TFW you miss a Muton when u had a 96% hit chance
March 4, 2016 3:44 PM   Subscribe

Procedural Snake Eyes is a blog post about varying outcomes and experiential feels in procedural generation, in particular in the tactical spycraft masterpiece Invisible, Inc. and the recently-released XCOM 2, by Rogue Process (gamejam demo!) developer Mike Cook.
posted by cortex (42 comments total) 18 users marked this as a favorite
Interesting article, but the thing that disappointed me post is that I was really hoping it would delve into a selection of examples of ridiculous outcomes.

Now I feel like I want some sort of twitter feed or tumblr that's focused on highlighting those "procedural snake eyes", the critical failures in level generation.
posted by evilangela at 4:04 PM on March 4, 2016 [3 favorites]

Nice! I recently got hooked on X-COM and Invisible Inc., so this essay interests me greatly. Thanks for posting it!
posted by Rustic Etruscan at 4:06 PM on March 4, 2016 [1 favorite]

That's XCOM, baby!
posted by NoxAeternum at 4:15 PM on March 4, 2016 [3 favorites]

I was really hoping it would delve into a selection of examples of ridiculous outcomes

Heh, me too. Maybe we can crowdsource some good meat there.
posted by cortex at 4:20 PM on March 4, 2016 [1 favorite]

You know, XCOM has made its mark by killing off your best characters with impunity since their first game. You could go in with the best equipped squad and lose them all to a well thrown hand grenade before you ever made it out of your ship. That has been the charm and the challenge of XCOM since it launched. Save frequently, swear often and expect to fail when the aliens attack all of your bases simultaneously when you are least equipped and most of your guys are wounded.

The series is released as a AAA game because of its fanatic fanbase, of which I am one (although the last I played was Apocalypse because they require too much time, trial and true luck to win and kids make that game a year commitment). I dig the series though... they'really brilliant episodes of random pain. It's like QWOP but a tactical RPG strategy - even masters of the game will lose their entire squad on a bad roll of the dice.

It ain't for everyone. It looks like it is, but it is for spread sheet masochists that don't want to deal with the sociopaths of eve online.
posted by Nanukthedog at 4:25 PM on March 4, 2016 [3 favorites]

Re: Ridiculous outcomes in XCOM 2:

On my raid on an Avatar project facility, this last game, I had a Sectopod spot my team in the initial drop zone, resulting in concealment being blown from the get-go. The plus side is the team had enough firepower to score the kill immediately (including an achievement, woo!), but you'd think the procedural generation would include -not- having aliens in position to spot the initial landing.

I haven't played Invisible Inc recently enough to have stories on that one right now.
posted by Archelaus at 4:40 PM on March 4, 2016

Nanukthedog - It ain't for everyone. It looks like it is, but it is for spread sheet masochists that don't want to deal with the sociopaths of eve online.

This really isn't my experience of it ... basically ever. Admittedly the older XCOMs were certainly a lot nastier about the ol' "alien shoots you on entry to level," but that very much hasn't been my experience with the newer two (which is why the previous anecdote stands out as an exception and not the -rule-). Then again, the most recent ones have vastly different base mechanics, only one field team at a time, and you're expected to maintain a large roster to deal with giving the wounded recovery time.
posted by Archelaus at 4:43 PM on March 4, 2016 [1 favorite]

For some reason I assumed this was going to be screenshots from somebody who had pegged the RNG to always return 1.
posted by phooky at 4:51 PM on March 4, 2016 [2 favorites]

Yeah, I will also completely disagree about the new ones being for spread sheet masochists. You can absolutely delve into that, I've seen some people discussing percentages and stuff, but that's a very small percentage of people.

XCOM 2 has sucked up hours of my life already and the procedural generation of the maps helps keep you on your toes. Each enemy type (I have lost a soldier to each and every one of them, yay) forces you to deal with them in a different way. The four classes and the composition of your squad also changes it. I love it, I just finished a mission and I'm about to dive right back into it.

Invisible Inc is a game I haven't put enough time into. I've tried to get into it three times but that game makes me feel completely incompetent. I can barely beat a SINGLE mission and I've tried at least eight times. I'll definitely try to crack into it again but it might be a game (like Darkest Dungeon) that I appreciate but never play.
posted by Neronomius at 4:54 PM on March 4, 2016

The pseudo random number generation being invisibly tweaked to make it feel more rational to players is quite common across many games. It turns out humans are very bad at judging truly random processes and don't actually like them: but randomness is a critical part of game design.

