FreeOrion

Forums for the FreeOrion project
It is currently Sun Dec 17, 2017 5:38 pm

All times are UTC




Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2, 3  Next
Author Message
PostPosted: Fri Nov 18, 2016 12:11 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4307
Location: Sol III
Opening a thread to transfer a design discussion that started on github to the design forum.

Corresponding PR: https://github.com/freeorion/freeorion/pull/1091

Geoff the Medio wrote:
Imposes limits on the size of meter increases from various techs that have population-dependent bonuses. This means that bonuses grow with population, but only up to a point. This should reduce the range of output levels over the course of a game, and reduce the benefit of having high population planets.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 12:12 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4307
Location: Sol III
MatGB wrote:
Not at all sure about the idea: I'd rather reduce the per-pop bonus for a fair few of them, which is what we've already discussed.

Getting techs and specials and boosts like Gaia, growth specials and similar are put of the fun of the game, fully maxing out a homeworld to get the best possible out of it is something I personally enjoy (and am very clear that at max the numbers are too high). Plus, we (well, I) have put a lot of effort in to balance the costs of improvements like terraforming and gaia transformation taking planet size into account of late.

If you implement something like this then the fun of trying to get your good planets truly great is reduced and there'd be a much stronger need to work out if a given improvement is worth it, in a way that I think might be detrimental: currently my decision on, for example, gaia transformation is whether the planet is already close to max output, but if it's at or close to max possible output then, unless that's VERY clearly displayed in the UI then working out if the building is worth it becomes far more complex/time consuming/micromangy/unfun.

Plus we did agree that we'd do a proper balance pass on these bonuses with/after Influence was implemented because doing it before then was basically a waste of time as it'd need doing again regardless.

There are definitely some elements that could be included for some techs, and I like the idea of the mechanic as a limiter for some things, but as a fairly blanket approach I'm not at all keen.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 12:13 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4307
Location: Sol III
Geoff the Medio wrote:
Perhaps it would be better to just add this to some of the early-game tech effects, then? The result would be that early techs initially have a population-dependence, making growing planets useful at first, but later the bonus is capped so that those techs work like the fixed bonuses. Some other later-game techs could be uncapped, so there would still be reasons to continue increasing planet population in the later parts of the content tree.

Part of the point of adding this now would be to get a feel for how it affects balance.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 12:17 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4307
Location: Sol III
LGM-Doyle wrote:
A fix that I have not seen suggested, is to use only the top M highest producing of all N colonies. If you set M = sqrt(N) then you get the following table.

Number of colonies contributing to empire production

  • 1 - 3 colonies in empire uses the best 1 planet's production.
  • 4 - 8 colonies in empire uses the sum of the best 2 planet's production.
  • 9 - 15 colonies in empire uses the sum of the best 3 planet's production.
  • 16 - 24 colonies in empire uses the sum of the best 4 planet's production.
  • etc

Its benefits are:

  • It is a simple to explain mechanic, you don't even need to mention the sqrt, just show the table.
  • The total production output is additive, no difficult mental math.
  • It steeply increases the cost of the next production increase later in the game, while not completely removing growth as a strategy.
  • It make taking large key planets more strategically significant.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 12:19 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12045
Location: Munich
I'm sure there have been numerous threads discussing this issue, though I haven't been able to find one in the time I'm willing to spend looking. The idea is that resource output grows far too much over the course of a game, and bigger more-advanced empires have too much more than smaller less-advanced ones. As well, it becomes problematic to display meter values on a planet with such a big range of possible values, from when +1 matters at the start, to when +20 is a small change at the end.

The point of this PR / change is to cap the population-dependent bonuses to various resource meters, like production or research. The bonus would increase with population up to a point, after which it wouldn't increase any more, and would effectively become a constant bonus per planet.

This would mean that very small population / newly colonized planets don't get a huge bonus just for existing, so they need to grow their population to increase their output. But it also means after a point, the bonuses to increasing population are capped.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 7:41 pm 
Offline
Programmer

Joined: Sun Feb 14, 2016 12:08 am
Posts: 359
Quote:
M = sqrt(N)

This leaves planets with nothing to do, may also increase planet management quite a bit.

Instead of the hard cap, what about a soft one from efficiency loss?
e.g. -X% to any production that exceeds the planets infrastructure.
Would prefer some exponential loss versus a set percentage.
Basing it on infrastructure is just for example, could take other forms such as some logistics office that might see tech upgrades.

In this manner there is still a gain from any species traits or other local effects.

