stakes and tragedy in games

I'm going to write down some thoughts I have on stakes and tragedy in FTL and Jagged Alliance 2. Hopefully you'll find some of it interesting even though I'm rambling on quite long to give you some context. Just my 2 cents, you have been warned.

One of the games that I have a love/hate relationship with is FTL. It has a lot of interesting choices to make, feels like you can get better at it through experience and I like that everything has consequences. I'd consider it a high-stakes game because you can lose members of your crew or even the whole ship quite easily and you can't just reload an old savegame. That adds tension and can make for some interesting emergent narrative that is unique to your experience. For example you could be in a situation where the oxygen generator room has a hullbreach, the system is damaged and the ship has dangerously low O² levels. You might have no choice but to send two people into the vacuum from which one might suffocate, but you need to make that sacrifice or else the entire crew will suffer that same fate. A little tragedy like this can make for an interesting part of the unscripted narrative that unfolds in your personal gaming experience. Moments like this have a higher chance to stick in your memory and become the things you'd tell other people about when you talk about a game. So much for the positive side of it. What I really don't like about the overall FTL experience is that due to the balancing and length of the game it can easily end up like this: You start another FTL run, play a few hours, all is going well, then either you suddenly encounter a ship that your loadout can't counter well, or you have bad luck with the RNG or you die at the final battle because it is balls hard, even on easy. I did finish the game 1 or 2 times on easy but usually my attempts at playing the game end with being super frustrated because my ship had just exploded. I even wondered several times why I'm coming back to playing this if most of the times I end up so frustrated.
There is something about the high-stakes in games that makes them engaging for me. I love the concept of FTL, but I think the way this is designed is far from optimal.

Let's look at another game: Jagged Alliance 2. I started playing it around ~15 years ago when I was still going to school. I've spent days on it, constantly quicksaving and reloading to avoid any tragedies like my mercenaries dying. I consider JA2 a high-stakes game at heart that gives you the option to play it as a low-stakes game by exploiting the savegame mechanic. I still think this is a valid playstyle. You keep rewriting your history until all your mercs are the typical 80's action heroes that never gets hit and kill hundreds of bad guys each. The resulting narrative may be satisfying but ultimately boring. The game is still pretty difficult by the way. I think I couldn't get past the tanks one sector before entering the final city. At that point the difficulty makes a sudden jump and I was not equipped to handle this. Ultimately I gave up and started another playthrough a few years ago.

This time I finished the game, was quite happy that I finally managed it despite the rather harsh difficulty. When I learned about the JA2 v1.13 community patch/mod a while later I gave that a go and finished the game once again. The mod makes inventory and weapon management a whole lot more complex (some might say unneccessarily complex) and that gave me some more complexity to play with that didn't interfere with my playstyle.

The next variant that I played was JA2 Wildfire. For those that don't know, Wildfire is still basically the same game, but everything is a little different. There are changes to balancing, leveldesign, items and most importantly: the RNG is seeded. So if you save, take a shot and miss, you can reload a 100 times, you will always miss. I've read several reviews where people said that they hated this about the game. Presumably because they all had the same "lazy and safe" playstyle that I had back then. But to my surprise I really liked how that changed the gameplay experience for me. I would save at the beginning of my turn, then try things out and reload if I couldn't get to another satisfactory beginning-of-my-turn situation. Because the RNG was seeded that meant I could repeat all the same actions that worked the previous time and then do something different were I screwed up last time. For me it was a rather satisfying experience, almost like playing a puzzle game where you don't have perfect information and not all the rules are clear to you. I think there is great potential in a third person puzzle shooter where you can stop and reverse time to find a solution for epic John-Woo style shootouts, but that's something for another thread.
When I finished Wildfire I went back to JA2 v1.13, can't remember if I've finished the game another time back then or not. I still liked that one a tiny bit more for the added complexity of inventory and weapon management and the wealth of usability features that were added. When I went on vacation 2 weeks ago I brought my laptop and intended another v1.13 playthrough but when I sat on the train I realized I had forgotten to copy it on my usb stick. I only brought vanilla JA2 v1.12 and I was disgruntled that I had to go back to the by comparison "boring and simple" version of JA2.

