How do you check what changes the AI makes between two consecutive ship design versions?

[Correcaminos]

I'm reviewing the auto-generated ship designs, and the AI often creates a “New Version” of a ship, but I haven’t found a quick way to see exactly what changed compared to the previous version (added/removed components, weapons, engines, etc.).

Do you use any practical method or a specific screen to compare two AI-generated designs?
How do you guys do it?

Thanks!

[Correcaminos]

I'm reviewing the auto-generated ship designs, and the AI often creates a “New Version” of a ship, but I haven’t found a quick way to see exactly what changed compared to the previous version (added/removed components, weapons, engines, etc.).

Do you use any practical method or a specific screen to compare two AI-generated designs?
How do you guys do it?

Thanks!

After digging into this for a few days, I think I’ve found the only practical way to check what changes the AI makes between two consecutive versions of the same ship design.

What I’m doing now is:



In the Ship Designs screen, I filter by the same ship type.

I enable the option to show obsolete designs along with the active ones.

This way, I can see all previous versions together with the new AI-generated version.

Then it’s just a matter of opening each design manually and checking the differences in components, weapons, engines, etc.

As far as I can tell, there is no built-in side-by-side comparison panel or summary of changes between versions, which would be extremely useful for understanding why the AI makes certain choices and for deciding which version to keep.

It would be great if a comparison tool or at least a “changes since last version” summary could be added in the future.
If anyone knows a more efficient method, I’d be glad to hear it — but so far this seems to be the only workable approach I’ve found.

[Xmudder]

This game spends far too much time upgrading and retrofitting ships and fleets. Further, the manual fleet creation only accepts the most recent models. Most of this is make work that wastes time and money and makes large fleets impossible to sustain,as they spend all their time upgrading.

Ships should be considered the same "generation" if they have the same hyperdrive and fuel capacity / range. So long as those are the same, they should be grouped together and not upgraded unless they have nothing to kill / invade (this included things too powerful to kill). Aquired ships should be grouped the same way - such ass when you kill a hive carrier, and get a bunch of pirate factions, they should auto group in to fleets with the same strategic characteristics.

Certain improvements should set off an upgrading binge, such as larger hull size, a different ( better) hyperdrive / engine, certain weapon improvements, or a new generation of energy that has different size requirements. The techs that will set off should be flagged in the research tree, and should include various weapon iterations. While designing new models are fine, upgrades should be postponed until they are relevant OR player requested.

Allow us to set the automation to always upgrade with new designs, never upgrade unless manually requested, and upgrade when certain techs are researched.

In my most recent game, I have 8 10k+ fleets, and 4 are retrofitting, 2 are refueling, 1 is loading troops, and one is moving to intercept the Hive. If I try to break them down into more smaller fleets so they actually kill things, the auto build fleet template will be useless, as it will ignore anything but the absolute most recent generation of ship, instead of grouping them by strategic characteristics, hyperjump speed and fuel.

I also don't want them upgrading until I reach "Torrent Drive" and then "Fusion Balance" (currently researching prerequisites) but that requires turning a bunch of automation off and then back on.

[Correcaminos]

This game spends far too much time upgrading and retrofitting ships and fleets. Further, the manual fleet creation only accepts the most recent models. Most of this is make work that wastes time and money and makes large fleets impossible to sustain,as they spend all their time upgrading.

Ships should be considered the same "generation" if they have the same hyperdrive and fuel capacity / range. So long as those are the same, they should be grouped together and not upgraded unless they have nothing to kill / invade (this included things too powerful to kill). Aquired ships should be grouped the same way - such ass when you kill a hive carrier, and get a bunch of pirate factions, they should auto group in to fleets with the same strategic characteristics.

Certain improvements should set off an upgrading binge, such as larger hull size, a different ( better) hyperdrive / engine, certain weapon improvements, or a new generation of energy that has different size requirements. The techs that will set off should be flagged in the research tree, and should include various weapon iterations. While designing new models are fine, upgrades should be postponed until they are relevant OR player requested.

Allow us to set the automation to always upgrade with new designs, never upgrade unless manually requested, and upgrade when certain techs are researched.

