Getting Started with SMAC & SMAX

[BitGamerX]

Hello all, I've just picked up the SMAC & SMAX bundle from GOG. I bought and played SMAC when it was originally released but I never gave it a fair shake and have regretted not getting back into it. My question is getting a patched up and stable release on a Windows 10 1809 machine with a 1080p monitor. I've been reading up on the all the great unofficial patches that the member of this community has created. From my readings I think what's listed below will give me the best starting place. Can you guys please look it over and let me know if I'm on right track or provide suggestions if I'm not on the right track.

First Install: SMAC & SMAX (GOG Version) - setup_sid_meiers_alpha_centauri_2.0.2.23
Then copy over and apply the files from: Yitzi's patch 3.5 (into the root of the Alpha Centauri install folder)
Then copy over the file from: Yitzi's patch 3.5d (terranx)
Finally apply: PRACX

[Vidsek]

   That looks like the drill to me.   

  Be aware that Yitzi's patch is also something of a mod.  I like it, and many others, but Yitzi has been off in RL for a while and his work is not currently being maintained or updated.

  Scient has made a patch that mostly deals with bugs, and is currently working on more decompiling of terran.exe's undocumented code plus more bug squashing.

  Induktio is also working on an .exe editing mod that is an early work-in-progress at the moment.

  bvanevery has created a very nice pure .txt mod (just alphax.txt and the faction files) called AI Growth Mod.  It addresses many of the problems with the AI without any code editing as well as totally rearranging the Tech tree and what you get from each Tech.  He has tried to balance the factions and sort out progression of when you access new stuff to give more choices of strategy and a more interesting, less frustrating game.  He's currently working on some final tweeks and adjustments.
  It is possible to merge AI Growth with the .exe edited patches/mods.  He's testing it with Induktio's work, and I'm (slowly) investigating a merge of it with Yitzi's patch/mod.

  There are also other slightly older .txt mods, many or all of which should be available on this site or have links to the sites where you can get them.

  That's a lot of choice and choosing, probably way too much for a sort-of beginner.  Best advice might be to start with just the more or less basic game, possibly with a bugfix patch like Scient's, to get a feel for it, then decide if you would like to try out the mods.
  Other folks will hopefully chime in with their opinions and suggestions, since I'm hardly the most experienced or knowledgeable person here.

  There are a lot of interesting threads in the forums, both new and old, which can be a fine way to waste um, spend time.  The WIKI is unfinished (work available, inquire within) but has some good info.

  If anything doesn't work with your install, folks here are always glad to help.

[Induktio]

Thinker mod also includes Scient's patch v2.0 so you can play both of them at the same time. It's been in development for many months so the features are getting pretty stable by now. It does not support Yitzi's patch however. Just remember to download the develop build because the latest release is pretty old currently.

[bvanevery]

He's currently working on some final tweeks and adjustments.

I know you mean well in your summary of my SMACX AI Growth mod, but I want to be sure the original poster is clear what's going on and doesn't get any inaccurate ideas about the quality or finished-ness of my work.  My work has been heavily tested for playability for the past six months of releases.  There's not been anything broken or desperately in need of tweaking or showing a lack of polish for quite some time.  Every single release of the past 6 months, would be a perfectly valid mod to try IMO.

Yes I think I ultimately did better or I wouldn't have made more releases.  But I think the quality of my releases has been pretty darned good for quite some time now.  As I've often said when making a public release, say to Reddit: "This isn't amateur hour.  This is professional grade game design."  I test my work, I don't just throw it over the fence and hope all is well.

What happens in the real world, is eventually someone else comes along with a perspective completely different than my own.  They say, "What about this?" and then I'll be like, oh, I never really encounter that in my own gameplay!  Let me have a look at that.  So for instance, the 1.29 release that I'm working on is mostly about nerfing the whole Condenser, Borehole, and Supply Crawler thing.