So I was thinking I'd do a little experiment to explore what it does to my gaming experience when I make a tiny modification to the parameters of my playthrough. I comitted to only reload when I did something that I really didn't intend to do and accept all deaths of my mercs as part of the narrative. I played with the settings: novice (still more than hard enough for me given the circumstances), realistic (no monsters in the mines), lots of guns and save anytime (also because I remembered the game to be crashy and I wanted to be able to quit playing anytime). I expected that I wouldn't like playing like this and I was all but certain that it is not actually possible to finish the game like this. I was very wrong on both accounts! It ended up being the most engaging and memorable playthrough of the game for me!
The first one to die was my IMP custom character and I was very sad that he died. At that point I had to fight the urge to reload really hard because he died in such a stupid way. It was the first time that I chose to withdraw from a sector because my mercs were wounded, outnumbered and outgunned. So I chose to retreat and leave the sector as soon as I could, failing to treat the wounds of my merc first. So even though he had health left he bled out on the journey because he wasn't bandaged. It might have worked out in a city where a sector change only takes a few minutes, but I was in the woods and it took over an hour. Well, I learned a lesson from it. Ideally the game would have had informed me of this problem before I left the sector. But with a game that lets you save and load at anytime a little less hand holding really is ok.
So I hired Lynx to replace my lost merc and had my next big setback when I attacked Alma, the city with the military facilites. Here Lynx died, Grunty was unconscious and bleeding out, Ira and Igor were badly bleeding, Barry was still on his feet but also wounded. With a lot of suppressing fire I bought myself some time and in theory could have saved all but Lynx. But I didn't bring enough medkits and mid-treatment I ran out of bandages. Ira and Igor were unconcious too now and the only one still able to leave the sector was Barry. I sent him back to safety in the neighboring sector, watched grunty bleed to death and then the game told me that Ira and Igor had been captured by the enemy. A spark of hope in this very tragic incident that could have ended very differently if I had just brought enough medkits. I realized at that moment that through the ruleset of no reloading I had unlocked a source of complexity that was inherent in the gamedesign since its creation, but that I never got to experience because the game also (unlike FTL or other Roguelikes) gave me the tools to effectively make a wide array of mechanics and items almost meaningless. At this point I'd say saving and loading is a very important game mechanic, one that I never put much thought into. The game gained a lot of emergent narrative for me through this new style of playing. In addition to the backstory already present there was now this story of the one lucky merc that survived the bloodbath but was depressed about losing his friends and hoping to one day rescue the two surviving ones from imprisonment (And the story about how the incompetent commander aka myself, has screwed up that mission through simple missmanagement ^^). And a while later I actually managed to find the prison, rescue them and they fought many battles together until Igor finally died when he was shot by a tank. Playing with permadeath gave every decision more meaning, made the overall experience way more exciting and when something worked it was also a lot more rewarding because now it meant something. Victory was no longer a given, it was uncertain. Of course I regretted many more deaths during the course of the game and it wasn't always easy to accept how I had again F***ed something up that I should have know better. Or when someone died because an asset looked like a merc should be able to look over it but in reality he couldn't (happend once in the hedge maze level). I reloaded once early on in the game when I left skyrider behind and he got killed in the swamps. That was one of the moments where I hat meant to sent him with the others, but the squad was full and I didn't realize it. If I had not been able to reload and fix that input mistake, it would have sucked pretty hard to no longer have a pilot for the rest of the game. So I can't really say that the golden solution for the save/load problem would be to just enforce every death as permanent. Another time I reloaded after wasting a bunch of explosives on an indestructible door (which imho should not exist in the context of the game), but that wouldn't have had much consequence either way.
What I didn't like about the whole experience was that once I reached the final city and had to fight tanks for the first time the game started to feel like I have to tear down a brick wall by smashing puppies against it. You have all these characters with backstories, friends, foes, emergent narrative that come from the gameplay itself and you start to put a price on their lives because those become the choices that you have to make. Who gets the next rocket launcher and tries to take out the tank? I never had found a good safe strategy against tanks because I had never played this way and now I was forced through a harsh trial and error process that cost many mercs their lives. And that is with my experience of having played through the game and variants of it multiple times before and playing on the easiest difficulty! Imho the whole balancing would not work for a beginner if the high-stakes gameplay was enforced as in FTL.

