what would you NOT put in a SMAC remake?

[Buster's Uncle]

https://apolyton.net/forum/other-games/alpha-centauri
https://forums.civfanatics.com/forums/alpha-centauri.27/
https://www.weplayciv.com/forums/forum/5-alpha-centauri/

-Speaking of AC archives research, the former two places go back to the late 90s.

[sthalik]

Like for instance in my case, approaching 2 decades of experience with this game, and plenty of other games in this genre... I know that unit pushing complexity has to be reined in somehow.  Or else nothing has been achieved, it will be the same old crap boring experience.  That most gamers don't even like.  Or they grow out of being able to make time in their lives for such things.  Most people got married, had kids, and took on more responsibilities in their jobs.  I didn't!  So I sit around posting the world's most incredibly long winded AAR for the sake of form, not even because it's that exciting to play the game or write up the game anymore.  But I figure, I should do at least one AAR like that, to see how it ends up.  Then it's gonna be been there, done that.  The bar will have been raised for what needs to justify the time expenditure on such things.

First off it takes balls to admit what you wrote, old chap. Mad respect for that.

Can we talk more about "unit pushing" and possible solutions? Otherwise can you direct me to an existing discussion?

One thing that's a certainty for me, an absolute requirement in any SMAC-similar game, is better control over terraformers.  Currently I work every single tile by hand.  The AI is way too stupid to entrust to such things.  I would like some kind of programmable or priority list specification for what kinds of terrain to develop.  C-Evo had some kind of priority list IIRC.  I'd like it to be fully programmable in some kind of scripting language.  Which would then ideally be compiled so that the decisions aren't bogging down a scaled simulation.  If compilation is too much, maybe I could come up with some kind of "condition mask and transition state" paradigm, where the player is merely selecting or deselecting option bits in a .txt file.

Making it fully programmable or even predicate calculus level won't work for actual end users. I had a program with polynomial coefficients as input. This had to be dumbed down into a linear slider. Users block out when they hear about anything remotely math-related. There's even research on how and why they block out.

There's a story where they taught secretaries to make text editor macros in TECO. The trick was to never say it was programming. They happily wrote the TECO line noise as long as they didn't know what it was.

FWIW compiling to native is a red herring. Just parse to an AST, validate it and naively interpret. Most of script slowness in games comes from running scripts too often and too many. You can only add a constant factor improvement for that.

[bvanevery]

First off it takes balls to admit what you wrote, old chap. Mad respect for that.

Well I just continue to try to make good on my unusual path through Life.

Back in the day, I tried to design a SMAC-like game that would be finished in 5 hours, for perhaps 200 turns of play.  That would be like 1 long game sitting, but not as long as the minimum 8 hours I thought was necessary to win a game of SMAC "by any means necessary".  I threw lots of concerns at that target number; for instance, the amount of time it takes the UI to zoom up to planetary orbit, or down to some local detail.  The 3d graphics would actually have to perform.  None of this "TBS games are allowed to be slow" crap!  4X TBS is a lot of heavy duty simulation with a lot of complexity, so it has to be fast.  Otherwise you will drown in screen scrollings, unit animations, crap where the player isn't actually doing anything and the game isn't advancing.  Those seconds per turn add up into a lot of hours of real wall clock time.  Much like audiences fall asleep or walk out of movies when too much time goes by, there are limits that the attention span of a working adult can be pushed to.

In more recent years, I've thought about serialization as a way to handle the time problem.  What makes me want to return to a game that I saved a day or two ago?  What doesn't?  Am I going to be able to remember what I was doing, or why I cared?  If I can't / don't, then I'm not going to pick up the game again.  I don't really have SMAC-specific answers at this time, I've just formulated the questions.  In some ways, my AARs are an exploration of this problem.  Because when I hit the point of saying "there's no way in hell I'm playing this game anymore" or "TNWIH I'm writing up this game anymore" then that's data about what is memorable or boring about a game.  My current heuristic is I've made a screenshot of anything I've found "interesting" about the game.  If there are no screenshots and a lot of turns went by without a writeup, that part of the game was probably dull as dishwater.

Can we talk more about "unit pushing" and possible solutions? Otherwise can you direct me to an existing discussion?

We can try.  My mind is too addled to point at any clearinghouse of unit pushing discussions.

Making it fully programmable or even predicate calculus level won't work for actual end users. I had a program with polynomial coefficients as input. This had to be dumbed down into a linear slider. Users block out when they hear about anything remotely math-related. There's even research on how and why they block out.

And you're willing to go to the bank on this being applicable to a 4X TBS SMAC-like community of gamers and modders?  I'm not, not offhand.  If you want to say the masses are formula averse, no argument.  But I expect sterner stuff from people who are still into SMAC!  And we don't need the dumbed down version of SMAC, we've already got Civ V, Civ BE, and Civ VI to serve those roles.

I have preferred rules in SMAC that have a more "basically scrutable" formula.  The Bureaucracy formula is, by contrast, tending towards the tedious.  I used to do math competitions.  I wasn't the best competitor as time went on, I wasn't "one of those", but I can basically calculate simple formulas in my head.  The Bureaucracy formula has enough terms that I cannot easily remember it or compute it in my head.  And it was a pretty important formula for me for awhile, when I was working on Huge maps on Transcend.  Until at some point I stopped caring, possibly because I just had enough "feel" for what a real problem was.  I think at some point, I accepted that stewing about the exact boundary of Bureaucracy wasn't a good way to plan an empire.  That I should just accept there would be some drone riots due to Bureaucracy and I could deal with them by building more facilities.  A more "continuous" rather than "integer step" view of the world, if you will.