In my most recent game, I have 8 10k+ fleets, and 4 are retrofitting, 2 are refueling, 1 is loading troops, and one is moving to intercept the Hive. If I try to break them down into more smaller fleets so they actually kill things, the auto build fleet template will be useless, as it will ignore anything but the absolute most recent generation of ship, instead of grouping them by strategic characteristics, hyperjump speed and fuel.

I also don't want them upgrading until I reach "Torrent Drive" and then "Fusion Balance" (currently researching prerequisites) but that requires turning a bunch of automation off and then back on.

You nailed it, Mudder.

While I was looking for a UI solution to compare designs, your point about the strategic inefficiency of the current system is the real underlying problem.

I couldn't agree more. We desperately need a "Retrofit Doctrine" or Policy setting that allows us to dictate when the AI should trigger an automated retrofit. It is a strategic catastrophe to have fleets spending 80% of their time traveling to drydocks for minor +1 damage upgrades and only 20% actually operating or patrolling.

It would be a game-changer if we could set specific Retrofit Triggers, such as:

Logistical Sync: Only retrofit if the new design changes the Fuel Range or Hyperdrive speed (to ensure the fleet can still fly together efficiently).

Structural Changes: Only retrofit if the Hull Size/Class increases.

Capability Leaps: Only retrofit for significant component tier jumps (e.g., New Sensor generation), ignoring minor incremental tweaks.

Operational tempo is key in a game of this scale. Right now, the constant incremental retrofitting kills that tempo.

[FirstPappy]

Amen!

[OrnluWolfjarl]

You can do two things to aleviate the problem (but doesn't solve the AI issue) :

1. Manually control design of ships (leave ship design automated in the policy level, but toggle the manual/auto design setting in the designs screen for each ship). You can do this on a per ship basis too, so you can leave civilian ships on auto and do military ships manually. That way you control the designs as well, so you can customize more efficiently your battle doctrine. This is the way I prefer. So you can do a design in times of peace, then let your ships to automatically head to a spaceport for retrofits when you are done.

2. Allow design to stay automated, but turn off the automatic retrofit at the policy level. Or turn it off on a per ship basis in the designs screen. This way the AI will pop out designs for you immediately as you research new components, but the ships you are interested in will not immediately fly off to be retrofitted. This is a bit more micro-intensive, but it will allow you to avoid long queues of ships waiting outside spaceports to be retrofitted.

Or you can do a combination of both options, but that would probably require you to maintain separate designs for fleet ships and other automated ships, to carry it out effectively.

[bbenham]

This game spends far too much time upgrading and retrofitting ships and fleets. .....

Completely agree. Even the most minor tech you don't care about forces fleets to come running in for a retrofit, leaving you high and dry. While I don't expect the A.I. to be overly strategic due to the complexity and processing cost, there are some easy "ish" adjustments/settings the developers could add in strategic spots to give the player more control of automated activity.

In my experience, I've found during war, it nearly impossible to track and respond to all the threats and fleets manually, much less do so and also manage strategy. It quickly evolves into a real test of multi-tasking and pausing rather than enjoyable planning and execution. I spend much of my time overriding what the A.I. forces on me because it doesn't really plan like I do. This becomes mind numbing quickly and forces me to just let the A.I. manage 95% of the fleets as I have no hope of duplicating that task. This then sacrifices any hope of directing a strategy myself, which ruins the experience. My hope is that future development will be more focused on automation improvements rather than simply adding yet another race and calling it done.

[OrnluWolfjarl]

I generally do all designs manually. Usually when I research what I want from a tech level. Auto upgrading is useful when learning the game, but it tends to get useless once you understand what components you should be getting for your opted strategy.

It might sound microintensive, but it's really not. You can upgrade the current components of a design with just a click, without adding anything extra. Then add extras when needed (generally first three tech levels, and when unlocking new hulls). Even civilians benefit greatly from this management scheme, because they keep your economy going instead of clogging up your shipyards and doing nothing (although, there's an argument to be had about the increased cash flow and positive resource pricing when civilians upgrade and retrofit on auto).