Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[GUNINANRUNIN]

That sounds like a fantastic idea.

[Steven Shi]

@heskey30

Thanks. I was definitely not trolling in my earlier posts.

I've seen too many devs add in wonderful features with the best of intentions but the end result usually adds up to be less than the sum of its parts.

At its most basic, every feature you add would require the AI to be equally proficient with the them or else the player becomes independent of the game environment. Gamers usually think extra features are cool. I cringe every time a non-core feature is introduced because I can just imagine a 1 year old (AI) having to learn another set of rules. And that becomes exponentially more difficult when the whole IFF, sensor, silent running, CR, pilot management, fleet mission, galaxy level behavior etc start to exert their influences.

Mount and Blade had a tight design by a 2-person team. Was it perfect? Can more meat be added? Of course! But the core gameplay was solid. In comparison, the combat portion of Starfarer is rapidly approaching AAA level in terms of features/polish while the other 2/3 of the gameplay remain unexplored.

Even if time/money is not an issue, the exponential rise in complexity in creating a final game where fighting isn't dominating 90% of gameplay by just Alex himself is incredibly daunting in my opinion.

[Alex]

SQW has a good point, he is obviously not trolling. We all want to see some nice dynamic industry/political stuff going on. Battle is already a strong point in the game. Officers might make it even better, but they also add problems - it is another thing the AI and fleet generation is going to have to match the player over before the game can realize its potential.

I don't think he's "trolling", but he does mention combat mechanics several times, so I do think there may be a degree of misunderstanding there, since none of the recent stuff involves that.

Sensors and abilities will do some to bring the campaign map to life, but they aren't really the emergent, meaty goodness of a solid dynamic world that many of us bought the game for. Not that we're sorry, but it is what we are most interested in out of all the upcoming features, and it still stays vaguely 'soon'.

I get what you're saying, but that doesn't just appear out of one or two big mechanics - it's built up piece by piece. The "emergent" properties in particular almost by definition arise out of multiple mechanics designed to interact with one another; sensors and abilities are two such.

They're not particularly high-level (i.e. it's not industry) but I think it makes sense to nail the "real" lower-level campaign feel first. It'll help drive a lot of the other design choices.

(probably because of less playtesting in the battle simulator )

Ha! The thing is, that stuff is very important to the campaign, particularly with officers being added. Being able to rely on the AI of allied ships is really critical, and most of the time I spent on it was focused on that - making your allied ships more reliable. I think I've made some pretty big steps forward on that front (nailed down some bugs, made some improvements in terms of general smarts), and I think that's time productively spent, even if we're measuring pure campaign impact

To SQW: I believe Alex did say he was going to change the AI behavior on the campaign map, so it hopefully the world won't behave 'pretty much the same regardless.'

Yeah, it's pretty close to being a complete revamp. It's also now officially moddable, with several modules that can be overridden individually.

Spoiler

I just had an idea. Suppose your fleet would automatically (and noticeably) be set to red alert before every hostile encounter. A good way to indicate that would be a klaxon sound at the beginning of the text-interaction with hostiles.
However, there would also be a way to engage in combat where on of the participants is not on red alert: when he has been surprised. I could imagine several ways for that to happen:

- Attacks by a friendly or neutral fleet
- Sneak attacks out of the cover of nebula, asteroid belts etc. - probably best to just introduce a minimum distance at which an enemy fleet has to be detected for their attack to not count as a surprise
- Maybe sudden attacks by a far smaller, inferior fleet (a way to boost the single ship playstile)
- Attacks by stealth ships

A fleet caught by such an surprise attack could not go to red alert before battle and would suffer a CR penalty.

Now, if the player is for some reason wary of such an attack, he could activate the red alert manually via abilities, with the same klaxon sound being audible. I guess it should be time limited with cool-down and cost a bit CR, but far less than being caught unaware.


I think that idea would open up a lot of potential for all kinds of trap mechanics and enable the cool red alert ability without introducing ways to abuse it.

[close]

Hmm. It's a neat idea! I don't know about not introducing ways to abuse it, though - I suspect there'd be any number of ways to get the AI to be ambushed, and that's something that'd have to be managed carefully. Also probably getting too involved, especially for now. The amount of playtesting/tuning that'll need to be done before things are playable in the "this is fun" sense is already pretty immense.

At its most basic, every feature you add would require the AI to be equally proficient with the them or else the player becomes independent of the game environment. Gamers usually think extra features are cool. I cringe every time a non-core feature is introduced

I hear you! Now and again something slips through (and sometimes I end up cleaning it up - there are actually a few campaign-level mechanics slated for that), but generally speaking, I do my best to stay on track and avoid adding random stuff just because it's "neat". Officers, sensors and abilities are all things that have been on the roadmap for, literally, years. They're all core to the experience - officers make larger fleets viable, while sensors and abilities form the foundation for campaign gameplay beyond "click here and wait to get there".