The rate of change has definitely slowed down.  I'm on a self-imposed minimum 1 month between releases now.  And unless I get a lot more feedback than I have to date, I expect I'll soon switch to a "quarterly maintenance" cycle, with almost nothing to do.  The concept of final tweaks and adjustments is meaningless.  The product is out the door, has been out the door, for quite awhile now.  Go play version 1.28 and be happy.  The rate of change just slows down over time, because IMO I've fixed lots of things.

In the coming year my energies will be focused on making a brand new game of my own.  Quite possibly a 4X TBS in the mold of SMAC, but no promises yet.  I've just invested so much design time in this game, that I think it might be wise to reuse a lot of that design.

[dino]

Both Ytzi and thinker can be used with https://github.com/DrazharLn/smac-in-smax mod, that allows you to play original Alpha Centauri with updated and patched executable from Alien Crossfire.
I'd recommend original for the first complete playthrough, since expansion is a bit bloated with features, unbalanced factions and modified setting is a bit cheap and inconsistent.

Ytzi patch is just a patch unless you edit alphax.txt to change its default config.

Thinker, doesn't change rules only improve ai, you'll probably loose against it even on medium difficulty setting.
On the other hand you'll be forced to learn basics properly faster, by observing what ai does.
Is it suitable for a new player? Probably depends on a player, but if you stick with the game long enough, you'll end up with thinker anyway.
Once you learn the game, there is no point playing singleplayer without it.

If you want to have freedom of doing whatever you want in your first games, because of incompetent ai, play Ytzi patch. If you want to understand the gameplay faster, play thinker.
Do not play mods that change rules untill you know vanilla game really well and have an idea what you would like to have changed.

[Vidsek]

   Ha!  Did I say other folks would chime in?

  My intro was a bit vague on the mods and patches since there is so much to them.  However, I agree that in stressing that Thinker and AI Growth were live projects with the creators still present and paying attention to them, I gave the erroneous impression that they were unfinished.  Which of course is not true.

  And thank you all for the clarifications and additional info.

[bvanevery]

Ytzi patch is just a patch unless you edit alphax.txt to change its default config.

Yitzi changed the encoding of 3 things in alphax.txt, so that they don't mean the same as they do in the standard game.  This can have gameplay consequences.  I can't remember if his tweak to "Stockpile Energy" will trigger a crash or not.  I should probably try it out with my mod sometime to see what really happens.

Thinker, doesn't change rules only improve ai,

There is 1 case where this is false.  It is substantial and worth mentioning.  Thinker AI freely ignores the faction compulsions specified in the various faction.txt files.  You will get Gaians who are Planned but chew you out for not being Green.  You will get a Hive that lambasts you for not being a Police State while being Fundamentalist.  They will go to war with you over this stuff.  This occurs regularly in Thinker games and it's highly annoying to a seasoned player.  It doesn't just break lore, it also breaks diplomatic gameplay.  I wish Induktio would reconsider this genericizing of factions and do it the way the original game does.  I do not personally feel that gaining minor benefits in faction performance is worth wrecking how diplomacy works, and he has many other tools in his toolbox for improving AI play.

On the other hand you'll be forced to learn basics properly faster, by observing what ai does.

It's going to force you into adopting a very particular style of gameplay though.  Infinite City Sprawl, Condensers, Boreholes, and Supply Crawlers.  These turn out to be pretty much the Golden Path through the game.  If you think games are inherently about minimaxing and abusing whatever the rules put in front of you, in short if you are strongly Gamist, then indeed this might be good advice.  But if you think games are supposed to also be about Narrative or Simulation, you really should just play the original game first a lot.  See how it is, what they actually gave us in 1999.  See firsthand why people recommend this game 20 years later.    Don't take other people's say-so about how it should be improved, use your own judgment.  When (not if) you become dissatisfied, try someone's mod.  As you can probably tell, people differ about what they think is important to do in a mod.

Once you learn the game, there is no point playing singleplayer without it.

