Public Review: Tech Tree I

Past public reviews and discussions.
Message
Author
Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#76 Post by Tyreth »

I have problems with this system, but I admit my contentions could be illusory.

If I'm researching, then I will probably try to maximise my projects so that the bare minimum of points is left over. Now what's left over is almost always going to be less than the minimum for another project. So I'll have only a small flow of RP going into refinements. Refinements should not produce advantages beyond full projects for (much) less cost.
That means practically that I'm not going to care about refinements if I'm always focussing on projects. After all, if I put all my points into full projects, then the leftovers in refinements will never produce an item in usefullness beyond what my current base projects give me.
The only option is to deliberately leave points left over for RP's so that they can have enough to improve by a useful amount. If I make this conscious decision, it's going to be because of a specific refinement, so I may not want to split the points. I may want to focus in a particular one. It seems to be introducing a new section unecessarily. We can force arbitrary restrictions, but essentially it is a separate area that we are trying to make look like one.

I would be just as happy if excess RP is automatically converted straight into money (if we have money) or some other commodity. That seems the simplest solution. In the status bar for the empire wherever that's located we show number of RP's being generated, and the amount of money that is produced from excess. That way the player can view quickly every turn what's not being used. Then we have refinements as being projects just like anything else, using the same system.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#77 Post by utilae »

Your right. It is a bit strange if you have to deliberately ignore researching projects to make refinements, rather than actively researching refinements.

Excess rp should just be put back into research. SO it goes into some kind of research pool, that is used to ensure that the rp going into projects each turn always meets the minimum rp needed. Though I can't ever see how rp flow will fall below the minimum, as the research system does not allow for excess rp to be put to good use. Why can't you just put all the excess rp into the current project, I mean you can afford to. Thats the downfall of this excess idea.

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#78 Post by Tyreth »

Aquitaine wrote:On category names:

My arguments in favor of them:

- Usability (not UI, but related). If I'm an FO player and I haven't been working on the project at all, and I think I really want to research corvettes or shields or a new monetary system, I have to guess which of Tyreth's categories those are in; and in many cases, there is more than one possible answer; perhaps most weapons are in a 'physics' category. What if I have one weapon that's a biological weapon? Should it be in phyiscs or biology? This 'organization by overlying scientific principle' seems counter-intuitive to me.
Such a problem becomes meaningless when we have an in-game encyclopedia like MOO2 or SMAC, where it is simple to see what tech a certain item requires. Of course, it will always be easier for the new player to find corvettes if there is a corvette category, but I'm not sure that this loss of navigation is really a problem. Part of the joy of a game like this is exploring the tech tree, and finding what leads to what. I find it more immersive than having specific titles.
This seems akin to naming weapons after their exact damage level rather than a nicer name. Eg, Laser IX instead of HyperDeath Laser. Players will learn what belongs to what.
- Modability. If I don't like part of the tech tree and want to completely redesign it, the use of specific categories in the parent FO design allows me a lot more flexibility. If we design our UI to support 5 or 6 very broad categories, then modders will have a more difficult time if they decide they want specific categories (between 8-12 seems to be our target number). This is not a reason to do anything in an of itself - obviously a modder can change the UI - but it is a consideration.
I'm not really following this argument. Are you saying that the dependancies of certain techs is a lot harder to follow under a general naming scheme? Whereas is a player wants to add a new tech then it's easier to find with specific categories? If so, I'm having trouble visualising that because the HoI tech tree seemed to have just as complicated dependancies as SMAC. Though I didn't play the game enough to say that absolutely.
As for modders using specific categories - I'd hope that if we use general categories, we do so because we decide it is a better solution. That means modders wanting specific categories would be doing so against what we believe is a better game design. The same problem would have the same problem converting the tech tree if we use specific categories and they decide they want simple ones.
To refute Tyreth's arguments:
1. It allows us to come up with creative advancement names like "Super Tensile Solids" rather than "Super-Surreal Laser". That way we can give the genuine feel that the player is advancing through theoretical research in a real universe, discovering its secrets. I really appreciated the effort of names and quotes that went into the SMAC tree.
2. Making theoretical advancements really about the underlying scientific advancement, rather than the direct application, lends itself naturally to a more intertwined tech tree. Some may not like that, I prefer it.
I think I already argued against #1 - we can name theories and advancements whatever we want. I don't see a 'game immersion' factor anywhere in here, because I don't see that choosing broad or specfiic categories ties us to using 'better' sci-fi names for things. I agree that Super-Tensile solids is a better name than Super-Surreal Laser. But I don't see that it's relevant to this discussion. But FWIW, I think the quotes in SMAC's tree were pretty great, and I'd love to do something like that. :)
Theory does not lead to a specific type of application. A single theory discovery may open up applications in multiple areas when using the specific categories. How do we decide whether a particular theory belongs in the ship research section, or if it belongs in social, if it produces an application to research that belongs to both fields?
Though I suppose that instead of having one theory directly produce an application in two areas, it can produce an application in the area which we place it, and open up a theory in the other that can lead to the application. Though this seems to be adding an arbitrary step.

