 |
| |
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Ah, I see. Well, thanks for clearing this up for me, Elmo.
About the system, I agree with you mostly. Many people propably find the year long games fun, although I think that some would, and that I am likely to be a part of the latter group.
So mostly people would propably play games like you described them - a few times longer than an ordinary Civ2 game.
But at the same time I think it would be cool if the games themselves lasted a long, long time, even though players would come and go. And I also think using clans (yes, they would only be an advanced version of password protecting your otherwise open Civ player slot) would be truly great.
So I believe a system much like the one we have thought of so far - realtime with a pause mode - would be the best for the game. Furthermore it could, with time, make it possible to speed up the pace and have the player only do general decisions, and so make really fast games (maybe just 4 hours or so to go from ancient to present day) possible.
So here is what I think the system should be like:
I believe it is much like what you have agreed upon, but just to make sure... 
Basically it would be a "slow" realtime, where 1 year might take 5 or 10 minutes or so (yes, this should be modifiable, and then each year (or more or less often, or not at all) the game would pause for x minutes, and allow players to think through strategies, give complex unit orders, manage economics, politics and other more time consuming things that is easier to do when you don't have to look at your units move around. This way, if we keep everything completely modifiable, the game can be made just as the players want.
The only problems now are the extended games, where 1 year could be 1 day, and when the players are not online at the same time. This simply does not work with realtime. But all we need is a mode where the realtime mode is set on "instantaneous", there is a pause every year (or whatever), and people just do everything they do in the pause mode. Then it just has to be playable by email, and we actually have simple, preplanned turns.
So with this system, which again is not at all my invention, but yours, we could have everything from pure realtime to pure preplanned turns, plus everything in between. And if anything is optimal then it has to be that.
Now the only problem is to make it work. Can we do that? Are there any design issues that we should think about? Any weak spots or potential bugs in the system? Otherwise the turns will just be up to the programmers now, which, I guess, should start working on them when we have a UI, and perhabs a map.
Anyway, I am home for easter from today untill monday. So if you guys would like to have a meeting with me then this is great time to do it. Can we arrange a time?
------------------
"I have an idea for a movie. It's about a killer robot driving instructor, that travels back in time for some reason."
- Homer J. Simpson
GGS Website
|
|
| |
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
The lack of responce here is sad.  Yeah, I know everybody are buissy. But it's easter! Surely you can't be working or going to school now?
Well, I've just thought about the system. And the only real flaw I see (apart from the fact that it will prove tricky to create an economic and social model that works in realtime) is that realtime doesn't work with play by email games. So I think that those who want to play 1 year a day or something, should just use the type where the realtime part is instant, and therefore use preplanned turns that they can download as a package, do their moves and send the whole thing back as a package, meaning they will only need two short online sessions a day to play the game.
And yes, there will be people who will want to play this way for a VERY long time. I have heard of people who are playing online games of Civ2 MP on year two and three, so they do exist.
About the memory usage, I think you guys should check out the site www.republictherevolution.com. It's a game by some guy who used to work with Peter Molineaux (the genious who made the most fantastic game since Civ - and it really is, buy it if you don't have it - Black & White), but who is now working on his own game. The game is about you trying to take the power in some former Soviet republic, while interacting with the people in it. Like B&W the game has individuals all with their complex daily rutines (They are talking about everything from voting in elections to picking up their kids from school), only in this game there will be 1 MILLION of them!!!!
What I am saying is that if you can have 1 million individuals in a game AND model every single leave on every single tree in a country you can have 1 million hexes in a game, and lots of populations which interact with each other. So I don't think we should worry too much about memory usage.
------------------
"I have an idea for a movie. It's about a killer robot driving instructor, that travels back in time for some reason."
- Homer J. Simpson
GGS Website
|
|
| |
|
|
|
VetLegion
|
|
hi everyone, happy Easter
Joker,
nice to see you ... good thing you are resting a bit from that long vacation you have (insert green - envy - smilie here, I dont know how to ; )
About the meeting, I cannot make it. I am cut off from internet a bit, heh, but not from computer. Perhaps I will do some work now instead of surfing porn sites all the time (ups, havent said that  )
chrispie,
I have your UI working on 366 Mhz finely. We never said the low specifications, but this may be it. We just divide it by two and say minimum spec is P II 183 and everyone will know what we are talking about. Hey, big corporations do it and get away
I dont think the system is bad itself ... perhaps there is a bottleneck or two we can find with pofilers and fix.
|
|
| |
|
|
|
VetLegion
|
|
I checked the site for the republica game. Black and White 2? Bigger village and no monster? I dont think 3D modelled cities "to great detail" is much of a strategy game ... check out Tropico instead, comes out by end of April from aouthors of Rairoad Tycoon 2. Sounds like very cool game.