I disagree, which is why I wrote SMACX AI Growth mod.  Perhaps you should play it more.  Much can be gained by redesigning the factions, the social engineering choices, and the tech tree, without changing the stock AI at all.  Yes, having done all of that for a year, I can still think of things I'd like the AI to do better.  There are limits to what one can accomplish in a *.txt mod.  However, the limits are a lot farther out on the design horizon than you might think.  If someone gives you a crummy unbalanced tech tree, and you swap it out for a good balanced one, much is different.  There are also a few actual inputs to the AI in those faction.txt files.  Tweaking those does produce better AI performance, if not Thinker mod levels of betterness.

If you want to have freedom of doing whatever you want in your first games, because of incompetent ai, play Ytzi patch.

I agree that Yitzi does nothing whatsoever to improve AI performance.  Additionally, it fixes bugs that I personally don't experience much or ever in single player.  Although I admire the amount of work he did, the end value to me personally is pretty close to zero.  I think Yitzi hunted targets of opportunity, as opposed to chasing down bugs that really really matter.  YMMV, maybe multiplayer needs his fixes more, but I just haven't seen the need or motivation for his patches in single player mode.

So I don't tell people, oh, go get Yitzi's patch, it's a patch, patches have to inevitably be good and important, don't they?  No, they do not.  I recommend the stock Good Old Games binary.  They did fix some bugs and in my experience it works.  That's all I really need SMAC to do, I need it to work.  I've been getting along just fine with the official binary, no unofficial patches, for years now.  Works for me, can work for you too.

If you want to understand the gameplay faster, play thinker.

OR you could read the vast quantity of After Action Reports I wrote up about the stock game, long before I started working on SMACX AI Growth mod.  They aren't going to teach you to play like Thinker, it's not my play style.  But they will teach you plenty about playing the game.

Do not play mods that change rules untill you know vanilla game really well and have an idea what you would like to have changed.

I pretty much agree.  I'd quibble about "really well".  I'd say "well enough that you actually know something about the game."  When you notice you are irritated about something, it's time to try a mod IMO.

Oh, and, uh, the unmodded Alien factions suck, m'kay?  I told you so, and I'm not going to be the only one.  The human factions in the expansion pack are goofy, like cartoons.  They're narratively weak compared to the original 7 factions.  I hate Svensgaard more than any other character in the game, particularly his totally un-Pirate like voice acting.  But game mechanically, they actually had some thought behind them and are ok.  The Aliens suck though.  Suck.  Really went into dumb space with all that, and their units on the map look like World of Warcraft orc rejects.  Think we need a new 4X that doesn't have dumb stuff in it.

[Vidsek]

  Whatever Yitzi did with Stockpile Energy, it now requires it to have a cost of 1.  Making it free (0) will cause the crash.
 That is with or without AI Growth.  Personally, I don't particularly care if SE has a small cost, but the crash on 0 is a bug.

  With Conventional Missiles he added in choices for how cost and power are affected by reactor type.  The options are:

Conventional missile rules:

This controls how the cost and power of missiles are affected by the unit's reactor.

1. They are not affected.
2. The power is multiplied by the reactor value.
3. The power and the weapon cost are both multiplied by the reactor value.
4. The power is multiplied by 2^reactor.  (i.e. doubled for each reactor level.)
5. The power and weapon cost are both multiplied by 2^reactor.
6. The power and weapon cost are both multiplied by 2^reactor, after which the weapon cost is increased by the
   (unmultiplied) power (causing the efficiency of missile weapons to increase with reactor, but fairly slowly).
7. The power is multiplied by 2^reactor; the weapon cost is multiplied by the reactor value.

  I forget how the stock game deals with this, but if you want the original style, pick whichever of these is closest/the same.

  Like you, bvanevery, I've totally forgotten what the third one is.  Since I play on Yitzi all the time, it must not be something that has made a big impact on my games.


