How does reduced Research Rate actually work?

[hawkeye]

I'm asking because I'm currently playing a Germany game, Super Large Fleet with 10% Research Rate and slow aircraft development.
It is 1938.
If I apply the research rate directly to the game, I should have technology about equivalent to 1904 maybe the first Dreadnought could have been laid down.

However, this isn't what really happened.

I designed my first dreadnought in 1915, commissioned in 1918 (30,000 tons, 23 knots, 10 x 12" guns)
In 1926, the first super dreadnought was commissioned (40,000 tons, 26 knots, 8 x 14" guns)
And currently, I'm building my 2nd generation super-dreadnought (57,500 tons, 27 knots, 8 x 15" guns)

The US has commissioned the first full-blown CV, but every nation is building their own except for me, since I can't yet (I _do_ have 8 CVLs though)

And aircraft _completely_ dominate the Mediterranean. 

Look at my research screen:





Now, keep in mind, this is a 10% Research Rate and Slow Aircraft Development game.
How can this possibly be in line with that research screen?

P.s. I had to buy torpedo- and dive bombers from the US, Japan and Great Britain and _still_ can't request designs for those from my own guys, but I got medium bombers in 1934.

[aeson]

How I understand the research system to work:

- You get a certain number of research points per credit spent on research.*

- Those research points are split out over the various research areas depending on priority and applied towards whichever technologies are currently being researched.

- Technologies are developed when you have invested their research cost worth of research points in them, give or take some minor random variation.

- Technology cost is scaled in some way depending on whether it'd be advanced for the current year (effective cost increased) or behind for the current year (effective cost decreased).

Reduced research rate presumably scales either the technology points generated or the technology costs, but since the powers' budgets grow over time spending a given percentage of your budget on research in 1915 gives you more research points than spending the same percentage of your budget on research in 1905, and since a c.1905 technology would be 10 years behind the nominal technology curve in 1915 it's also easier to develop in 1915 than it is in 1905. Thus, reduced research rate delays technological progress, but not linearly.

Also, it's worth mentioning that some technologies become easier to develop once other powers have them, and technological development can also be boosted by building ships in yards that have access to something you don't have, by buying technologies when the opportunity arises, and by stealing technologies through espionage, which can help push the rate of advancement along since it means that if for example one power develops advanced machinery while another develops advanced hull design & construction techniques or something else like that then they can pull each other (and others) forward so that both(/all) of them have advanced machinery and hull design & construction.

* Higher spending produces more research points, but to my understanding it is not a linear relationship, so spending e.g. $200 on research isn't ten times better than spending $20 on research. It's still enough to be noticeable when comparing technological progress for wealthy and poor powers at the same percentage-investment in research.

[hawkeye]

Thanks for the reply.

I get that buying and stealing tech can speed up development (then again, I'm pretty sure that stuff happened historically too, so it shouldn't have such great of an effect, now should it?)
But even then, I kinda doubt that it's effect would be _that_ large.

Reduced research rate presumably scales either the technology points generated or the technology costs, but since the powers' budgets grow over time spending a given percentage of your budget on research in 1915 gives you more research points than spending the same percentage of your budget on research in 1905,

Maybe I'm stupid, but that does't make a lot of sense to me.
If either the cost of technologies or the tech-points generated are cut by 90% in a 10% research rate game, this should line up perfectly with a regular 100% RR game, since in such a game, the economy increases just as much (percentage wise) as in said 10% RR game, thus you would _always_ research stuff 10 times slower, no matter the year.
If in 1920, your economy is twice as large as in 1900, you gain twice as many RP, no matter if in a 10 or 100% RR game.

and since a c.1905 technology would be 10 years behind the nominal technology curve in 1915 it's also easier to develop in 1915 than it is in 1905. Thus, reduced research rate delays technological progress, but not linearly.

Wait a second, are you saying that because a technology was supposed to be developed in 1905 and you haven't done so in 1915, it becomes easier to research that technology and that this time-table is used for all games, no matter the research rate?
Because if that is the case, shouldn't those time-tables be adjusted for different research rates?
I mean, like, normally, steam turbines are supposed to be developed around 1905, but in a 10% research game, no-one would even dream of steam turbines before, I don't know, 1930 or so?

[aeson]

Oct 1, 2020 15:07:43 GMT -6 hawkeye said:

Reduced research rate presumably scales either the technology points generated or the technology costs, but since the powers' budgets grow over time spending a given percentage of your budget on research in 1915 gives you more research points than spending the same percentage of your budget on research in 1905,

Maybe I'm stupid, but that does't make a lot of sense to me.
If either the cost of technologies or the tech-points generated are cut by 90% in a 10% research rate game, this should line up perfectly with a regular 100% RR game, since in such a game, the economy increases just as much (percentage wise) as in said 10% RR game, thus you would _always_ research stuff 10 times slower, no matter the year.
If in 1920, your economy is twice as large as in 1900, you gain twice as many RP, no matter if in a 10 or 100% RR game.