In general, a lot of game phenomena would be easier to represent if they were made continuous in floating point, rather than contingent upon integer steps.

FWIW compiling to native is a red herring. Just parse to an AST, validate it and naively interpret. Most of script slowness in games comes from running scripts too often and too many. You can only add a constant factor improvement for that.

True, a bytecode representation may be enough.  But I reserve the right to prematurely optimize because it's a career hallmark for me.   

[bvanevery]

https://apolyton.net/forum/other-games/alpha-centauri
https://forums.civfanatics.com/forums/alpha-centauri.27/
https://www.weplayciv.com/forums/forum/5-alpha-centauri/

Thanks for those links; although, going through the old part of those archives, has way too many posts to do it linearly.  I'm not sure what search terms I'd try to glean "what ifs" about desired game features, etc.  There's also a certain point at which I become saturated and cease caring.  I just digested everything in this site's "The Theory of Everything", which goes back to about 2012, and seemed to have the most discussion circa 2013.  Maybe like a boa constrictor I need to lie around for 6 months now.

The new parts of those archives, are quite inferior in volume to this site.  Which is why I'm here; it seemed like the extant critical mass for SMAC on the internet nowadays.

[Buster's Uncle]

Oh yes; there's no SMACX activity to speak of anywhere else anymore.

I understand, BTW -and this is probably true of AC2, too, as all forum internal search engines seem to suck, but we're the only one still using our original forum software, as well as the youngest, which should both help- that to search for something at, say, Apolyton, you're better off googling "Apolyton.net: [search term]".

[Elok]

I'm not sure if this counts, but Re:OP, a lot of SMAC does seem dated, most notably the human genome project, but also the part where all vehicles are human-crewed and thus vulnerable to mindworms.  Even today's armed forces are increasingly automated, with drones and the like.  Of course, mindworm combat was an integral part of game balance and all that, but it's odd in terms of lore.

[sthalik]

I'm not sure if this counts, but Re:OP, a lot of SMAC does seem dated, most notably the human genome project, but also the part where all vehicles are human-crewed and thus vulnerable to mindworms.  Even today's armed forces are increasingly automated, with drones and the like.  Of course, mindworm combat was an integral part of game balance and all that, but it's odd in terms of lore.

A possible mind worm lore fix for a remake -- Attempted robotic cleanup of the Fukushima disaster failed due to the amount of radiation. Even heavily shielded hardware couldn't stand the EM radiation. Humans don't like ionizing radiation either and large amounts are almost-instantly lethal.

I love the original's retro-future vibe, i.e. how people from the nineties imagine the future, with obligatory poor-quality prerendered 3D videos. If SMAC was based on millenial-future, we'd have "Hologram Theater" replaced with "Twitter relay node".

[bvanevery]

The Human Genome Project was dated even when the game was made!  It was being worked on in real life at the time.

Yeah they didn't anticipate flying drones.  And a sin that just about every sci-fi franchise has made, is they don't deal with personal weapons that automatically hit things.  That's because they keep wanting to write stories about flesh and blood actors who wield firearms.  Westerns in space.  Star Trek and Star Wars are both ridiculous about that, but at least the original material was a lot closer to the era of the Western.  We'll have to forgive their franchises for surviving so long and bringing with it all the baggage.

I'm not seeing how piles of radiation fixes anything?  If radiation kills both robots and humans, that doesn't lead to manned vehicles.  It can lead to training mindworms to go into hazard zones, but I think then you've got a story about "radiation worms", not mindworms.

Anyways, I don't personally care about and am not married to mindworms.  I don't consider them, or Planet, to be an essential part of the game.  I always perceived them as a Dune ripoff, and I don't like ripoffs.  Any environmental hazard can take the place of mindworms, game mechanically.  Could be radiation, could be planetary winds, could be lava, whatever.  Mindworms offer a way to use the environment for combat, and those things can be done with radiation, storms, and lava too.  Just make the planet "inhospitable" somehow and you'll have your "Man vs. Environment" theme.  Humanity will tame it somehow or die trying.  If there needs to be some horrible fate in store for humanity, the whole atmosphere can go up in a super caldera or something.  Stargate Atlantis did a pretty good episode along those lines, featuring Rodney's hubris with thinking he could make some ancient geological energy source work.  Whole planet melts, they barely fly out of there.

I never liked the Voice of Planet either.  Sounds German to me.  I have a lot of trouble with the idea that Planet is German.  "Ya goot, zie sprocken witz ze tiney knickerbockeren, ya ya?"  It's like the Monty Python sketch about the most lethal joke in the world, with the British troops uttering it in German without any understanding.

I can do without a lot of the low fidelity rendering in the videos.  I've come to appreciate that someone was probably just banging them out best they could, and couldn't afford to spend much time on them.  But I'd still up the level of cheese from what it is.  Especially the "oh, so you know how to do wipes in a video editor" stuff.