You will get more players being interested in performing an activity with a 20% chance of producing 10 units of currency, than performing an activity with a 100% chance of producing 2 units of currency. Hence the popularity of, say, lottery tickets or gambling. So randomness is important but you also don't want to hurt the player experience.

Hearthstone for example - a free to play Blizzard digital collectible card game - has players opening "booster packs" of 5 random cards every 2 days or so if you play consistently, say if you play two to three 5 minute games per day. On average, players open a "Legendary" card every 20 packs, so say a 5% chance. However, they have a built in "pity pack" algorithm that smoothly increases the chance of a legendary card appearing every time you fail to open a legendary card - increasing up to 100% chance at your 40th pack. So if you buy a pack of 40 boosters at once, you are guaranteed at least one legendary card - it would really suck if you dropped a bunch of money on packs and didn't get at least one - you might never spend money on that game again!

DOTA2 uses a similar mechanism for dice rolls within the game (chance to block attacks, etc)

I'm playing XCOM2 on Commander Ironman (hard difficulty, no ability to reload games and do over any part that went badly). I don't really have much experience with the game yet and I'm making a whole bunch of avoidable mistakes. But all that IS part of the game. Randomness is part of the game - sometimes you'll miss a shot or the enemy will crit you and you will lose a soldier. The ability to reload an outcome you don't like deletes a huge part of the game - if you reloaded often enough, you might as well have all your soldiers at nearly 100% accuracy, with full prescience of enemy positions in the map, and all enemies have nearly 0% accuracy. I would prefer a save / reload mechanic similar to Edge of Tomorrow but that's something brewing in the back of my mind.

Also big part of the charm in playing the game for the first time is not knowing what the enemy aliens do and then losing soldiers to them.

I'm on mission 15 and have lost two soldiers so far (while on Kevlar armor) and just got to powered armor which will hopefully reduce my casualty rates! As it stands a soldier standing out of cover in Kevlar usually gets one shotted by an enemy critical attack (enemies have, what, 40% crit chance on soldiers who are flanked).

It's like the cheese holes theory - every bad outcome lining up and then people die. Trying to ambush one pod and getting ambushed yourself by another. A Faceless appearing out of the crowd. Being outnumbered 2 to 1 because two pods are hitting you and they got the first strike and disabled 2 of your soldiers. Only 4 turns left until your dropship leaves, whether you get your soldiers on board, because enemy interceptors are arriving. Half your squad out of ammo because you were rushing forward to save the VIP and didn't have time to reload. The VIP dies anyway because in the hail of gunfire some of it accidentally missed and struck his vehicle which exploded. You meet every unexpected obstacle with more and more contingency actions until you have nothing left and just turn and run and the last guy to the chopper gets shot in the back because he was just short of 1 movement tile to get to safety.

That's the heart of XCOM2.
posted by xdvesper at 4:59 PM on March 4, 2016 [6 favorites]

Quit telling me the shitty things this game can do to kill your players and make you cry. I'll be ponying up $60 to steam and screaming take my money if this keeps going on.

Though, bummer about the spreadsheets....
posted by Nanukthedog at 5:10 PM on March 4, 2016

Nanukthedog you can live vicariously through my Commander Ironman playthrough instead (well, I certainly prefer watching some games now instead of going through the effort playing them), I'm progressively uploading the missions, cutting the non-combat sections out. I may or may not win this game but we'll see, a loss would be more entertaining than a win. Youtube playlist link
posted by xdvesper at 5:21 PM on March 4, 2016 [2 favorites]

It's like QWOP but a tactical RPG strategy - even masters of the game will lose their entire squad on a bad roll of the dice.

What does this make Fates Conquest?
posted by pwnguin at 5:22 PM on March 4, 2016

This really isn't my experience of it ... basically ever. Admittedly the older XCOMs were certainly a lot nastier about the ol' "alien shoots you on entry to level," but that very much hasn't been my experience with the newer two (which is why the previous anecdote stands out as an exception and not the -rule-).