_________________
Any content posted should be considered licensed GNU GPL 2.0 and/or CC-BY-SA 3.0 as appropriate.


Top
 Profile  
 
PostPosted: Fri Nov 18, 2016 8:00 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12045
Location: Munich
dbenage-cx wrote:
Quote:
M = sqrt(N)
Instead of the hard cap, what about a soft one from efficiency loss?
e.g. -X% to any production that exceeds the planets infrastructure.
Increaing industry or any other meter with increaing infrastructure is probably better expressed as an increase, which is fun to get, than a penalty, which is not.

This also seems like it would better fit in threads about what to do with infrastructure. If the limit can be increased than it won't help much with the issues being discussed / addressed.


Top
 Profile  
 
PostPosted: Sat Nov 19, 2016 11:09 am 
Offline
Programmer

Joined: Mon Feb 29, 2016 8:37 pm
Posts: 205
A great point about Geoff's PR that I want to highlight is that the approach will allow decoupling of the per population bonus and the population cap.

This will allow the creation of separate population based technologies targeting tall empires (large colonies) vs. wide empires (many colonies). Tall technologies would increase the population cap on an existing population dependent bonus. Only colonies larger than the old population cap would see benefit from a tall growth technology. Wide technologies would introduce a new population dependent bonus or increase an existing bonus. Tall empires would research a single wide tech and then all of the tall tech that improves that wide tech, while wide empires would research only wide tech, since few/none of their colonies would be larger than the population caps of the initial wide tech.

Geoff have you played your PR into the late game to see how it affects the feel of the race to expand?

Geoff the Medio wrote:
This would mean that very small population / newly colonized planets don't get a huge bonus just for existing

We could also do this be adding a minimum cap as well as a maximum cap. If the minimum cap were 5 then a population smaller than 5 contributes nothing and after 5 its starts contributing from 0.

My own suggestion is really independent of the PR, since it could also be applied with population caps in place. It would allow another type of wide population based techs that increase the fraction of the colonies that contribute to a empire production/research tally.

dbenage-cx wrote:
Quote:
M = sqrt(N)

This leaves planets with nothing to do ....

Having low impact colonies contribute nothing was intentional.
1. It changes military strategy. Using the table shown, if an empire has 24 colonies and only 4 contribute to production, then to reduce the empire's production you need to capture/blockade 1 of the 4 colonies or capture/blockade 8 other colonies. It increases the impact of surgical strikes.
2. It increases the coarseness of the production/research steps. Even late in the game you have key colonies that your entire empire depends on, not 40 faceless colonies that are the same except for a scaling factor.
3. It limits the late game strategy of switching all colonies to production after you have researched all of the technology that you want technique.

dbenage-cx wrote:
, may also increase planet management quite a bit.

What management specifically are you referring to? I thought of adjusting the planet foci after losing a planet, but the penalties to adjusting focus are already severe enough that this isn't usually beneficial.

Your question did make me think about whether N should be all the colonies in your empire (which is what I intended) or N all the colonies set to a particular focus. If it is the later, then I can see why that would cause micromanagement.


Top
 Profile  
 
PostPosted: Wed Nov 23, 2016 9:55 am 
Offline
Programmer

Joined: Sun Feb 14, 2016 12:08 am
Posts: 359
Quote:
What management specifically are you referring to?

With (e.g.) 4 production planets, 4 research planets, and 8 "other", I would expect a strong urge to shuffle the species of specific planets around.
It is a fun dynamic, but I also feel it would be highly aggravating to new players.

Played a couple of game with the intended PR. Just as a general feel, it does stem a good bit of the snowball while still allowing a sense of progression.
e.g. I wasn't close to researching transcendence shortly after producing the first flagship.
I do still agree with the general sentiments of retaining growing "tall or wide".

Is the issue with meter display mostly for the horizontal bar graphs?


Top
 Profile  
 
PostPosted: Wed Nov 23, 2016 4:24 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12045
Location: Munich
dbenage-cx wrote:
Played a couple of game with the intended PR. Just as a general feel, it does stem a good bit of the snowball while still allowing a sense of progression.
After max -> min swap you suggested in the pull request?
Quote:
Is the issue with meter display mostly for the horizontal bar graphs?
That's a major part of it. Having large ranges of numbers can be difficult to display well in various contexts, though.


Top
 Profile  
 
PostPosted: Wed Nov 23, 2016 6:50 pm 
Offline
Programmer

Joined: Sun Feb 14, 2016 12:08 am
Posts: 359
The first game (prior to the comment), research went pretty quick ;) The above was after swapping to min.