But that kind of opens up a dilemma. How can you get a player to go for the richer experience that has more frustrating moments but more than makes up for it by the added complexity, engangement and "fiero" moments when he is victorious? I think this is something worth having a discussion about because on the one hand Roguelikes or Roguelites (or whatever you want to call them) have a lot going for them but the concept also has a lot of drawbacks.

What I find especially good about how stakes and tragedy in JA2 are handled compared to FTL is that in FTL it is very likely that you end your play session at the point where you are the least happy about your progress. In JA2 you are much more likely to stop playing after you fought a hard fight but in the end won a new sector. Every death of a crewmember in FTL has very little emotional impact and doesn't do thaaat much to enrich your narrative, but in terms of winning the game it is a significant loss because you can't always just buy new people. In JA2 the death of a merc has a much higher emotional impact because everyone has a backstory and character, everything has context and the narrative that can emerge from the death of a merc can be more interesting imho. Often you will be faced with the choice to let one merc bleed to death or risk the life of one or two mercs for a chance to save him. In terms of winning the game it's not that big a deal, at least for the first dozen or so people that you lose. Yet while playing to me it feels in JA2 the stakes are a lot higher and I'm more engaged because you more frequently experience permanent loss during combat, whereas in in FTL most damage you can suffer is reparable except for dying crew members or your whole ship exploding.

I'm not yet sure what exactly they are, but I feel like there are a few valuable lessons to be learned from these two games and I wonder how I could handle permanence, loss and tragedy in my own gamedesign the best way. I want the constant tension of having something of value to lose permanently, but I don't want the sudden game ending loss that comes as a frustrating surprise more than anything else. I want everything to be well integrated and have a context that enriches the emergent narrative and player experience, but I'm not sure yet how to handle it. Also now I really want a metagame as JA2 has it, where you don't fight your battles just because they are missions from your todo list, but because they are the key to certain goals and you have choices on where and how to engage.

I'm very interested to hear your thoughts on this.

 

Interesting stuff (and yes, I read it all the way through!). I haven't played any of the games you mention, but I have played quite a lot of Roguelike games, as well as X-COM (both the original DOS game and the more recent iPad version), which sounds similar to JA2. I also played a great game on the Palm Pilot called Space Trader (or something like that), which has similar aspects (more on that in a moment).

And yeah, this is a hard issue. Permadeath increases your engagement, and save-scumming can turn a game into a grind. But when you lose a game after investing a huge amount of time & effort into it, it doesn't feel like an accomplishment; it's the direct opposite. It's frustrating and makes me regret the time I spent.

One aspect that relates directly to this is how replayable the game is meant to be. A good roguelike game should take, at most, a few hours to win — and more typically, a half hour or less to lose. So when you lose, you haven't lost that much, and this cycle of play/replay is something you quickly come to understand and expect. It's not fundamentally different from arcade games, which are typically won/lost in a matter of minutes, and then you play again. Contrast this with most modern arcade games, which are considered "too short" if they don't have at least 20 hours of play, and you're generally expected to only play through it once (or if you're a real fan, maybe several times). So, if you lose a game like that, it is a serious blow.

So, one thing that I think helps is if you can't actually lose. Some modern "Roguelite" games use this mechanic: when you lose, you pass on something (items, skills, stats, whatever) to your next incarnation/offspring/whatever, so you are not completely starting over; you can retain some sense of accomplishment, even as you mostly start over. Space Trader had things you could buy (insurance and an escape pod), once you had enough money, that would not only save your skin when your ship was blown up, but pay enough money to buy a new one, so again you wouldn't have to start over (unless you died in the early game, but then you haven't invested that much time yet anyway).

So perhaps that's the general trick — instead of permadeath, have some near-death consequence which means mostly starting over, but not entirely. Let the player keep trying, and retain some sense of accomplishment even when they fail.

 

In this case, it's not an alien mechanic to board game users (from which the roguelike genre sprang from), but an expected thing. These games aren't designed to be played too long, and thrive on the permadeath mechanic. Part of the fun is still winning with less optimal loadouts. In general, most well-designed roguelikes will suffer from around 10% of impossible to win scenarios, and these are pretty much impossible to design out without protracted development times. Generally you find that it was a decision you made in the past, that made it impossible to win, a bit like chess in that respect. These games are deeper than people realise, and often that's the draw.