Yes, I actually find the modern XCOMs quite intuitive. If it looks like your shot would be hard to make because there's stuff between you and the target, then it probably is. Same thing in the opposite direction if they're standing in the open or if you run one of your guys into a flanking position. The only iffy mechanic is "pod" activation -- i.e., the moment at which enemies become aware of your troops and take hostile action against you. But that's not a spreadsheet or randomness thing, it's strictly based on distance between your soldiers and the enemies (unless the soldier in question is "in concealment", in which case the enemies have to actually see them to activate).

The soldier stats in the game aren't something you interact with directly, since the only thing that increases them is ranking up (i.e., leveling). All the perks and gear upgrades offer transformative abilities rather than stat bonuses. Even the scopes, which used to increase your Aim stat, not give a percentage increase to your hit chance instead. XCOM 2 is not very spreadsheety at all. The thing that comes closest is planning your base construction and research on the highest levels, because you won't have the resources to build all of the rooms and the order in which you build them is very important.

I think the comparison between Invisible, Inc. and XCOM 2 is interesting for a lot of reasons, but the difference also feels intentional to me. Invisible, Inc. challenges you to plot out a flawless heist. There's very little room for error, and brute forcing a mission is extremely difficult. Recovery is not impossible, but getting discovered usually goes very, very badly for your team. In XCOM, however, the game does its best to ruin your plans, because the fun part is trying to recover a bad situation. This comes through not just in the use of randomness but in the enemy designs. All of them but the most basic grunts have at least one way to break the "rules" you've been fighting under, whether by mind-controlling one of your soldiers or blowing up their cover or summoning a horde of zombies or whatever else.

Finally, here's an interesting interview with Jake Solomon (Firaxis' head honcho for the XCOM games) concerning the game's use of randomness.
posted by tobascodagama at 6:14 PM on March 4, 2016 [1 favorite]

As my life goes on I find that people who jack off at extremely randomized difficulty and gameplay that forces you to spend hours on resources that could be lost in an instant have way more time than I do to play games. Something like dark souls respects gaining skill, you can be good enough at the game and roll through it at level 1 in starting gear, but stuff like x-com and darkest dungeons added layers of "haha fuck you we're so hard, don't know you random chance maaaaaan" are the exact opposite.

It feels like the games don't respect my time, I'm not a teen with endless hours to blow anymore. I don't mind a procedural game where the stakes are low like binding of isaac or FTL, but when you start getting into games where you can lose hours and hours of effort on a dice roll, at a point it becomes a statement of "look how much time I have to play this"

The response I often hear is "well just save scum!" That's even less respect of my time and the reason why I roll my eyes at people who act like fallout 2 infinitely better than the modern (highly flawed) incarnations . Those were games where you spent the first half of the game with 12% chances to hit, and expected you to reload over and over to get through scenarios.
posted by Ferreous at 6:31 PM on March 4, 2016

Dude, if it's not for you, it's not for you, but in the case of both Darkest Dungeon and XCOM, it's less random than it appears. In both cases, the right tactics mitigate a -lot- of the apparent randomness.

Unfortunately, what is "the right tactics" does necessitate a certain eye towards the situation and what you bring, so I suppose it's something of a matter of spending the time to learn that. Still, if you don't have that time to put in, that's hardly the fault of the -game-.
posted by Archelaus at 6:40 PM on March 4, 2016 [4 favorites]

ugh, now I have to go buy Invisible Inc. FINE.
posted by GuyZero at 6:54 PM on March 4, 2016 [1 favorite]

Seriously, buy it. It's GREAT.
posted by tobascodagama at 7:17 PM on March 4, 2016 [1 favorite]

Oh, but when Invisible, Inc. tells you that you really ought to play on Beginner difficulty for your first playthrough, believe it.
posted by tobascodagama at 7:49 PM on March 4, 2016 [4 favorites]

but when you start getting into games where you can lose hours and hours of effort on a dice roll, at a point it becomes a statement of "look how much time I have to play this"

I feel like I want to address this, philosophically ...

First the appeal of Hardcore vs Softcore in Diablo 3 - Hardcore mode has permanent death, so you lose the entirety of your character and all their gear when you die, and obviously this game has no save points. People say, well, they like to play Softcore because their progress is permanent, and will never be wiped, so you will always be making incremental progress.

But in reality, your progress is always effectively wiped in Softcore. 2 months after you finish playing the game you will not care about your character any more. 4 months later the season ends and the items are changed and rebalanced, making your gear obsolete. The next season paragon experience scaling is changed. You will stop caring about your character and that's when it dies a true death.

Hardcore has a way of increasing the stakes, making you more interested in your character, making it more real. In some philosophical sense, a character can only feel alive if it has the possibility of dying.

Second: these games are designed to be played realistically (actual tactical combat) as opposed to Rambo style jump in and shoot everything. There's a place for the latter, this just isn't it. I've played hardcore characters in Diablo 3 who have never dropped below 90% hp in their entire careers. In XCOM2 I've never been close to a team wipe. It's a game about risk management, and if you are at a point where you might, I quote, "lose hours of effort in a single dice roll" you screwed up AGES ago. A long series of mistakes had to be made before you got to that point.

1. The mission briefing gives you an idea about the difficulty of the mission so you know roughly how many enemy pods there are lurking in the map.

2. Do not engage more than 1 enemy pod at once. You will almost never lose that way. You do this with a combination of scouting ahead with a concealed Ranger, using battle scanners, hugging the edges of the map, blowing up walls to see enemies behind them and moving safely through cover. Plan your advance with the knowledge of how many turns you have been given, in some maps, the distance to your objective is too long, there are too many enemies on the map, the terrain is not favourable with too many potential spots for being ambushed, you do the smart thing and cancel the mission and call for evac. It is never worth the coin flip of losing your entire squad just to gain the reward of a successful mission - seriously want to risk everything for 150 supplies or recruiting an extra Engineer? Everything can be recovered except the death of your soldiers.

3. Despite precautions you will accidentally engage two pods - every few missions. That's why you bring defensive items and skills: Specialist with aid protocol, mimic beacons, flashbangs, grenadiers with suppression. Stay in cover, use your defensive abilities and weather the incoming fire from the enemy then strike back.
posted by xdvesper at 8:06 PM on March 4, 2016 [7 favorites]

The only times in XCOM 2 that I've found things unfair is when I've activated three pods (groups of enemies) and two heavy turrets at once with a single, short movement. It's happened once (and I restarted the mission because my squad got totally decimated) and never again.

In fact, they seemed to have really decreased the amount of unfair situations from the reboot. I always found Thin Men to be pretty cheap in their tactics, the ease at which they flanked you and how they dropped out of the sky. I also found the entire Interceptor portions cheap and frustrating. Both of these are gone from XCOM 2.

A lot of stuff that seems like it could be unfair has a solution. You have so many options and counters. There are a variety of grenades and bullet types that are all useful (except for smoke grenades).

I have a few actual real gripes with the game. The first is how badly optimized it is. Load times are way too long and it slows down quite a bit especially when you're in the Avenger. The second, and this one really, REALLY bothers me, are the lack of voice packs. There are no Asian languages. No Mandarin or Cantonese, no Japanese, no Hindustani, no Arabic. The map is of the WHOLE world, most of my soldiers are from places that do not speak any of the languages included. There are voice pack mods but yet to fill that gap.
posted by Neronomius at 9:32 PM on March 4, 2016 [1 favorite]

Chance find, in passing: very different game for the most part, but also on procedural generation there's a nice look at how Spelunky manages object interactions on RPS today.

A lot of stuff that seems like it could be unfair has a solution. You have so many options and counters. There are a variety of grenades and bullet types that are all useful (except for smoke grenades).

This is the thing I like about both XCOM/XCOM 2 and Invisible, Inc., really. And I like both games very much; while I agree with Cook's take on the difference in feel between Inv Inc's sure-thing actions and XCOM's roll-the-dice shooting odds and how that intersects with the procedural generation of levels, I'm a player who really likes the weird uneven bullshittiness of the chancy shooting odds as a core part of the game play.

But both games definitely shine in how they offer options, offer you a bevy of potential approaches to difficult, sometimes seemingly hopeless odds. And the more you play of either, the more you understand what those options are, and just how many there really are, and it makes you a better player and helps both games bloom.

And I think Cook's post gets right to the heart of the difference in how the two games function mechanically, even aside from the question of how unfair-or-not XCOM may feel (which is I think more a matter of setting expectations than anything). Because they are different in a really interesting way.

Invisible, Inc. gives you more or less absolute control over the next moment. Actions can't fail; a taser will always taze, a hack will always succeed, a guard's patrol path (if you spent the action to peek at it) will always be where you expect it to be, the line of sight from that guard will always stick to the confines of his cone of vision, etc. With the exception of insufficient recon, you know what's coming in the next turn; it's up to you to make a plan, and if your plan is sound it will succeed. The challenge is entirely in learning the rules, and learning how to project to future turns. You take risks at the strategic level in Inv Inc, and execute with absolute confidence tactically.

XCOM offers no such certainty. A plan is not a blueprint, it's an aspiration and a fragile one at that. If you need to make every shot you take on a turn, that turn is almost certainly going to end poorly for you. You won't go six for six, even if the odds are pretty good for all six shots. You might go five for six, or four for six. You might do worse. Shots that hit might not do full damage because of a modifier. A plan that involves moving over to a flanking position to improve shot odds might end up revealing a new pod of enemies. Nothing is set in stone, nothing is wholly reliable. And so you make contingencies. You come up with a plan, and then you ask yourself: what do I do if this bit goes wrong? How do I change things up if this 90% shot misses? If I can't kill this particular alien, how do I minimize the exposure of my squad to its counter attack? What can I do to save that one stranded soldier, and how much can I risk trying to do so? Is it better to risk a goodish chance of hitting that alien with my sniper rifle, or to toss a grenade that's sure to hit but won't damage it as much? If one of my squad needs to run past the alien who is waiting to take a reaction shot, who has the best chance of dodging the incoming fire, or who if they don't dodge has the best chance of surviving the hit, or who in any case is the most disposable if they don't survive? It's contingencies stacked on contingencies, it's hoping for good luck and planning for catastrophe.

Invisible, Inc. gives you the toolset to play a tense, studied game of chess in hostile corporate offices, and sets you up for one big sigh of heist-accomplished relief when you teleport out at the end of a level.

XCOM gives you the odds, hands you the dice, and sets you up for a lot of moment-to-moment careering between garment-rending and celebratory fistpumps.
posted by cortex at 10:54 PM on March 4, 2016 [9 favorites]

The first is how badly optimized it is. Load times are way too long and it slows down quite a bit especially when you're in the Avenger.

Tell me about it - I'm running this on an i7 with 16GB of memory, and I have to wait *several minutes* after each mission to get to the post-mission screen. What are they doing with all this time, simulating the emotional reaction of each individual alien to the mission outcome?
posted by Dr Dracator at 11:24 PM on March 4, 2016 [2 favorites]

Dude, if it's not for you, it's not for you, but in the case of both Darkest Dungeon and XCOM, it's less random than it appears. In both cases, the right tactics mitigate a -lot- of the apparent randomness.

I ask everyone who complains if they use hunker down which confers a massive defense and dodge bonus. They just blink.

End of conversation.

XCOM 2 is easier than XCOM 1 it just is more lethal on each side. That means you have many more options to swing it decisively in your favor, and in fact the whole concealment mechanic means you don't have to engage suboptimally often!

Now I still get stuff like a grenade on cover reveals 3 aliens chilling out during a firefight RIGHT THERE that will now join. That was a heroic ass mission.

edit: Also you cannot lose an entire team in a single turn in XCOM 2. You just can't. If you did... I want to see that replay to marvel at the bad luck. Most missions allow you to even call in an evac helicopter and get out as a free action. So you can sprint to it. I've saved so many people from almost certain death with it. Scrapping a mission (I play impossible ironman) or just cause they're shot to hell. Get out when you can sometimes. It's not a stomp the aliens, it's a fight for life and turning the tide.
posted by OnTheLastCastle at 12:09 AM on March 5, 2016

Invisible Inc, Darkest Dungeon and XCOM 2 are all excellent games I should add.

Of those, only Darkest Dungeon will absolutely not give any fucks at you wanting to turn down the difficulty. As far as I know, you can't.

XCOM 2 and II can both be turned down pretty far.

None of those three games is going to wipe your entire game in a single move at all. It would take a dedicated series of them. DD is actually the most forgiving to a party wipe because you can grind back from any situation (enter dungeon w/ no supplies, loot then run away and repeat). The other two you can fail outright if you lose all agents.
posted by OnTheLastCastle at 12:17 AM on March 5, 2016

I really need to make some time to finally play Invisible, Inc. again. I played it relatively early in the Early Access phase, had a good time, and haven't touched it since.
posted by egypturnash at 12:18 AM on March 5, 2016 [1 favorite]

Endless mode calls you, my friend. To die in the cold sterile corridors of this or that corporation that is.
posted by OnTheLastCastle at 12:23 AM on March 5, 2016 [1 favorite]

Also you cannot lose an entire team in a single turn in XCOM 2. You just can't.

No, but you can and do get irrevocably fucked in a single move - overextending is a good way to do this - and then spend the rest three or four turns watch your team getting wiped out with very small chances of salvaging the mission.
posted by Dr Dracator at 1:11 AM on March 5, 2016

Sure, but that means you messed up. If you don't get fucked if you mess up it means the game is too easy.

It's like when some people can't get past a boss on Dark Souls and you're like "you gotta block his first hit and then dodge around behind him" and they're like "block?" and you just stare at them.
posted by Justinian at 1:35 AM on March 5, 2016 [1 favorite]

I roll my eyes at people who act like fallout 2 infinitely better than the modern (highly flawed) incarnations . Those were games where you spent the first half of the game with 12% chances to hit, and expected you to reload over and over to get through scenarios.

Sounds like you were playing Fallout 2 wrong - you should never be in a position of relying on odds that short.
posted by Dysk at 2:39 AM on March 5, 2016 [3 favorites]

Darkest Dungeon, XCOM, and Invisible Inc will all absolutely crush you if you play poorly (although not immediately; it requires a sequence of bad plays to take serious losses of time). You have to learn the systems of each if you want to win, which means a lot of the gameplay is about learning and exploiting those systems. Whenever you see someone throwing a fit about how the games are too random and they love to fuck you and anybody who likes them is just a masochist who jacks off to randomness (which so obviously isn't a real thing that it's bewildering that anyone would bother to say it since nobody's going to believe it), what they're actually saying is "I don't want to have to learn systems and I think I should never suffer any setbacks. Everything I do should be successful." I can understand how someone could enjoy that, even if I personally find it quite boring, but I've never understood why those people seem to hate people who don't feel the same way.

... you can and do get irrevocably fucked in a single move - overextending is a good way to do this - and then spend the rest three or four turns watch your team getting wiped out with very small chances of salvaging the mission.

This is not a real thing. If you get into a position where your team is just going to get picked apart with little or no hope of victory, you evac. As long as there's a 3x3 area on the map that's within sprinting distance of all of your soldiers, you can evac instantly. Also, obviously, don't overextend, use battle scanners and phantom rangers to stay apprised of enemies that you aren't yet engaged with, don't move in directions you haven't scouted unless a lot of your soldiers have actions left, etc.
posted by IAmUnaware at 2:45 AM on March 5, 2016 [1 favorite]

Sure, if you don't want to rescue the VIP or whatever, or have enough turns left to proceed cautiously without ever exposing anybody to danger.
posted by Dr Dracator at 4:34 AM on March 5, 2016

Oh, good lord. The time limits are -not- that restrictive, and pretending like they are is a tired, tired meme. I expect better than that forum meme of the blue.
posted by Archelaus at 7:12 AM on March 5, 2016

ugh, now I have to go buy Invisible Inc. FINE.

I picked it up last month, and it is fantastic. One piece of advice: never give up, no matter how fucked your situation looks. I've been in levels where my team is trapped in a dead end, needs to cross the level to get to the exit, and have guards swarming; and some patience, along with careful collection of information, has allowed me to find a path for the team to sneak past everyone and get out. I've also had agents cross entire floors to recover bodies of team members and get back just fine, unseen.

As soon as you start feeling confident, you'll get a situation where you can't seem to find a solution - I've got one right now in my current game, where there are 3 intersecting guard patrols; I found a hole in the pattern that would allow me to sneak one agent at a time from one room to the next; only to discover that there is a fourth guard patrol that I had no clue about previously that comes into that room and immediately has the drop on my agent, who doesn't have enough movement left to get to any cover after dashing between rooms. I'm hoping to find a solution because the game has been going awesome right up until this.

That being said, I would love a bit of insight into how the game makes decisions about guard movement once they've been alerted; they converge on an "interest point", sure, but after that they wander off in some strange and surprising directions - 3 or 4 of them piling into the same dead end room, or scattering into the far corners of the level. It's not a bad thing, because you can observe & predict, and the weirdness of it both creates and limits opportunity - it's just some of the decisions are...weird.

These guys are doing a playthrough, with the player being someone who hasn't played Invisible, Inc before, and his buddy someone who has. It's a lot of fun to listen to, because the guy playing is doing really, really well (at least through the first 7 episodes) and it is frustrating the heck out of the experienced guy.
posted by nubs at 8:14 AM on March 5, 2016 [1 favorite]

My favorite XCOM:EU moment was toward the end of the game. My squad loads in, but one of my snipers glitched and loaded inside a billboard. He had a great vantage point, but he couldn't move. The rest of my squad proceeds down the long and narrow map, and since my stuck sniper had Squadsight, he could see and shoot almost anything my squad could see, and he was just one-shotting everything with his Plasma Sniper Rifle, and getting bonus actions for the kills. It was a bloodbath, and I was cackling like a mad supervillain as the mission went on. Squad got to the end of the map but for some reason, the mission wasn't over. Out of nowhere, a Thin Man appears way back at the start of the map, and just goes *boop!* and my heroic, tragically stuck-in-a-billboard sniper was no more. :(
posted by xedrik at 8:33 AM on March 5, 2016 [4 favorites]

People have different ways of playing games and different degrees of patience with various game mechanics, game mechanic alterations, and trial-by-fire introductions to those mechanics, and that's okay! I think the timer discussion is interesting as a design and experience thing, and as much as I can sympathize with maybe being tired of it if you've been wading hip deep into XCOM 2 discussions elsewhere, I don't see any real upside to getting short about it in a MetaFilter thread where folks are just kinda having a discussion about. Maybe let's just keep having a discussion instead!

Like: the prevalence of fairly restrictive time limits in XCOM 2 are an interesting change from XCOM. And they are fairly restrictive. Not impossibly or unfairly or game-breakingly so, but enough so that it disrupts some of the other totally valid and reasonable strategies of conservative play that XCOM taught players the hard way to adopt. You have to be more constantly on the move and less deliberate and conservative in your scouting and overwatch, and that requires a real change from a kind of by-the-book approach that has itself been a really normal part of the series.

So people saying "the timers are too short / too common / too punishing" aren't being dinguses. They're reacting pretty understandably to a significant design change and the effect it's had on messing with their expectations, as fans of turn-based tactics in general or of XCOM (or good ol' X-Com) in particular. My feeling is folks who feel that way aren't getting as much out of XCOM 2 as it has to offer, because I think the design move here is really a pretty interesting one for the sequel, but I totally get the feeling.

What I think the timers offer is, specifically, a tactical and strategic nudge that helps XCOM 2 not just be XCOM Again. By forcing the player to assess the distance to the target, the likely enemy resistance on the way, and the time it'll take from reaching the target to dealing with it successfully (whether that's blow something up, keep something from getting blown up, activating and escorting or bodily hauling a VIP to a drop zone, or so on), the game creates a new strategic layer to the mission that would have been absent before. And that new layer requires new strategies, new risk assessments, new methods of establishing contingencies.

And that's a pain, initially! It fucked with my playstyle, for sure, and has lead to some seriously panicky time-crunches and a couple of genuinely bad outcomes where I favored playing the turn-to-turn safe and so ran out the clock on the actual mission goal. But! I kept working on adjusting, and I learned some new things tactically, and what I've ended up doing is having interesting new XCOM experiences that are in their own right pretty thrilling and satisfying when they work out, even if they only barely work out. The timer focus added that. And it did something else: gave un-timed missions a more distinct slow-burn feel; whenever there's not a timer, I'm a lot more aware of it. Blacksite stuff tacks on a kind of added import just by virtue of feeling different, etc.

So, I'm a fan of the timer. I'm a fan of the pressure that design choice makes on gameplay exploration. I think they made a good, interesting call with it. But I'm also deeply sympathetic about the frustration it can create. I think working through that frustration yields up some nice benefits, but that's easy for me to say as someone who has come out the other side on it in a good mood.
posted by cortex at 8:50 AM on March 5, 2016 [1 favorite]

A couple more thoughts on time-crunch stuff, actually:

1. Terror missions (aka Retaliation missions in XCOM 2), where there's no timer but your goal is to rescue at least a minimum number of civilians from the aliens currently marauding through a town or rebel encampment, are still functionally a kind of timer-based mission, and were in XCOM as well. The distinction here is one of withheld information vs. an explicit timer: you are definitely under pressure to move quickly, take bigger risks, favor positional advancement over careful scouting. Because if you don't do that, the aliens will just wander around killing civilians until the mission is a bust. So here you have a time constraint combined with not even really knowing when the timer is gonna tick down next, and by how much. (Also, I love that pressure to overextend to save civs collides with some civs turning out to be shapeshifting monsters. Contingencies!)

