What to do with the stockpiling techs

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: What to do with the stockpiling techs

#61 Post by Ophiuchus »

Vezzra wrote:AFAICT, the problem of this "focus dance" is the possibility to gain a substantial advantage in certain situation by micromanaging a lot of focus switching between resource output foci (research or industry) and the stockpile focus. The underlying problem is that the costs for the focus switching are too low to prevent that exploit. Furthermore, as far as I have understood, how serious this issue really is is still somewhat of a theoretical consideration.
For me it is real not theoretical, happens the enemy gets me on a bad foot when i'm pressing outwards.
Anyway, i agree the cheap focus switching (cost/time) is the underlying problem. Personally wouldnt want to merge influence cost with planet focus switching though.

What do you think of the timer-based or slow-then-exponential local growth ideas? That were my counter for the cheap switching cost and short-term planning. Too complicated? Valid?
(rereading i think you ment those as well)
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: What to do with the stockpiling techs

#62 Post by Oberlus »

Vezzra wrote:regarding the global stockpile empire meter idea [...] while this idea prevents the exploit of temporarily switching many colonies to stockpile focus to get an immediate large increase to the extraction limit without loosing too much of resource output of the switched colonies (because you switch back after only a few turns), it opens up another one: if you want to e.g. boost your industry for a few turns you could temporarily switch your stockpiling colonies to industry. The many local industry meters will in total give you a larger PP output boost than your loss of extraction limit, which is determined by the per turn drop of one single empire meter. Which again gives you the potential to gain an advantage if your willing to do the required micromanagement.
Consider this: for industry/research meters, the focus change includes a penalisation (it goes down 3x faster than up, until that tech that makes going down 1.66x faster than up), and that seems enough to avoid that problem you mention here for the cases of industry-research switching. It's intend to make the stockpile meter work the same, hence the problem for {research/industry}-stockpiling switching should also be solved, allegedly.
Vezzra wrote:Making increase and drop rates of the empire stockpile meter dependent on an empires number of colonies isn't a good idea either IMO: that is just a more complicated and less intuitive/transparent [1] way to achieve the same that is now done by doing the increase and drop on the local (planetary) meter level. If the intention is to just slow down the amount by which an empires stockpiling capacity can grow, this can be more easily achieved by just reducing the rate at which the local stockpile meters grow/shrink [2]. Much more in line with how all the other meters work, and therefore much better understandable for the player [3].
[1] Maybe that is subjective. I don't think "change rate of current meter depends on difference between max meter and current meter of each individual colony" is more readily understood than "change rate of global meter depends on the number of colonies" (or my preferred "empire's population").
[2] No, the intention is to avoid the focus dance to speed up reaching the desired maximun MaxStockpile (at which point you switch back many/some of your stockpiling colonies). And that effect is unnaffected by the increasing or decreasing change rate. It is solved, partially, by Geoffs idea of decoupling MaxMeter increase by making it empire-wide, problem then is you need a new way to mimic the industry/research meter's behaviour ([3]).

Vezzra wrote:AFAICT, the problem of this "focus dance" is the possibility to gain a substantial advantage in certain situation by micromanaging a lot of focus switching between resource output foci (research or industry) and the stockpile focus. The underlying problem is that the costs for the focus switching are too low to prevent that exploit. Furthermore, as far as I have understood, how serious this issue really is is still somewhat of a theoretical consideration.

Therefore, my proposal is as follows: For 0.4.8, remove stockpile boni from the growth focus, and remove the restriction of the stockpile bonus to certain colonies. If necessary, adjust the boni the stockpile meter gives to prevent too high extraction limit levels [4]. That change should be reasonably simple and doable for 0.4.8. This should provide us with the amount of playtesting feedback we need to determine where to go from here.
I agree with this proposal, but also advocate for the suggested changes to be implemented and playtested after 0.4.8 release as a serious candidate for solution.
[4] That shouldn't take too long.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: What to do with the stockpiling techs

#63 Post by Vezzra »

Ophiuchus wrote:Personally wouldnt want to merge influence cost with planet focus switching though.
Why? It would be a simple and straightforward approach to counter any and all mass focus switching micromanagement exploits. That's a fundamental issue of the whole focus mechanic, currently addressed by having an extra reduction of the meter on the turn the focus is switched away from that meter.

