Cogmind progression thoughts

[Suslik]

I know that good roguelikes often take some time to get into. I instantly fell in love with cogmind's graphics, setting and lore and decided to try really hard to like it. And for first 10-20 hours I literally forced myself to play because I did not like the process at all but I knew it might be worth it in the end. I have written a negative review for the game at 20 hours mark or so.

Long story short: yes, after about 40 hours I started enjoying the game and after 80 hours I could not think of any other games anymore -- I was far too busy theorycrafting builds in cogmind. Unfortunately I was playing offline and Steam decided to wipe my offline playtime, currently I have 100+ hours and I'm finally properly hooked. I'm clearly not the greatest player in the known universe, but I can reliably now progress to level -05..-03 every run, still figuring lategame and all the branches. However, I was very close to dropping the game multiple times during first 40 hours.

My major complain that is still valid after all this playtime is: early game is too boring but it blocks access to all the interesting stuff. An example: I really want to make a build around melee weapons with 5 kinetic spears, microactuators, adv melee suites, actuator arrays and all the good stuff, but.. But to even start experimenting with it, I have to get through the same exact earlygame boring nonsense: fighting hordes of Pest's without a grenade launcher, looking for lrg storage units that sometimes forget to spawn, getting screwed with a mid-sized garrison at level -08 instad of a small one that refuses to open and leaves me without RIF and 99 others ways to make earlygame super-annoying.

Earlygame melee weapons are hands-down bad. Earlygame energy shields block 25% of damage which's not even worth wasting a slot for. Earlygame weapons just marginally differ from each other making them boring. Earlygame armors block as much damage as a weapon does. Earlygame flying propulsion is just not viable. Eearlygame accuracy is so low and you have so little options to counteract it, that fighting operators, watchers and swarmers is just super-annoying.

Apparently there's no simple way of fixing this problem(I definitely see it as a problem), but I really think it might be worth moving more build-enabling tech earlier into the game, making it more accessible. Instead of starting interesting tech at -07, make it available at -10 and then make it slower to advance. Replace bad starter equipment with its more useful versions: sml storage units are just trash, sml batteries, sml matter containers, starter shields and armors can be just replaced with later versions making earlygame less boring. It might be worth adding more interesting enemies like rogues at starter levels to counter-balance these changes.
Another concern: earlygame has too little options, but late-game might feel overwhelming with too many options. Again, maybe just starting at power level of -07 would counteract this. Maybe give players more equipment slots in early game and give them less slots on the last floors? For example give players +3 starter slots and give them 1 extra slot during last three levels instead of 2.

And last one: I really like scripted events that give you essential equipment like rocket launcher in caves. How about making them more reliable? Maybe add a way to get access to essential equipment by adding fabricators at starter levels (with possibility of Schematic(Lrg. Storage Unit)) ?

I really want to know what other players and the developer think of these.

[DDarkray]

Originally posted by Suslik:

early game is too boring but it blocks access to all the interesting stuff.

I also believe the early game is the biggest drawback in Cogmind so far. It used to be worse in the past, but it's slowly getting better. Still, there are a few things that could be improved. For example, as you've mentioned, some items are just really bad as if they're disguised as newb-traps. I know certain items are made specifically as enemy's parts (like the weak Sensor Array, Sml. Matter Pod, and Flight Unit) which is fine, but they probably shouldn't be part of the stockpile on the floor and could be replaced with something more useful or interesting.

Originally posted by Suslik:

Maybe give players more equipment slots in early game and give them less slots on the last floors?

It's an interesting idea, but I can see it as a problem of impacting the mid-game, making it a bit too easy, but I like the idea of making the early game more fun.

If you haven't noticed, one of the challenges in Cogmind is called Devolution which gives you a whopping 20 slots at the beginning (?!?!), but you lose 1 slot every time you ascend a floor (making it really difficult to beat the game). So yeah, your suggestion did partly come true.