If you always had a way to win regardless, where would the fear be? also, you probably could have seen this coming, but again it takes many hours of play to get that good. The key to FTL is learning how best to prepare yourself with what RNG came your way. I guess it's a similar mechanic to gambling in a sense. A lot of poker players could've won if they didn't play a bad hand a few rounds previously.

At what point does it cease to be a roguelike and something else entirely?

 

I would say an element to stakes is figuring out where on a spectrum of "totally your fault" to "completely expendable" the stakes need to be. Actually there are probably a lot of variables that come into play: random versus determined, expendable versus coveted resources, partial versus complete information, singular versus repeated playthroughs, and a few more I'm sure.

A lot of it just comes down to what the player is in the mood for/will tolerate, I think. Look at the difference between Fire Emblem and X-COM. Both use perma-death, but in FE it purely serves as an annoyance and a drain on resources (exp mostly), while in X-COM it's not insurmountable and can be overcome. Add that most people playing FE are expecting to make a perfect run, so you end up with one that gets reset at the drop of a hat and the other with consequences that will usually end up being played through.

 

It's also down to personality of the gamer. Some people will always assume that things are unfair, and it's basically everyone else's fault. This kind of person is prone to everyday accidents (repeatedly) and generally is quite rubbish at realising the problem is generally between chair and keyboard.

These guys hate roguelikes.

On the other hand you have a kind of player that understands cause and effect aren't necessarily immediate neighbours in time, and something they did a while ago could have far reaching consequences. These people don't generally have accidents. They predict what could happen and plan accordingly. I'm guessing they like roguelikes too.

2 quite extreme examples, but illustrates why the genre can only ever really be a niche and something you can either embrace - target your audience - or deliver an experience neither type of personality enjoys.

I like roguelikes but I also like the kind of game that everyone else likes: games where I can make mistakes and muddle my way through to an epic conclusion. I'm just not sure I'd enjoy a game that did both.

 

I'm happy my wall of text didn't scare off everyone and you guys are raising very interesting points!

@JoeStrout: I'd say for what we are talking about X-COM and JA2 are pretty similar. I am not sure though if in X-COM new soldiers are a finite resource or if you could keep hiring new guys and throw them into the meat-grinder as long as you have the money for it. In JA2 there is a finite number of them and I don't know what happens once all are dead. I'm assuming it's either an implicit or explicit game ending failure-state. Another difference is that as far as I remember you can only hire inexperienced soldiers in X-COM and they become elite soldiers over time. In JA2 this is far less pronounced and right from the start you have the choice to focus on a few expensive elite mercs or hire a bigger number of less experienced mercs that cost a fraction of the price. They will get better with experience, but you can't turn the worst guy into the best over the course of the game.

Perma-losing a game that I have spent a lot of time on could make me regret the time I spent with it as well. I'm all with you that that isn't the way to go. The point about Roguelikes generally being rather short makes a lot of sense. The thing is I haven't really made up my mind yet whether or not I want to make a Roguelike or something else. I think I lean towards something that has a meta-progression that persists beyond the player death. And I'm looking for ways to introduce stakes and loss that don't just involve "boring" resources like money or time investment through grinding. Ideally the game would be played because it's fun and engaging and not because the player has to grind X hours to get the next Y thingy. I'm also a bit averse to the idea of having to play more of the game as a "punishment".

I haven't found too many fixed points for what I want to make yet. For example I'm sure I want the player to control a single mech from an isometric (orthogonal) perspective, walk around an urban/industrial environment and shoot stuff that shoots back at you. And I want to have some metagame where you customize the loadout of your mech and acquire components/unlocks from successful missions. One thing that came to my mind is that a risk/reward balancing choice could be given to the player for example by allowing to either have more armor (be able to take more damage) or more cargo space (be able to carry back more loot from a mission). Borrowing from your example with the space game there could also be a choice for an escape pod that saves the pilot if the mech is destroyed or maybe an additional weapon instead.

