If he doesn't answer inmediately to stop the slow freezing process (which takes say 60"), he will be frozen for a max of 10 minutes.
When frozen, he can do everything except moving units, any unit. You can then safely perform your attack, but after 120" frozen, he can already push the "defrost" button (which takes another 60") which will hurry you to finish your attacks. So you have a minimum of 3' to perform it.
You can freeze the same player only once per turn. Freezing can't be performed around TC (1 hour before/after), or when the other player is freezing a 3rd one. You can freeze as many players as you want at once, but you can't freeze them consecutively (you have to wait 5 minutes between freezing sessions). This will allow enemies to freeze you instead.
Same for the frozen player: enemies have to wait 5 minutes between each freezing performance against the same player, so he can start freezing them instead.
You can only freeze players with which you have contact. You can't freeze allies. Frozen/freezing players will not be shown to 3rd players. And there should be also some cost for discouraging freezing, but I can't think which type.

For cheaters there could be many workarounds (or ways to take advantage) of the freezing concept, but the limitations I thought of (and wrote above) should  take care of most of them.
I think it could be easy to implement it.
Maybe we can experiment it with a "soft" version: the frozen guy movements are not blocked, there's just a banner who "invites" him not to move units, or that asks for confirmation before performing them.

An easier option could be the immunization. You can ask once per turn for a immunization (10 minutes) during which you are immune from enemy attacks, and can perform your attacks with ease. I would not let more than 10 minutes cause some well coordinated guys could immunize consicutively and "defend" cities  with stacked international units. To limit this, the immunization could also be preceded by a "infection": a randomly-long period (from 0 to 50% of immunization period) in which you can still be attacked, giving a small window to perform attacks against multiple-defended cities or stacks.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: LT31 will start soon

This discussion is all nice and good, and hopefully patches will be ready just in time for LT40

Any rule we want to apply to LT31? If we want to use the "don't be logged in a long time before pressing turn done, don't move aggressively after pressing turn done" rule then we should decide it now. We can trust on players to spy on each other and rat their enemies out to the admins in order to enforce the rule

I think we must be sure if we want RTS or not. Any kind of loggers, autoattackers and so on or not. Then take measures.

This is already clear in the rules, though we should formulate it more clearly: we don't want RTS, autoattack is a serious offense. Logging is allowed.

From the LT rules:

It is common courtesy to log out and let your enemy do his moves in peace, after you have done yours.
Many players have local client modifications (patches). This is acceptable, as long as automatic attack, moves or similar functionality is not implemented.

Modeemirotta

Player

Posts: 14

Re: LT31 will start soon

How does logging work if it is common courtesy to log out? I don't see how you could have something logging stuff from server with out keeping the client open? Don't see any contradictions here?

I don't think the idea in logging is to record stuff happening during the hour you do your moves, but to see stuff that happend when you were afk.

I my self don't like logging / don't keep the client open 24/7 / but like to spend time browsing my cities and the map (taking 2-3 hours of spare time)
(just to show my opinion on these things...)

wieder

Administrator

Posts: 1,863

Re: LT31 will start soon

@Marduk

It's very understandable that people want to do their moves without any RTS.

Is it reasonable if I keep the client open AND tell beforehand if I'm going to make moves? Would you consider 10 minutes ok?

The only patch/whatever I would like to see in the next game is the random time for TC. Someone once told us that it would be very easy to implement. I don't know about that but if it is, that would help us avoiding the TC RTS. That seemed to be a big issue in LT30.

And stuff...

Davide

Player

Posts: 1

Re: LT31 will start soon

I'm working on an interesting fork of Longturn. (info: http://freeciv-mundi.org/forum/viewtopic.php?t=233)
I may work on the patch for random TC if it is a fast task. Just... please tell me what TC means. Thanks.

Modeemirotta

Player

Posts: 14

Re: LT31 will start soon

I don't think the idea in logging is to record stuff happening during the hour you do your moves, but to see stuff that happend when you were afk.

Okay, found out that it is also used to see what a city has in stack during an attack. (logs the defender if you ain't fast enough to see it)

Kryon

Administrator

Posts: 370

Re: LT31 will start soon

Just... please tell me what TC means.

