Post reply

Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.

Note: this post will not display until it's been approved by a moderator.

Name:
Email:
Subject:
Message icon:

shortcuts: hit alt+s to submit/post or alt+p to preview

Please read the rules before you post!


Topic Summary

Posted by: Froggiest1982
« on: June 29, 2022, 07:45:21 AM »

I also think that the C# version still lack features we had in VB for multiple faction gameplay to make it smoother and fully enjoyable. In VB you could easily change technology, ships. wealth and parts between empires in a way you can't in C#. You could for example have one empire build an advanced sensor and then sell that to other empires or have them license build them for a price for example. Doing the same thing in C# is much more complicated. You also can't edit Wealth in C#, not that I know of... which is a bit weird.

Why you want to kill me reminding this?  ;D
Posted by: Jorgen_CAB
« on: June 29, 2022, 03:22:58 AM »

To be fair Steve have been developing both the tactical and strategic AI quite substantially in C# over VB version of the game and continually enhanced and tweaked it. So saying he does not care for it that much is in my opinion not fair.

I do however agree that it would be fun if the AI would split their fleets if they are being kited in beam combat, that is a pretty valid tactics. This is also why having a homogenous fleet speed in practical terms not always is as good as many people think it is. Lower speeds usually means more room for weapons or other components and having various speeds on ships also make it much harder for the opponent to have a homogenous speed to, which ships are they targeting... your slower or faster ships. If they target your fastest ship it might become very expensive for them, not a strategically sound option most of the time.

In any way... Steve probably could tweak the AI as he go and probably will over time.

I do, however, agree... playing multiple factions is the best way of enjoying Aurora. But this is a niche way to play and not everyone are able to enjoy it that way. In the biggest game I played I had a total of twelve faction, while fun it did take an awful amount of time to set it up and execute it and was quite complicated.

I also think that the C# version still lack features we had in VB for multiple faction gameplay to make it smoother and fully enjoyable. In VB you could easily change technology, ships. wealth and parts between empires in a way you can't in C#. You could for example have one empire build an advanced sensor and then sell that to other empires or have them license build them for a price for example. Doing the same thing in C# is much more complicated. You also can't edit Wealth in C#, not that I know of... which is a bit weird.
Posted by: nuclearslurpee
« on: June 07, 2022, 02:00:00 PM »

Whether or not Steve can improve the AI (probably) or whether he wants to take the time and effort to do so in a passion project I don't know.  I can say that it would probably be a huge investment in time to make meaningful improvements to the kiting AI. 

I've seen kiting work against AI in expensive AAA titles. 

Steve also seems to prefer the kinds of stories where wars are won on the strategic and logistical scale rather than by superior tactics. That is, ensuring that the player's forces have the advantage in every engagement, avoiding engagements if there is not an advantage, and avoiding wars if the enemy has the advantage. So while improving the AI is an ongoing project, advanced tactics are not so far something that has been very interesting for Steve as the programmer it seems.


Quote
As for the AI adjusting their design philosophy on the fly in response to the player?  I can say with lots of certainty that this doesn't happen.  I'm pretty sure each AI race has a design philosophy set at race creation.  And adheres to that until they die.

This is correct. There is currently no allowance in the DB or game code for the NPR to change its design doctrine. To a certain extent the NPRs will gain intel on the player's ships and learn a few things about how to counter them, but this is purely on a tactical level and not to the degree of advanced formation tactics.

It is unlikely such a change would help anyways, at least unless the NPRs were strategically able to keep up with the player (i.e., economically) to build large enough empires that a single war of conquest will not wipe them out in less than a year - and the diplomacy features extended to support negotiated peace. Right now, most player-NPR wars are decided in a pretty short time as the player crushes the NPR fleet and invades or annihilates their colonies; most of the challenge is on the strategic and logistical side, or in not provoking a higher-tech opponent (or else building super-cheesy designs to exploit the NPR, if you're a boring person  :P).  Otherwise, I think long-term campaigns with multiple player races are the way to go for people who want serious intrigue, at least until the diplomatic toolkit and AI are significantly expanded.
Posted by: brondi00
« on: June 07, 2022, 01:46:11 PM »

Whether or not Steve can improve the AI (probably) or whether he wants to take the time and effort to do so in a passion project I don't know.  I can say that it would probably be a huge investment in time to make meaningful improvements to the kiting AI. 

I've seen kiting work against AI in expensive AAA titles. 

As for the AI adjusting their design philosophy on the fly in response to the player?  I can say with lots of certainty that this doesn't happen.  I'm pretty sure each AI race has a design philosophy set at race creation.  And adheres to that until they die.
Posted by: nakorkren
« on: June 07, 2022, 09:39:46 AM »

I'm not sure how Steve programs the AI, but I would love to see some tactics added to the AI's repertoire that complicate player tactics, specifically kiting.

If you have a single beam ship which outranges and outruns the enemy fleet, you can pick them apart over a long time just by kiting. The AI could evaluate when they have gotten into that position and perform one of several options.
  • First, they could split their fleet into one or more groups and try to encircle the kiter, potentially even having the active target(s) draw back while the other fleets press out and around.
  • Second, if some of their ships can "tank" the damage on shields, they could have that part of the fleet continue to pursue, and have the others retreat to a planet with STO.
  • Third, they could sacrifice some ships to continue chasing and being destroyed and pull the rest back to a planet with STO.
  • Fourth, if some of the ships in the fleet are faster than others, they could split off the faster ships to either attack the kiter (if faster) or run for a planet with STO (if slower). That would include fighters, if that AI uses them.
 
Also, does the AI currently adapt its ship design templates to the player's ship designs and tactics? I.e. it would be really fun if the AI could recognize a player's reliance on missiles, fighters, kiting, etc and react by changing their preferred ship templates.

I suspect these kinds of behaviors are much harder to achieve in practice than they are to just write up, so I understand if they're not feasible, but it would be a big boost to players who prefer to play against the AI.