Top
 Profile  
 
PostPosted: Wed Nov 23, 2016 9:02 pm 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3295
I've been ill, and thus not contributed to this thread thus far. I'm going to keep this brief.

Any change to pp output requires an AI update. If the change also affects the relative value of different sized planets, it requires a bigger AI update. The AI team is already behind, busy and concentrating on the Fighters branch as well as, for example, catching up with the existing supply propagation mechanics change that the AI isn't coping well with.

Any work done before we introduce Influence on PP output and similar will need to be redone as and when we introduce Influence as the entirety of planetary output meters including PP and research will need rebalancing at that point. In addition, as Influence is likely to largely replace existing colony upkeep and fleet upkeep mechanics (thankfully), we have already agree that, when we introduce influence we will also be massively reducing PP bosting techs/buildings, etc. But we've also discussed that, due to the way fleet upkeep currently works, you basically need large bonuses to deal with fleet production.

I, quite frankly, don't want to do a minor balance pass now that will need to be redone soon anyway. I explicitly don't want to rethink basic mechanics in a way that undermines a fair amount of the balancing work I've been doing in the last few months recosting effects that improve planets to take into account the extra bonuses you get from large planets unless we spend a significant degree of time talking through the full implications, and we've already agreed that this release cycle will a) concentrate on Fighters as far as scripting/balance/AI is concerned and b) be relatively short.

Here are some previous discussions and relevent quotes Vezzra in topic - Spending PP on Industry Meter Growth
Quote:
With the current proposal of introducing influence and influence maintenance costs, as well as the suggested major revision of the resource output boosts to counter the current excessive growth curves I'd say lets see how those changes will affect game dynamics, before thinking about such drastic measures
Me in topic - My Feedback
Quote:
I'm basically thinking that, once Influence is in the game at roughly the same time we should half all the bonuses (so that it's 0.1, or 1 per 10 population), once we've taken fleet upkeep out of production it'll be far easier to balance things better, and I would really like it to be each extra planet is a real bonus once it's gotten going and not, as currently, another production centre churning out stuff.
and then Geoff in topic - Roadmap for the next release
Quote:
To avoid another year-long release cycle, it might be good to make v0.4.7 primarily about fighters.
A lot of that thread is relevent and goes over stuff I'm repeating here that had been discussed before as well.

Short version:This PR should be tagged as "After next release" and worked on alongside/after Influence is ready for testing, PP output is a known issue we've already been discussing solutions to and agreed to include in the Influence work, I don't have time to do any work on other issues, and especially don't want to spend time on something that will need redoing next Cycle.

And that's just addressing the issue of whether we should be workign on this at all, let alone discussing whether this specific idea is even a good one or one that works with the other things we're working on. For that, the summary opinion is: I REALLY don't like it and explicitly several of the numbers are way off before we even get started on whether this is the right approach.

_________________
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.


Top
 Profile  
 
PostPosted: Thu Nov 24, 2016 12:58 am 
Offline
Vacuum Dragon

Joined: Wed Aug 26, 2015 6:15 pm
Posts: 507
Mat brings up something that concerns me. I consider the AI development to be behind a lot of other stuff already. Have the developers working on that been apprised of these proposed changes to fundamental game play? Human players are going to be able to adapt much faster to changes than the programmed play of the AIs.


Top
 Profile  
 
PostPosted: Thu Nov 24, 2016 12:11 pm 
Offline
Graphics
User avatar

Joined: Tue Jul 01, 2003 8:27 pm
Posts: 708
MatGB wrote:
Short version:This PR should be tagged as "After next release" and worked on alongside/after Influence is ready for testing, PP output is a known issue we've already been discussing solutions to and agreed to include in the Influence work, I don't have time to do any work on other issues, and especially don't want to spend time on something that will need redoing next Cycle.

And that's just addressing the issue of whether we should be workign on this at all, let alone discussing whether this specific idea is even a good one or one that works with the other things we're working on. For that, the summary opinion is: I REALLY don't like it and explicitly several of the numbers are way off before we even get started on whether this is the right approach.

Couldn't have said it better. I agree on all points.

_________________
If I provided any images, code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0.


Top
 Profile  
 
PostPosted: Thu Nov 24, 2016 4:36 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4307
Location: Sol III
As I get the feeling some would like to go ahead and get this proposal in I want to voice my opinions/concerns too:

First of all, I completely agree with Mat that this will introduce such a major change in dynamics that we should not merge anything into master now. I'm very much in favor of sticking with the plan to concentrate on the Fighter/Carrier mechanic for this release cycle, and leave other major things for later. Smaller additions/improvements, and fixes of all kind are of course perfectly fine, but Geoffs reasoning why we should not try to do Fighter/Carrier and Influence within this release cycle has been very sound, and should apply to proposals like this too.

