FreeOrion

Forums for the FreeOrion project
It is currently Sun Dec 17, 2017 4:19 am

All times are UTC




Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 104 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7  Next
Author Message
 Post subject:
PostPosted: Thu Jun 10, 2004 1:25 pm 
Offline
Designer Emeritus

Joined: Thu Jun 26, 2003 8:07 am
Posts: 935
Quote:
Those PP's can be routed to planets which produce something.


I haven't seen a good system for routing PPs to needy planets yet. In my mind such as system should a) be intuitive: obvious to the player exactly what is transpiring b) automatic, requiring no sliders, prioritization, or other inputs.

Plus I just don't like the idea of having 50 different parallel queues to keep track of. If build projects are like research projects (x resources for y turns) then slapping everything into one queue would seem to make organization easier for the player.

btw, if queues are serial and industry is exported to needy worlds, then we end up with the same problem as the slider split queue. The best strategy would be to build just large one project at a time, since completing it faster will maximize the number of turns the empire recieves benefits from the project.


Top
 Profile  
 
 Post subject:
PostPosted: Thu Jun 10, 2004 2:13 pm 
Offline
Juggernaut
User avatar

Joined: Fri Dec 26, 2003 12:42 pm
Posts: 872
Location: Germany, Berlin
drek wrote:
I haven't seen a good system for routing PPs to needy planets yet. In my mind such as system should a) be intuitive: obvious to the player exactly what is transpiring b) automatic, requiring no sliders, prioritization, or other inputs.

I thought of something based on effectiveness. As intrastructure defines how much a planet can produce, all planets which have less PP's get additional PP's equally after that PP's are transfered to planets which use them best (lower discount by overdriven production). Maybe some clever mind finds a far more better solution :-)

drek wrote:
Plus I just don't like the idea of having 50 different parallel queues to keep track of. If build projects are like research projects (x resources for y turns) then slapping everything into one queue would seem to make organization easier for the player.

Me, too. But i think research projects and building project aren't of the same kind. Reseach project gain advantage of all scientists all over the empire as they are virtual connected. Build project are based on what workers actually do. I prefere a model where build projects are limited by a single planet. To me it's not the same if you have 100PP on one planet or on ten. This way you can't pump out build projects. For example, you can't produce a starbase at a frontier world even if your empire has enough PP's to build it in one turn. It's limited by the planet infrastructure itself.


Top
 Profile  
 
 Post subject:
PostPosted: Thu Jun 10, 2004 6:21 pm 
Offline
Space Kraken

Joined: Tue May 25, 2004 3:15 am
Posts: 157
Sorry to be OT, but I think this whole empire building queue is about one thing: reducing mm.

The main inconvenient of having everything built locally is that you have to manage the developement of all your colonies. Also, a minor inconvenient is that you cannot rush things.

A solution would be a helpful, nonobstrusive AI (not like moo3). Such an AI would only help doing macro-orders, effectively enabling you to do your management in two phase: macromanagement and micromanagement.

A macro-order would have you select a list of planets, apply filters (in a sort+select/remove way) and then order construction of a building/ship/etc. The macro-orderer could generate a note to self or automatically propose upgrade when the place is ready.
Here is an example to illustrate:

Your research a enabled you to build research labs mark IV. Those are a no brainer structure: you need them on every planet doing research. You go in the macro orderer: you begin by sorting your world by safety (a user set, very low for border world, high for you capitol), you select the world that are safe enough for your taste and add them to the list (which was empty beforehand). Then you proceed to cull world you don't want the structure built upon: you sort by focus, and remove research inefficient focus (I don't know, farming and industrial for example). Then you remove low pop world. Now you have a list of every world you want a mark IV research facility built upon. You set a priority (a place in the queue so it doesn't disturb more pressing matters) of 50% (medium). You press a set button and the macro-orderer submit you a batch of order you can review and approve. You can remove specific planets there when there are exception.
To this point, the main difference with MoO3 dev plan is: you decided on the conditions to build the facility, it didn't obfuscate the underlying micromanagement, it just made it less tedious but you still see the details. It is also a punctual event in time: a building won't be built behind your back without you reviewing and approving. It is player empowerment without being annoying, which is very good in my book.
Next the macro orderer ask you (it could be a checkbox) to remind you later, when a planet change status and would have been in the list. In the example, the macro-orderer tells you in a SitRep that a world is now fit to build such an facility and the reason (what have changed): Spaceball would be ready to build a mark IV research facility because it as reach a population of 15. Build it (yes/no/ignore this macro-order for x turns).
It would be a list of all the macro-order that would apply be applied to worlds (that don't already have/produce it and fulfill the marco-order's conditions).