I'm very interested in ways to integrate all the choices nicely within the context of the game. For example apparently (I never played it, only read about it) in ZombiU you have permadeath and when you die your character turns into a zombie that still carries all the stuff around in his backpack. And you start as a new character in the same area that either can start scavenging again or you can take the risk and go where your last character died, kill the zombie version of him and take the backpack with all the stuff you had. I imagine this could be a mechanic that helps immersion and engagement without being too punishing in terms of gameplay and time investment. It makes sense in the world of the game and I'm trying to find such things that could work in my own setting, or ways to tweak the setting to make them work. One thought I had is that there could be a choice to either play a mission as a simulation to get to know the level and try out different tactics to see if they work before you enter the mission for real and have the risk of losing something but also the possibility to get real loot and progress from your victory. It's plausible in the context of a sci-fi setting but I don't think it would make the game better (probably the contrary). What could make sense though would be a little firing range level where you can try out new weapons. This could also serve as a tutorial level.


@hippocoder: I didn't know Roguelikes had their roots in boardgames, but it makes a lot of sense. The personality type extremes are an interesting point that I never really thought about, considering that I'm mainly designing for myself here. I'd say I'm on the planning ahead side of the spectrum and manage to avoid most accidents. But I'm still not a huge fan of Roguelikes in general. I think I much prefer to have immediate feedback on bad decisions instead of feeling the consequences 1 or 2 hours later. I don't consider myself a strategist and I tend to be rather bad at strategy games too. I just like to make tactical choices based on the current situation without putting too much thought into past or future. I think that is the problem I have with FTL. It takes too long for my bad choices to finally screw me over so that I don't learn as much as I should to avoid it the next time. I very much like that in JA2 the punishment usually comes quick as a "You shouldn't have done THAT specific thing, now your guy is dead".
I remember Prince of Persia: The Two Thrones/Rival Swords had a mechanic where you could reverse time if you screwed up, but you couldn't do it as often as you want. I found that to be an interesting middle-ground between starting over a bigger portion of the level and having the ability to make a few mistakes but not too many. In a way the typical regenerating health AAA shooter does the same thing. You can survive as many bullet hits as you want, but not too many in a short time. You may screw up a bit and get your immediate feedback in form of a red screen and sound cues etc., but if you overdo it, you get the harder slap on the wrist that says "No! Go back to the checkpoint and try again.". That seems to be the most widely accepted form among players: immediate feedback, low stakes, two different levels of gravity of mistake and no long-term consequences of your actions. The other end of the spectrum are full on roguelikes. Do you have any examples of games that try the middleground but end up not working out for either target audience? When you put it this way the theory makes sense but I'm not sure if I know any examples of that and whether the root of the problem there lies in the muddied concept or the implementation.


@RockoDyne: You're right, there are a whole lot of variables to the equation and not everyone is looking for the same experience. Do you think the solution is to give players more choices at the start of the game on how punishing they want the ruleset of the game to be? In a way JA2 and X-COM do it by offering modes that don't allow saving during combat. I'm sympathetic towards the idea of giving that choice to the player to maximize his enjoyment of the product and since I'm aiming for procedurally generated content it shouldn't be too much extra work to implement.

 

One important aspect is to always try to describe your game to potential players so they know what they are getting into. I think much of the problem with perma-death ( or rogue-like) games is players begin playing thinking they are something else, like a typical RPG.

I also think developers need to balance the length of the game based on the RNG factor involved with permadeath. The easier it is for RNG events to spiral your game to permadeath, the shorter the overall game needs to be. I was playing an online game recently, was probably 15-20 hours into it, and I got attacked by AI. The incoming attack was a couple hours away so I spent all that time ignoring progression and instead turtling as much as I could, trying to eeck out every possible defensive advantage. In the end, nothing matter, the AI attack was so high on RNG that nothing I could have done for the entire 20 hours of playtime could have allowed me to survive that attack. That is horrible design. If that same thing happened in a game meant to last an hour, it would be fine.

 

Like everything else in this topic, choice is situational. In a game that's aiming for high replayability, choice makes a lot more sense. A more typical roguelike could easily get away with a bunch of different classes that each impact the difficulty uniquely: the warrior could have an easy start but hard finish, the wizard could have a hard start but sweep the endgame, while the thief is barely competent throughout. Those aren't just a simple difficulty slider, but a complete rebalance of the game's systems.

In games focused on a single playthrough, I don't think it makes much sense. Part of it is because I feel those games are aimed a bit more at creating a consistent, shared experience. I can't say stakes are even beneficial to these cases. Fire Emblem is probably the worst offender as it's a pretty standard tactical jRPG, but they then tacked on perma-death and never seriously thought about how that affects the formula (and these were games where exp was a limited resource too). The other part to this is it hints to the design being flawed, like you need a casual and hardcore mode for before and after you accept the bullshit the game is going to throw at you.

X-COM is somewhere in the middle. It's got replayability, but that first playthrough is pretty important (and I'm still on that, it's been slow goings). I feel like some of it is couching the design, but it's also there for the vet's.

 

I just noticed this article on Gamasutra that might be relevant: Designing difficulty in role-playing games.

 

There was a game for SNES called Cannon Fodder. The best part about the game -- or at least, for me, anyway -- was the ability for your men to die. After each completed mission, they'd get promoted, but if they were killed, they were gone forever and would be replaced by fresh recruits.

To drive this point home, in between missions you'd get a shot of the countryside where all the young men were lining up to go off to war, as well as see the graves of the men you've lost.


It didn't affect gameplay, really. It was just a small emotional connection to your team that really made you suffer when you'd see one of them get hit. This is so difficult for games to achieve, I think. Games like The Last of Us really work hard to make you feel an emotional attachment to its characters, but in the end, it's like watching Star Wars for the 100th time: I know the main heroes aren't going to die...

It's also why I love the Song of Ice and Fire book series. Nobody is safe, so the reader actually feels the tension of the constantly high stakes. And when tragedy strikes, it affects you.

So I'd say if tragedy and loss are core themes of your game, having the player suffer some sort of permanent loss is a super effective way to get there. Obviously you must be careful with the tactic, though, as affecting them too strongly may make them choose to not play the game at all.

 

Good stuff.
I think the punishment for death is an important consideration.

Make it too harsh and players WILL reload if the game allows it. This creates annoyance for the player and completely circumvents the mechanic, so i'd say it should always be avoided (of course what is "too harsh" is another consideration, imho losing one dude in XCOM is no big deal).

And it has been absent from games for a long time (until some time ago). One of my way-out-there theories is that MMOs became popular because players couldnt save and reload while they'd do it in every other game.

As an aside, a run of FTL shouldnt take a couple of hours, as you mentioned in the OP. I think you exaggerated there. Oh boy, i played FTL to death back then. The RNG and balancing is a problem, yes. But with the strong ships, i beat it pretty consistently on normal.

It depends on the player what they prefer.

I like the FTL concept a lot. Others might not.
In that case, you can look at how MMOs handle death because that is big consideration for online games i'd imagine and there has been a lot of thought spent on it.

Just dont make death super punishing but allow players to save and reload at any time. Nothing good will come of that. Saving and reloading is part of your game design and directly interacts with the death penalty.

 

Very interesting read and highly relevant. Thanks a lot for sharing!

I like this idea. It has some emotional impact without being a gameplay annoyance.

This, and after having seen your character's death animation for dozens or hundreds of times you get apathetic towards it. I found the game pretty hard and it often felt like a real struggle (which fits well with the theme) to advance at all at certain points (started my second playthrough on the hardest difficulty but didn't finish). I wonder how the experience would have felt differently if it was balanced to have a harsher punishment for death, but be easier at the same time. I lean towards it not being a good idea, but I'm not 100% sure. The problem is that playing a larger part again would likely just be annoying because all the dialog and cutscenes repeat and you might die again at the same encounter. This might be something that works better with procedurally generated content so that at least everything you replay isn't 100% the same.

I'm curious about handing a risk/reward balancing mechanic to the player. I vaguely remember that "Severance" had a save system that let you save anytime, but based on the intervall of how often you save you got rated on your bravery. And the first Alien vs Predator game had a difficulty setting where you only had 2 saves per level if I remember correctly. The choice of when to use them was highly meaningful for being able to finish the level at all and it made the game quite tense as a marine.

I usually pause the game a lot to play optimally and I read all the text which takes some time since I'm a slow reader. I just finished a "quick" run without reading all the text and using less pauses in combat. Took me about 1 1/2 hours and I got destroyed when facing the final boss for the 3rd time. Again, it was super frustrating because I had what I consider top gear for that encounter and I'm not sure I ever had a chance to win this fight. I had the hermes missile launcher and 2 burst lasers on the osprey ship, that had the fully upgraded artillery beam and maxed out shields. I've entered the battle with full hull, had fire extinguisher and weapon precharger systems. So many of my shots missed and so many of the other ship hit me that I didn't stand a chance. Do you think this is the wrong loadout or did the RNG screw me?
I also noticed that in the Highscore I have beaten the game 3 times (with kestrel, torus and osprey), but all of those were in the basic edition and since I'm playing advanced I wasn't once able to finish the game.