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Great to hear from you guys again!
Now all we need to hear from are our two Finish friends, and Heardie (who is excused because of his exams).
Chris,
I am free for a meeting tomorrow (hey, nearly all my friends are travelling the globe or something, so I don't have much of a life when I'm home), so I think a meeting would be a great idea. How about around 9ish? I am buissy from 8 and around an hour ahead, so 9 would fit me well (and of cause I am talking about 9 in the evening ). I can not promise to be 100% on time though, but around 9 I will be there. It's still GGS on Undernet, right?
And yeah, B&W is truly fantastic. Especially the creature is great, and obviously the graphics are stunning. I do, however, agree with you that I have not been caught by the gameplay in the same way as I was with Civ2.
Your UI worked good for me. I am using a 450 Celeron something PC, which my parents have bought without consulting me, so it's overpriced and rather poor. It's ok, though.
Vet,
LOL@porn sites! 
I disagree with you about Republica, though. To me it seems to be a game where the stunning B&W type graphics are put to use in a real world, in stead of the cute eden world from B&W. It seems the game will be really complex when it comes to interacting with the people of the world, something B&W is not. So I think Republica, if it is as good as it seems, could be just my type of game.
I have heard about Tropico, and in fact I did some quite extensive research on it when looking for inspiration for the government model. It too looks great. But what bothers me about it is the scale of it. The island you are on starts with 20 inhabitants, and I think the max amount is around 500. So in stead of modelling flocking and factions the game engine can just model individuals on a much lower level. And that is not as impressive as having 1 million people all playing part in society.
Anyway, to keep this a bit on topic, what do you guys think of the turn model I suggested (but not at all invented)? Do you think it's good enough for us, needing some work or something? Let's GGS!
------------------
"All that monsters fear is the day."
- Kashmir
GGS Website
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
21.27, and noone here yet.
I'll go offline now, and be back in about half an hour. So if anybody shows up (come on, Chris!) you don't go nowhere, you hear?
------------------
"All that monsters fear is the day."
- Kashmir
GGS Website
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Back online. And still noone. I'll just stick around, waiting for someone to come. Sure hope you're coming, Chris!
------------------
"All that monsters fear is the day."
- Kashmir
GGS Website
|
|
| |
|
|
|
amjayee
|
 |
Jyväskylä, Finland
Oct 1999
time: 07:14
|
|
I'm gone for some days, and the forums gets flooded. Gosh. Well, not flooded actually, except relatively speaking. About UI, on my P3-700 it runs fine, though I think the graphics card is the one that decides how smooth it is. My card id bad, so that cuts down on the frame rate. But I think we will manage to optimize it later... and I think messaging or anything will not make the performace much slower, drawing graphics will create the largest overhead for the processor, sending messages is nothing compared to it. It's hard to realize how powerful the current processors actually are. I agree it would be good to have a mouse-driven UI, though in this kind of game it might be a little tough... I think we would be wise to go with a menu system until we get so much game ready that we have something to actually make a UI for. About the UI demo, I really hope to get some results soon. My school work is at the final lap, so perhaps in a few days I will have time... but I really can't say. Then our 2d-artist Elmo should also get some work, I think...
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Nice to see you Amjayee!
Me and Chris had a meeting allright. It was mostly catching up, and talking about games and stuff, so not much work was done. It was, however, a great feeling to finally attend a meeting (even such a small one) again.
I agree with you Amjayee that we will need menus and stuff. GGS would not work with a B&W type menuless enviroment. But I think we could combine clicking and dragging with a menusystem. We should try to keep the game completely mouse driven, though.
What do you think about the RTS/TBS system I described, Amjayee? Do you agree that this is what we should have? And could you make such a system?
Well, I am off now. Good luck with the UI creation. I will be back home in a few weeks. Untill then I will make some sporadic posts now and then.
Bye!
------------------
"All that monsters fear is the day."
- Kashmir
GGS Website
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Yeah, you are both right.
Everything can be done anytime. No restrictions here of any kind. The pause mode will just allow players to plan things without having to worry about what else is going on.
I think later on we could also make the realtime length of the realtime mode dependent on the amount of units in the game, and the amount of "action" going on (battles etc). This way in the early turns the realtime phase could be over in just a few seconds (1 year), and so could the pause mode (since players would hit the "end turn" button fast due to little for them to do). This would make sure games didn't became boring when the player had little to do.
------------------
"All that monsters fear is the day."