Which IMO is not an optimal solution, as depending on the type of meter and tech level that can translate into a very severe or almost negligible cost. Having focus switching cost influence provides a catch-all solution that should address all such potential exploits.

Also, we are currently considering policy card activation having a influence cost. Policy cards, in terms of game mechanics, are essentially something like a focus setting, but on the empire level.
What do you think of the timer-based or slow-then-exponential local growth ideas? That were my counter for the cheap switching cost and short-term planning. Too complicated? Valid?
(rereading i think you ment those as well)
Yep, meant those as well. So far we've managed to stick with a meter growth rate that is more or less consistent across all kind of meter types. Some meters might have to grow/shrink at different speeds than other meters (to account for differences in scale, see supply, troop, industry and planetary shield meters). But the growth/shrink rates are constant, not dynamic, which makes thing a lot simpler, transparent and easier to comprehend for the player.

Which is why I want to avoid making exceptions to that unless there isn't really any other solution. As I said in my prior post above, we had such an exception, and it got ripped out. Although, to be honest, the pop meter still is the exception to that rule, but at least it does not depend on how many turns after a focus switch has happened, but on the current level of the meter, which is somewhat easier to comprehend.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: What to do with the stockpiling techs

#64 Post by Vezzra »

Oberlus wrote:Consider this: for industry/research meters, the focus change includes a penalisation (it goes down 3x faster than up, until that tech that makes going down 1.66x faster than up), and that seems enough to avoid that problem you mention here for the cases of industry-research switching.
That penalty is only applied on the turn of the focus switch, so that shouldn't be that much of a cost. And anyway, the same applies to the "focus dance" exploit, isn't it? That penalty should have a similar significance in both cases.
I don't think "change rate of current meter depends on difference between max meter and current meter of each individual colony" is more readily understood than "change rate of global meter depends on the number of colonies" (or my preferred "empire's population").
Maybe, which is why I'm not fond of either approach.
No, the intention is to avoid the focus dance to speed up reaching the desired maximun MaxStockpile (at which point you switch back many/some of your stockpiling colonies). And that effect is unnaffected by the increasing or decreasing change rate. It is solved, partially, by Geoffs idea of decoupling MaxMeter increase by making it empire-wide, problem then is you need a new way to mimic the industry/research meter's behaviour ([3]).
But isn't that the point? The decoupling should prevent that switching many colonies to the stockpile focus gives too much of an increase to the global extraction limit within a turn. If the growth rate of that global meter depends on the number of colonies (which means, more colonies, faster increase wihin a turn), doesn't this defeat that purpose...?
Vezzra wrote:Therefore, my proposal is as follows: For 0.4.8, remove stockpile boni from the growth focus, and remove the restriction of the stockpile bonus to certain colonies. If necessary, adjust the boni the stockpile meter gives to prevent too high extraction limit levels [4]. That change should be reasonably simple and doable for 0.4.8. This should provide us with the amount of playtesting feedback we need to determine where to go from here.
I agree with this proposal, but also advocate for the suggested changes to be implemented and playtested after 0.4.8 release as a serious candidate for solution.
Well, that would allow us to get 0.4.8 out sooner, as the proposed adjustments certainly would delay things further. Not by much, but still. OTOH it means to put out a stable release with a partially broken mechanic (growth focus providing stockpile bonus being the main concern here). Admittedly, both options aren't particlarly compelling...

