Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Xii]

I've long had this feeling about Cogmind that I'm finally starting to place accurately, due to the recent explosion of study of randomness, expressed in GDC talks and various prominent game design YouTube channels. They call it "input vs output randomness", and the feeling I have is of extreme frustration.

You see, Cogmind is brilliant fun to me always up until the point where some specific part gets destroyed by randomness. It is impossible to carry spares for everything, and multiple slots are required to work together to bring a complex build to life; therefore, the build rapidly starts falling apart.

Of course, I could keep going. If I were playing professionally, in a tournament, streaking, or for some other such reason obliged to stick with it, I probably would. But I'm not. I'm playing Cogmind for fun. When the fun ends, I stop playing. Make sense to me.

This causes a psychological regression into the safest builds. I start ignoring the cool, interesting parts because I know they'll last for all of one floor before some completely uncontrollable source of randomness kills it. It's the random critical hit against the player. Almost universally hated by players in all games present.

The introduction of RPGLIKE mode highlights the issue - it's much more fun! I'm actually excited to find unique, rare parts because I know I can design my build around them! But this experimental mode also makes other changes, like XP allocation, that aren't very Cogmind.

The whole thing is neatly outlined with input vs output randomness. Input randomness is a source of much fun and variety. It's the principal component of joyful randomness in roguelikes - level layout, loot caches, branch exits and surprise encounters. Output randomness is the stuff of nightmares. It's the thing that says "anything you do could backfire catastrophically at any moment".

Worst of all it's completely unnecessary. All instances of output randomness can easily be translated to input randomness. Let's consider Cogmind's core niche: PART DESTRUCTION. At the very heart of this concept is simply the notion that everything is consumable. This makes the game about resource management. But as it currently stands, the consumable nature of everything comes from output randomness - things last indefinitely until randomly they don't. Contrast this to input-random consumables, where it is upon the player to choose whether to use and thus consume an item or save it for later. In essence, part integrity would degrade on use. Every step you move, propulsion integrity -1. Every volley you fire, weapon integrity -1. Every terrain you scan, integrity -1. And so on.

Another example, terminal hacks. Each hack has a number next to it. The higher the number, the more powerful and costly the hack. When player selects a hack, randomly disable other available hacks up to the number, proportional to each hack's number. So a high hack will disable many others, while a low one is likely to only disable one. This creates the strategic trade-off of whether to pick many small advantages or one big one. If the number exceeds the sum of other hacks present, trigger negative feedback. Hacking gear reduces these numbers. Numbers above a certain threshold might be inaccessible altogether, requiring hacking gear.

Maybe this isn't the game you set out to make. Maybe it's too gargantuan of a change to undertake. But it's the game I wish I could play. RPGLIKE comes closest, but I don't like the XP mechanics.

[Kyzrati]

Yup, like any game its gameplay will only appeal to a subset of gamers, and Cogmind was designed with extreme intent. I must say it's working exactly as planned.

As I understand, you're the type of player who'd prefers more standard CRPG roguelike mechanics, in which you have a more reliable and consistent build and use that to overcome challenges, but that's not at all what Cogmind was designed to be. Hence the RPGLIKE experiment/alternative mode (the design of which I described on the blog, for anyone interested in reading more about it and its background).

But this experimental mode also makes other changes, like XP allocation, that aren't very Cogmind.

Neither is keeping your build xD. The whole RPGLIKE mode is very un-Cogmind, as described in the article.

things last indefinitely until randomly they don't. [...] Contrast this to input-random consumables, where it is upon the player to choose whether to use and thus consume an item or save it for later.

Skill is the key here, though, because there's generally less randomness than it seems. You have a good degree of control over the variables that influence when your build will change, or various options to maintain a relatively consistent build. Treating all parts as "consumables", you can indeed save them for later using your inventory, or if they're currently attached (and important enough) you can opt to protect them via armor, shields, part shielding, or otherwise taking care to manipulate your coverage distribution.

There will of course be times when so many variables are stacked against you and your plans will fall apart (especially more likely before you've had time to accumulate a stable and effective build), but when you're a good player this will almost never stop you from winning the run anyway, it just changes how you have to approach further challenges. And that's how Cogmind is designed--every run is winnable, though not necessarily in the way you planned.