TC = Turn Change

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: LT31 will start soon

How does logging work if it is common courtesy to log out? I don't see how you could have something logging stuff from server with out keeping the client open? Don't see any contradictions here?

I don't think the idea in logging is to record stuff happening during the hour you do your moves, but to see stuff that happend when you were afk.

I my self don't like logging / don't keep the client open 24/7 / but like to spend time browsing my cities and the map (taking 2-3 hours of spare time)
(just to show my opinion on these things...)

If I understand it correctly there's no way to easily detect if players are logging or not, so banning it doesn't seem useful. But as soon as other measures against RTS work, logging will be almost useless and banning it wont be necessary.

I also prefer anti-RTS measures that still make it possible for people to be logged in without doing moves, for example to chat. Most measures proposed in this thread are actually compatible with this.

Kryon

Administrator

Posts: 370

Re: LT31 will start soon

I think the best solution would be to impose a per-turn time limit for moving units. Logging in should still be allowed to chat and to manage cities. For example a 2 hour time limit should be enough to do all the moves. We can do this in 2 ways:

1) The 2 hr countdown starts as soon as a player logs in after each turn change and stops when he logs out.
2) The 2 hr starts when the player hits a button (we might call it "Move Units" button) and stops when he hits the button again. That way, the player can make plans (e.g., chat or city management) before moving units but he is still limited to 2 hr of moving units.

We can have a vote to decide on the Move Unit time. (e.g. 1 hr instead of 2).

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: LT31 will start soon

I'm working on an interesting fork of Longturn. (info: http://freeciv-mundi.org/forum/viewtopic.php?t=233)
I may work on the patch for random TC if it is a fast task. Just... please tell me what TC means. Thanks.

Hi Davide, welcome to the forum!

Always good to see other people developing new ideas. I also spotted your 3D replay on youtube (here), very cool!

We're always open for new ideas, including radical ideas (for example we've been toying around with an idea for dynamic maps). Problem is that we're understaffed so it's hard to implement new features. Would be cool if we can combine our ideas and energy.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: LT31 will start soon

I think the best solution would be to impose a per-turn time limit for moving units. Logging in should still be allowed to chat and to manage cities. For example a 2 hour time limit should be enough to do all the moves. We can do this in 2 ways:

1) The 2 hr countdown starts as soon as a player logs in after each turn change and stops when he logs out.
2) The 2 hr starts when the player hits a button (we might call it "Move Units" button) and stops when he hits the button again. That way, the player can make plans (e.g., chat or city management) before moving units but he is still limited to 2 hr of moving units.

We can have a vote to decide on the Move Unit time. (e.g. 1 hr instead of 2).

Sounds good, but we'll need someone to develop such a patch and make sure it doesn't cause side-effects. If the patch is there I'm all for trying it out.

Marduk

Administrator

From: Rotterdam, Netherlands

Posts: 151

Re: LT31 will start soon

Random TC: we've already used this in several previous LT games. It helps to prevent "double-turn attacks" (move your units just before TC and again just after TC, so you can make two moves before the enemy has time to respond; entire nations can be destroyed in minutes). But to use this to address RTS would require that TC becomes very random (e.g. plus or minus 6 hours or more), otherwise people can still RTS except for an hour before TC. Very random TC can cause bonafide players to miss turns because they cannot log in at any time of the day.

Unit-waiting time was developed to replace random TC, but only as a solution to double-turn attacking.

sokrat

Player

Posts: 24

Re: LT31 will start soon

About that interesting discussion about RTS and PBM... and issues that will appear with moves and orders only realised at TC :
If i remember well, i suggested many years ago that combats should be resolved at TC, all units on a tile against all units on the other tile.
I think that would be more realistic, and would bring a little more strategy, with real battles, and not just only I-send-tons-of-howitzer-to-the-city before the defender sees it.
And that would also let units runaway if they don't feel like fighting (Running away could be used to trap followers).
And permit to defend a city with canons, which is indeed very appropriate.

All of this would need to be combined with appropriate rules about the attacking moves, like no more attacks 6 hours before TC.
(Another rule I used to suggest : No move in a foreign territory 6 hours (+-) before TC).
Another rule to go along : once an attack order has been given to a unit, no cancel is possible, and no other move.
A message should be sent in Game (and by email if offline) to the defender when an attack has been ordered against him.