Geoff, what's your opinion/preference on this? Leave the IS implementatiob as it is and release 0.4.8 with things in place we already know we'll rip out afterwards, or make the adjustments to the IS I proposed and further delay 0.4.8? I really can't decide which I dislike more... ;)

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: What to do with the stockpiling techs

#65 Post by Ophiuchus »

@geoff
Vezzra wrote:Geoff, what's your opinion/preference on this? Leave the IS implementatiob as it is and release 0.4.8 with things in place we already know we'll rip out afterwards, or make the adjustments to the IS I proposed and further delay 0.4.8? I really can't decide which I dislike more... ;)
one way or another i made another PR-2130 for this (focus cleanup and no counter for the focus-dance), so to reiterate the design
Ophiuchus wrote:After taking in the feedback once more a suggestion:

* keep growth and stockpile distinct: remove the effects of growth focus to stockpile completely
* free choice: allow all planets to set stockpile focus
* homeworlds are cool: setting stockpile focus on homeworlds give a +0.04 per population of that species (or boost that species one stockpiling level bad to average, average to good, good to great, great to hmm ultimate?)
* focus has benefit: on all planets stockpile focus gives some (fixed/population based bonus) and maybe doubling the species trait bonus
* counter the focus-dance: by messing with the local stockpile meter growth - locally wait for five turns for the effect to kick in, then give maximal bonus +0,+0,+0,+0,+0,+max,+max,+max
Last edited by Ophiuchus on Mon May 28, 2018 10:36 am, edited 2 times in total.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: What to do with the stockpiling techs

#66 Post by Oberlus »

Vezzra wrote:
I don't think "change rate of current meter depends on difference between max meter and current meter of each individual colony" is more readily understood than "change rate of global meter depends on the number of colonies" (or my preferred "empire's population").
Maybe, which is why I'm not fond of either approach.
I was talking (in bold) about the way Industry and Research works currently, a mechanic that you seemed to support. But here you say you don't like it. I may try and find other alternatives to control meter change rates if FO if that's what you mean.
Vezzra wrote:The decoupling [planet-meters for MaxStockpile, empire-meter for Stockpile] should prevent that switching many colonies to the stockpile focus gives too much of an increase to the global extraction limit within a turn. If the growth rate of that global meter depends on the number of colonies (which means, more colonies, faster increase wihin a turn), doesn't this defeat that purpose...?
Nope. Player can't move up the number of planets his empire has. Because we are talking here about raw number of colonies: you have more colonies, your global stockpile meter increases/decreases faster; but just "colonies", not colonies with certain conditions (focus, species... nothing). Raw number of colonies or raw number of population (empire-wide values).

Vezzra wrote:Therefore, my proposal is as follows: For 0.4.8, remove stockpile boni from the growth focus, and remove the restriction of the stockpile bonus to certain colonies. If necessary, adjust the boni the stockpile meter gives to prevent too high extraction limit levels [4]. That change should be reasonably simple and doable for 0.4.8. This should provide us with the amount of playtesting feedback we need to determine where to go from here.
Oberlus wrote:I agree with this proposal, but also advocate for the suggested changes to be implemented and playtested after 0.4.8 release as a serious candidate for solution.
Vezzra wrote:Well, that would allow us to get 0.4.8 out sooner, as the proposed adjustments certainly would delay things further. Not by much, but still. OTOH it means to put out a stable release with a partially broken mechanic (growth focus providing stockpile bonus being the main concern here). Admittedly, both options aren't particlarly compelling...
I don't understand. If we decouple growth-focus and stockpiling-focus (straightforward, just commenting out a line in a few tech and building files) prior to the 0.4.8 release, why would it be still tied after the release?

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: What to do with the stockpiling techs

#67 Post by Vezzra »

Oberlus wrote:
Vezzra wrote:
I don't think "change rate of current meter depends on difference between max meter and current meter of each individual colony" is more readily understood than "change rate of global meter depends on the number of colonies" (or my preferred "empire's population").
Maybe, which is why I'm not fond of either approach.
I was talking (in bold) about the way Industry and Research works currently, a mechanic that you seemed to support.
Um, I'm confused. Since when is the rate at which the industry/research meters grow/shrink dependent on the difference between the actual and max/target meter values? Industry/research meters grow/shrink at a constant rate (of 1 or 3/5 once you research a certain tech IIRC). As do all other paired meters - the grow/shrink rate is always constant (albeit different in certain cases). The only exception to that rule that works as you describe (growth rate dependent on current and max/target meter value) is the pop meter.

