Let's Play Dominus Galaxia

[Jeff Graw]

Huh.   Don't you have to worry about planets at max pop with or without OC mod in place?  I personally don't play that way (pedantic about pop growth) so perhaps you have a point.  But if you're obsessed with moving population around at 99/100 or 109/100 -- is there a difference?

When you're at 109/100 and hardly have any opportunity cost it really is a no-brainer to send that excess pop whenever you can. Without the OC mod you at least have a trade off, in that you're speeding up expansion at the expense of lowering production for some time from the worlds you send from.

I think MoO3 experimented with automatic population spread where nearby colonies would send their excessive growth to neighboring (not maxed) planets in some organic/behind-the-scenes way?  Would we see something like that to fix the mathy problem?

Yeah, we also had an idea like that. Partially implemented it even. It didn't seem like it was going to work well, so we fell back to a MoO 1-esque system. I wouldn't be surprised if we played around with things some more in the future, maybe in terms of opportunity cost (maintenance penalty for each transport fleet enroute?).

Trying this again so it's a little clearer than mud:  What matters is the difference between the abilities of each side to infer how the battle may, can, or will end.  The simpler and especially the more intuitive the combat system, and the more nearly deterministic (or more nearly pure-dead-random; it's easy to predict the probability range of a coin flip) the easier it is for humans to get their brains around it.  The more you add weirdly-interacting random factors, the more the human is left shrugging and saying, "I have no idea," barring a deep understanding of the rules behind the system and a spreadsheet with which to let a computer do the work instead - which means that even a rudimentary calculation by a computer will do a better job and have a big advantage.  Even oversimplifying massively to a binary outcome (which is a mistake) 100% certainty isn't much of an advantage over 95% certainty.  80% certainty has a HUGE advantage over 30% certainty.  Using random numbers to make calculations harder doesn't bother computers, which are literally calculating machines, but it can really tie humans in knots unless you come up with a way to make it somehow naturally intuitive.]

Sure, I get what you're saying, and you do have a point. However, the rule itself still has the premise of "simpler and more intuitive." It's entirely possible to have a combat system that is deterministic yet complex for humans to figure out. Keep in mind that I'm a devil's advocate by nature, so when someone proposes a rule I'm compelled to try and poke holes in it

Approaching things from the practical angle as opposed to purely theoretical, and therefore maybe slightly off topic albeit interesting, another factor is that if we had a carbon copy of MoO 1 combat, with or without random factors, the AI would be able to calculate the outcome (or likely outcome) of every single possible encounter and gain precognisance/spooky prediction ability. However, the DG combat system is computationally complex enough compared to the MoO 1 equivalent that if we were to run every possible encounter through a proper simulation we'd slow down turn processing to a crawl, so we only use a broad abstraction.

This doesn't matter to the game, but it annoys me because either we're speaking completely different languages or it's strictly wrong.  You can't even theoretically memorize information to which you don't have access, so with no database of all possible moves in a game, it isn't memorizable.  Even if there were a database of every possible position, you'd have to calculate how to get from each position to the next, which is math, not memorization.  Also note the best Chess computers use "memorization" only for opening moves and endgames, and as I understand it the best Go computers barely rely on "memorization" for anything, if at all.

I think you're speaking the language of the player, and I'm speaking the language of the programmer. There also might be some conflation between practical manifestation and fundamental state (I'm not saying that Chess isn't practically strategic, far from it, practically speaking it's mega, extremely strategic). And no, if you did somehow attain a database of all possible positions, you would use pointers that link to the next position until the game is finished. There's not much in the way of math there, although you could make the argument that everything computers do is math. And yes, I'm aware more or less of the current state of Chess and Go AIs, and I don't mean to make any kind of assertion that we're even close to being able to solve either one. Maybe in 100-300 years we can solve chess, at least discounting obviously absurd or losing strings of moves, and it's unlikely we'll ever solve a full Go board.

The UI I saw in the video would be a good addition to MoO's, but is it meant to be a replacement?  When the number of planets with spaceports gets large, I feel like it could be daunting.  I'm also not convinced that removing the ability to grow pop helps, especially in combination with the interface as I understood it from the video: You have to remember which half-dozen (or more) planets you want to use to send a little pop each so they'll stay in the sweet spot of growth/factories/etc. and that seems way more micro-intensive in the later game than what I do in MoO:  Pick a nearby planet or two whose production I've decided I can spare, and have them hurl a big raft of transports over to the new planet and regrow immediately.