It's a good point about overall scope; Starsector is... ambitious. I'll say this, though - I finally feel like it's all starting to come together.

[Steven Shi]

Thanks Alex for clearing things up a bit.

I classified the last few blog updates under combat because I can't see how officers, sensors and abilities being awfully integral to non-combat scenarios. Personally, I see them as ways to make either initiating combat or participation of combat more engaging.

Maybe because there aren't a lot going on in the galaxy at the moment so I'm seeing any new additions in the context of what exists at the moment...which is all about fighting. =)

Hope to see Starsector gets more fleshed out as a whole in the coming months.

[Histidine]

EDIT: redacted, Internet argument doesn't need continuing

[Alex]

I classified the last few blog updates under combat because I can't see how officers, sensors and abilities being awfully integral to non-combat scenarios. Personally, I see them as ways to make either initiating combat or participation of combat more engaging.

Maybe because there aren't a lot going on in the galaxy at the moment so I'm seeing any new additions in the context of what exists at the moment...which is all about fighting. =)

I can understand how one might see it that way, but you could take that line of thinking further. Suppose Industry is about building ships - this would in turn power a different combat playstyle. Or suppose Industry isn't about building ships, but is about, say, helping prop up the Sector economically and defending your assets - which then gives combat more meaning. Just about everything - well, not everything, but most things - comes back to it in some way, even if it takes a roundabout way to get there.

It'd be kind of strange to spend this much time building it and then not take advantage of having it while building the campaign. The goal isn't to build two completely separate games, you know?

[SafariJohn]

The goal isn't to build two completely separate games, you know?

The more interactions between the campaign and combat, the better.

I can understand how one might see it that way, but you could take that line of thinking further. Suppose Industry is about building ships - this would in turn power a different combat playstyle. Or suppose Industry isn't about building ships, but is about, say, helping prop up the Sector economically and defending your assets - which then gives combat more meaning. Just about everything - well, not everything, but most things - comes back to it in some way, even if it takes a roundabout way to get there.

Hopefully we can also flip this analogy around to fit any other major parts of the final Starsector. For example, saying combat, trade, and etc. all come back to supporting industry. Or combat and industry eventually come back around to matters of trade. And so on.

All interconnected and important, whichever way you look at it.

[Alex]

Hopefully we can also flip this analogy around to fit any other major parts of the final Starsector. For example, saying combat, trade, and etc. all come back to supporting industry. Or combat and industry eventually come back around to matters of trade. And so on.

All interconnected and important, whichever way you look at it.

Hmm, that's an interesting way to look at it. Sounds like a useful thing to keep in mind going forward - not necessarily in any "immediate practical application" sense, but in terms of being another tool for doing design.

[Megas]

I view trade as a means to fuel the war effort.  After all, if about 75-80% of all endgame XP (from all bounties and shortage events) comes from trade, I will build toward it so I can have a stronger warlord.

I see industry as a possible way to declare independence from all of the factions, so I can wage war against all without depending on their resources.  Of course, if trade remains the way to powerlevel, I probably will play nice (grudgingly) with factions so I can get the most power.

[Clockwork Owl]

I see industry as a possible way to declare independence from all of the factions, so I can wage war against all without depending on their resources.  Of course, if trade remains the way to powerlevel, I probably will play nice (grudgingly) with factions so I can get the most power.

My playstyle is quite opposite... I'd gladly get along with NPCs.

[Gothars]

Hmm. It's a neat idea!

avoid adding random stuff just because it's "neat".

Ha, I see what you did there ;P


I will just bring it up from time to time in the hope that it will fit in eventually.
Just like with operation time. And battle time. And sinking ships 

[Alex]

Fair enough "Sinking ships" is an especially cool one, if I ever manage to find time to add "breaking up ship hulks", I'll definitely look at that alongside it.

[SafariJohn]

operation time

As an aside to that, Alex, could right clicking on part of the UI bring up the expanded info (in addition to or instead of F1)? I think it would be very convenient and feel natural.

[Clockwork Owl]

Fair enough "Sinking ships" is an especially cool one, if I ever manage to find time to add "breaking up ship hulks", I'll definitely look at that alongside it.

Is there a way to break the sprite up procedurally in game? Would be much better with some inner greebles(like FTL did) tho...

[Alex]

operation time

As an aside to that, Alex, could right clicking on part of the UI bring up the expanded info (in addition to or instead of F1)? I think it would be very convenient and feel natural.

I don't really see doing that. Right-click is too useful for other things to take away across the board like that. Just one example: you use right-click on an ability button that's already got an ability assigned to it to bring up the dropdown to pick a different ability.

Is there a way to break the sprite up procedurally in game? Would be much better with some inner greebles(like FTL did) tho...

That's one of the tricky parts. Theoretically might be possible to tesselate using the bounds as a guide, but that might, in turn, put some extra requirements on how bounds should be specified. My gut feeling is it wouldn't end up being simple and lots of weird edge cases would come up. Then there's performance to consider.