I thought you were referring to one of those ideas where a dynamic change of the growth rate of the local stockpile meters to counter the "focus dance" had been suggested.
Nope. Player can't move up the number of planets his empire has. Because we are talking here about raw number of colonies: you have more colonies, your global stockpile meter increases/decreases faster; but just "colonies", not colonies with certain conditions (focus, species... nothing). Raw number of colonies or raw number of population (empire-wide values).
Ah, that's what you are getting at. Change rate of global stockpile meter depending on the number or colonies (or pop) which is independent of how many colonies are set to stockpile focus, which is different to sum of local stockpile meters which is dependent on number of colonies set to stockpile focus.

Looking at it that way it makes more sense to distinguish between these two approaches (one global vs. many local meters), but still... it's not a very elegant solution (IMO). I definitely like my influence cost for focus switching idea better. Simpler and covers all potential similar cases (of micromanaging mass focus switching). ;)
I don't understand. If we decouple growth-focus and stockpiling-focus (straightforward, just commenting out a line in a few tech and building files) prior to the 0.4.8 release, why would it be still tied after the release?
Never mind. I misunderstood your statement:
Oberlus wrote:I agree with this proposal, but also advocate for the suggested changes to be implemented and playtested after 0.4.8 release as a serious candidate for solution.
I thought you meant to agree with the "contents" of my proposal itself, but objected to the proposed timing (to include it in 0.4.8 ), but to make these changes afterward.

Still, making these changes will require at least a moderate round of playtesting to see how much the dynamics and balance of the IS are affected, and adjust the boni if needed. Not that much of a big deal, but would still mean another (albeit moderate) delay. Which is why I want to hear what Geoff has to say about this - Geoff?

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: What to do with the stockpiling techs

#68 Post by Oberlus »

Vezzra wrote:Um, I'm confused. Since when is the rate at which the industry/research meters grow/shrink dependent on the difference between the actual and max/target meter values? Industry/research meters grow/shrink at a constant rate (of 1 or 3/5 once you research a certain tech IIRC). As do all other paired meters - the grow/shrink rate is always constant (albeit different in certain cases). The only exception to that rule that works as you describe (growth rate dependent on current and max/target meter value) is the pop meter.
Sorry, my wording was rather ambiguous, and in fact what I was thinking about is not really true: I meant that only planets that are not at their max meter will provide growth to the global meter. But in fact, when you change N planets to a given focus, you know for sure that at least for the first few turns you'll be getting a +1 (or +3) xN increase of your global meter, so my argument is wrong and both formulas are as easy/hard to understand/explain.
but still... it's not a very elegant solution (IMO). I definitely like my influence cost for focus switching idea better. Simpler and covers all potential similar cases (of micromanaging mass focus switching). ;)
I agree! I like more the influence thingy. For simplicity, every kind of penalisation/upkeep should be payed with influence.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: What to do with the stockpiling techs

#69 Post by Ophiuchus »

Oberlus wrote:
Vezzra wrote:
but still... it's not a very elegant solution (IMO). I definitely like my influence cost for focus switching idea better. Simpler and covers all potential similar cases (of micromanaging mass focus switching). ;)
I agree! I like more the influence thingy. For simplicity, every kind of penalisation/upkeep should be payed with influence.
Not sure how feasable it is .. but as long as we dont have influence .. we could "pay" (stockpile) focus changes with PP from imperial stockpile?
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

Re: What to do with the stockpiling techs

#70 Post by Geoff the Medio »

Vezzra wrote:Still, making these changes will require at least a moderate round of playtesting to see how much the dynamics and balance of the IS are affected, and adjust the boni if needed. Not that much of a big deal, but would still mean another (albeit moderate) delay. Which is why I want to hear what Geoff has to say about this - Geoff?
This is all post-release stuff.