(There is some super-micro-intensive stuff in MoO, like the pop-growth super-optimization Thrawn talked about in our SG thread, and getting rid of forced-pop-growth kills half of that, but that would still leave the transport half on the table unless - as I'd hope - DG uses a growth formula that works a little more smoothly and so isn't as easy to game ... which should solve both halves of that problem, I would think.)

Yes, in terms of micro we're targeting players like Thrawn. There's a ton of micro that I don't do when I play MoO 1 because I can't be assed to, but that's different for every person, and I feel that having more micro does hurt the overall design of the game. I mean, we could only worry about the majority of players, but then the majority of players are just fine with many 4X games you or I wouldn't touch with a ten foot pole

And no, it's not meant as a replacement per say. I intensely dislike MoO 1's transport mechanics. We just weren't able to come up with anything better, so settled on an improved implementation of what is in MoO 1. We didn't eliminate the micro, but made it much less painful. Of course it gets a bit more daunting when you have more worlds to send from, but the ability to sort the list by either pop or distance, with the ability to see all pertinent information like fertility fairly easily, makes it pretty simple to manage. I haven't had any issues with it, but I tend to only play on 100 star galaxies.

As far as pop growth goes, the formula is very similar to MoO 1, except waste also has an effect on pop growth (keep in mind waste is represented as a percentage. There is no accumulation). The higher level of factory automation you have, the less of a haircut you take on production when population dips below the level needed to control all factories, at the expense of slowing down population growth a bit. You could theoretically have dedicated "population farm worlds" with few factories, but with the current balance at least I'm skeptical it would be worth it.

[RefSteel]

Keep in mind that I'm a devil's advocate by nature, so when someone proposes a rule I'm compelled to try and poke holes in it ;)

Now, who does that remind me of? It's on the tip of my tongue, probably someone who yammers on and on about all the things that might go wrong with crit mechanics or something, dunno, I'll think of it eventually...

Approaching things from the practical angle as opposed to purely theoretical, and therefore maybe slightly off topic albeit interesting, another factor is that if we had a carbon copy of MoO 1 combat, with or without random factors, the AI would be able to calculate the outcome (or likely outcome) of every single possible encounter and gain precognisance/spooky prediction ability.

That's surely true. Although the question is what it would do with that information.

However, the DG combat system is computationally complex enough compared to the MoO 1 equivalent that if we were to run every possible encounter through a proper simulation we'd slow down turn processing to a crawl, so we only use a broad abstraction.

Makes sense to me. The key question is really how intuitive it is for the human player, which isn't really something I can guess based on one video.

And no, if you did somehow attain a database of all possible positions, you would use pointers that link to the next position until the game is finished.

Until you attain that database though, complete with pointers, you're solving a problem, not memorizing. (For a database of all positions without pointers, the problem is basically creating the pointers.) For a human, it's a spacial problem because that's something we're good at. For a computer, it's a math problem because that's what it's good at.

Maybe in 100-300 years we can solve chess, at least discounting obviously absurd or losing strings of moves, and it's unlikely we'll ever solve a full Go board.

It's interesting: How do you discount obviously absurd or losing strings of moves? On the other hand, if Western civilization keeps up the current pace, I wouldn't bet against computers solving Chess sooner than that - or even Go!

Of course it gets a bit more daunting when you have more worlds to send from, but the ability to sort the list by either pop or distance, with the ability to see all pertinent information like fertility fairly easily, makes it pretty simple to manage.

The sortability and readily-visible information sound great, although I'd want more sorting options, and I do want to be able to do it from the map to get a visual sense of where the transports are coming from and where they're going.

The higher level of factory automation you have, the less of a haircut you take on production when population dips below the level needed to control all factories, at the expense of slowing down population growth a bit.

This is a mechanic of which I'm skeptical, but I haven't seen it in action enough to be sure how best to use it. My sense right now is that having unmanned factories would still be a mistake, especially because of the way Alex said factory costs increase. I don't know the system in detail though, so I can't be sure.

You could theoretically have dedicated "population farm worlds" with few factories, but with the current balance at least I'm skeptical it would be worth it.

