Conventional Start, slow research

[serger]

If I understand correctly, the suggestion by "diminishing returns" is that the more labs a project uses the fewer RPs each lab contributes to the project?

Yep.

I don't think this actually reflects how large scientific collaborations work at all, so I don't think there is a realism aspect to this.

As far as I know it's actually exacly how large scientific, or engeneering, or nearly any other collaborations work.
To make the project in time, you need more funds, yet the more funds you are investing - the less efficiens they are. It's still profitable to make, because you need to make it first, or else it will be just a zero return.
(A bit of simplification, yet any game mechanics have to be a simplification too.)

On the other hand, this creates an optimal gameplay path of using only very few labs per project and I don't think this is good for gameplay; forcing a single strategy to be optimal does not create interesting decisions for the player.

Nope. Urgencies (and consequently the need to flood the project with resources) are inevitable in-game the same as in real life. It's just currently the mecanics makes the opposite: the obvious best R&D strategy is a sequence of urgencies + 1-lab projects to train researchers. Nothing like a real life at all, with too much micromanagement at the same time.

A better solution would be to have some amount of time when a new project starts or is resumed during which research efficiency ramps up to 100%. However, this would be complicated to implement

Not only harder to implement, but even more micromanagement-provocing.

In my games I don't think of the RP numbers as research production, rather as funding allocated to projects by government grant agencies which happens to return the desired results after 5k or 10k or however much investment. It is not a complete, perfect reflection of the game mechanics but I think it works well enough.

Again - it's completely possible to patch any game with your own imagination, if you can just rewrite what you see on the game UI with what you want.
It's just an absolutely universal (and so pointless) reply to any suggestion.

[Jorgen_CAB]

Diminishing return of investment is clearly realistic as that is how things work in the real world. You generally have more of an S curve when it comes to funding research. You will get faster results up to a point and then diminishing returns in the investment.

If such a method was implemented it has to be in such a way you feel that investing more is worth it even if it is less efficient. Some project just are important timewise, more so than others. In times of peace and general stability you want to maximize the efficiency. For example the admin would be more of a soft cap rather than a hard cap. You gain full efficiency up to the cap and then diminishing return of you add more.

Now you have the option to hurry some specific technology but at higher cost and something else taking longer.

[AlStar]

You're suggesting to make another tedious-to-implement home rule to negate one of the factors that are making a game mechanics awkward.
Sure, it's possible to pile up one home rule onto another, you can even make some entirely another game by making enough of these home rule epicycles. I just think there are more efficient solutions.

You specifically called out characters dying or retiring as being disruptive to your narrative process; so I pointed out how to stop your character from dying or retiring when you don't want them to.

You found that too be too "tedious" and/or "home rule" for your taste... but I don't see what the happy medium between these two points are? Either the characters last forever/until you remove them, or they die/retire when the game dictates - what's the third option?

Diminishing return of investment is clearly realistic as that is how things work in the real world. You generally have more of an S curve when it comes to funding research. You will get faster results up to a point and then diminishing returns in the investment.

If such a method was implemented it has to be in such a way you feel that investing more is worth it even if it is less efficient. Some project just are important timewise, more so than others. In times of peace and general stability you want to maximize the efficiency. For example the admin would be more of a soft cap rather than a hard cap. You gain full efficiency up to the cap and then diminishing return of you add more.

Now you have the option to hurry some specific technology but at higher cost and something else taking longer.

Now this is something that I could get behind - I like the idea that you could "overload" a scientist beyond their abilities for an extra push; although I'll admit I can't think of any time in my playthroughs where I specifically thought "if only I could finish this project a couple months faster, it'd really help" - generally the build times on warships are far enough out that your current fleet is mostly playing with the last generation's toys.

[Jorgen_CAB]

Now this is something that I could get behind - I like the idea that you could "overload" a scientist beyond their abilities for an extra push; although I'll admit I can't think of any time in my playthroughs where I specifically thought "if only I could finish this project a couple months faster, I'd really help" - generally the build times on warships are far enough out that your current fleet is mostly playing with the last generation's toys.

In multi-faction games I can see this happening since competition and time is a much more important factor. Factions will be closer to each other and have allot more intelligence on each others scientific progress etc...

[Noriad]

Diminished returns per lab is highly realistic.

First, you need to get your lab-workers up to speed before you can do any actual research. If you have 8 labs, you need to start 8x as many lab-workers on the job. If you have 1 lab, you only need to start 1 set of lab workers, and after that, they can simply continue until the research is done. So 1 lab will still take longer than 8 labs, but not 8 times as long.

Also, the bigger an organization, the more time and overhead is spent coordinating everybody.

Furthermore, discoveries are often made in sequence. Experiments take time. Once an experiment is finished, new insights are gained that are needed to design the next experiment. Experiments often take time that can't be sped up by going parallel. It's like how a woman can grow a baby in 9 months, but 9 women can't grow a baby in one month.