This and the ability to manage your macro-orders would allow players to crunch much more micromanagement. Of course, not everybody would want to develop such macro-order, but kit of ready-made one could be available for separately or with the main distribution.


Would somebody point me in the right direction (forum) to bouce this idea.


Last edited by vishnou00 on Thu Jun 17, 2004 5:33 am, edited 1 time in total.

Top
 Profile  
 
 Post subject:
PostPosted: Thu Jun 10, 2004 6:56 pm 
Offline
Juggernaut
User avatar

Joined: Fri Dec 26, 2003 12:42 pm
Posts: 872
Location: Germany, Berlin
Hmm, i don't thing it's off topic, at least not completely. Maybe you are somewhat late because as i understand "public review" means focus your already made suggestions but add nothing new.

the majority would say that a >no brainer structure< is handled by infrastructure and i also read that building a mark IV research lab will be done at the same time as it is discovered.

Over all, i go into the same direction as you do. Using local build queue and adding some marco tools to reduce micro. But others said that avoiding mirco in the first place is better. IMO that's in general true but when it's comes to buildings it isn't. In the end if we get those limited slots, we don't have to mirco that match at all.


Top
 Profile  
 
 Post subject:
PostPosted: Thu Jun 10, 2004 9:03 pm 
Offline
Space Kraken

Joined: Tue May 25, 2004 3:15 am
Posts: 157
Then I think the whole imperial build queue has only one goal: reduce the source of micromanagement. But is it the right solution? The bad thing is not the source of micromanagement (having infractucture), but the fact that it is no fun to manage that infrastructure.

Still, some of the infracture is no fun to manage because these building doesn't provide new gameplay elements: a library in a civ game doesn't add anything to the gameplay, it takes an elements (research done in a city) and change it a little (+50%). There is nothing new with 6 research point from 4 research points. I think all those building that simply modify a value should be abstracted in a number (like 5 research labs units). Tech that would enable more research would simply rise max research labs units. The buildings that fit in that category (the way I view it)(overly entensive list, I hope there won't be more than 4 or so):
-labs (+ to research points)
-factories (+ to production points)
-mines (+ to mineral extraction)
-farms (+ to food production)
-recreation centers (+ to happyness)
-military training fields (+ to military units experience)
-courthouse (- to corruption)
-spy agency (+ to spy ops)
-housing (+ to max pop)
-terraforming facility (+ to environnement)
-market (+ to money based economy)
-defence turret (+ to planetary defence)
-temple (+ to relogious anything)

The only reason to have pretty names (temple, colloseum, cathedral, market, banks, stock market, library, university, research center) is to present improvement in a non-abstracted form to make player confortable with the game concept (as in a Sid Meier game). If it is clearly explained somewhere that a "factory unit" (whatever the name) is an improvement to increase your production capacity, everybody is happy. Being abstracted that way (a number) it becomes feasable to devise simple and efficient macromanagement tools (as Stars! autobuild) in queue.

Buildings I don't fit in this category are building that enable new gameplay elements: a shipyard enables you to build (bigger?) interstellar ship, something you couldn't before. A mass driver or stargate (in Stars!) allow you new gameplay possibilities: hurling mineral packets at dangerous speeds or teleporting ships to the other corner of the universe, respectively. In this category, things should not be abstracted in numbers and each building keeps its individuality.

I don't think it is a new idea, it is basically slots (farm, lab, mine...) that can be limited (by focus?) but with infinite slots for not-abstracted-in-numbers, special-gameplay-feature-enabling buildings.

I'd say that the imperial-building-queue scheme is to building micromanagement what the DMCA is to copyright infringement : an easy way to deal with a problem (reduce number of building -- stopping research efforts in encryption reverse engineering) while avoiding to solve the real issue (eliminating the no-fun micromanagement of buildings -- stopping pirates).

To summarise: it is proposal 1a (that I desguise a bit as proposal 2 :roll:) and build methodology != 3 (I would go for build method. 1, provided the right macromanagement tools).


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 6:19 am 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 27, 2003 2:35 am
Posts: 383
Location: Texas
The imperial build queue is useful for one thing for sure. Building Ships. If we are going to have fleets of hundreds of ships we are going to have to build those hundreds of ships. We also need a central location to refrence for everything we are building. Thus the imperial Build Queue.

For the most part we have agreed that buildings will be rare. Thus there is no need to implement a new planet system such as a local build queue. There is nothing for hte player to do in an individual planet. You can't "tweak" the production of an indivdual planet, you can't rearrange population units, you can't "rush" build certain planets. There really isn't anything to do in the planet screen except look at how pretty your planet is and change the focus. Now lets say you need to build a Shipyard. Click planet click build shipyard, and viloa the shipyard appeaers in the imperial build queue, all planets spend their production on the shipyard (modified by their infrastructure rating and the planets infrastructure rating). A few turns later a shipyard is built. Seems simple to me.

When we start getting into complexities of multiple planets building multiple buildings and ships and wonders etc. We will have a problem about how to distribute production. This will be fairly difficult and complex but it's not an inherent weakness in our system of an imperial build queue.

Also I think the positives outweigh any of the complexities involved with refining it.

Possible Remedies:
As Drek has presented, X industry for Y turns. I don't like this as a hard and fast rule as far as production is concerned but it is viable.

A planet Radius: The planet/shipyard must be within X distance to contribute. If two projects are within range the closer one gets the production or a ratio. If equidistant than split the production?

FIFO: First in first out. The first thing in the planet queue gets all production until its done, then move on to the second and so on.

All of these (and more) are viable problems to the Imperial Build Queue. If you are still skeptical bring up your concerns and perhaps offer some alternatives.

_________________
Aquitaine is my Hero.... ;)


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 7:25 am 
Offline
Space Kraken

Joined: Tue May 25, 2004 3:15 am
Posts: 157
PowerCrazy wrote:
Seems simple to me.

The issue is not simplicity : both planet/system and imperial production model you order, it is built. The only difference is origin and destination of the production and one is no simpler than the other (from everywhere to anywhere and from somewhere to the same place)


A few year from now, we are playing a terrific FreeOrion PBEM game. With scanners I can see 75% of worlds. We have a truce and are waiting for the other to make a move, preparing to call allies for support. With the imperial build queue, you can pump ships from all your worlds in a few turn and their "rally point" is cleverly out of my scanner range (in the 25% of the planet I don't see). Then when you are ready, TADA, a mighty fleet pops into my scanner range without any warning and THEN I wonder "why can someone divert all his empire's production to a single world? I really should have convinced them in that building thread back in 2004!"

With a local build queue and efficient macromanagement tools here is how it may have happened: you call up the micromanagement tool, selecy a rally point and all the production sites (planet or system) you want to use. Then you order your ships: the tool calculate ETA to rally point of all the ship types and then produce ships at every planets for (time_to_finished_production - ETA) turns.

Returning to my hypothetical scenario: now, I can use my scanner intelligence, I can see a pattern, I have time to prepare, counter your simple ship pumping tactics, find weaknesses in your production web, strike the weak point ... in short: I can play a strategy game!

You can't always dumb down gameplay to reduce micromanagement. If coordinating the production and operation of hundreds of ships is a simple matter of ordering 100s of ships at one spot, we should just go back to tank rush à la C&C.


The solution I propose is based on a known technic: macromanagement tools. They are still very clunky, but we have a good idea of the do and don't. With an imperial building queue, we have a bag of tricks we haven't balanced and we don't know all the implications on the gameplay (but I don't like what I foresee). So, the inherent weakness of our system of an imperial queue is that we really don't know about its consequences on gameplay. The local building approach is tried and true, it is only a hell to micromanage with the wrong tools. I see the problem in the tools (the interface), not the gameplay.

The only positive I see is that it dumb down production/coordination to the point of needing no micromanagement, and the complexities to refine it are still there.