So, while discussing ideas is of course very welcome, I strongly propose to postpone anything beyond that for later, at least until we implement Influence (for the reasons Mat cited).

Now to the proposals. We're actually dealing with two different ideas, in this post I want to deal with Geoffs idea, which started this discussion.

The idea to have boni which aren't just flat (like NAI and AA) or purely pop-dependent (as the vast majority), but kind of something in between sounds interesting and has a certain appeal, BUT: I definitely share Mats sentiment that this should not be applied to all or even most of the pop-dependent boni, for exactly the reasons he cited. As an extension to the currently two kinds of boni (flat and purely pop-dependent), yes, as a replacement for the purely pop-dependent boni (or, for that matter, the flat boni), no.

My main objection is, while I've always been in favor of giving more importance/value to small planets and trying to avoid mechanics that make the larger planets too advantageous, this will take things too far. I mean, what's the point in trying to get large and huge planets, and increasing the max pop of your colonies beyond a certain point, when doing that doesn't earn you sufficient benefits? And just keeping some non-capped pop-dependent boni so it's not completely pointless to increase your pop is not sufficient IMO, given how extensive our tech tree in the growth department is, in addition to that the growth specials, etc.

It reminds me a bit of the old mining/industry mechanic of ancient times. Originally, to get PP you needed two resources: minerals (which you got by switching a colony to the mining focus), and industry, which took the minerals and converted them into PP. Which meant you basically had to try to balance mining output against industrial capacity, as an imbalance always meant a loss of efficiency. If your mining output was below your industrial capacity, the surplus industrial capacity was wasted, if your mining output was above industrial capacity, the surplus in minerals was wasted. Which resulted in some micromanagement to keep these two matched, which finally had been deemed a mechanic not fun enough to warrant the micromanagement involved, and therefore mining had been removed.

Basically, what you had were two separate caps you needed to juggle, without really offering sufficiently interesting choices for the player. Applying a cap to all/most of the pop-dependent boni introduces something similar. You have two caps - the pop itself of course, and the cap up to which a certain bonus increases with pop. Increasing your pop beyond what most of your boni can increase to is a waste of investment, as is increasing the caps of your boni beyond the pop levels of most of your colonies (by whatever means). You need to keep them roughly in sync, everything else would mean playing a sub-optimal game.

And for what? Because if the purpose is to limit resource output, this can be done just as well by reducing the boni to pop we currently have. Or making it more difficult to achieve higher pops. I don't see the point of introducing what basically amounts to having several caps which do the same for the same thing.

However, if the purpose is to add another kind of bonus, to give us means e.g. to more fine-grained balancing of high and low pop planets, then I can see the appeal. Where flat boni make pop levels completely irrelevant, and purely pop-dependent boni make pop levels all that matters, capped pop-based boni give some importance to pop, but not so much as the purely pop-based ones. The way I see how this can work is: flat boni are very low compared to the other two (maybe 1 or 2 for easy to get, and maybe 3-5 for those harder to get), purely pop-based give a low bonus per pop, but enough to substantially exceed the flat ones already at medium pop levels, and capped pop-based boni could give higher per pop, but low enough so that the purely pop-based ones are still substantially higher at high pop levels.

Then balance things so that high pop planets offer still sufficient benefits over low pop ones that it's worth fighting over them because of their supreme value, and improving them is worthwhile, satisfying and fun, while making low pop planets valuable enough so that acquiring them is worth considering. The "tall" vs. "wide" empire dynamic idea LGM-Doyle pointed out should be achievable by that, although I think to really get that working we need to introduce the "tech cost increase" mechanic dicussed elsewhere (basically apply the current the more ships you have the more expensive new ones get to techs: each tech you have researched increases the cost for researching new techs). Because that can make following certain research paths more efficient than trying to cover everything. Meaning you have to decide e.g. if you want to go for the techs that make it more ease for you to build a tall or a wide empire.

ALl that said, I'm not fully convinced that capped pop-based boni are needed to achieve this kind of balance. Maybe flat and purely pop-based boni are already sufficient to do all that, balancing high vs. low pop planets, and allow for "tall" vs. "wide" empire approaches. In that case capped pop-based boni would add complexity without adding anything that isn't already there/possible.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 32 posts ]  Go to page 1, 2, 3  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:  
Powered by phpBB® Forum Software © phpBB Group