That's where the key lies: do you want a game that allows you to play the same stable build throughout? OR one that requires you to adapt, sometimes significantly, to solve problems? A strong focus on adapting is central to Cogmind's vision. Regardless of how experienced you are, some degree of forced adaptation is the experience.

I generally don't like determinism in my roguelikes, and it shows

To me, using your understanding of a roguelike's world and its mechanics to stack variables in your favor and therefore increase the chances of a beneficial outcome is an important and fun skill to develop in a roguelike.

In Cogmind we've seen that skilled players can be very successful with pretty much any build type (but many players stick to one or two subtypes they're better at), so the skill ceiling is quite high, and at the same time we have a clear bell curve with regard to difficulty overall, so that much is working as intended, too.

[Xii]

[...] that's how Cogmind is designed--every run is winnable, though not necessarily in the way you planned.

That's where the key lies: do you want a game that allows you to play the same stable build throughout? OR one that requires you to adapt, sometimes significantly, to solve problems? A strong focus on adapting is central to Cogmind's vision. Regardless of how experienced you are, some degree of forced adaptation is the experience.

If planning is not to be the core tenent of gameplay, then what is, exactly? I struggle to see the fun in a world where planning has minimal value. What's the takeaway here? Nihilism?

I can win (almost) every run. But there is a point in some runs where the fun stops entirely, through no fault of my own. It's not like Getting Over It which is technically 100% skill-based. Cogmind literally throws a die at me and says "look at all your pathetic planning. worthless. go back to square one, for no other reason than because I say so." What game is made better by continuous, spontaneous, hostile entropy invalidating your every decision but the simplest, safest ones? It reduces the design and play space into a tiny speck where any deviation from the base is akin to gambling.

In my last run, a random explosive trap blew off a wall between my stealth build and a squad of swarmers. They proceeded to shoot off all of my parts. Where's the fun in here? Who enjoys a random punch in the face every now and then?

[Kyzrati]

If planning is not to be the core tenent of gameplay, then what is, exactly? I struggle to see the fun in a world where planning has minimal value.

But it doesn't have minimal value, it has just as much value as ever, it's just a question of what you're planning for.

And yes sometimes your build might hit a challenge that causes a setback, even a major one, and the idea is how to position yourself to come back from it. Adapting and survival is key, not tricking out your build so that it's completely unstoppable no matter what is thrown at it (or even just being able to see every single challenge you might face, although most all of them can be detected in advance if you want to focus on intel gathering).

That's a different type of game, one where you have a permanent character build and the mechanics are fully designed around that style of play. A game like that would have many systems designed very differently from what we have in Cogmind.

In Cogmind the idea is that you should be able to come back from setbacks, sometimes even in a different form, so if you really want to emphasize doing only one kind of build or without a doubt achieve your specific goal for a given run, you are going to be disappointed for sure.

Cogmind is a game that punishes inflexibility, and rewards adaptability and being able to naturally ride the ebb and flow of the game.

Ever since Cogmind was first advertised in 2013, the main short description I use for it (to this day) includes the following line:

The situation can quickly change as you lose components and rebuild yourself from enemy remains.

That is what Cogmind fundamentally is, and although skilled players can do much better at keeping a specific build together despite all odds, that is the baseline.

I can win (almost) every run. But there is a point in some runs where the fun stops entirely, through no fault of my own.

Honestly, I think you should probably stop playing if it's not fun, since the point of playing a game is (generally? always?) to enjoy it, and if you're not, then I'm sure there are other games you'll find fun!

Cogmind is actually designed to be significantly less random than most roguelikes, and it is, so I imagine you might enjoy other genres more. Or go for puzzle roguelikes where determinism and often perfect knowledge are the norm.

Like I opened with in my first reply: Some people enjoy this type of experience, some don't. There's no right or wrong, just people who enjoy it and people who'd prefer something else.

We're just going to continue talking past one another if you insist on Cogmind giving you something that it's not intended to provide in the first place