Yeah, this is an interesting balance point surrounding the Spaceport mechanic. Poor and Ultra-Poor worlds (if they still exist) would be good candidates for pop-farm worlds except I imagine their penalty would also apply to the spaceport, pushing them toward becoming pure research colonies. The truth is there's probably a findable "sweet spot" of population at which to build factories and/or spaceport at any given world though (even if it's beyond the world's max size for some/all UPs and 1 for the likes of URs).

[Jeff Graw]

Now, who does that remind me of?

Heh.

Makes sense to me.  The key question is really how intuitive it is for the human player, which isn't really something I can guess based on one video.

In case I accidentally implied something along the lines of "DG combat is so complex it slows down the computer!" about 99% of the performance cost is down to path finding.

It's interesting:  How do you discount obviously absurd or losing strings of moves?  On the other hand, if Western civilization keeps up the current pace, I wouldn't bet against computers solving Chess sooner than that - or even Go!

One way is to calculate a decision until completion, and then throw out all of the positions that lead to non-optimal outcomes. This is assuming that the bottleneck is storage as opposed to processing power. It's unlikely that humans will ever solve a 19x19 Go board. There are significantly more possible positions than there are atoms in the visible universe squared. The kind of technology required to process and store that would leave the likes what we see in Star Trek to shame.

The sortability and readily-visible information sound great, although I'd want more sorting options, and I do want to be able to do it from the map to get a visual sense of where the transports are coming from and where they're going.

You can also click on the name of each planet in the list to select it and get further info. We could also add a sorting option based on how full each world is.

This is a mechanic of which I'm skeptical, but I haven't seen it in action enough to be sure how best to use it.  My sense right now is that having unmanned factories would still be a mistake, especially because of the way Alex said factory costs increase.  I don't know the system in detail though, so I can't be sure.

In most situations, yes. But they're less of a mistake, which weakens the allure of "I should micro this to get optimal results." One idea we're playing around with (eg. heartily debating) is using a momentum system similar to what we do in research that would mean any change in planetary production allocation comes with a haircut. Theoretically, the combination of the two might drastically reduce micro.

Yeah, this is an interesting balance point surrounding the Spaceport mechanic.  Poor and Ultra-Poor worlds (if they still exist) would be good candidates for pop-farm worlds except I imagine their penalty would also apply to the spaceport, pushing them toward becoming pure research colonies.  The truth is there's probably a findable "sweet spot" of population at which to build factories and/or spaceport at any given world though (even if it's beyond the world's max size for some/all UPs and  1 for the likes of URs).

It's not that difficult to build a starport on poor worlds assuming you're able to pour in enough stimulus. Stimulus in DG, unlike MoO 1, does not take into account mineral richness or other attributes, since otherwise it's fairly obvious to stim rich and artifacts worlds, and also obvious not to stim poor.

About the sweet spot, yes and no I think. Increased factory building cost pushes out the date at which the investment becomes positive further and further out into the horizon, but given a long enough time scale, building factories should be the right move, assuming there are no other factors. There are other factors though, especially other empires (which is not a constant), so when to stop building factories is going to be fairly situationally dependent.

Keep in mind that, at least for now, factory doesn't become too significant until later in the game.

[RefSteel]

The kind of technology required to process and store that would leave the likes what we see in Star Trek to shame.

On the other hand, especially as you go to earlier versions of Star Trek, we have tech now that puts some of theirs to shame.

You can also click on the name of each planet in the list to select it and get further info. We could also add a sorting option based on how full each world is.

Cool.

One idea we're playing around with (eg. heartily debating) is using a momentum system similar to what we do in research that would mean any change in planetary production allocation comes with a haircut.

One unintended consequence, unless something is done about transports too: Grow a given planet to a good place in the growth curve, building facs until you have just enough for that population level, then change production allocation (just the once, to minimize the penalties you receive) and ship transports away every turn to keep the planet at the "right" population level.

About the sweet spot, yes and no I think. Increased factory building cost pushes out the date at which the investment becomes positive further and further out into the horizon, but given a long enough time scale, building factories should be the right move, assuming there are no other factors.