The crux of my argument is that the nature of a theory->application model lends itself more naturally to general category names. General categories are areas of scientific research which make breakthroughs, and those breakthroughs result in applications. Specific categories are more akin to applications, and these applications lead to theory in order to produce them. Seems like backwards.
Though this does remind me of the fact that people once invented out of a need, and now we invent because we discovered something new. Specific categories lend themselves to an older view of invention, "We need a way to defeat our enemies in war, so lets find a method to make our swords sharper", whereas now we tend to follow the opposite process, "We found a way to store light inside of diamonds, how can we apply this?".

I must admit after saying that, that either method is just as good. I feel though that general categories lends itself just a little better to the theory->application model.

Aquitaine
Lead Designer Emeritus
Posts: 761
Joined: Thu Jun 26, 2003 1:54 pm
Location: Austin, TX

#79 Post by Aquitaine »

I think you are improperly predicting how the 'specific categories' would develop.
How do we decide whether a particular theory belongs in the ship research section, or if it belongs in social, if it produces an application to research that belongs to both fields?
This rarely, if ever happens in HoI, and when it does, the distinction is obvious; I also can't think of any instances where in application in category A depends on a theory from category B. It is, as far as I can tell, always the cass that an application in category A requires a theory in category A and (possibly) an application in category X (which, of course, requires you to have the parent theory from category X, but that's not quite the same thing).

Now, we don't have to hold to this and could tie theories to each other however we want, but I don't see the problem you mentioned as being specific to this system. If all ray guns are in 'physics' and we have a biological ray gun, does it make sense for it to go in biology?

Also, FWIW, even with specific categories we would not have 'corvettes' as a category - we would have 'Light Capital Ships' or something to that effect.
Such a problem becomes meaningless when we have an in-game encyclopedia like MOO2 or SMAC, where it is simple to see what tech a certain item requires.
I disagree entirely. My first rule of usability is 'don't make me think.' You're saying 'in order to figure out what category corvettes are in, you must load up the encyclopedia and find corvettes, then go to the tech screen and click on the category.'

You are also presuming a feature we've never discussed, which, I'll grant, may be a fair presumption (I'd certainly like one). But I don't think it refutes the argument. If you have a 'light capital ship' category then the only thing you have to know is whether a corvette is a light capital ship or a starfighter or a heavy capital ship. With general categoies, you have to figure out if it's physics or construction.
This seems akin to naming weapons after their exact damage level rather than a nicer name. Eg, Laser IX instead of HyperDeath Laser. Players will learn what belongs to what.
I think your notion of our specific categories is too specific. :) I don't like Laser IX any more than I liked your last example of an overly-specific theory. I encourage you to download C.O.R.E. for HoI and look through their tech tree -- the actual progress of science that I think is what you're after is very much present there. It's just laid out in a manner so that it's pretty easy to find what you want, and the tree is large enough that you still have a lot of fun exploring it.
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!

Aquitaine
Lead Designer Emeritus
Posts: 761
Joined: Thu Jun 26, 2003 1:54 pm
Location: Austin, TX

#80 Post by Aquitaine »

Tyreth wrote:I have problems with this system, but I admit my contentions could be illusory.
I think so, but I need to be convined as well. I'm currently in the camp of 'refinements as spare RP is good' but this is a good discussion to have.
If I'm researching, then I will probably try to maximise my projects so that the bare minimum of points is left over. Now what's left over is almost always going to be less than the minimum for another project. So I'll have only a small flow of RP going into refinements. Refinements
This is where I think we're getting off track. If you're researching and there is no benefit to having spare RP, then yes, of course you'll maximize (and also maximise) your RP allocation. But if we set this system up the way we've discussed, we can't do it in a way that makes players go 'oh, I have 9 RP left over. I wonder if I can refine anything,' but rather one that makes them think 'I must consider allocating some RP to refinements for X reason,' so that is part of their decision. It is really less 'allocating your spare' than it is 'dividing up your RP' - all your RP are always going somewhere, so they're always maximized. It's basically just a slider without the actual slider.