[serger]

You specifically called out characters dying or retiring as being disruptive to your narrative process

Nope, never did it.
Researchers dying or retiring are disruptive to managing the project, not to the narrative itself.

In Aurora, R&D projects are dissolving immediately with the death/retirement of it's admin. It's the player who needs to keep track on every project and restore them in every case of admin death or retirement. Then it's an inevitable point were you slip and it's a narrative break, indeed. Yet not because the chars are dying, but because the R&D projects in Aurora are made as if they are always in emergency mode.

There is no such awkwardness with ships or production - ship crews do not forget their tasks in case of their commander death or retirement, planetary production don't stop in case of planetary admin death or retirement. It's specific to R&D projects in Aurora.

[serger]

A small suggestion from inside the current real war:

Make ECM and ECCM techs affecting accuracy / fortification modifiers of the GF.

It became real, and more so - it became ubiquitous and obligatory. There's just no attack or strike any more without drone spotting and fire control here. EW is not anymore focused on air and naval operations, the ground forces became completely involved and dependent on the radio bands warfare.

[Marski]

I also like to play slow starting always conventional but leaving all the other options as default when we speak about research speed.
Believe me, in the beginning you probably makes a lot of discovery because the low cost and that gives you the feeling of a fast game, but as long as you progress in the game, as long as you do not construct dozzillion of labs, the amount of researches slow down due to the increase cost and the need to reseach always new designs if you do not want be oblitered by aliens and keep your fleet in order.

Maybe Steve could reconsider the research cost in terms of wealth making it more expensive, thus avoiding the buildup of laboratories; in real life the tech research is a very expensive part of the budget expecially the basic one.

I'm always a proponent of using diminishing returns to automatically balance things. So I want to suggest something similar.

Research funding setting.
Basically you can set research funding (wealth cost of running labs) between for example 10% and 500% and it modifies your research speed acc to for example SQRT ([Research Funding %]/100)

This means you would get the following:



If that is too steep impact the exponent can be changed to cube root or any other value that seems appropriate, but the main purpose is to allow you to easily scale down and up research funding as your wealth budget and priorities allows. Most player would probably want to run at higher research funding levels to speed up research so the wealth cost would naturally increase.

Another potential outside the box suggestion to make researching expensive is considering if running research facilities should consume any TN minerals in addition to wealth.

Old 4x space game called "Sword of the Stars" handled research very nicely; you selected research project, you allocated funds and it presented you estimated completion time. However the actual completion time varied and if you just poured money into it, it suffered a very realistic and close to real life fate; after initial burst the progress starts to rapidly slow down until eventually it will take longer than if you'd just given 20-40% of what you allocated in the first place.

[kks]

A small suggestion from inside the current real war:

Make ECM and ECCM techs affecting accuracy / fortification modifiers of the GF.

It became real, and more so - it became ubiquitous and obligatory. There's just no attack or strike any more without drone spotting and fire control here. EW in not anymore focused on air and naval operations, the ground forces became completely involved and dependent on the radio bands warfare.

I think that should belong to the suggestion thread? It kind of looks off-topic in here.

[serger]

I think that should belong to the suggestion thread? It kind of looks off-topic in here.

Ermm. This thread is in Suggestions subboard actually.

[kks]

I think that should belong to the suggestion thread? It kind of looks off-topic in here.

Ermm. This thread is in Suggestions subboard actually.

Yes, I just thought it might get overlooked in this particular thread and has better visibility in the pinned thread "Suggestions for 2.x". As most posts here are discussion posts.

[MinuteMan]

@steve

I like the ideas and discussions present already.

But I want to propose some new ideas. Maybe it can be a source of inspiration / other ideas.
To be honest, it's a larger change. And not fully defined.

Instead of having "general" research labs which are used for every field.
You could change it as follows:

* Introduction or research campus
** A research campus is one or more labs.
** A research campus is focused on a specific "field"
** A research campus can research techs not in it's field. But forgo's (or diminished) any research bonusses.
** A research campus can be "refocused" on another field. TN and wealth cost.
** A research campus is assigned a "research administrator", which determines upper research lab limit.
*** If the administrator is replaced with a lower cap, the research bonus is reduced and no new labs can be added.
** A research campus can be expanded with x labs.
*** Adding additional labs can have their own cost curve. Promoting the idea of possible multiple campuses.
** Ancient construct bonus applies to whole campus
* Researchers are assigned to "Research campuses"
** Multiple researchers can be assigned to a campus. If a campus has 10 labs, max 10 researchers.
** Each researcher can have an assigned tech to research.
** Research bonus is a combination of campus bonus + researcher bonus.
...