  I don't totally detest *everything* about the Alien factions, but I never play with them either.  One of these days I'll get around to replacing them with a couple of the many, many modded factions available on this site and others.  If I rename the replacements Caretakers and Usurpers the game might not even notice they're gone 

[bvanevery]

  I forget how the stock game deals with this, but if you want the original style, pick whichever of these is closest/the same.

The style is not the problem.  The problem is he also changed the value for a Conventional Missile's attack rating, because he changed a formula inside the game binary.  So encodings of "12" or "18" don't mean the same things to a stock binary or a Yitzi binary.  This is a screw-up and it's not easy to resolve without poring over the binaries, or doing a fair amount of empirical testing to reconstruct how the behavior has changed.  I tried to find documentation on this once upon a time and I failed, it just doesn't seem to exist anywhere.  That wore me out and I declined to go on to the next stage of reconstructive research.  I just said someone with a lot more enthusiasm or concern for Yitzi's patch would have to figure it out, and I'm not going to support this sort of "change of meaning" shenanigan myself.

  Like you, bvanevery, I've totally forgotten what the third one is.  Since I play on Yitzi all the time, it must not be something that has made a big impact on my games.

I suppose I'll go look up my old article on the subject.  Found it:

• the firepower of Conventional Missiles is changed from 12 to 18
• the minerals cost of Stockpile Energy is changed from 0 to 2
• the tech requirement for Ascent to Transcendence is changed from "Thresh" to "None"

[BitGamerX]

I really appreciate everyone's comments. There's lots of food for thought. It sounds like to get started with a beginners vanilla playthrough start with SMAC patched with 'scient's v2.0 patch' and 'PRACX'. After that check out SMACX AI Growth or the thinker mods.

[bvanevery]

Personally I'd start with PRACX only as it does not affect gameplay in any way.  And frankly you don't even need that.  If you put DirectDraw=0 in your .ini, it's going to run full screen resolution.  On Windows 10 the GOG version is not bug ridden.  Look unless I'm choosing to test Thinker mod, I play on a totally unpatched game.  I don't do any of this patching "because I'm worried about bugs" stuff.  There are no bugs worth worrying about in single player.

There used to seem to be some bugs related to video playback on earlier versions of Windows 10.  On my machines, nowadays they seem to have gone away.  The historical workaround is to turn off Secret Project videos.  Recently I turned them back on and everything was working fine.  Microsoft probably fixed something on the sly when we weren't looking.

If your game crashes, look for patches.  Otherwise there's no need.

[dino]

I'm a bit of a thinker fanboy and got carried away. I agree thinker is probably not the best idea for a fresh player.
When wonder of discovery vanish and you start to treat it primarily as a game, less like an experience, then it will be the time to try thinker.

I think there is no point in not using Scient 2.0 even if fixed bugs are minor enough, that unexperienced player wouldn't notice them probably.

Btw, how does Scient 2.0 affects gameplay in any way that developers not intended ?
The only thing that comes to mind is workaround for an exploit, that involves some minor reduction of production queue functionality, but a new player probably won't be using queues anyway.
I don't remember what it was exactly, because I never use production queue and it's a change to user interface not to gameplay itself. Is there anything else ?

With pracx you get smooth scrolling, mouse scroll zooming, hotkey to hide forest/fungus/jungle to see the base terrain and potential tile yields on the base screen may be informative for a new player.
But there is small glitch with not important interface text, easy to get used to, so there is that against it, but pros outweight this tiny visual glitch.

They are both absolutely non essential, but I see no reason to not use them.

[bvanevery]

Btw, how does Scient 2.0 affects gameplay in any way that developers not intended ?

No idea.  I don't think I've ever played even a single game with it.  I've merely noted that based on diffs of its alphax.txt compared to standard, Scient didn't do anything that REQUIRED a particular alphax.txt.  His syntactic additions to predefined units, that let you specify a Fusion reactor or whatnot, are completely optional.  His changes of descriptive text are helpful, but can be completely overwritten and ignored without any consequence.  In short, he did a good job of not invading alphax.txt and so I mention this fact in the COMPATIBILITY section of my mod's documentation.