If you don't have anything to refine (or don't want anything), then we could funnel the excess RP back into the economy as money, with a penalty.

I guess I don't see it as 'deliberately ignoring research projects' any more than I see increasing your production in HoI as deliberately ignoring research since you're just re-allocating things.

However, all of that said, I would be fine with a system where refinements are simply applications. What I don't want is what utilae's suggestion (sorry!) of a 'research pool' that is basically a gargantuan exception to our regular rules, because they're two different scales. Think of it this way:

Project A costs 20 RP for 30 turns. Your empire produces 29 RP. You have 9 RP/turn free. Project B costs 15 RP for 30 turns.

How do you send this 9 RP into a pool that is useful? Do you 'accumulate' RP? This seems to go against the whole notion of 'fixed cost per fixed time.'

So, personally, I think we should either make everything fit into the theory/application model and dump excess RP back into the economy, or use the previously discussed method of excess RP going into refinements. The second will 'feel' better to the player as there is hardly ever any RP wasted, but this is only a minor bonus, although I like the seamlessness of it.
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!

tzlaine
Programming Lead Emeritus
Posts: 1092
Joined: Thu Jun 26, 2003 1:33 pm

#81 Post by tzlaine »

I have objected before to making two research systems when we don't need to, and I restate my objection here.

Can someone please explain to me the advantage of having a per-project yes/no model for researching thoeries and applications, and having a semi-automated category-queue percentage-spending model for refinements? I have asked this several times now, and no one has responded.

I'm not trying to provoke anyone. I really just don't understand the motivation for two systems.

If the reason is decreased micromanagement, I believe this is almost a non-issue with regard to research. Micromanagement becaomes an issue when you have lots of things to micromanage (like 100 planets), whereas there is only one technology screen. I have also proposed a reasonable queue-based UI to make all the research projects more easy to automate, not just refinements. See previous posts of mine in this thread for details.

What I'd like to see instead is that we handle refinements just as any other research item, though refinements are sure to be shorter in time and/or lower in cost than other research items, like theories and applications. This is simple, and makes sense to me.

As for excess RPs, I favor one of two nearly-equivalent systems for dealing with this:

1) You can store up excess RPs, and when you have enough for a project you can start spending on that project at the normal rate. So it will still cost X RPs for Y turns, but you just happen to have all X * Y RPs at the outset of the project.

OR

2) You can start one extra project that you can't quite afford, and every time you accumulate X RPs, one turn of progress is made on it. So in essence, you can partially fund one project to eat up the excess RPs. It will just take more than the usuall Y turns that it takes to research the project at full funding.

In both of these scenarios, you are able to suspend, cancel, etc., the "excess" project.

This way of doing things keeps the model simple and consistent. KISS. It also avoids wierd conversions of RPs to PPs or cash or whatever. Allowing such conversions opens the door to gaming the system. You could potentially make more money than you were ever intended to by building up your research infrastructure. This makes our jobs more difficult as game-balancers.

Shadow Knight
Space Krill
Posts: 2
Joined: Wed Mar 03, 2004 7:38 pm

#82 Post by Shadow Knight »

Hey my first post here. First I wish to put lavish praise on all the people working on the game, whether designers, artists, programmers, etc., excellent job. I have been lurking on and off for about a year and am most impressed. Okay enough of the praise.
emrys wrote: FO is going to have a tech model based mostly on HOI. So far this means things like:
  • Research projects will take a number of research points for a number of turns, not just a total number of RP's. (already passed)
  • More than one project will be able to be researched at one time. You will (probably) have to have the required number of RP/turn available to start a project. (already passed)
  • We need to think what to do about the spare RP's left over each turn (either another use for them (is this the place for refinement?) or bear in mind that they will need to be easily redirected back into the economy somehow.) (open question)
As one of the modders for TGW, (The Great War) for HOI over on Paradox's forums, I was one of the team members to work on the tech tree for that mod, so I completely agree with your decision to use it.