- Kashmir
GGS Website
|
|
| |
|
|
|
Well... I don't like the system. Having a real-time mode will make lagging a bigger problem than in turn-based action, and there are the problems of synching all the clients properly. From a technical viewpoint, I'd say that leaving out the real-time mode would be the best thing to do. Instead I advocate turn-based, simultaneous movement (as I always have), with lots of flexibility in turn lengths (from minutes to days). Another thing that bothers be with RTS/TBS hybrid is that the game is no longer uniform... the player is practically forced to alternate between moving units and doing other stuff. It doesn't matter that all the options are available at both modes, since moving units is easier in real-time. Or, more accurately, the threshold for moving units and adjusting their moves is lower.
If there was only a turn-based mode, then the player would be encouraged to make big enough decisions instead of micromanaging units. Isn't that what we're aiming for?
"Leland"
|
|
| |
|
|
|
I admit, I hadn't understood it thoroughly. Luckily chrispie explained the system to me in the last meeting, and now I think I have a lot better view of it. I still don't agree though . In an attempt to clarify this issue so we can move on, I will now first explain how I understood this RTS/TBS abomination, then list the problems I see with it and finally present the system I think we should use, formulated by amjayee.
Dual-mode of play
There are two modes of play, a real-time phase where units move more or less continuosly, and a planning phase where the world is frozen. A turn consists of a real-time phase followed by a planning phase, though it could be configured that one of them lasts zero seconds and effectively doesn't exist. In both modes all options and orders are available to the player, but in the real-time phase there is the advantage of being able to react to whatever situations arise. To make it simplier to manage, real-time phase would consist of "ticks" that are kind of like light-weight versions of turns. Within ticks, some minor things could happen, e.g. unit movement, but more demanding, empire-spanning events would only happen when the turn changes.
(This is pretty much how Joker described it, except for the notion of ticks. Here, they are just a neat way of managing the movements. Each turn could consist of 10-20 ticks, for example. Most of my critique doesn't involve ticks in any way, and I actually think that they are a very good way of internally simulating unit movement.)
What's wrong with it
The game would become a micture between two types of play. On the other hand, there would be the real-time aspect of moving units around, looking at them crawl on the map and if necessary, manually guiding them. While on the other hand there would be the context of turn-based events which would happen only at the end of each turn. I think that this discrepancy will violate our goals of realism and micromanagement-reduction as well as make the game more confusing.
Realism would be compromized because there would be an artificial boundary between events that happen real-time and events that happen once every turn. In real world, everything happens all the time. And I don't think that unit movement is a special case compared to, say, resource extraction, production, taxation, economy, migration, riots, administrative changes (creating new regions and such), infrastructure improvements, scientific development, climate, natural disasters or politics. In fact, moving units around is a very small part of the game. However, it was pointed out to me that the real-time phase could include other things besides unit movement, but I believe that my point remains as long as there is any artificial distinction between turn-based and real-time events.
I can imagine a scenario where two nations would be at war, both struggling to build new armies as well as fighting for their lives. It would seem off if you could only get units at a certain time of the year (the end of turn), after which you'd move them in a real-time mode until the next year arrives. I think that's unrealistic. Okay, then you could of course add army production to the real-time mode; that way, armies would be built all the time, but there would be the problem of extracting resources where to make the tanks or whatever the player is using. Resources would apppear magically at january first of each year, and the players would build as many tanks as possible in the first quarter, waiting rest of the year for the new resources to arrive. Okay, so let's add resources to the real-time mode as well. Then you have a problem of people whom you use in the army... population growth is calculated at the end of the turn, so you could get a million new soldiers overnight while the remainder of the year crawls with the existing crew. And even if you had real-time population growth, you still might run into trouble with trade, politics, civil disorder, ... you name it.
What I'm trying to say with all this is that everything is interconnected: we cannot have some aspects of the game work with a totally different turn order than the rest. Well, technically we can, but it would be unrealistic and confusing.
Another point is that I don't think we really need the kind of detail that is related to real-time mode. Adjusting moving orders for your armies in real-time is exactly the kind of micromanagement that I thought we're trying to avoid. Also, it would not be sufficient only to leave the option of adjustment to the player: if real-time adjusting is beneficial, then those players who can't or don't want to do it are being punished. Not to mention that it ruins the excitement of having unpredictable results (I'm sure S. Kroeze would agree with me here). On the other hand, if the game is rigged so that units with adjusted orders have no benefits from the real-time adjustments, then I see no point in having such phase at all, it does not add anything to the game.
As was discussed earlier (way earlier), I think that using relatively few armies with large-scale orders is something we should aim for. Given this goal, I don't see what good a real-time hassle would bring to the game, except maybe attract C&C fans to try the game out. Why should we add unnecessary complexity to our already bloated plans? What value does the real-time mode bring into the game?
