live game vs multiplayer

[gyboth]

i'm still working on the live game in 1.9, but also planning the server/client multiplayer features. now there's a slight clash between the two: i don't think i can implement a real live game if i choose server/client architecture. so there would only be recordings of live games (which would look more or less like the live game you've already seen in the cvs version). the drawback of this is obviously that you can't influence the game while it's happening (substitutions, playing style change etc.).

now my question is, what would you people prefer? a real live game + no multiplayer, or live game recordings without live influence + multiplayer? (of course as i gain experience with server/client programming i might be able to make the live game really live even in multiplayer, say in 1.9.5 or so ).

i'm tending to prefer the second possibility because i believe that multiplayer would be a big boost to the game's popularity (but of course it also means that 1.9 will take a lot longer to develop because i know nothing about server/client programming).

gyözö

[vector]

now we are getting into the nitty gritty.
I have one thought only and that is that for me, if you are referring to a multiplayer, internet based. game you have one problem.

When does it take place?

sure if its a little league amongst some friends on a sunday arvo its "controllable" but a "world wide" league with all the time frame problems it produces would, I feel be un controllable.

the one on one games (me in my own little world Vs the computer) would be great to have to at least half time substitue changes etc.

Are you saying that you cant design a system that can be 1v1 and meVworld?

Sure, not having the ability to "change" things on the fly in a live game is a compromise.. but I hope, would only be disabled in multiplayer or even worldwide games.

erm.. so do we actually have 3 situations?

1) me V computer
2) me V "local" multiplayer
3) me V a world wide collection of bygfoot gamers

state of live game changes in above
1) enabled. changes can be made at any time
2) enabled
3) disabled. or preset
preset = Maybe we could allow a script "complication" if loosing at half time. I wish a substitution of 6<>15 8<>17

hmm /me starts to go off on a complex tangent of being able to create a bygfoot scripting language. Maybe ver 7

Maybe we need to decide first wether people want
a type 2 or a type 3 mulitplayer game.?

[gyboth]

When does it take place?

sure if its a little league amongst some friends on a sunday arvo its "controllable" but a "world wide" league with all the time frame problems it produces would, I feel be un controllable.

that will be configurable. the chap who starts the server will be able to specify how long the clients may take to send their data. so in world-wide games that'll be a day or two, or even a week. in sunday arvos (whatever that means) you could specify five minutes. when the time is up, the results are calculated and the clients may fetch their updated teams and the results and the new transfer list and whatever else, and a new deadline is set.

Are you saying that you cant design a system that can be 1v1 and meVworld?

yes.

Sure, not having the ability to "change" things on the fly in a live game is a compromise.. but I hope, would only be disabled in multiplayer or even worldwide games.

that'd be awkward programming. if i do server/client architecture, it'll be used for every type of game, even the 1v1 type.

1) me V computer
2) me V "local" multiplayer
3) me V a world wide collection of bygfoot gamers

this is all the same. in 1), you start a local server and a local client and play away. in 2) you start a server on one of the computers in the network, and those who want to participate start clients. in 3), a server somewhere on the internet is started, and those who participate start clients on their home computers.

state of live game changes in above
1) enabled. changes can be made at any time
2) enabled
3) disabled. or preset
preset = Maybe we could allow a script "complication" if loosing at half time. I wish a substitution of 6<>15 8<>17

nope, this won't be good coding. i can't make such differentiation. it's all or nothing in this case.

Maybe we need to decide first wether people want
a type 2 or a type 3 mulitplayer game.?

well, this is obsolete, as i pointed out.

gyözö

[vector]

im not explaining myself well.

my concern is the live game "substitutions".

I cant see how you can handle this in a multiplayer situation?
imagine 100 people world wide all waiting for each other to make substitutions. The "round" may never get finished.

ok you can add a timeout. but I cant see the point in waiting a day for someone to make a substitution at half time. when its supposed to be a LIVE game.

that will be configurable. the chap who starts the server will be able to specify how long the clients may take to send their data. so in world-wide games that'll be a day or two, or even a week. in sunday arvos (whatever that means) you could specify five minutes. when the time is up, the results are calculated and the clients may fetch their updated teams and the results and the new transfer list and whatever else, and a new deadline is set.

so this sounds like no substitutions allowed?

3), a server somewhere on the internet is started, and those who participate start clients on their home computers.

erm but im 10 hours diff.. im asleep!!

I still see only one way out.
local games allow me to make changes during the match, multiplayer version dont.
multiplayer versions just simply work by sending text file info

I hope thats clearer.. ??

[gyboth]

I cant see how you can handle this in a multiplayer situation?
imagine 100 people world wide all waiting for each other to make substitutions. The "round" may never get finished.

you're right i hadn't thought of this. it really seems that the two are mutually excusive then.

so this sounds like no substitutions allowed?

right, no subs.