I liked the way HOI just had the extra go back into the IC (or in this case PP).
[*]Tech will be split up into theoretical advances (mostly serving as level markers with limited practical benefit) and applications (and refinements).
Theoretical advances will unlock groups of applications (i.e. each level has a bunch of applications). (mostly passed - refinements still open)
[*]Theoretical advances mostly depend on the theory before them.
Practical applications will have prerequisite applications which must be developed before them, often from different branches. (already passed)
[*]We would prefer a broad tech tree with multiple viable routes through.
Categories should be as directly related to gameplay and particularly strategy as possible (i.e. no 'mathematics' category, but 'Ships' might be o.k.) (open question)
I agree with the broad tech tree with multiple viable routes. As for things like 'mathematics' that could fall under something like a Science category, where math would be a theory level and various types of math would be applications (such a algebra or geometry for example), and more advanced mathematics could be found further down the tech tree as their prerequisites are researched.
[*]Research cost will scale with galaxy size. (open question)
As other posters said, perhaps a scaling factor that can be chosen at game startup, but not tied into the galaxy size.
[*]Randomly block out different techs each game or not?
How about randomly blocking out those refinements that are being discussed. So every game has your core tech tree, but the refinements available changes (where refinements are a much smaller RP cost and smaller time to research than a normal application). This way the player can always get say their plasma cannons, but not necessarily the MkII Plasma cannons that they owned with last game (note: I have no idea if a MkII Plasma cannon would actually own).
[*] Many categories (10+) or fewer (6-8 )?
I would not go above ten categories. Here are some off the top my head that I would consider:
Science(This is kind of the same as gneral research that offers real small bonuses to most things, but is more for pre-requisites)
Biology (agriculutre improvements, genetic engineering, etc.)
Engineering (This is where you would research ship systems, better armor plating, maybe vehicles for ground combat, Spy equipment)
Capital Ships (cruiser class and up)
Smaller ships (everything up to and including destroyer class vessel) Industry (Research that effects industrial efficiency, and mining efficiency)
Weapons (both for ship and ground combat)
Doctrines (Same concept as HOI, implemented differently of course)
Goverment/Social (Improvements to government, social tech changes)
[*]Will it be possible to develop applications without knowing the theory, if all the prerequisites have been met, maybe at some kind of penalty?
This seems to be bypassing the rules of the tech tree so I would say no. However, I would say an event such as: "You have found an ancient ship wreck" could bypass this, or if tech capture is in you can acquire a more advanced application (not theory) by conquering an alien world. But there should be a penalty to using the technology (refinements can't be researched for it till the theory is, etc.)
[*]Should we have techs that reduce the cost or time of research generally (or for particular categories). ? (Nobody ever suggested we shouldn't in the game design thread, and it'd be silly to legislate against it. The answer to this question is 'yes, as game balance needs it, to be determined later. The possibility for techs like this will be built in.' If anyone really disagrees, post here, otherwise this is simple enough that I don't think we need further debate on it.)
Why not, but don't put them all in one tech category. A good trick in HOI if you were a great power was to race down the electronics tree so you could get the computers to reduce the cost and time to research new techs.
[*]Do we want to repeat the outline of the tech tree to extend the research game?
If you make the tech tree big enough (i.e. unlikely for but the most powerful/largest empires to research everything in one game.) this should be superfulous.
[*] Should we try to hardwire choices with race bonuses to particular techs, or try and make it dependant only on strategic position.
Well you could have some start off with hard wired doctrines (such as race A prefers carries, thus fighters, over huge gunships like battleships) that give bonuses to the preferred way of fighting. Also some races like the Psilions from the MOO series could have built in modifiers that reduce the time or the cost of research (one not both in my opinion).
[*] Do we want to achieve multiple paths by balancing all techs perfectly, or by deliberately introducing several 'prefered paths' or tech families? AND How can we produce genuinely alternative gameplay styles and support this with the tech system? AND How do we do this whilst keeping things relatively simple and fun?
I would try to balance the techs as mcuh as possible. Thus if I choose the path of big ships, but the other player chooses the path of swarms of fighters. Neither should be inherently better, but have built in pros and cons.
[*]What happens to research on projects you can no longer afford, or want to cancel? [/list]
If you cancel a project then all the work you have put into it should be lost (unless you can somehow save the data that you started a project and researched it for ten days, and then if you start-up the research again perhaps get a small deduction in the time it will take to complete. But I don't know if that would be worthwhile.)