With pracx you get smooth scrolling, hotkey to hide forest/fungus/jungle to see the base terrain and potential tile yields on the base screen may be informative for a new player.
But there is small glitch with not important interface text, easy to get used to, so there is that against it, but pros outweight this tiny visual glitch.

They are both absolutely non essential, but why not ?

Because they are not needed and it builds neurosis in new players that they "must have" patches.  It's simply not true.  At least, not for single player.  I don't know about multiplayer because I never do it.

I remember the main sales pitch of PRACX, once upon a time, was that it overcame the ALT-TAB task switching limitation of the GOG binary.  Well subsequently I learned that CTRL-ALT-TAB would work on Windows 10 and even some earlier versions of Windows.  It's a little clunkier to use 3 fingers to do it, but I'm long since used to it.

PRACX looked like more of an advantage when Windows 10 was breaking Secret Project videos somehow.  But Windows 10 doesn't seem to break that anymore.  It's been several years since it was released, guess Microsoft fixed what was broken.

PRACX has some visualization features.  Someone might like those.  I don't care about them.  I've been playing this genre of game for 20 years. I don't need no steenkin' visualizations!     I never had them then, and I don't need them now.

If you know you want some feature of a patch, if you know why you want it, by all means use one.  But don't neurose about how you might be "missing something" and don't have "the best" version.  It's not like that.

If I wanted to change a bunch of global warming parameters, I'd use Yitzi's patch.  But I personally don't need that.

I push back on the mantra of "patch, patch, patch, patch, patch" because it undermines the inherent stability of the original game.  At least the GOG version, it's not broken.

Every once in awhile you also find out these patch authors introduced new bugs as they sought to fix old ones.  Does it matter much?  Probably not.  Are consequences minor?  Probably.  I haven't played enough Yitzi to know what I'd be burned by, or saved from.

Pretty sure the Believers' music got broken in SMACX, and that Yizti fixed that.  That sort of thing, is probably a lot more important to a brand new player, than it is to me.

[dino]

I"m not aware of any bugs in Scient 2.0. He already red about the patches and installed them, so the "damage" is done. Throwing them away now would be neurotic.
Flipflopping from advertising your mod, to "only vanilla game from 1999" is neurotic.

***
You can disable opening movie in Alpha Centauri.ini btw, from the in game options choose pause at the end of turn.

I'd play with direct research, bvanevery will disagree, but there are some completely essential techs in an early tech tree, they are essential on a level that does not exist in any Civ game.
But like in Civ games tech cost increases with every tech developped, so if RNG God grants you few useless techs early, it can delay getting the techs you want/need by 50+ years.
Completely frustrating and unfun way to play imo, there are way better options to increase difficulty.

Blind research can work with an experienced player, that has memorised tech tree and knows exactly what techs each military/science/build/explore focus can give at any stage of the game.
If you have tech tree memorised, then blind research only adds a slight randomisation to your preffered direct research order, so every game research patch is not exactly the same.
But as a fresh player you would not be able to consciously experiment, with what works in what stage of the game and be in control to play the way you want.

[bvanevery]

I"m not aware of any bugs in Scient 2.0. He already red about the patches and installed them, so the "damage" is done. Throwing them away now would be neurotic.

That's fine for you.  The sales pitch for Scient's patch has never articulated anything that I personally need, badly enough that I'd require Scient's patch when deploying my own mod.  The only feature I'm aware of that I would consider of real value, is the ability to specify the reactor size for a predefined unit.  That's a nifty feature, it really would help my work, but by itself it's not worth requiring users to install a binary patch to make my mod run.  All my mod needs is the official game binary.  Or an unofficial patch that doesn't mess up how alphax.txt is interpreted.  Scient's patch is one of those, so I tell people hey go ahead and try it if you want.  But that doesn't mean I care about it, only that I know they can do it.

Flipflopping from advertising your mod, to "only vanilla game from 1999" is neurotic.