To counter-balance this, defend bonuses could be seriously reduced.

So, to resume my thinking :
- no restriction on moves, except for attacking moves, that should be ordered n hours before TC, and could not be canceled.
- battles resolved at TC
--> ACTIVE DEFENSE instead of a passive one.

elrik

Player

Posts: 108

Re: LT31 will start soon

First of all we didn't have the real random TC until now. By now it was solved with changing TC lenhgt randomly each 4 minutes between 23 and 25 hours, changes wasn`t hidden. It means that:

1. TC was moving as with turn lenght between 23-24 hours
2. If someone has enough time could be online in this timewindow and waits for 4 minutes to the TC without change to make moves:)

To fix that turn lenght must be hidden:)

What about our other problems... There is already a path with allows to be online only for some time(in previous ex game it was 1 hour) so it is some base punkt, the other is observer feature. What if player enter observer after logging and then can switch to "owner" for some time and then after that time is gone can be observer only? or maybe 1 hour owning, 3 hours observing?
- it solves problem with in game chat
- it uses mainly existing features and should not be so hard(i hope it is not like always with programming problems: the easier it seems to be the hader it is...;P )
- still there is a problem that someone can sit with observer and use string view to be a super spy - which i don`t like
- i don`t know how hard is limiting obserwer feature for one nation only

wieder

Administrator

Posts: 1,863

Re: LT31 will start soon

Truly random TC would solve a lot of issues with TC RTS.

Before we can have all these nice features, it would be nice to know what will happen if someone does RTS while it's forbidden? And the very good question: Is RTS forbidden in LT31 or is it just recommended that the opponents should be allowed to make their moves without RTS?

How is RTS defined? Attacking enemy units? Moving more units to support a city under attack? Moving workers inside homeland to block enemy movements?

Is it RTS to move units into mountains (inside homeland, invisible to the enemy at the time) in an attempt to block enemy attack? Is it RTS if you do that only after you see some suspicious maneuvers performed by the enemy when the actual attack has not yet begun?

I'm asking this because it would be really good to know where to draw the line. There probably can't be hard limit to what is considered as RTS but there just might be some guidelines.

And finally, what happens if someone does RTS? In LT30 it was extremely hard to prove if someone did RTS or autoattacks. What should a player do when this stuff happens?

elrik

Player

Posts: 108

Re: LT31 will start soon

Ehh... as always nothing will be clear enough without server side regulations:) The border between what is and what is not RTS is very thin. The point is to avoid sitation in which:

1. someone is online all the time monitoring movements inside vision range and reacting for any enemy movement as soon as possible. - can you see a difference between that and normal real time fight Wieder? If you are lucky and you caught enemy during attack, well... shit happen, but when you can catch and react EVERY move its not right.
2. Someone can be online the whole time waiting for every the minute enemy is off.
3. Using ANY kind of scripts - the hardest to check - but it can kill fun for everyone else
4. Someone can just gain great adv just from the fact he can wait for TC and use it(TC fortress, attack before, def with new units just after etc.)

I can live with fact that someone is just better than me and i think that everyone here can, i just want to avoid the situation in which someone can earn so great advantage just from f.e. being online all the time that my skills are meaningless.

wieder

Administrator

Posts: 1,863

Re: LT31 will start soon

I'm not trying to argue that one should be allowed to be online and react to every movement made by someone else. If people think that's a bad stuff then it's bad stuff. Clearly many people consider it as something you shouldn't do. All I'm trying to do is to find out what can be done and what is not allowed to be done.

It doesn't seem possible to have all these nice patches available for the LT31. That's why talking about it seems to be the only thing we can do.

1. I see the difference. However I would like to chat and plan the game without logging out too often. Chatting and planning is a major part of the game. It's a huge part of the game when one turn takes 23h. I can live without RTS during that time. I never used any logging features and I don't think it gives that much advantage if it's not used to wake you up for RTS.

2. Very true.

3. Automated actions should not be done because it's not what this is about and everyone just can't do that.

4. Yeah yeah and yeah. A truly random TC would fix that. If it can't be done, well..