As for projects that you can no longer afford, but are still in your que it should be done similiar to HOI. They remain in your que, but being unfunded the time to finish does not count down. And as the techs above it finish being researched the unfunded tech moves up the que and if funds are available resumes research. A put on hold button would be nice also so it freezes the tech at its current stage, and can be resumed later. This is useful in the case of where you are researching two techs of moderate cost, then one finishes and has opened up an expensive tech to research. By putting the other moderate one on hold you can now research the expensive one, and then resume it later when the expensive one is done.

Aquitaine
Lead Designer Emeritus
Posts: 761
Joined: Thu Jun 26, 2003 1:54 pm
Location: Austin, TX

#83 Post by Aquitaine »

tzlaine wrote:Can someone please explain to me the advantage of having a per-project yes/no model for researching thoeries and applications, and having a semi-automated category-queue percentage-spending model for refinements? I have asked this several times now, and no one has responded.
The best reason I've seen for it is that you are never wasting RPs. However, it is possible to use this argument in favor of the opposite, since if you funneled RPs back into money or some other part of the economy, you'd still never waste them, but it's not as effective; this way, all RP is always being used for some kind of scientific progress.

It's not really just 'micromanagement' - it's more of a containment mechanism so the research system doesn't spill over into the other systems. I don't personally see that kind of spillage (funneling) as a bad thing, necessarily, but I do consider the containment to be more elegant than simply converting your excess RPs into something.
1) You can store up excess RPs, and when you have enough for a project you can start spending on that project at the normal rate. So it will still cost X RPs for Y turns, but you just happen to have all X * Y RPs at the outset of the project.
But this is also a wholly separate system and violates KISS at least as much as what you're arguing against. We are saying 'our research system involves items that have a fixed cost of X RP per Y turns because you can't just buy it all at once.' If we did as you suggest with #1, we would have to keep track of your spare RP through a separate mechanism and allow you to spend it through a separate mechanism.

I think, if we do as you initially suggest and simply agree that parallel systems are not worth the trouble, it's best to just convert RP back into something else (perhaps money.)
2) You can start one extra project that you can't quite afford, and every time you accumulate X RPs, one turn of progress is made on it. So in essence, you can partially fund one project to eat up the excess RPs. It will just take more than the usuall Y turns that it takes to research the project at full funding.
I like this better than #1, but it still complicates the system and the interface and it's not as clean as saying 'all RP you don't spend goes into this system; all RP you do spend goes into this sytem' since you still have to pick a project to partially fund and it will be one thing on the entire research screen that is different than every other thing.

I guess what I'm saying is, in my book, if we establish two systems with simple rules, that is breaking KISS less than one system with rules we then break.
This way of doing things keeps the model simple and consistent. KISS. It also avoids wierd conversions of RPs to PPs or cash or whatever. Allowing such conversions opens the door to gaming the system. You could potentially make more money than you were ever intended to by building up your research infrastructure. This makes our jobs more difficult as game-balancers.
I don't agree. I think this brings up images of 'housing' and 'trade goods' from MOO2, but we'd simply penalize this backwards-conversion enough so that you just didn't really feel bad when you had 15 RP not going anywhere, rather than actually letting it be a source of income.
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!

tzlaine
Programming Lead Emeritus
Posts: 1092
Joined: Thu Jun 26, 2003 1:33 pm

#84 Post by tzlaine »

Aquitaine wrote:
tzlaine wrote:Can someone please explain to me the advantage of having a per-project yes/no model for researching thoeries and applications, and having a semi-automated category-queue percentage-spending model for refinements? I have asked this several times now, and no one has responded.
The best reason I've seen for it is that you are never wasting RPs. However, it is possible to use this argument in favor of the opposite, since if you funneled RPs back into money or some other part of the economy, you'd still never waste them, but it's not as effective; this way, all RP is always being used for some kind of scientific progress.
I feel that the problem of wasted RPs and the means of doing research are orthogonal. You can have 50 different ways of doing research and still have some wasted RPs, or have a unified way of doing research and have no waste, etc. The fact that someone thought of an idea that happens to consist of two ways of doing reseach and no wasted RPs doesn't mean that to prevent waste we need to have two research systems.