3), a server somewhere on the internet is started, and those who participate start clients on their home computers.

erm but im 10 hours diff.. im asleep!!

well, as i wrote, a world-wide league should have timeout 1 day at least. so you'd have at least a couple of hours to manage your team.

I still see only one way out.
local games allow me to make changes during the match, multiplayer version dont.
multiplayer versions just simply work by sending text file info

so you vote for real live game and if possible multiplayer-via-email, more or less.

gyözö

[vector]

Yes now we understand each other

With no subs, a multiplayer game via email file upload or simple ftp or ssh or some such device would be great.
An available replay of the live game "a bonus." If possible of course!
Probably just a file that the user grabs and plays it client side. that way he can choose the commentry level, settings etc. and play it over and over again if he wishes

All the above posts are now correct, since we have understood that subs and live games just wont work with muliplayer. 1-2 days to get your teams "setup" onto the server site etc all fits, works and is manageable.

so you vote for real live game and if possible multiplayer-via-email, more or less.

Yep hole in one !.

PS just to add another thought. If people really really wanted the ability to have substitutions in multiplayer mode. I dont think it would be too hard? to allow a half time only sub. (using the multiplayer 1-2 day setup rule) ie you could send in your team setup before the kickoff deadline. the next day the first half is played and result displayed, 24hours later the second half is played. ie you had 24hours to make a half time sub.
This might just work. But we have to keep in mind that there maybe cup games as well during a "real week" so you would run out of "real" time if it was more than 2 games in a week and you tried to expand it to more than substitutions at half time only.
to be honest i just dont think its worth it! Id prefer to have say:

1) mon - thur. Send in my team setup. Setups are not viewable by other users. Other users may however view my teams player listing, with the same sort of view and information that is currently available in the game.
1a) allow a different team setup for cup and league matches
2) thur-sat server Locked. No mods allowed.
2a) The teams setup is now viewable on the server to other users
3)sat the server plays the games
4)sunday the results are posted and
the live game replay file is made available

Note: why stage 2a)? Afterthought really. I thought it would serve as a kind of "preview" and some users might like to speculate on the outcomes.


And a Note to readers.
HEY we can see your reading these posts so please click the vote buttons

[gyboth]

so in the end you vote for having a 1.9 version like 1.8, mainly for single players, with real live game etc., and with the possibility of adding multiplayer-via-sending-files (which would require some 'merging' tool).

gyözö

[vector]

yes

[arnaudus]

I'm sorry, but I am not sure I understand why we have to choose between multiplayer and live games.

Look at this virtual dialog between the server and two clients :

Code: Select all

Server                                           Client 1                  Client 2
ready_to_play                            
                                                        i_m_ready
msg "Pl 1 is ready"
                                                                                         i_m_ready
msg "Pl 2 is ready"
start_game
min 1 "xxx"
min 2 "yyy"
min 3 "zzz"
min 4 "Injury John Smith!"
min 5 "aaa"
min 6 "bbb"
                                                     chg J.Smith J.Doe
msg "Change : Smith->Doe"
min 7 "ccc"
min 8 "ddd"
                                                                                       all_out_attack
msg "Pl2 : all aout attack"
min 9 "eee"
...
min 45 "xxx"
min 46 "zzz"
msg "Half-time!"
ready_to_play
...

Could you explain to me where is the problem? A team can play a few minutes with 10 players before a change, for example.

[gyboth]

the problem is mainly that the two users would have to be online and connected to the server simultaneously. this is a concept that suits a shooter game fine, but a manager game would be a lot of fun if complete leagues would be managed by users all over the world. and you can't get 20 users to be online and connected absolutely simultaneously.

gyözö

[arnaudus]

Mhh, to my opinion, playing only one game every week is only an example of a multiplayer game. But there are certainely lots of other possibilities.

If you want to play such a multiplayer game, just add some options to the serveur : livegames = false; gamefrequency = week; for example. But you can also imagine a multiplayer game with "livegames = true, gamefrequency = minute", and you can play one season each evening. I'm not sure that it is a good idea to decide "how to play bygfoot" instead of the players : people will choose the options following the way they want to play.

Moreover, I imagine that a single player game against AI is exactly the same thing than a multiplayer game with less than 20 (teams)x 5 (leagues)x 10 (countries) human players : in a big network multiplayer game, artificial players will play the "missing" teams instead of humans. But if you are the only human player, you just play a "single player game"... and the server is hosted by your own computer.

BTW, if you publish a simple standard protocol to communicate with the server, I think I will be able to make an AI player. And I would be very interested by AI games against another programs

[gyboth]

If you want to play such a multiplayer game, just add some options to the serveur : livegames = false; gamefrequency = week; for example. But you can also imagine a multiplayer game with "livegames = true, gamefrequency = minute", and you can play one season each evening. I'm not sure that it is a good idea to decide "how to play bygfoot" instead of the players : people will choose the options following the way they want to play.