2. Invisible, Inc. uses time pressure in a different, also interesting way that is really key to its character as a game as well: instead of having to beat the clock in explicit "do it in x turns or you lose" terms, it uses the clock to keep ratcheting up pressure the longer you stay in a given mission. Hacking gets more expensive; extra security cameras come on line; more guards spawn; security systems start identifying agents' exact positions. And so playing a mission is both a base matter of identifying the core goals—where is the macguffin, and where is the exit, and what's the good route from one to the next—and making a risk assessment about how much extra time you're willing to spend on the level to handle a given situation safely vs. quickly, and how much extra time you're willing to spend going after extra goals. Do you snag that extra stash of credits? Scrounge up the system power to hack that corporate terminal? Backtrack an agent to check out that by-the-wayside nanofabrictor? There's no fundamental timer in the game; if you're willing to keep cat-and-mousing on a level once security is at its max with lots of nasty guards wandering around, you can. It's just not exactly a good idea. But bad ideas can be profitable!

The contrast between Inv Inc's time pressure via ratching up of the difficulty over the course of a mission and XCOM's time pressure via explicit timers or civilian headcounts is really pretty interesting as a design question, and I think they both work really well in very different ways for, again as Cook's blog post notes, games that have so much basic level design concepts in common on paper.
posted by cortex at 9:12 AM on March 5, 2016 [3 favorites]

This thread is great. I feel like I'm among my people.
posted by JHarris at 10:13 AM on March 5, 2016 [3 favorites]

So I bought Invisible, Inc last night and played a few hours... Tons of fun. Very much like X-Com with less shooting. It's going to get SO. HARD.
posted by GuyZero at 2:52 PM on March 5, 2016

Also I see why you like it so much cortex... These goddamn levels are straight up roguelike.
posted by GuyZero at 2:52 PM on March 5, 2016

Got a recover-the-item mission. Found myself in an awkward position when an enemy MEC patrol unexpectedly climbed to a roof my Sniper was alone on. Accidentally activated a second pod just before I finished them off. Meanwhile, I was sneaking my Ranger forward to scout around the objective. Rounded a corner and activated two additional pods at once: one pod had an Archon with two Codices, the other had a lone Sectopod.

Needless to say, I aborted the mission, but I loss three soldiers in the process, including two Captains and an EXO suit.

Thing is, though, it was totally my own fault. I made 100% of the shots I took, I was just an idiot trying to rush my Ranger ahead rather than using a Battle Scanner or other more sensible scouting options. I also made the huge mistake of going on a Very Difficult mission with my B-team.

It's definitely possible to get screwed by the RNG, but most of the failures I've experienced can be directly traced to things I did or didn't do correctly.
posted by tobascodagama at 4:32 PM on March 5, 2016

I really love Invisible, Inc., not least because its strategic layer reflects its tactical layer: Both force you to consider the best way to proceed through a randomized challenge, and both become more challenging as time passes. The decisions you make at the strategic level affect the decisions you make at the tactical level, too.

For instance, in one campaign, one of my two starting agents was Sharp the cyborg. He begins the game with the maximum capacity for augmentations. His starting augmentation makes his melee attacks knock out guards for more turns the more augmentations he has. On the strategic level, I had a strong incentive to break into cybernetics labs regardless of their security level, because each of them has two free augmentations. On the tactical level, I had a strong incentive to break into all the safes I could so I could buy augmentations for him at the basic kiosks you usually find in all facilities.

I did manage to fill up all his augmentations, but I failed the campaign on the final facility. I didn't have strong enough equipment to deal with the well-armored guards there. This happened because I had to waste a mission rescuing an agent I'd lost in the field. Fortunately, I'd lost her on a mission in which I rescued a third agent, so I didn't have to break into her prison with only one agent, but it didn't do any good in the end. The last facility's guards and cameras were placed so that I had to walk into their view, and my team got slaughtered.
posted by Rustic Etruscan at 8:53 PM on March 5, 2016 [1 favorite]

« Older A Virtual Cycloid Drawing Machine   |   Can you smell what The Rock is cooking? Newer »

This thread has been archived and is closed to new comments