It's not really just 'micromanagement' - it's more of a containment mechanism so the research system doesn't spill over into the other systems. I don't personally see that kind of spillage (funneling) as a bad thing, necessarily, but I do consider the containment to be more elegant than simply converting your excess RPs into something.
I agree completely that spillover is a bad thing. I'd like to keep the research system self-contained, if at all possible.
1) You can store up excess RPs, and when you have enough for a project you can start spending on that project at the normal rate. So it will still cost X RPs for Y turns, but you just happen to have all X * Y RPs at the outset of the project.
But this is also a wholly separate system and violates KISS at least as much as what you're arguing against. We are saying 'our research system involves items that have a fixed cost of X RP per Y turns because you can't just buy it all at once.' If we did as you suggest with #1, we would have to keep track of your spare RP through a separate mechanism and allow you to spend it through a separate mechanism.
This is not a wholly separate system at all. To start a "normal" research project, you need to have X RPs of per-turn capacity, and spend that same number of RPs for Y turns. To start an "excess" research project, the only difference is that instead of needing the RP capcity per-turn, you need it all up front. You still spend the same number of RPs per turn, just as you would whether the project is "normal" or "excess". The place that the RPs are drawn from changes from per-turn capacity to your excess RPs, but the project can be suspended and/or resumed at any time, just like any other project. Adding the "excess" projects becomes just like adding "normal" ones -- after you've allocated all your per-turn RPs, you then allocate all your excess RPs. The difference to the user is largely trivial, whereas the proposed two-model system is vastly different in both conception and execution.

I think, if we do as you initially suggest and simply agree that parallel systems are not worth the trouble, it's best to just convert RP back into something else (perhaps money.)
I disagree, because I think conversions are a bad idea. However, I would be fine with just letting excess RPs go bye-bye.
2) You can start one extra project that you can't quite afford, and every time you accumulate X RPs, one turn of progress is made on it. So in essence, you can partially fund one project to eat up the excess RPs. It will just take more than the usuall Y turns that it takes to research the project at full funding.
I like this better than #1, but it still complicates the system and the interface and it's not as clean as saying 'all RP you don't spend goes into this system; all RP you do spend goes into this sytem' since you still have to pick a project to partially fund and it will be one thing on the entire research screen that is different than every other thing.
I agree that under #2, one item will be very different from the others. This is actually why I favor #1, which seems far simpler to me.

I guess what I'm saying is, in my book, if we establish two systems with simple rules, that is breaking KISS less than one system with rules we then break.
I disagree. Two completely different ways of specifying your research goals are not simpler than one way with one easily-understood exception.
This way of doing things keeps the model simple and consistent. KISS. It also avoids wierd conversions of RPs to PPs or cash or whatever. Allowing such conversions opens the door to gaming the system. You could potentially make more money than you were ever intended to by building up your research infrastructure. This makes our jobs more difficult as game-balancers.
I don't agree. I think this brings up images of 'housing' and 'trade goods' from MOO2, but we'd simply penalize this backwards-conversion enough so that you just didn't really feel bad when you had 15 RP not going anywhere, rather than actually letting it be a source of income.
The question then becomes "How big a penalty to we impose?". If the penalty is too small, it remains possible to game the system. If the penalty is too large, we might as well let the RPs go to waste, because with either wasted RPs or a large penalty, people will still be encouraged to micromanage their RP usage, which is ostensibly what we're trying to prevent.

I would argue that any arbitrary conversions between RPs, PPs, food, money, etc., make the game harder for us to balance, by introducing ways for people to jack up one of these attributes to unreasonable size by using more than one of them and converting.


To summaraize, I don't think that excess RPs are a problem. If you don't like waste, don't be wasteful. If that leads to micromanagement, a decent interface should fix things. I should add that even if it doesn't, micromanaging research is not that big a deal, because it's just one thing. Micromanagement becomes a big issue when you have 100 planets or whatever.

However, if everyone insists that there be no waste, there are ways of ensuring zero waste that do not involve completely conceptually different research models. I have suggested one (#1) and so has noelte (#2).