I like randomness to create a situation for me that I then can test my skills in, without randomness being a factor anymore. When I fail I prefer it to be my fault instead of bad luck. If it is skill based I can learn from it or practice more and get better next time. If it is luck based it is out of my control and I don't enjoy that. That might be an atypical point of view. Afaik I read that TF2 has some slight randomness in its mechanics, because they found through playtesting that people on average enjoy it more than a purely skillbased game. I don't like TF2 either ^^. But that won't be caused only by a tiny bit of randomness.



I spent some more time going over ideas for my game and came to the conclusion that in spite of all the enthusiasm I have for JA2 not all of what I like about that game is a) neccessarily a good fit for my game or b) for myself as a designer. I play few strategy games and I realized that I'm probably not able to come up with a really compelling and balanced metagame. So after careful consideration and remembering that my original mission statement was to make an "immersive HUD-free arcade game about blowing things up with a mech" I feel like a got sidetracked too much. So now I lean towards making the game a rather short Roguelike with sequential levels where the decision making happens in the missions. This is probably a lot better fitting for my abilities to balance a game. I intend to start off with a little-more-than-MVP version that starts you off with a random loadout in a single big procedural level that hopefully ends up being 20-30 minutes of gameplay for a playthrough. If I can make that fun I'll try and expand on the idea by making the game a series of levels that are smaller and take something like ~10 minutes each. From there it would be easy to offer different "campaigns" that play with the parameters of the progression and procedural generation. E.g. one could be only 3 big and hard levels with ~1 hour per playthrough and another could be 10-20 smaller and easier levels with ~2 hours per playthrough for a more casual and paced out experience.
So in short: I think I should try to make a roguelike that I enjoy more than some of the roguelikes that I've played so far by balancing RNG factors and skillcap differently.

 

The loadout sounds pretty nice. I guess those were 3 damage lasers?
The two lasers wont penetrate tier 4 shields reliably due to evasion but you have the missile for those, normally. The flagship has a defense drone in one of the stages however which will stop the missiles until it's destroyed by a stray shot. That would leave you with 6 laser shots against 4 shields with an evasion of 35% or such, which isnt much. Might have been that.
Bombs arent stopped by drones.

A cloaking system is very nice to dodge the power surges and with a boarding party you can conveniently destroy the missile launcher. A defense drone of your own is nice and dont ignore your engines (evasion).

Of course one probably wont have all of those modules but cloak and drones give a big defensive boost.

 

One laser was 3 shot and one only 2 shot. Also had a 2 shot hull laser but I didn't have the power to use that and I'm not sure there were any empty rooms anyway. Are engines really worth upgrading? I usually just maxed out shields and pretty much ignored engines. Maybe that's what I should reconsider? I tried playing with drones but didn't really like that playstyle and found them to be very expensive. Can't remember if I ever tried cloaking device, but I'll give it a go on the next run. Thanks for your advice!



I have installed "Sir, you are being hunted" yesterday and gave it a quick go. Imho that game really markets itself wrong. I always thought this was some wacky last stand survival carnage similar to Killing Floor. But it really feels a lot more hardcore and serious than the graphics and trailer look.
In this game you can only save at a certain stone monument and at boats it seems. That really adds to the tension while being "out there" trying to survive, but I don't like that I can't stop playing anytime.

Do any of you have a good link about saving and loading current game state (especially with procedurally generated content) at hand? Originally I was hoping I could avoid bothering to invest time into that functionality if levels are short enough but maybe it's worth it?

 

Sir is definitely closer to a roguelike (barring the default permadeath and dungeons of the genre). I'm really hoping more games will try to be fully fleshed out 3D roguelikes.

Unless the typical playtime is seriously short, 30 minutes to an hour tops, it's probably not a bad idea to have some kind of "suspend play" saving option. Most roguelikes do it, and FTL and Isaac do it. Sir has the advantage of the scene transitions that make for natural points to save anyway (purging garbage and all of Unity's usual dirty bits).