Originally posted by Suslik:

Maybe add a way to get access to essential equipment by adding fabricators at starter levels

I don't think that would work here, partly due to lore reason and partly due to over-complicating the early game for new players. Trying to figure out the basic terminal machine is already hard enough, so having the fabricator in early game would probably push it a little too far.

I also want to say that I've been avoiding Mines a lot, simply because I hate the infestation event. I like the idea of having a lot of RNG events there (to spice up the early game), but I had times where I lose really easily even when I try to play carefully, and it's becoming a bit discouraging for me. If there's a way for players to deal with infestation more easily, I would certainly try out Mines every game. :)

[zxc]

I'm super glad you found your way to enjoying the game.

I think everyone agrees that the early game is by far the weakest part of the game. Materials is pretty tactical but it has a low impact on the rest of the run, so long as you survive it. Random out-of-depth item caches were added to Materials to spice it up a bit, and the first Materials floor (-10) was made extra small to be quicker (and easier) to complete. This is still probably not enough, however.

A few notes on Materials: fabricators used to exist there but it was a bit weird because you almost never had schematics that early, so it was suggested to remove them. Every Materials floor has a guaranteed storage unit cache, which contains two of each of the inventory storage items (lrg, med and sml). Sml storage isn't entirely useless - it is pretty 'affordable' for flight builds, though mostly people stick to med or no storage that early.

I agree that a *lot* of early game items suck. Mostly, the best ones are storage units and high integrity + coverage items which can soak up hits. Small targeting bonuses etc are negligible. One suggestion made a while back was to outright remove one Materials floor, so you would start at -9. I still think this is the way to go, and it would also mean you start with two extra slots and 100 more core integrity than now. The developer made the first Materials floor smaller instead.

I don't think many items are 'essential' to grab before Factory. Mostly, you need to get to Factory in one piece and with some items equipped. Probably the most important is lrg storage units, but otherwise, you will quickly find much better items in Factory. A launcher is probably the next most useful item, as you pointed out.

Have you been to any midgame branches off the caves? Since you're into the lore, I'm sure you'd love that.

[Suslik]

@DDarkray

Trying to figure out the basic terminal machine is already hard enough, so having the fabricator in early game would probably push it a little too far.

One [s]newbie trap[/s] developer trap I really wanted to point out: Oversimplifying starter levels will make the game LESS accessible to new players because they'll have less controllable opportunity to learn essential mechanics. Having 2 utility slots will NOT teach new player what the game feels like. Besides, a new player is very likely going to randomly bump into a bunch of level transitions and will find themselves randomly at -05 or so without any clue what to do and without any room for experimentation because one mistake = death at that point. So I really think it's best to introduce players to more mechanics in the beginning of the game instead of lategame.

I also don't consider easy modes a solution in any way, because easy mode teaches strategies that will not be viable on normal difficulty and it also feels humiliating to play them.

@zxc

I don't think many items are 'essential' to grab before Factory. Mostly, you need to get to Factory in one piece and with some items equipped.

Storage, RIF and launcher are all pretty much obligatory for Any early game build. Getting RIF is admittedly an awesome challenge that I love completing over and over(although getting screwed with a big garrison on -08 is no fun at all), but finding storages and launchers with 2 utility slots to play with is just a chore and not fun.

Have you been to any midgame branches off the caves? Since you're into the lore, I'm sure you'd love that.

I always do all side areas without exceptions. I'm not interesting in finishing the game skipping some of its parts. And all the zion/data miner/warlord optional stuff is probably what I love the most about cogmind's lore.

[zxc]

Neat, I hope you've visited Zhirov as well, my favourite out of the midgame.

RIF is not obligatory. It's very much just an option (considering it comes with downsides as well). RIF was actually just added this last update, so you've had the unique experience of learning the game with it in existence whereas everyone prior to now has never had the option. Storage and launcher is close to obligatory, but storage is guaranteed on every Materials floor so there's a very good chance you run into it even if you don't poke around looking for it. Launchers are the only real question mark.