Except that you're spending for example a 1915 research budget on developing a tech that's meant to be developed using a 1905 research budget, so if your 1915 research budget is twice as large as your 1905 research budget in absolute terms you have significantly more research points going into developing that technology each turn than you're "supposed" to, and on top of that since it's 1915 and the tech was supposed to have been developed c.1905 it's ten years behind and so cost-scaled to be easier to develop.

Wait a second, are you saying that because a technology was supposed to be developed in 1905 and you haven't done so in 1915, it becomes easier to research that technology and that this time-table is used for all games, no matter the research rate?
Because if that is the case, shouldn't those time-tables be adjusted for different research rates?
I mean, like, normally, steam turbines are supposed to be developed around 1905, but in a 10% research game, no-one would even dream of steam turbines before, I don't know, 1930 or so?

Ship templates care more about year than technology, or at least did so until Fredrik did something in a recent patch to make them work better with reduced research rates - there's at least one thread somewhere around here about the computer building ships that none of the powers in the game have the technology for. I don't see any reason to assume technology cost-scaling is handled differently, especially since that's a part of the game that was probably carried over from RTW1 and the patch notes for RTW1 suggest that that option was only added in v1.30.

I also wouldn't be terribly surprised if this were at least partly working as intended, because a game where everything takes ten times as long to develop as normal is a game where the only things that really change are the size of your docks and the size of your budget; there's little reason not to lay down ships of the same class ten years straight because there's no new technology making it worthwhile to actually change things.

[hawkeye]

Oct 1, 2020 20:15:27 GMT -6 aeson said:

Oct 1, 2020 15:07:43 GMT -6 hawkeye said:

I also wouldn't be terribly surprised if this were at least partly working as intended, because a game where everything takes ten times as long to develop as normal is a game where the only things that really change are the size of your docks and the size of your budget; there's little reason not to lay down ships of the same class ten years straight because there's no new technology making it worthwhile to actually change things.

But, that is kind of the whole point of playing a 10% research rate game, isn't it?
I mean, if I want (just as an example) steam turbines from around 1910 onward, I'd reduce the research rate to 50%, doubling the time to get them.
If I want to play a ridiculous "pre-dreadnoughts forever" game, I'd go with a 10% research rate and field Pre- and Semi-Dreads up into the 1940s

As it is now, the reduced research rate feels a little bit like a "Mogelpackung" (I don't know if there is an English word for this German expression from consumer protection, when companies create packages that contain a lot of air so they look bigger. Literally: Package-Cheating).


Also: Thanks for explaining the 1905 tech in 1915 thing, I get it now - even though I had to create a table and count accumulated tech-points for it to finally click  .


Anyway, thanks for explaining all of this. I guess I just have to live with things as they are, which means I'll have to shelf any game after reaching the late 30s (handling of a/c simply is too much of a bother for my taste - any chance we can get a "no aircraft" checkbox for the game setup screen, btw? )

[tbr]

If you want to linearily stretch out game time by e.g. a factor of ten, set research speed to 10% and, at the end of each year, reset the calendar year in the savegame file back one year until you have "played" that year ten times (it helps to generate a separate text file with a "repeat counter"). Restrict yourself with shipyard expansion with a "house rule" (the AI does not build its yrads too fast) and start the game with medium or smaller fleet sizes since the GDP/budget growth will be overwhelming with larger fleet sizes.

[hawkeye]

Good idea, I might try this.
Just have to remember doing this every January - I mean, it isn't that I forget to keep expanding my shipyards every second year or so. Nope, that never happens!

[tordenskjold]

By the way, how does a lower research rate affect aircraft development? Are the obvious research areas ("Naval aviation, heavier than air" and "Shipboard aircraft operation", respectively) available later, or are they available as early as in a 100% game and then only researched slower?

[sjpc302]

Oct 3, 2020 7:29:58 GMT -6 tordenskjold said:

By the way, how does a lower research rate affect aircraft development? Are the obvious research areas ("Naval aviation, heavier than air" and "Shipboard aircraft operation", respectively) available later, or are they available as early as in a 100% game and then only researched slower?

This is a question I'd like answered too. Aircraft's performance depends on the year the designs are generated, and I think slower aircraft development tweaks this table to make aircraft come to prominence later. I'm unsure if it tweaks the specific research as well, or related research like AAA or fleet tactics.
Here is the unofficial wiki's info and tables - unsure if this entirely accurate but you get the general idea.

This could cause issues playing with lower reasearch rates without slower aircraft development. If it works like I think, if you were to research dive bombing in 1944 on slower research rates, your first dive bomber will be a 1944 irl-era dive-bomber as defined by the tables, so something like a helldiver or destroyer. I would at least like to know at what rate slower aircraft development changes things, so people could pick the research setting that best fits it to have a balanced slower game.