Discussion about unitwaittime and its length

Planning and discussing Freeciv Longturn gaming
User avatar
Corbeau
Elite
Posts: 622
Joined: Mon Jan 13, 2014 11:13 pm

Discussion about unitwaittime and its length

Postby Corbeau » Thu Oct 05, 2017 10:59 am

There is a warr currently in Discord about this issue, it's a mess as usual when a lot of people are participating at the same time, so this would be a nice place to copy some of the more coherent arguments.

(a) people who can only log on once per turn are screwed either way (unitwaittime or not)
(b) Freeciv is a turn-based game, not an RTS game
(c) double-turning throws the rest of the game's balance out the window, because it allows performing an attack from a safe place with next to no way to stop it. It makes killstack and the fact that AttackStrong units have low defense redundant, since defense outside of fortified positions will never happen
(d) the only real problem currently mentioned is that coordinating allied operations (around the turn change) is more difficult
Also a thought experiment I did:
Let's divide players into three groups: Firstly, the people who log on once a day for an hour or so. Secondly, the players who log on once in the morning for maybe 15 minutes, once at noon for maybe 30 and once in the evening for maybe 3 hours. Thirdly, the players who are able and willing to be online at any time.

Now, let's see what happens without unitwaittime: The first group – let's call them "casual players" – can always make their moves. They don't get any special advantages or disadvatages, except when they get to late-game when their units and cities don't fit into their hour anymore. The second group – let's call them "moderate players" – can do their turns normally. They don't get any special advantages or disadvantages, as they should generally be able to properly do their turn throughout the different parts of the day. The third group, however – let's call them "powergamers" – will focus all their actions on the turn change. They will be able to perform double moves and react in real-time, gaining a huge advantage over the first two groups. Additionally, when double-turning, the current balance of most units doesn't work that well anymore. Both casual and moderate players will be overwhelmed by their attacks, unless their time window just happens to be around the turn change.

So, what would happen with a 1 hour unitwaittime? Well, the casual players would occasionally lose their turn. The moderate players wouldn't be affected much, as their activity is spread throughout the day. The powergamers would encounter difficulties when double-turning against other powergamers (or players who happen to be online right now), since the other side can always react as long as they're online. However, most of the time, they will still have an edge against the causal and moderate players, since they will be offline for more than an hour, so if the turn change falls into their offline period, they can't react.

Next up: 6 hours unitwaittime. No difference for the casuals as compared to 1 hour; they'd still lose one in 24 turns. Not much of an issue for the moderates, as they still have their move spread over the day. However, the powergamers will now have a harder time double-turning against moderates, since the unitwaittime might include both their morning and noon logins, allowing them to react to anything around a turn change inbetween. Double moves would be less common, the powergamers' advantage would be slightly diminished.

And with 10 hours, it still wouldn't get any worse for the casuals, wouldn't really get any worse for the moderates – they might sometimes have to wait, but it shouldn't be a problem since they have enough time – but the powergamers would lose their edge. Even if the turn change is at night, they wouldn't be able to effectively double-turn, since the end of a moderate's evening activity and theeir morning login are close enough to one another that only one move can be done inbetween. Thus, the problem of double moves is solved (except against casuals), and really, the only one directly suffering are the casuals, who probably (a) didn't stand too much of a chance anyways and (b) don't take their losses that seriously anyways.
So...

In summary: A short unitwaittime isn't really better than no unitwaittime. A long unitwaittime makes it so that moderate players (who have a life other than the game) stand a chance against powergamers – which is a good thing IMO, because we should be encouraging people to do things other than the game – working, studying, going outside, and most importantly: getting a full night's rest.


A turn lasts 23 hours. You are prohibited from moving a particular unit in the FIRST 10 hours of that turn IF and ONLY IF you moved that unit in the last seconds of the previous turn. If you moved that unit earlier, then you have to wait less or none at all.

If you want to move that unit at the start of the turn, then PLAN AHEAD or MAKE A RESERVE, and DON'T MOVE THAT UNIT IN THE LAST 10 HOURS OF THE PREVIOUS TURN.
Last edited by Corbeau on Thu Oct 05, 2017 11:13 am, edited 1 time in total.

Caedo
Elite
Posts: 376
Joined: Sun Feb 10, 2013 10:21 pm
Location: Germany

Re: Discussion about unitwaittime and its length

Postby Caedo » Thu Oct 05, 2017 11:08 am

Funny how you put in the things I said and the things you said, but not the things Lexxie said. For the sake of completeness:
Lexxie wrote:1 hour wait will reduce about 60%-90% of the disadavantages made by double movers, because that one little hour exposes their units to risk during that time period, advertising what they're about to do, putting them in vulnerable spots on the way to their hill or mountaintop destination, etc., and there is no certainty the other won't log on in that hour, making a smart player totally avoid the risk. Except in 10%-40% of the time where it's no risk, but most of those cases aren't "exploits" or sneak attacks.
1 hour wait, UNLIKE 5 or 6 hour wait, also reduces by about 60-90% the disadvantages of unitwaittime. A 5 or 6 hour wait leaves all the disadvantages of unitwaittime in place, unfixed.
The disadvantages of unitwaittime are many and not properly compiled or listed yet, and this is a project currently taking place. You may list thoughts on that in public and also please send me a PM if you want to make sure I don't miss putting it in the list. This also goes for arguments in favour of unitwaittime