[GJ]

My own experience has been that Materials has always been fun and interesting to play out. The fact that Mines is actually difficult (sometimes) gives meaning to all those fights with Swarmers and Grunts that you can't really die to as such, because core integrity can still matter. Playing it "properly" almost certainly calls for careful play like making use of a Mining Laser that you acquired in Scrapyard. At the same time, while Mines does matter, it's rarely significant. There's much more impactful branches and decisions in the game, so if you're not in the mood for playing very carefully from the first second of a run, it doesn't feel like you have to do Mines, you can just breeze through Mats pretty quickly. You don't have to reach Factory with more than 2-3x Lrg. Storage Units, which can be found on a single Mats floor at times (for -10, just getting 1x Med. from a Hauler is sufficient). Launchers aren't necessary for dealing with Swarmers: combat can comfortably deal with them using just 3x Lgt. Assault Rifles, flight in Factory does like to have a launcher for them, but melee (accuracy bonus!) in a doorway can work.

I've always considered Cogmind to have an impressive transition from the early game to midgame to lategame. Stuff like the basic sensor array and targeting computer has fairly niche value, but from the perspective of evolving slots, it's good to know such items exist (and will likely get better). When I started playing the game, the targeting computer I'd found represented valuable information with regard to what I wanted my first evolution to be. Finding these items isn't particularly impactful once you already know they exist, you don't have to hunt for them (though they're not worthless), but the main emphasis of -10 is just teaching anyway, you do Mines if you want intrigue, or you just run through it quickly if you want to get to the meat of the game.

These kinds of items represent something fairly similar to the 1x Hunter that you can see in -8, or the 1x Programmer that gets sent after you you in -7. To me, the game does an excellent job of giving a new player hints of what to expect in the later main floors, without being too obvious about it. You still have to think it through how you're actually expecting the challenge to evolve. I liked that a lot when I was new to the game, still do. Branch maps have somewhat unique/specific dangers, which given you can identify that an exit is a branch before taking it... I like that design too. Of course you won't know what a branch is starting the game, but learning via death is something of a roguelike standard, it's at least somewhat implied in the manual nowadays, that Mines is a special type of map more dangerous than the next set of Mats, somewhat discernible from the name as well.

Also, I really don't get the fairly common claims at this point that weapons only differ marginally. Direct upgrades only have slightly better stats, because they're just the same type of weapon you're already running, except appearing later in the game, but different damage types and weapon types are... very different. In the early game you can occasionally see Flak and high-crit kinetics in addition to the already notable differences between the various assault rifles and autoguns, any form of EM is significant because of frying and engine explosions (particularly EM Shotguns), thermals have variety in terms of Spread Lasers (low range), Adv. Beam Cannon (high crit) and just midrange "do damage at a decent range" weapons that are substandard, use only in a famine situation. Melee and launchers are obviously very different weapon types, and there's notable differences between EM and EX launchers.

Admittedly, I don't have much to say to your actual main point, just wanted to elaborate on my love of the early game given how often it gets trashed. It's true that certain builds only come into play later on, because committing to them outright requires a lot more slots than you start with. I don't personally consider this a problem, I don't think you have to be capable of playing all types of builds from the very start, I still play these builds because something later on makes it very attractive to go for such a build. I don't think I would enjoy the early game being significantly easier even if it would enable more playstyles at that stage of the game. In particular, Factory being even easier than it already is outright removes nuance from the game. I like having excuses for not doing every single instance of Caves while fully exploring them, the risk of attrition and losing key items makes you actually consider such decisions. If the challenge disappears, doing that every time... is just the thing to do, the nuance of sometimes wanting to hurry, sometimes wanting to explore efficiently rather than thoroughly, is diminished.

[Kyzrati]

(I'm out of town right now for the Roguelike Celebration so haven't been as available to engage in discussion, just getting to this thread today.)

Originally posted by Suslik:

And last one: I really like scripted events that give you essential equipment like rocket launcher in caves. How about making them more reliable? Maybe add a way to get access to essential equipment by adding fabricators at starter levels (with possibility of Schematic(Lrg. Storage Unit)) ?