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#85 Post by utilae »

Aquitaine wrote: However, all of that said, I would be fine with a system where refinements are simply applications. What I don't want is what utilae's suggestion (sorry!) of a 'research pool' that is basically a gargantuan exception to our regular rules, because they're two different scales.
It seems to me that the excess rp going into refinements is the best idea. I think you need to make sure that the player knows whats going on. They need to know how much excess rp they have in refinements. Whether excess rp this turn is going into refinements, etc.

Now, refinements. How does everyone see refinements.

A)
'Level 1 laser' and 'Level 2 laser' are both seperate application techs. 'Level 2 laser' is a refinement of 'Level 1 laser'. So refining 'Level 1 laser' makes the application 'Level 2 laser' become available.

or

B)
'Laser' is an application, that starts at level 1 in the tech tree. The 'Laser' tech can have its level increased through refinement. So refining 'Laser' increases its level from 1 to 2.

Now which way of dealing with refinement does everyone think is better. And I would also like to know from a programming side which is better.

Now I think option B is better from both sides. See, from a programming side level would be a variable in the tech object. So the tech object laser could have its variable 'level' increased as the tech is refined.

We can do some cool things with this too. Like, at a certain level the tech 'laser' would learn an ability. So at level 10, lasers have 'Armour piercing'. That would be cool. Also as the levels increase, other stats (damage, range, etc) should increase, but they should have the same pattern as their intial values. I mean if damage is rated 5 points higher than range, well a few levels later, their should still be a 5 point difference.

Aquitaine
Lead Designer Emeritus
Posts: 761
Joined: Thu Jun 26, 2003 1:54 pm
Location: Austin, TX

#86 Post by Aquitaine »

The exact details of what refinements do aren't really part of this review. Sorry. :?
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!

Sandlapper
Dyson Forest
Posts: 243
Joined: Sat Nov 01, 2003 11:50 pm
Location: South Carolina, USA

#87 Post by Sandlapper »

tzlaine wrote:
This is not a wholly separate system at all. To start a "normal" research project, you need to have X RPs of per-turn capacity, and spend that same number of RPs for Y turns. To start an "excess" research project, the only difference is that instead of needing the RP capcity per-turn, you need it all up front. You still spend the same number of RPs per turn, just as you would whether the project is "normal" or "excess". The place that the RPs are drawn from changes from per-turn capacity to your excess RPs, but the project can be suspended and/or resumed at any time, just like any other project. Adding the "excess" projects becomes just like adding "normal" ones -- after you've allocated all your per-turn RPs, you then allocate all your excess RPs. The difference to the user is largely trivial, whereas the proposed two-model system is vastly different in both conception and execution.
You say "Adding the "excess" projects becomes just like adding "normal" ones"; if it's that easy to reassign excess rp's, why can't an "excess" project equal a refinement?

Have refinement simply be a finite branch attached to the application's tree. The refinement branch can be just one or a dozen refinements of any specific application. As discussed, refinements would be a fraction of the application's initial cost and time. Excess rp's simply rollover to refinement project until complete. Have a sitrep report completion, either micro next refinement, or if no response, AI choses next available refinement in category.

I don't see why a refinement should be seperate from the tech tree, just treat it as another application with a previous application(albeit a specific, like application) as a prerequisite, but is fractional in cost and time.

User avatar
utilae
Cosmic Dragon
Posts: 2175
Joined: Fri Jun 27, 2003 12:37 am
Location: Auckland, New Zealand

#88 Post by utilae »

Sandlapper wrote: I don't see why a refinement should be seperate from the tech tree, just treat it as another application with a previous application(albeit a specific, like application) as a prerequisite, but is fractional in cost and time.
Or a refinement could be an application (the same that you researched) with its level increased by 1. :o

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#89 Post by Tyreth »

I'm not terribly attached to general categories, but I still do favour them. I'm happy to go with more specific titles if that's the way everyone else is thinking. As you've explained, it can work just as well.

Regarding refinements, I still have problems. I now understand your position better - you were expecting a balancing between full projects and leaving RP over to build refinements. I was still envisaging refinements as a solution for the problem of excess RP, and therefore assumed players would only want to leave excess RP to them.