These suggestions do the following (game balance):
* A Wealth and TN minerals commitment to certain tech fields.
** So it becomes a conscious decision to go down a certain path.
** Same argument for "refocus" of campus. (Like retooling a shipyard)
* Protecting research colonies becomes even more important
* Losing a specific researcher with a large bonus is still hurtful, but possibly reduced by a good research campus admin.
* Slower to snowball in specific research fields without lacking in others. Currently you can move over all your labs between research fields / tasks.
* ...

Goal: Slowing down the research lab snowball we have assigned to a single research task. And being able to move labs to another research field in an instant.

[Froggiest1982]

...

I will start by quoting Nuclear on our previous discussion:

There is a piece of wisdom about research rates: since tech costs in Aurora scale exponentially (roughly 2^x), there will always be a 'wall' you hit after which advancing in tech is a very slow process. Changing the research rate speed only changes the point in the tech tree where that wall is hit. With the default 100% that wall might come at, say, solid AM tech level, while on 20% research the wall may be at MP Drive or Magnetic Fusion tech levels. However, there will be a wall either way, all the tech slider decides is what tech level the wall is at. If you want to play 100 years at NPE tech, then you can set research to 5% as long as you are comfortable with the implications RE: NPRs and spoilers.

The research will inevitably slow down at some point if you adhere to the admin cap, as there is only a limited amount of labs and RP that can be generated per year, even by the most gifted scientists. Additionally, this is a benefit that can only be enjoyed during the later years of their lives, making it a temporary bonus.

I think we should aim to distribute research evenly or at least gradually, with some key technologies being more expensive and others slightly less so as a direct consequence of initial efforts. Unfortunately, this would require the creation of several intermediary technologies, and the Aurora tech tree would need to be rebuilt from scratch, something Steve may need to outsource the design to a forum group or panel as it cannot be redone in-game during testing. You have this partially due to components, engine/power plants connection, and some other intertwined techs and branches. Perhaps it could be expanded? I seriously doubt he would consider it anyway, as the amount of work for little to no gain would likely not be worth the trouble. Also, the changes could involve several other part of the code, making it even more difficult to implement.

Regardless, I am happy that research is getting back on topic and the usage of TN resources is on the table. Personally, I agree incorporating minerals that have had less utility into the equation adds an enjoyable twist to our challenges without causing too much micromanagement!

I would also support using wealth on a larger scale in the process. To elaborate, in a recent campaign's early stages, I have 48 million scientists generating 120 wealth per year per million. My research expenses during this period were approximately 8.5k. Essentially, the cost of my research labs and efforts to taxpayers was less than 3k ((48 * 120) - 8500).

Maybe it's just me, or perhaps my equation isn't correct, but it seems too inexpensive compared to how much R&D functions in the real world. Additionally, as Wealth generation technology advances, the impact of wealth on actual research efforts may become completely irrelevant, which is likely an overlooked side effect of the current wealth balance. I'm not suggesting we break the financial model, but if some are concerned about slowing down or unintended consequences from overly ambitious research, financing could be a solution, along with the already suggested mineral usage.

Finally, I would also welcome random events that could advance or degrade the project by a random percentage from time to time. This will make the research timeline even less predictable when combined with always possible unexpected deaths.

Just on a personal note, I have seen many great ideas, but please, while formulating them, bear in mind that everything still needs to function with the NPRs' AI as well, or we may encounter larger issues than expected.

Of course, Steve knows this better than we do.

[kks]

I would also support using wealth on a larger scale in the process. To elaborate, in a recent campaign's early stages, I have 48 million scientists generating 120 wealth per year per million. My research expenses during this period were approximately 8.5k. Essentially, the cost of my research labs and efforts to taxpayers was less than 3k ((48 * 120) - 8500).

Maybe it's just me, or perhaps my equation isn't correct, but it seems too inexpensive compared to how much R&D functions in the real world. Additionally, as Wealth generation technology advances, the impact of wealth on actual research efforts may become completely irrelevant, which is likely an overlooked side effect of the current wealth balance. I'm not suggesting we break the financial model, but if some are concerned about slowing down or unintended consequences from overly ambitious research, financing could be a solution, along with the already suggested mineral usage.

As far as I know, in 2.x wealth is not generated simply by pops but by installations which "produce" something. I could not find which buildings this are, the mechanics post just says "TN installations". I was under the impressions that it are mostly mines, financial centres and factories which produces wealth.
I would be quite surprised if labs large carry most of their own expenses, as research for my empires is most often the biggest wealth expenditure, taking up to 50% of wealth.

[serger]

As far as I know, in 2.x wealth is not generated simply by pops but by installations which "produce" something. I could not find which buildings this are, the mechanics post just says "TN installations". I was under the impressions that it are mostly mines, financial centres and factories which produces wealth.
I would be quite surprised if labs large carry most of their own expenses, as research for my empires is most often the biggest wealth expenditure, taking up to 50% of wealth.

That's what marked as TaxableWorkers in the DB: