Victoria 3 - Dev Diary #27 - Technology

[lachek]

I remember it being mentioned before that new military tech just increases the original production quantity (i.e there are no Rifles, just small arms), which was a bit disappointing, but I assume from this it means that I discover rifling, my factories produce more of the same small arms after I change its production, but then my military changes its "production" to "Rifles" which increases its combat power at the cost of more small arms. Then when I discover bolt action rifles, it's the same deal; I expend more small arms, my units consume more small arms, but use the production type "bolt action rifles" and fight much better on the battlefield with the same quantity of manpower. Is this line of thinking correct?

You discover Bolt Action Rifles, your Arms Industries can now make more effective firearms, meaning your existing army becomes cheaper or you export the excess and make your Arms Industries richer or you upgrade your infantry or specialist corps (which might require other techs) to use more firepower, making them more effective.

 

[lachek]

Would it be possible to mod the game to remove direct research, and handle all tech spread through osmosis?

Absolutely. Remove about 20 lines of script, done.

 

[lachek]

Can we cue research? This would help to avoid having a 'save' up research points mechanic and would generally make life easier.

At present, no - this would actually be a pretty sizeable feature since it'd have to recursively compute which techs are prerequisites (since the trees can get quite complex), manage and display the queue, alter the order of the research queue automatically if certain techs become cheaper to research (due to e.g. tech spread), different UI considerations (since you can select any tech, not just the ones that "light up" as available), etc.

It's also of lesser importance to have a save-up-points mechanic since excess Innovation goes towards the techs spreading to you, so you're not wasting your output if you don't immediately select the next tech.

 

[lachek]

The Eureka concept in the Civ 6 tech system is interesting, since it boosts research in logical things that the civilization excels at.

Have you considered a similiar mechanic, making for example research in steel production technologies more effective if you have a strong steel industry?

We actually considered and discarded the Eureka concept for V3 very early on. It'd be trivial to do in our engine, what we didn't like about it is that it felt like putting the cart before the horse, and constrained creativity: rather than me inventing the Bessemer Method because I need more Steel, I build more Steel Mills so I'll get a discount to researching the Bessemer Method, so I wouldn't be "wasting" my research. Presumably the increased focus on Steel I get from this process now makes it cheaper for me to get some other Steel-production-related tech, so now it feels bad to go research something in a different field where I don't qualify for a discount, etc.

The connection between what a country knows and what a country does already comes naturally from the other direction - if I need more Steel, I research a better way of making Steel, then I implement that. My Steel Mills are now more effective and more of my economy will start to rely on Steel. Maybe my next invention will be in Trade so I can leverage this Steel power globally, or in Artillery production so I can make use of the Steel in my domestic armaments production. I prefer if players are free to think in those terms rather than in what hoops they now have to jump through to get the best bonuses next time. The main issue is with the AI, ensuring it doesn't make too harebrained decisions and research garbage that doesn't benefit them and has nothing to do with their strategy, but that issue should be dealt with in AI programming - not short circuited via artificial reward systems - in my humble opinion.

 

[lachek]

Would it be possible to mod current literacy as a modifier?

Current literacy is already a "modifier" in the sense that progression towards Education Access is gradual by definition; the higher the current Literacy the longer it takes to drop to a lower level, since it's modeled via birth and death rates (meaning also that the higher the Pop's Standard of Living, the longer their life expectancy, and therefore the longer current rates stick around). Sure, we could make it adjust Education Access as well, but all that'd do in practice is add more inertia and make the values that factor into it harder to understand.

 

[lachek]

It will entirely depend on how fast it actually drops in practice, but if it goes from 100% literacy to 4% literacy in a generation(somewhat insinuated but could also just be generalized language in the image caption) that seems like too fast of a drop, even for the poor and propagandized. And I don't feel like wealth fully captures the ability of the already educated to pass that knowledge on.

Thanks for the answer!

Ah, poor choice of words on my part! The next generation of Clerks - i.e. all Pop growth from this point on - would indeed have only 4% Literacy, but this Pop growth happens gradually over time, so the length of time required to make the entire Pop only 4% Literacy would be equal to the time required to replace every individual member of that Pop - so maybe 30-40 years or so at least? About 1/3rd to 1/2 the total running time of the game. Longer than that if at higher Standard of Living (even disregarding the boost in Education Access that would inherently provide) and shorter if lower.

 

[lachek]

Vicky 3 unironically has one of the worst looking tech trees I have ever seen in a video game.

We're aware that the tech tree is a bit visually underwhelming right now, and we have set aside time to work on the visuals further before release. The explanation for why it looks the way it does at the moment might be a little contentious, but getting a peek at how the sausage is made is the whole point of dev diaries, so I assume it might at least be interesting even if you end up not agreeing with our reasoning.

When developing the tech tree, we did not start out the way many other games do, by creating a visually pleasing and symmetric empty layout and connections that meshed with the pacing we wanted the game to have, only to later plug in the content we thought might fit well there. Instead we started in the other end, by working out the kinds of techs we wanted to represent in the game and then create the logical connections between them, which then generates the layout and visual connections automatically.
This machine-computed layout is not going to be as pretty as a human could potentially make it, and we briefly considered moving to a model of static positioning to improve it. But we resisted and eventually discarded this idea in favor of making incremental improvements to the visuals of a machine-computed layout instead, for two very connected reasons:

1. We don't want layout and visuals to constrain game design or production - if we think it would improve the game to wedge another tech into Era 3 with a couple of additional prerequisites, we don't want that idea blocked because it'd be expensive to redraw the tree and the art team thinks it'd be hard to make it look as good. We want a development flow where a designer can decide a new tech is needed, spend a few minutes scripting that tech in, add a placeholder icon for it, and start playing their prototype immediately, with the new icon being the only additional requirement which can be added whenever the art department gets around to it.

2. We wanted to ensure the tech trees were as easily moddable as humanly possible, since tech has enormous potential for modding which is often hamstrung by UI constraints. This flexibility also helps us in expanding and maintaining the game long-term.

Visual appeal does take a backseat to those concerns (within reason, of course), but as mentioned we have every intention to continue improving the look and feel of the tech trees beyond what is in the game currently.