Not only it is unrealistic (but this is irrelevant) but it provides no better gameplay.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 7:34 am 
Offline
Designer Emeritus

Joined: Thu Jun 26, 2003 8:07 am
Posts: 935
Quote:
you can pump ships from all your worlds in a few turn and their "rally point" is cleverly out of my scanner range (in the 25% of the planet I don't see).


The amount if industry in the queue wouldn't effect the speed that projects are completed in, rather the number of projects that can be worked on at once. So, pumping out a whole new fleet would probably take more than a couple of turns.

The "rally point" would have to be an expensive shipyard.

I suggest that having a bunch of ships pop up all over the empire and having to move them to a single system in order to form the fleet is much more annoying than having a few expensive shipyard "rally points."


Quote:
and efficient macromanagement tools


No such thing. Name one game with *effective* macromanagement tools. Despite valient efforts by experienced teams of designers/coders, these sorts of tools have never met expectations, imho.

In the arena of macromanagement tools, expect failure. We can't depend on them.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 7:53 am 
Offline
Juggernaut
User avatar

Joined: Fri Dec 26, 2003 12:42 pm
Posts: 872
Location: Germany, Berlin
drek wrote:
The amount if industry in the queue wouldn't effect the speed that projects are completed in,...

IMO, that's not true. If you produce something big on a single planet it will take longer than producing the same on empire level. That's true because planet production is based on infrastructur and you are limited in hurry production. if the project need 100PP and the planet only produce 10PP's you need 10 turns. On empire level you might be able to produce it within one turn.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 8:13 am 
Offline
Designer Emeritus

Joined: Thu Jun 26, 2003 8:07 am
Posts: 935
erh,

Not following.

I'm thinking something like research projects. It's take 5 turns to build ship foo costing 10 minerals/industry per turn. It's going to take 5 turns regardless of how much extra industry is produced. Industry would be a capacity, so with 20 industry/turn you could build two of these ships at the same time.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 8:38 am 
Offline
Juggernaut
User avatar

Joined: Fri Dec 26, 2003 12:42 pm
Posts: 872
Location: Germany, Berlin
maybe i wasn't clear. I'm talking about producing f.i. ships on a planet vs. on empire queue.

say the ship needs 500PP's (maybe limited like research projects to 100PP's per turn). you have ten planets each producing 10PP's. this way you can produce that ship in five turns. if you produce the ship on a specific planet you need 50 turns. in this example after 50 turns both using empire queue and planet queue would have produced 10 ships, but with empire queue you got one ship every 5 turns and with planet queue all 10 ships after 50 turns. That's the difference!


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 8:49 am 
Offline
Space Kraken

Joined: Tue May 25, 2004 3:15 am
Posts: 157
A fleet is not a project, a ship is a project (unless I'm mistaken). A fleet is composed of many ships, so having the whole empire producing ships would effect the speed that the fleet is built. But this isn't the issue. The issue is that the production is doing faster than light (in fact instantaneous transportation) to the rally point. If you simply add a delay proportional to distance, there is still the problem of absolute stealth. It doesn't matter if it takes 5 or 50 turns if I only see it when its done.

In my example, I didn't address the "new borderworld ship gateway" issue, but the "just out of scanner range well developed outpost" issue. It is normal it has an expensive, it should be a place where you can build ships.

I don't say the "rally point" is annoying, I say it is providing lesser quality gameplay. As for the annoyance of having to move a bunch of ships to a rally point, it's then that you love and use the macromanagement tools.

I can't say I've seen it, but I can show you it has been done for other types of games. Take the simple problem of converging multiple units to a single point. In 4X games, you select every fleet/unit/stack and instruct them to go somewhere, one by one. What was the last RTS that forced you to this process? Dune II. Every RTS after (starting with the original Warcraft) had multiple unit selection and ordering, and their situation was worst than 4X: they had to plot path of colliding units in real-time (<50 ms). True it wasn't perfect the first iteration, but we have the same problem, multiple pathfinding with very little constraint: units don't collide and we have no hard time constraint.

