Do the devs have any plan to fix ship design automation?

[Beric01]

As a Distant Worlds 1 player the most infuriating aspect of the game to me was the fact that ship design was all or nothing in regards to automation. Either you left it in the AI's hands completely and the AI produced suboptimal designs (that perhaps didn't take advantage of your empire specialties or tech advances), or you did it entirely manually and thus had to update all ship designs manually after every tech advance, which is incredibly tedious and a dealbreaker to me.

I am interested in this game, but have been looking at reviews and it seems this issue is still not fixed in Distant Worlds 2, a year and a half into release. So I am just curious if this is ever going to be fixed or I am just going to need to pass on this game.

[soldier6661111]

I wouldn't bother upgrading a ship class due to one technology, but rather a collection of technologies. Makes no sense to retrofit for one component. Needs to really be a step up.

Upgrading is made easier though. Click upgrade and outdated components get replaced with newer ones. (I don't think that was a thing in DWU?)

[Beric01]

Originally posted by soldier6661111:

I wouldn't bother upgrading a ship class due to one technology, but rather a collection of technologies. Makes no sense to retrofit for one component. Needs to really be a step up.

Upgrading is made easier though. Click upgrade and outdated components get replaced with newer ones. (I don't think that was a thing in DWU?)

That's interesting. So you're saying if you make a manual ship design and then if you update it down the road it will keep the same design except with better components?

I might be able to live with that, though I would still prefer an AI design system with overall directives I can give.

The main struggle I have in these types of games is I don't want ship design to be completely hand-off, but I also don't want to spend hours designing ships.

[Nightskies]

Upgrade does replace older components with newer ones...for the most part... but this isn't always desirable. If you get your hands on the vastly superior Quameno reactor (forget the name atm), you won't need anything else until you get Hyperfusion. It will replace it with even the fission reactor, though, just because its 'newer'.

An easier alternative to something unnoticed botching your design is to "copy as new" instead of "upgrade", allowing two things: to change only the component or two to the new improvement, and to mindfully set the upgrade path of the old design before making it obsolete after making the new design.

This *can* be tedious, but you really only need to do it a few times. A lot of ships are just fine to hit 'upgrade' or let the AI handle it by being selective to only design the ships you care about. Have fun. Anything you don't enjoy doing- don't do it!

Besides, if you control everything, one of the challenges becomes making a challenging galaxy :3

[Prince Metternich]

It's important to note that you can set ship design to automated but select individual ship designs to manual. The AI generally does pretty well at designing civilian ships and is satisfactory at designing bases, but it isn't very good at designing warships because it will select sub-optimal weapons suites, defenses, and sensors. Personally, I'm a veteran player of terrestrial ship-design games, so I find manually designing warships enjoyable, but there are a couple of excellent guides to ship design on here.

[jorgen_cab]

To be honest I don't understand the mentality that designing ships is tedious. I actually design all my ships manually all the time and I spend almost NO time with the ship designer during an entire game. Upgrading ships is usually only a few clicks worth of job per design.

Just don't upgrade them every time you research a new component, it is not worth it to have your ships refit that often. This goes as much for civilians as for military ships.

At the very least you should design your military ships and that certainly is worth a few minutes now and then to do.

I still would like to have better tools to guide the AI designs from a modding perspective, I do want the AI to build more solid designs in general.

[AK_icebear]

There's a lot of handwaving in these responses about why the OP shouldn't care about what they care about, but the direct answer is this: the system is still all or nothing. You can fully automate a design and upgrades, or have it fully manual.

I am one of those people closer to the OPs view - manual designs with manual upgrade is very tedious. Yes, the upgrade button makes it slightly less tedious, but it still is awful busy work to click upgrade one by one on two dozen designs, double check to make sure the weapon type you prefer persisted (usually ok), and that crew, power, fuel all are roughly in the correct amount given marginal changes from components.

There is nothing intelligent to these decisions, it is pure tedium. I remain flabbergasted the issue isn't addressed by devs. For me it's the biggest weakness in the game.

[Nightskies]

Ah... AK, its very much not all or nothing in this. If you set to automatic, you can create or specify specific designs to be manually handled.

[jorgen_cab]

They also have added some automation assistance in that you can designate what weapon systems you prefer and so on... will have some impact on the designs.

[AK_icebear]

Nightskies, jorgen - all respect as you often have invaluable understanding of the game systems.

But here you are essentially agreeing with me. You're saying that my preference, to manually design once per hull, for all hulls, and then have a simple automatic-upgrade (not automatic-design), is not possible.

Instead you suggest only manually designing some ships to manage the volume of manual upgrades later. Or using a preference feature of the automatic upgrade that applies to all designs (not hull by hull). Neither is anywhere close to satisfying my preference. From my perspective, I am still stuck with an all or nothing approach.

The design and research preferences are a nice idea. But they should be applied per weapon slot on a ship design template - not to the entire set of designs.

I'm trying to be clear here as again, I think your understanding of this game as it is is excellent. But sometimes there are questions or requests in these forums that are more about how a player would prefer the game to be. Your responses sometimes come across as finger-wagging "the game is fine as is, it's you that's wrong".

[Nightskies]

I get it. You get us too, though? Nothing we said implies or suggests the system is impeccable nor that you're wrong for having feelings or expectations. You are correct that the game could benefit from more specific quality of life features with regard to designing, as covered in nearly identical topic threads which is still active, so repeating its content feels inane.

Really, about the all or nothing automation- that's your choice in dealing with it as it is, but these suggestions are just ways of handling it with less frustration as it is. Isn't it kind of rude to say "I don't like your suggestions because it isn't giving me what I want" ?

[AK_icebear]

Sure, suggest away. My read of the OP was they have the same frustrations in DWU as I have in DW2 wrt designs, so I answered with that emphasis. Your suggestions do not alleviate my frustrations with this system at all, so I thought that worth flagging - it may be worthwhile to OP, if not you.

[Beric01]

Originally posted by AK_icebear:

Sure, suggest away. My read of the OP was they have the same frustrations in DWU as I have in DW2 wrt designs, so I answered with that emphasis. Your suggestions do not alleviate my frustrations with this system at all, so I thought that worth flagging - it may be worthwhile to OP, if not you.

I appreciate the feedback. Looks like I'll keep this one on the back burner in hopes the devs address it.

[jorgen_cab]

One of the problem is the open sandbox and components being highly moddable... there need to be a way for the AI to know and understand what module it should upgrade too... let's say one module now have two newer modules unlocked that is split in new research into two types, they also might have different weight but roughly the same function. How will the automation know which module to upgrade it too. Maybee the the weapon now is classified as stand-off and not close-in so it is not recognized as a new upgradable module. How will it know if you like this new module or not?!?

When you research a new hyperdrive for example... how will the automation know which one to upgrade too if you got three different ones in the same technology. This is perfectly legal through modding for example.

Then you have all the changes in size, energy use, static of combat draw, speed, fuel, operational range etc... exactly how components should be upgraded is very difficult to know and everyone might want to do this very differently.

Maybe you have a new module you unlock that is both a reactor and a fuel cell... should it upgrade all your reactors and fuel cells or not?!?

In the current game the "new" module are based on a few parameters, such as research cost where in the research tree the module are situated, type and they also introduced weapon family group too as an additional measure to guide the process. In addition to this there is a weighting system in the race policy files for what components the AI should try and use before any other.

When you get down to the nitty gritty it is much more complicated than just upgrading from one to the next, especially when you include how diverse the system is and how modding it could severely break any automation system and also would limit what they could do themselves without adapting it.

In my opinion this is way more complicated than some people make it out to be.