I like the two as Zach listed, but I prefer #2 to #1. It seems simplest. Assume we allow beginning production of projects that have no RP available to begin, just like HoI (I know we may not do it this way). Now as it stands, in a game of 50+ planets, the chances of RP fluctuating are very high. More often than not it will be getting higher. Either way, it's very unlikely to stay static for more than one turn, especially later in the game. I don't want to have to visit the research screen every turn.
If we allow queueing, then I can stack up projects for a number of turns ahead. If we use #2, then I know that my excess RP's are not wasted, but are going into the next project on the list without needing to think about it. I know if my research goes down I don't need to worry about it. If my research goes up, I know that it will automatically allocate the RP to the right project.

I don't like #1, mostly because I don't like the idea of stockpiling RP. #2 seems cleaner. There wouldn't need to be extra UI to handle option #2 either. If a project has no RP going into it, then highlight the RP's needed each turn in red. If it's only getting part of what it needs, highlight it orange. If it's getting the full quota each turn, green. That way you can see at a glance, and you know where your RP's are going.

emrys
Creative Contributor
Posts: 226
Joined: Fri Oct 24, 2003 3:44 pm

#90 Post by emrys »

Tyreth wrote:
Theory does not lead to a specific type of application. A single theory discovery may open up applications in multiple areas when using the specific categories. How do we decide whether a particular theory belongs in the ship research section, or if it belongs in social, if it produces an application to research that belongs to both fields?
The way I thought HOI model we were following essentially worked was as follows:

Each category has a straight line path of theory through it. E.g. if we have a category called 'Physics' then Physics theory 1 opens up physics theory 2 , which opens up physics theory 3 etc. obviously with better names! (we might want to complicate this a little by having a 'mini tree' of theories within each category, but probably not)

A theory opens up one/many applications. If more applications are in a single different category than are in this one, then the theory is in the wrong place. It is quite possible for certain applications to require theories or applications from more than one category as a prerequisite, but it is rare for a theory to need much but the previous theory in the same category for it to open up, and if it did, these would likely to be items from several levels earlier in the other categories (limited interdependance).

This allows you to research one category at the expense of the others for a fair distance before being held up by lack of breadth, even if you don't get all of the benefit of each level you go up without that breadth (i.e. not all the applications from each theory open up).

The choice between specific and general categories has nothing to do with the names, and all to do with the gameplay effects. It essentially comes down to this, if I say that your opponent has the following levels in each tech category:

A 10
B 12
C 17
D 11
E 5

Do you say

1) hmmm, level 17 in C, but behind in E, that means my best tactic for dealing with him is to ... (something like "attack first before he can use that industrial advantage" or "avoid fighting his ships because they're good but try to bombard his planets because he has poor defenses" or "send in the sabateurs, because he has poor counter espionage, and avoid war at all costs because he'll wipe the floor with me". If so, then you have specific categories with limited crossdependancies.

2) so, he's somewhere between level 5 and 17, averaging about 12, he's proabably a bit more advanced than me, so maybe I'll avoid him for a bit. If he's got level 17 in that category, he might have thing X, so I'll steer clear / send in the spies. If that's all you can get out of this information, you have general categories with limited cross dependancies.

3) "how the hell did he get to level 17 in C without at least 13 in everything else?" - in this case you have strong cross dependencies (and so the difference between general and specific categories is irrelevant, since). You might as well have no categories at all because they're pointless because all you can really tell about someone is that they have an average tech level of whatever (a complaint that probably applies to option 2 as well) and you're probably playing a game where research is about as much fun as it was in Moo3 :wink: .
Specific categories lend themselves to an older view of invention, "We need a way to defeat our enemies in war, so lets find a method to make our swords sharper", whereas now we tend to follow the opposite process, "We found a way to store light inside of diamonds, how can we apply this?".
Indeed, that is the whole point, I want players you to be able to say, "I need to beat him at this game, how can I do it, and what do I need to research to get there?" not just "more money to research! Why? Who cares why, just do it because that's what you do in these games!". The probable fact that that's not how it happens in real life, is not a reason not to do it in the game.

(As a side point, in fact there are good examples in recent real life, of when there was an occasion when a strong need arose that technology was seen as being able to solve, and this is exactly the kind of way research did happen, for example Enigma decoding and Manhatten Project in the second world war)

Locked