[my unfactly vision of 4X development]
4X and strategy game is a niche market and because of it, developers don't have a lot of ressources. UI innovation isn't their primary focus, what they do is addings layers of complexity to the gameplay (for the best). That is where experienced teams of desigers/coders invest their valient efforts: developing and supporting (doing AI, interface) those new gameplay elements. Their various gameply AI helper are a desperate attempt to appeal to the mass market by taking loads of gameplay elements off the player.
[/my unfactly vision of 4X development]

They are not macro management tools as I see them: to me, Mm tools are converter. They have a simple macro order for input and they spit out micromanagement orders, they don't do non-trivial decisions, they only do path, filter by certain criterions, evaluate quantities (ETA, fuel consumption), do simple predictions and apply orders to unit/planets. I can't name a game that did it entensively but can't name one that attempted it.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 1:48 pm 
Offline
Lead Designer Emeritus
User avatar

Joined: Thu Jun 26, 2003 1:54 pm
Posts: 761
Location: Austin, TX
I'd like to put the brakes on this thread before it turns into a rehash of the DESIGN thread.

Before you post any further, identify exactly what it is you are either clarifying or asking to be clarified, and make sure it hasn't already been done. This is the point at which we start repeating pages 2-4 in pages 7-9.

Over the next few days, I am going to go over this thread and either figure out which proposal had the most support, or glue the proposals together into a quasi-proposal.

_________________
Surprise and Terror! I am greeted by the smooth and hostile face of our old enemy, the Hootmans! No... the Huge-glands, no, I remember, the Hunams!


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 3:19 pm 
Offline
FreeOrion Lead Emeritus
User avatar

Joined: Thu Jun 26, 2003 6:23 am
Posts: 885
Location: Australia
Two issues I see with this review:
1. When choosing what to build and where, will we
a. more often know what we want to build rather than where it should be built (ie, is the building more important than the location) or
b. more often answer where something is going to be built at the very same time we answer the question of what is being built
Working out which of these is the common factor I think will help answer whether an empire queue is the best answer, and how such a queue would work.

2. Tied in with the first one, is the problem of instant appearing buildings
a. Buildings appear instantly at their location once placed from an empire queue, keeping opponents in the dark and causing instant defences
b. Buildings have a delay, giving opponents a chance to gather intelligence and prepare an invasion
This is a question of how the empire queue would work if we decide that's what we want.

Though I feel like there are a great deal of small differences in how we understand parts of each other's vision of this portion of the game which is preventing us from answering these questions properly.


Top
 Profile  
 
 Post subject:
PostPosted: Fri Jun 11, 2004 5:38 pm 
Offline
Space Krill

Joined: Fri Jun 11, 2004 2:46 am
Posts: 2
Location: Grants Pass, Oregon
Maybe it is a little late for my vote, but I would vote for proposal 1a with build methodology 1 or 2.

The way I would implement the developments plans is to have them be simple build lists, similar to what Stars! had. You could setup as many of these plans as you want. They would simply be a list of what you want constructed and in what order. You could have one list for your core industrial worlds, a different list for your farming worlds where you don’t need all that industrial stuff, and so on. When you colonize a planet you select what plan to use, or if this is a world that you really want to micromanage, you don’t select a plan.

When you research something new, you would get prompted, asking if you want to add this building to your development plans, then you could select the plans you want it added to and you’re done.

A world would only go to the development plan if it had nothing in its queue. This way if you had a dozen worlds with one plan, but you wanted a starbase at one of them, but not the rest, you just go to that planet and add the starbase to its queue. When the starbase is done, the planet will just go on producing what ever was next in the development plan.

For handling ship production, I would have an empire wide ship construction queue. When you want a ship constructed, but don’t care where, you put it in your ship queue, and the next available world, capable of constructing that ship would take it out of the empire queue, and start building it. If you want a certain ship made by a certain world, you simply go to that world and place that ship in its queue. I would also have some kind of check box for each world to tell it if you want it to produce ships or not, and maybe some way to tell a world, that if it cant produce a ship in X number of turns, don’t take it.

Brian


Top
 Profile  
 
Display posts from previous:  Sort by  
Forum locked This topic is locked, you cannot edit posts or make further replies.  [ 104 posts ]  Go to page Previous  1 ... 3, 4, 5, 6, 7  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group