There are always other factors. In particular, whatever you could be building instead of factories (tech, war ships to go a-conquering, etc.) repays its investment at a harder-to-calculate-but-nevertheless-real later date. If factory break-even dates get far enough away, they actually never break even due to these opportunity costs. Early techs to allow unworked factories to be marginally less terrible seem like junk techs to me since the slightly-less-terrible return is still not something you want to use. And by the time you get high-end techs that allow unworked factories to approach the value of worked ones, if you still have planets that aren't fully populated and factoried, any extra factories built to take advantage of the new tech probably have a payback horizon somewhere beyond the end of the game.

Those techs can have value if you have fully-maxed planets and want to launch an invasion of somebody, but without forced growth and with increasing factory costs, I have doubts about the viability of maxing planets in the first place - apart of course from special worlds from which you wouldn't want to send transports anyway. There's an argument that if many of your planets are approaching full populaton, you should be expanding more, even if that just means conquering somebody who put production into top-expense factories instead of a fleet. Honestly, nearly all the mechanics changes I've seen so far suggest that constant offensive war is probably the strongest play ... which might just mean that you'll need to balance this out with strong value available from positive diplomacy.

(Another thing on transports: I think you mentioned the idea of per-turn maintenance for them somewhere, but I don't think that helps with the micro. You already have the per-turn cost of having population doing no labor in space. One thing that might help is a cost for each planet that sends transports to a given world on a given turn: The cost of launching a raft of X transports from a given planet is XC + K, where C and K are constants and K is large in comparison with C.)

[Reformations]

Thanks for the comments on overcrowding. As I said, I don't play the micro game when it comes to planetary management (why I like MoO1 better than 2) so it is good to have feedback from that point of view. I think giving it as a racial bonus may be too intense for the micro-er's out there. However, there may still be potential as a planetary bonus for players to get excited about discovering (instead of just riches). Natural fertile/gaia are somewhat nice but of course mitigated by technology. What could be a strong planet-level population modifier that players would get excited about discovering? Just high growth (farms)?

---------------
To discourage transport micro, what about something that increases the cost of each transport when done in consecutive turns? Something like:

Cost = max( 1, 3*(5 - #of turns since last sent transport))

Sending back to back transports, the latter costs 12 per unit (still cheaper than would-be cloning)
Waiting 1 turn in between brings cost down to 9.
2 tuns in between is 6
3 tuns pause means cost is tripled of next group.
Finally, if you send ever 4 tuns in between the cost remains 1.

The formula above allows 2 parameters to customize.

Of course this still can result in a solvable math problem for the microers to figure out. Maybe it would be easier to just not allow transports to be sent within 3-4 turn windows? Then the player has to keep track of all the 'available' planets? Maybe the gui could do that... (Micro-er's are kind of annoying to cater to )

-------------------------------------
Would empire wide edicts (after researching terraforming or RC) also be penalized by your anti-micro proposals? If the goal is against micro, it seems that those types of events could be 'freebies' in terms of penalizing the deltas. I *think* I would support a mechanic that kept people from checking decimal points every turn. (Yes, I know that this lowers the skill ceiling)

[RefSteel]

Wanted to point this out, re: computer-aided battle predictions and combat terrain, since it came up in the succession game we're playing: Even a completely deterministic combat system is not perfectly predictable before the battle begins if terrain (e.g. the presence of asteroids) is randomized and its presence and location can affect the outcome (even though deterministically). That is: As with the galactic map, the random factor can be made part of the landscape to which players have to respond instead of being built into ongoing mechanics. Of course a perfect AI would know exactly when it was safe to attack and when it had to retreat (given a set of strategic objectives) for a given asteroid/nebula/whatever-terrain layout and an assumption of perfect play by its opponent once the fight begins, but first of all this isn't necessarily a bad thing, and second of all even that isn't the case if information about the map and/or opposing fleet is hidden by some version of Fog of War.

[Jeff Graw]

Apologies for dropping off the radar for a bit, but I've been super busy trying to get things ready for a potential crowdfunding campaign. We added a whole slew of features recently, including a tutorial, neutral forces, and pre-combat hailing.

In fact, with the latest improvements I'm pretty happy with where things stand. As such, I've sent out ten beta invites to prominent RBers!

[Alhazard]

Will this be PC only?

[Jeff Graw]

Most likely, at least for launch. Touch mostly works well though, so I won't say never to an IOS/Android tablet version.

[Alhazard]

ok thanks. Let me ask a 2nd question. Any plans for an OSX version?