Simultaneous preplanned turns
I've copied an earlier post by amjayee here, he summarized it pretty well:
quote:
Generally I think this would be a pre-planned turn system. Players would plan the coming turns, they would not change orders "real-time". The orders you give would be buffered, and at the beginning of the next turn they would become effective. This would simplify the system, and reduce the hassle.
So, my idea is; time goes by at constant speed. Units and regions have general orders they follow. Player watches the game as it goes by, and changes the orders. The new orders are buffered, and they are updated at the beginning of the next "turn" or update cycle. The orders could be sent to server and buffered there, or buffered at the client machine and sent in one package at the beginning of the update cycle - the first option would reduce the traffic of course. When there is no pause between the turns, we have a real-time system with pre-planned turns; you plan the next turn while you watch the current turn go by. The original pre-plan turns system could be got by having a pause mode between turns. The players could change orders during the turn execution phase, but that would in effect be "planning the next turn". So, if your units screw up during the turn, you can't change the orders until next turn... which is only realistic I think, the leaders don't have very much they can do in the finer details.
So this system would combine real-time and pre-planned turns. It would be very flexible; the turn could be one month, one year, ten years, anything. The turn speed could be any desired speed. You could play real-time, pre-planned, extended mode, quick action-games, anything, with the same system. What would this system sound like? I think it's the best compromise to obtain all the ideas presented here. Or is this a completely foolish idea? |
I agreed, and I still do. What exactly is wrong with this idea? Amjayee, do you still think this is a valid basis for a model, and if not, I would like to know why. It seems that this baby can do all the tricks Joker's model can do, except that the players would not be able to change orders in the middle of the turn.
Elmo, I am not sure what you mean by The game is representated to the player as tbs with simultaneous turns. Although technicaly it is rts? If a player can see the armies moving on the map, and can affect the movement at any time, I think it's pretty much an RTS.
It would be nice to have this thing settled. Once we do that, to one way or the other, I'll write a new proposal of design doc and maybe the first version of an official turn order model. Throughout the conversation there have been several changes to plans, many of which are simply due to everyone having their own slightly different vision, and because the "model" at the moment consists of numerous dispersed posts. Writing these ideas down in a coherent document would free us from going through the same things over and over again. But first, let's hear some opinions! Did I misunderstand Joker's idea? What about ticks, how should we incorporate them? Is this post total rubbish?
"Leland"
|
|
| |
|
|
|
I re-read some of the posts above, and realized that this confusion appeared when Joker presented his impression of the system, and then you and amjayee agreed with him. I was then confused about your confusion, and seriously thought you had all changed your minds while I was gone! Well, I'm glad too that this was settled. Now I can probably start thinking of revising the design doc.
So, to clarify once more: we are using preplanned simultaneous turns. This is how the game carries out the orders internally. The real-time aspect of the game is restricted to the way in which players receive information on what has happened: the server may not send it all immediately (due to time it may take to process a turn), but let it trickle over some short time period (3-15 minutes, i see no point in longer real-time phases). This "trickle" could either be synchoronized so that the units move with constant speed (it could look nice, and be more fair for players with slow connections), or it might be as fast as the slowest bottleneck in the system (server speed, network on server/client).
Hmm. To make this more clear, we should probably start calling these two officially an execution phase and a planning phase to avoid confusion. We are not making a real-time game, and it is really confusing if we use the term real-time in this context. That's how we got into trouble here in the first place.
"Leland"
|
|
| |
|
|
|
Hmm.. I suppose you're right. "Planning phase" should really be called "non-execution phase" because that is the only thing defining it. Better yet, we could stop using the term planning phase altogether because the whole game is in a planning phase and the execution only overlaps it. Like this:
<--------- planning -------->
<- execution -><--- left --->
Here, "left" really means nothing except that the execution phase of that turn is over. I hope this clarifies this somewhat. I apologize for my sloppiness.
"Leland"
Last edited by on 28-05-2001 at 22:18
|
|
| |
|
|
|
The Joker
|
|
Copenhagen, Denmark
Aug 1999
time: 06:14
|
|
Boy it's been a while since I've been here.
Everything is all changed, and I can't even recognize it any more.
Well, after reading this thread, and the (rather pointless) discussion here I just wanna say that I agree with Leland.
My idea for a more RTS system came since I thought that was what the rest of you wanted. And I thought of a system that incorporated RTS in a way that I would like. But honestly, I have always been more fond of the preplanned turns that we have decided a long long time ago.
So you have my complete support, Leland, to go on with a Design Doc where preplanned turns (or whatever we choose to call them) are explained and used.
|
|
| |
|
| |
All times are GMT. The time now is 05:14.
Apolyton Time is 00:14. |
top of page
|
|
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|
12-04-2001 04:04
|