*sigh* it is really easy to imagine a perfect football manager. it is easy to imagine a lot of fancy options and superb flexibility. 'hey, how about having all sorts of multiplayer possibilities, live and not live, hotseat, weekly, hourly, whatever?'. but how about coding it? the more complex and ambitious your concept, the more likely it is that you will never finish it because it's a huge task and you don't have a lot of motivation to work on it if you can't see the end of it.

you suggest the players should decide how they want to play? this means, you want me to code a manager that offers every possibility to the players, so that they can choose what they like. but hey, i'm not getting payed for writing the game and i don't have a lot of time for it, and most importantly, i'm not a programming professional. i'm not sure i'd be able to code all the things you'd like me to.

Moreover, I imagine that a single player game against AI is exactly the same thing than a multiplayer game with less than 20 (teams)x 5 (leagues)x 10 (countries) human players : in a big network multiplayer game, artificial players will play the "missing" teams instead of humans. But if you are the only human player, you just play a "single player game"... and the server is hosted by your own computer.

the biggest problem with this concept and the reason why i'm (again) tending to make a normal single player game out of 1.9 is that it is so different from what i've done till now. bygfoot has been a completely different game until now, without servers and clients and such. so i'd have to start a whole new game were i to write server/client stuff. i'm not sure i have the energy for that. i don't even have a lot of energy for developing the single player version currently.

BTW, if you publish a simple standard protocol to communicate with the server, I think I will be able to make an AI player. And I would be very interested by AI games against another programs

unfortunately this server the protocol of which you're talking doesn't exist, except in your imagination so i can't offer you a protocol right now

i hope you're not offended now, it's just that your ideas are nice and all, but you seem to ignore completely that they'd have to be implemented, too. a lot of projects never came out of their fancy shells because they were too ambitious. have you ever heard of LINA? that's a football manager for linux. they have a great concept (a lot better than bygfoot). they've had that concept for years. sounds very nice. unfortunately, they haven't released a single version yet. but hey, i'm sure their first version will be great. in 20 years or so. (they had a nice homepage, too, but i can't find it right now, not even with google; you won't have problems finding the bygfoot homepage with google.)

gyözö

[arnaudus]

*sigh* it is really easy to imagine a perfect football manager. it is easy to imagine a lot of fancy options and superb flexibility. 'hey, how about having all sorts of multiplayer possibilities, live and not live, hotseat, weekly, hourly, whatever?'. but how about coding it? the more complex and ambitious your concept, the more likely it is that you will never finish it because it's a huge task and you don't have a lot of motivation to work on it if you can't see the end of it.

I never told that it could be simple. You know how it works : people only see what they dislike, and you will never receive dozen of congratulation e-mails. But you just have to consider that more the users want new features, and more they want to play bygfoot again and again.

the biggest problem with this concept and the reason why i'm (again) tending to make a normal single player game out of 1.9 is that it is so different from what i've done till now. bygfoot has been a completely different game until now, without servers and clients and such. so i'd have to start a whole new game were i to write server/client stuff. i'm not sure i have the energy for that. i don't even have a lot of energy for developing the single player version currently.

Yes, we love you, and we love bygfoot If you remember, I spoke about a multiplayer bygfoot version not before version 2.0, since this needs to rewrite all the code.

Everyone can understand that bygfoot is not all your live, and that you can't implement all the features needed. But you're the boss : since you are the only one who wants to code, you have to decide what you want to do.

But if you need help on bygfoot development, you have to help people to help For example, I don't know C language, and I don't want to learn it because it is a very very old language, etc. I don't know gtk too. Since bygfoot is not modular, it seems very difficult for me to help you. But, as I proposed just above, I think I'm able to develop small modules, such as AI players. Of course, for the moment, I can only say "please make it modular!". Do you want I define a simple communication protocol? I can do it with you. No problem.

[gyboth]

just another thought... how about hotseat multiplayer? that wouldn't be hard to implement. on the other hand, i'm not sure whether people would play with others on the same computer.

gyözö

[gyboth]

But if you need help on bygfoot development, you have to help people to help For example, I don't know C language, and I don't want to learn it because it is a very very old language, etc. I don't know gtk too.

i can't help you there... i don't know c++ (and hey, c isn't that bad, especially combined with glib). but gtk is simple. after some basics, all you need is the API reference and you're ok.

let me sum up a bit: the easiest way for me to proceed is to make a single player version and add mp-via-email-or-upload later. this would offer the benefit of real live games for single player games (ie. if you're playing at home and not in a mp game).

the second possibility would be a lot harder, namely to make a server/client architecture. still, i could do it, but only without real live games, because those would add a difficulty i'm not willing to overcome yet. maybe in later versions.

the third possibility is to make server/client and real live games. this is the one i'm not going to take, because it's too hard. you wouldn't see 1.9 released in years.

so this all boils down to the poll question: do you want real live game, or server/client mp? the first one is better for single player games, whereas the second one would be the best solution for mp games (better than email-or-upload).

gyözö