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: What to do with the stockpiling techs

#71 Post by Vezzra »

Geoff the Medio wrote:
Vezzra wrote:Still, making these changes will require at least a moderate round of playtesting to see how much the dynamics and balance of the IS are affected, and adjust the boni if needed. Not that much of a big deal, but would still mean another (albeit moderate) delay. Which is why I want to hear what Geoff has to say about this - Geoff?
This is all post-release stuff.
Ok, that settles the matter.

Which means, 0.4.8 will be released with the IS as it currently is (with stockpile focus restricted to certain colonies and the growth focus giving stockpile boni). All further revision work on the IS is already post release stuff. I know, not ideal, but we have to cut corners and make some final decisions now.

So, we're proceeding with the release as planned: only bugfixes go into the release branch, no change of existing or adding new stuff. Which essentially means, please concentrate your efforts to get that last piece of IS related content finished (issue#2080 and the corresponding PR#2115).

User avatar
Vezzra
Release Manager, Design
Posts: 6095
Joined: Wed Nov 16, 2011 12:56 pm
Location: Sol III

Re: What to do with the stockpiling techs

#72 Post by Vezzra »

Ophiuchus wrote:Not sure how feasable it is
The biggest challenge I foresee when making this change is adjusting the AI. For a human player that should be no problem, but I expect the AI evaluation of the costs of a focus change to be non-trivial... (I can already hear Dilvish rejoicing :mrgreen:).
but as long as we dont have influence .. we could "pay" (stockpile) focus changes with PP from imperial stockpile?
Well, influence is going to be added next release cycle, so I don't see the necessity of a stop-gap solution.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: What to do with the stockpiling techs

#73 Post by Ophiuchus »

Vezzra wrote:
Ophiuchus wrote:Not sure how feasable it is
The biggest challenge I foresee when making this change is adjusting the AI. For a human player that should be no problem, but I expect the AI evaluation of the costs of a focus change to be non-trivial... (I can already hear Dilvish rejoicing :mrgreen:).
If the AI doesnt do focus dance in the wild, it probably doesnt really matter. i actually think AI is not using the focus at all. @dilvish :?:

I thought about burning 3PP (or maybe 5PP) in the first turn you use stockpile focus. I am not sure how to communicate that though. Just taking it from the stockpile sucks if it is not immediately visible in the (stockpile and build) predictions (have to check).
Else there would be workarounds: e.g. maybe you have to build a local stockpile building first (so you have to spend PP and time), which gets used up if you actually use the focus?
If it doesnt get used up and you need to decide how many buildings you build in advance would at least mean some long-term planning and cost scaling with the number of planets (although a micromanagementmaster could probably build one on all planets). The cost/time prediction with the building approach is at least quite understandable i think :!:


Or more generally you have a limited number of any focus switches for free (tracked as a planet's special's capacity) and you could recharge it using buildings. Thats convoluted :shock:
And AI would suffer.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Dilvish
AI Lead and Programmer Emeritus
Posts: 4768
Joined: Sat Sep 22, 2012 6:25 pm

Re: What to do with the stockpiling techs

#74 Post by Dilvish »

Ophiuchus wrote:If the AI doesnt do focus dance in the wild, it probably doesnt really matter. i actually think AI is not using the focus at all. @dilvish :?:
Right-- the AI currently only gets those focus stockpiling boosts coincidentally via the Growth focus, which it decides about purely on the basis of the standard Growth effects. I.e., the AI currently does not use the stockpiling focus at all. As it stands the value vs decision complexity (and corresponding unreliability) ratio doesn't seem high enough to me for it really to yet rank in my AI development priorities.
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0

User avatar
Oberlus
Cosmic Dragon
Posts: 5715
Joined: Mon Apr 10, 2017 4:25 pm

Re: What to do with the stockpiling techs

#75 Post by Oberlus »

Dilvish wrote:the AI currently does not use the stockpiling focus at all.
So the AI plays like me :)

Post Reply