User avatar
Corbeau
Elite
Posts: 622
Joined: Mon Jan 13, 2014 11:13 pm

Re: Discussion about unitwaittime and its length

Postby Corbeau » Thu Oct 05, 2017 11:20 am

Caedo wrote:Funny how you put in the things I said and the things you said, but not the things Lexxie said. For the sake of completeness:

Do I have to do everything myself? :P

Lexxie wrote:1 hour wait will reduce about 60%-90% of the disadavantages made by double movers, because that one little hour exposes their units to risk during that time period, advertising what they're about to do, putting them in vulnerable spots on the way to their hill or mountaintop destination, etc., and there is no certainty the other won't log on in that hour, making a smart player totally avoid the risk. Except in 10%-40% of the time where it's no risk, but most of those cases aren't "exploits" or sneak attacks.

This 60-90% estimate is completely arbitrary. It may reduce their advantage to other players who are constantly online, but if you log in, say, 3-4 times a day, do the calculation yourself. it's far less than 50%. Actually, I'll make it for you. If you log in 4 times a day for, say, 15 minutes, the probability that you'll catch a glimpse of a particular hour is 25%.

1 hour wait, UNLIKE 5 or 6 hour wait, also reduces by about 60-90% the disadvantages of unitwaittime. A 5 or 6 hour wait leaves all the disadvantages of unitwaittime in place, unfixed.
The disadvantages of unitwaittime are many and not properly compiled or listed yet, and this is a project currently taking place. You may list thoughts on that in public and also please send me a PM if you want to make sure I don't miss putting it in the list. This also goes for arguments in favour of unitwaittime

Caedo, the point of copy/pasting this part this is...?

User avatar
AndreasR
Elite
Posts: 686
Joined: Thu May 02, 2013 10:26 pm
Location: Norway

Re: Discussion about unitwaittime and its length

Postby AndreasR » Thu Oct 05, 2017 2:32 pm

This is a friendly reminder that the process for changing the ruleset or settings for LongTurn Web X is as follows:
1. Show that you have community support for your change.
2. Submit a pull-request to Github.

Settings
Ruleset

LongTurn Web X is starting on friday!

wieder
Hardened
Posts: 165
Joined: Thu Jul 27, 2017 10:43 am

Re: Discussion about unitwaittime and its length

Postby wieder » Sat Oct 07, 2017 5:07 pm

10h unitwaittime is quite good because it usually allows the players to sleep/do work without allowing the enemy to double turn. It's not perfect, but 6h or less would pretty much always allow double turn attacks while sleeping/working if the tc happens at that time.

10h will limit hc players and those who login 2-4 times a day. Players who connect only once / day may have issues if they try to play both turns at the same hour and connect only every other day. I guess there are not too many of those however.

Canik
Posts: 26
Joined: Tue Jul 18, 2017 4:26 am

Re: Discussion about unitwaittime and its length

Postby Canik » Sun Oct 08, 2017 2:46 am

Canik's Con List. I'm only doing 'Cons' because any 'Pros' are just a 'Con' of the other. Keeps the list shorter, easier to read while still giving a full picture.

*****

Cons of No UnitWaitTime:

Double-move attacks will become a thing. Resulting in:

a) Surprise blitzkriegs being far more devasting. An enemy can be in a 'dark tile' then suddenly be 2 cities deep. You can adjust your defensive strategy to counter it to an extent but it would still be annoying to deal with. Wars will likely be more lopsided since a more devasting initial attack will be possible. Most new players will be have a considerably harder time defending themselves and learning to defend themselves, I imagine.

b) Being online at the turn change becomes very important. People who work/are students are punished by this as they will often not be able to be online for turn change. We all lose the convience of being able to log in once a day at a regular time of our choosing or else we risk being hit by a devasting blitz. If too many people start all trying to move at the time it could destablize the server as well, potentially.

c) Game balance could be disrupted by double-moves. It wasn't balanced with that in mind. Certain strategies, units, governments, etc would become stronger than before. For example; mobile units become much more important. Their range effective doubled if you only move every other day.

A less balanced game is a less rich game.

Cons of 10 Hour UnitWaitTime:

If a unit is moved it cannot move again for 10 hours. Resulting in:

a) People having to be aware that if they move within 10 hours of turn change they will not immedately be able to move those units again. New players may be confused by this initially.

b) On occassion, the rarity of which hasn't been scientifically determined but I imagine this scenerio to be uncommon, a person may only get a short window within a 46 hour period to move and if the move is around the update they may not be able to make both turns when they would otherwise, without the 10 hour wait time.

c) You won't be able to double-move often before your enemy gets online to react.

*****

If I forgot a Con post it and I will add it if I agree with the logic on it.

As it stands I think 10 Hour is clearly the better option. It may be a little annoying at times but generally, it should not be an serious issue for most people most of the time.

No Wait on the other hand could cause serious issues for virtually everyone.