Apples and oranges.  My mod only changes *.txt files.  It is not basically possible for me to introduce game crashing bugs that way.  When a *.txt file isn't interpreted correctly, the game simply doesn't work, and the *.txt file has to be edited.

Maybe it's not entirely clear for the home audience, but various people who have edited binary code in the game binary, have mistakenly called their work "mods".  They are not mods, they are hacks.  When you hack binary code you stand a high chance of breaking something, absent a lot of testing.  I used to write assembly code for a living.  You can think you know what you're doing, get something wrong, and not have it revealed until months later by some other engineer.

"Mods" generally means replacing text and art assets.  Not hacking and digging around in the original code absent source code.  I mean it's a small sin to use terms like that incorrectly, but it hardly makes me neurotic to point out that my methods of game modding, are vastly safer than hacking binaries.

You can disable opening movie in Alpha Centauri.ini btw, from the in game options choose pause at the end of turn.

Yes, that's how I survived the "seems to be crashing stuff" era of Windows 10.

I'd play with direct research, bvanevery will disagree,

It's cheating.  You are learning how to play a baby game, like someone just handed you a pile of extra tanks.

but there are some completely essential techs in an early tech tree, they are essential on a level that does not exist in any Civ game.

The correct solution to that IMO is to mod the tech tree, so that it doesn't suck like that.  Any problem - if indeed there is even a problem, and not simply a matter of perception - goes away.  A noob isn't going to know why there's a "problem" for awhile in any event.  Noobs have lots of things to learn about what is or isn't so important in the tech tree.

This is not a reason for a noob to learn how to play the game wrong, with a crutch from Day One.  Just learn to play the game right, as it was designed to be played.  Directed Research was given as a baby option for whiners who don't like anything random happening to them, who couldn't let go of how they did things in Civ II and understand this is not Civ II.  We'd get the same crap in Axis & Allies, people whining about how the dice are so unfair.  Instead of basically seeing the game as being about managing randomness.

But like in Civ games tech cost increases with every tech developped, so if RNG God grants you few useless techs early, it can delay getting the techs you want/need by 50+ years.

Solved by making sure techs are balanced and worth something.  I mean yeah, getting Optical Computers in the original game is pretty useless.  It doesn't do anything, it's an empty speed bump in the tech tree.  A sign of development pressure that they didn't get their work done, back when they were making the game.

Completely frustrating and unfun way to play imo, there are way better options to increase difficulty.

Like playing a real mod, if it's bothering you so much.

Blind research can does work with an experienced player,

FTFY.  And how does one become an experienced player?  By gaining experience playing the game the right way.

that has memorised tech tree and knows exactly what techs each military/science/build/explore focus can give at any stage of the game.

Well heck if the point is "muh game, too deefeecult, too noob" it's not a sin to start with the absolute easiest difficulty level and just increase when you've mastered something and ready to move on.  You don't have to play the game wrong to make it easier for you.

Yeah, I hate people who say things like there's no wrong way to play a game, and how dare you stop a player from having their fun doing whatever they want.  That's not a game mentality.  That's a software toy mentality.  What do you think anyone ever said about Chess over the centuries.  "Oh, muh Knights move too funny, it's haaard!  Makeum move straight, plzzz."  Yeah right.  The game has rules, the rules describe the intended way to play, what the game actually is.

With complex games you get into vast systems of rules, which can make people wonder about why this particular rule is even in there.  But the overall design intent of SMAC is pretty clear in this regard.  They printed Explore Discover Build Conquer on the box for a reason.

If you have tech tree memorised,

And if you don't, that's what the Datalinks are for.  Or back in the day, we actually read the paper manual while playing the game.  It's a 20 year old game, it's a bit old school like that.  It's not unfair today, this is still the 4X TBS genre.  People who can't swallow the gobs of details, probably aren't going to be happy.  But it's not like SMAC has a vastly different system than all the other 4X TBS games out there.  It's genre.