Dungeon of the Endless

In Dungeon of the Endless your goal is to explore a set of rooms to find the elevator leading to the next floor, while fighting off monsters that emerge from the darkness. The action is real time, but generally big things only happen when you open a door to a new room. That's when you get resources from your generators and new enemies appear. It includes elements from roguelikes and tower defence games - you can spend resources to place turrets and boosters, research new or improved structures, or level-up and instantly heal your heroes. Which structures you can research, which additional heroes you discover, and which items you find or merchants sell are all random. The only save games are when you quit. There is a lot I like about this game, but every time I lose a run, I find myself wondering if I really have the energy or inclination to try again.

This is the basic starting room, the Escape Pod. I haven't unlocked any others yet because I suck too much at the game...

At first it wasn't too bad. I was still learning the mechanics of everything, and knowing it was a perma-death affair I was steeled for a few failures while I got to grips with the rules. Those first few games were enjoyable, as I gradually progressed further each time and reached higher levels. I got a better feel for the kinds of items and structures that were available, and how best to balance the high cost of resource producing modules against the likely number of doors left to open (since the modules only produce when you open a door, if you buy a new one and then find the next door you open is the last in the floor, you've just wasted a lot of Industry).

Early days. You start with two heroes of your choice, and can recruit others that you meet with Food.

I adjusted my tactics and felt like I was getting better. I went from dying on the first level to making it to level three, then made a massive jump with a game where I made it to level 11 (of 12). This is all fine and good, and I'm still learning as I encounter new modules to place, meet new heroes, and unlock more artwork for the album you can access from the menu, which includes descriptions of what monsters actually do. They visually stand out from each other, and you gradually come to realize key details about them (those ones run straight for your crystal, those ones attack your turrets first, etc) but according to the album, some apparently weaken heroes or explode on death, which was never clear to me while playing the game. They were obviously doing something, but it gets lost in the mass of monsters swarming through the corridors. Your heroes have a couple of active abilities each, with one or both of them being unlocked as they level-up, which I found useful but not generally game changers. Normally when things were going well I didn't need them, and when I did need them they weren't enough to salvage the situation.

The elevator between levels. I have quite a lot of Industry saved up here... so I really should have bought a few more defence modules, yes. Easy to say in hindsight. Plus, I didn't exactly save all that Industry by throwing defences down all over the place!

This is the problem I have. Things swing from 'going well' to 'imminent disaster' with the opening of a single door, and I'm not always entirely clear why. I'm still not sure if the number of monsters which spawn depends on the number of dark rooms (you have limited power, so must decide which areas to light up), or if it only affects where they come from - much of my strategy so far has been trying to light enough rooms to create chokepoints where I can focus my heroes and defences against the monsters, but I'm starting to think this just forces them into an unstoppable horde. I probably need to spend more on defences - but if I do this too much, then I won't have resources for when I really need them. And it isn't a case of the monsters getting gradually more numerous so that I can tell when my defences are starting to come under strain and I'd better shore them up, it seems to be fine fine fine and BAM, suddenly my heroes are dying and/or they're streaming through my defences too quickly to possibly kill them all before they tear my crystal apart.

This is a crystal exploding. As you can see, my heroes are fine this time (I spent a lot of Food upgrading them), but it wasn't enough - they didn't have enough firepower to kill everything fast enough.

It's not so bad when you're only 3-5 levels in; it's not like you've lost a great deal of progress. But I really don't fancy replaying the whole game every time I want to try improving my level 11 strategy (or level 9, which was as far as I got in my last game). I have a feeling I know what went wrong - sometimes they start banging against unopened doors, and if you don't react they will open those doors themselves - I got into a kind of chain reaction where doors were forced open, revealing more doors with monsters bashing them, and after a few opened in succession there were too many spawned to handle. I think more spread out defences and perhaps a more proactive approach to the threatened doors might work. But there's a whole lot of stuff to do before getting to that point again. It's not boring as such - the random nature of the game means you are genuinely exploring in each new game, and might develop different strategies depending on the modules, items and heroes you find. But it still takes time, and usually there's not a lot of real danger... until there is. It doesn't seem to be choreographed that well, as you'd expect from a game with many random elements. But this isn't an action game where a surprise ambush leads to an exciting gunfight, it's a strategy game where being suddenly overwhelmed can feel cheap and disappointing. There are no heroic last stands here in my experience, just the numbing weight of numbers and watching your important UI bars shrinking.

This situation is actually much better than it looks. I managed to light a clear path to the elevator, preventing any monsters spawning there, so just had to run along it, shooting the odd fast monster coming at us from behind. We're about to escape unharmed.

I think I'll deal with it in Dungeon of the Endless, for at least a few more games, since I do feel like there are more strategies I can try, and because enough of the game isn't random that I feel empowered most of the time. Resource income is constant (you get base Industry, Science and Food income even without any production modules) and the outputs from spending it are dependable, except in explicit situations that you can choose to ignore if you so choose (a cryo-pod costing resources to open, which might have a hero to aid you, or a monster, or nothing).

This is in sharp contrast to FTL: Faster Than Light, which a whole lot of people loved, but which left me frustrated after a few games. You can probably argue I didn't give it enough of a chance, but it felt too random to me, and too many games ended in fights with enemies that I didn't even feel close to being able to beat. There's supposedly the option to engage your jump drive to escape combat, but even when trying to activate it from the very start of any suspect battles, I was either going to win without too much trouble, or be blasted so quickly that the engines would get knocked out well before I'd have any option to jump. Unlike Dungeon of the Endless, I don't remember any reliable way to spend or earn resources - it all seemed to depend on what random encounters I had in each zone, so I didn't feel like I could plan or aim for anything, and didn't feel like I was really learning anything.

The useful overview map. You can see I've powered everything north of the crystal, so no monsters will appear from there and my tech guy (robot actually) can work in peace.

There's a tricky balance between leaving some mystery in games and giving the player enough information - for instance, I don't think Dungeon of the Endless would be improved by some kind of 'enemy strength' counter to warn me about strong upcoming attacks. At least, it would change the game radically, and not for the better. Instead of trying to gauge the security of your position and manage your resources carefully, you'd just save up until a big wave appears, then lay out everything you've got to make sure you get through. Perhaps a sort of checkpoint system would be interesting - so that when you reach certain levels the game notes how many resources you have and how much food you've spent levelling up heroes, and then you can choose to continue again from that point with the same resources. Of course that changes the game so that you're playing to get to the next checkpoint with the most resources possible... but that was sort of what you were doing anyway. The real issue would be deciding how to handle items and techs you've acquired - if you let people keep them from the checkpoint, they'd just be using the same strategies every time, but if you randomize them you probably end up with a lot of useless combinations, or players starting new games until they something close to the combination they wanted anyway.

Stepping into a dark room, finding a nice haul of Dust (used to power rooms).

The game doesn't take that long to play, and earlier levels are quicker - so the devs have a good setup really. To put it in perspective, I'm complaining about maybe 1-2 hours of 'average' gameplay before I get to the interesting challenging stuff, and even during that time I can be trying different tactics. That's not an enormous length of time, and it's not just busywork, because of the randomization. Nonetheless, I would have appreciated more exciting endings when I lose a game - even just something like the monsters having to destroy a security shield before the crystal itself takes damage could provide an additional step between 'fine' and 'totally screwed' to make it less of an extreme transition when the game takes a negative turn. Especially when you're basically guaranteed to be seeing those negative turns more than a few times - to put it mildly. Losing can be fun (Dwarf Fortress, anyone?) but in Dungeon of the Endless, I don't think it is.