I don't really want Cogmind to be about starting with a reliable "build X" right away from the start. If that were the goal, the game would be designed quite differently overall. It's meant to be more a game about adaptation, but as it turns out some players both prefer, and are skilled enough, to create stable builds that last throughout much of their run to the end. And yes this is what Fabricators were added for in the first place, as a way for players to specifically tweak their build more reliably amidst all the randomness, though they weren't, and still aren't, meant to be a way to create and maintain a full build.

Actually, early versions of Cogmind did have all machine types early on, including Fabricators, though they were removed later on as players didn't think they were worth using. Schematics were also removed from the early game, as were Scanalyzers--all of these were player suggestions since they didn't fit the pacing and weren't all that meaningful until Factory anyway.

One could say that makes the early levels less important, but they still provide interesting opportunities for tactical challenges, despite being a stretch that technically doesn't last too long if you're bent on just getting past the first few floors while you set up the rudimentary parts of your build, perhaps enjoying a few fun unexpected encounters along the way.

Note that the early game is still getting content added to it (including a lot in the very next release, Beta 8, which is focused on the early game).

Originally posted by Suslik:

Earlygame melee weapons are hands-down bad. Earlygame energy shields block 25% of damage which's not even worth wasting a slot for. Earlygame weapons just marginally differ from each other making them boring. Earlygame armors block as much damage as a weapon does. Earlygame flying propulsion is just not viable. Eearlygame accuracy is so low and you have so little options to counteract it, that fighting operators, watchers and swarmers is just super-annoying.

I disagree with a lot of the specific comments in here, though.

Early-game energy shields can be useful for certain builds (especially in the mid-game), just not yours, apparently. I've heard from players who are happy they became a thing (low-percentage shields with low energy cost didn't exist before--they were added in the shields overhaul a while back).

Early-game weapons may not differ much in damage, but in terms of damage type their differences are quite meaningful, and if you're not taking advantage of that then your effectiveness is suffering for it.

I'm not sure what you mean by "Earlygame armors block as much damage as a weapon does." The absolute weakest armor you can take from Sentries you meet early on can absorb an average of
9 shots from the weapons enemies are using at that depth. (And of course it doesn't take all that damage consecutively--it's only hit by a percentage of shots, those where it's covering/protecting other parts from damage, doing its job.) It's perfectly effective and something I use on my combat builds alot, if I'm not going for an offensive build (which among other strategies stack weak weapons to take out enemies very quickly).

Like many parts, early-game propulsion is primarily something that other bots need, so sure it'll be available to you if you want it, and some players prefer to stockpile it in the early-game for a quick and safe transition to flight later on, but in general there are a ton of parts lying around and in Cogmind you're not meant to be faced with a huge range of parts that will all be viable for you and your build. Sometimes you are forced to use substandard gear, and have to deal with it.

Regarding accuracy there are a lot of ways to deal with that and make it easy to take out any particular target(s), the main exception there being Operators, which are intentionally not easy to quickly take out, though they aren't even armed so don't pose an immediate threat.

Originally posted by Suslik:

(although getting screwed with a big garrison on -08 is no fun at all)

For Beta 8 it's changing such that the -8 Garrison is always the smallest one.

Originally posted by DDarkray:

If there's a way for players to deal with infestation more easily, I would certainly try out Mines every game. :)

Just bring a Grenade Launcher, done! Don't have one, don't go! That's the easy way to ensure survival in any Mines.

Originally posted by Suslik:

I also don't consider easy modes a solution in any way, because easy mode teaches strategies that will not be viable on normal difficulty and it also feels humiliating to play them.

Yep, this is why I've been strongly considering how Cogmind handles difficulty modes for the 1.0 release. An easier mode will become the default (but will be renamed), and the current standard "roguelike" mode will become something that players must choose to activate. It's all a matter of perception.

[Qua3st0r]

When I had played 50 hours, I made a suggestion to unlock quickstart levels ( permalink ) for reasons similar to those described by the OP here: when you want to focus on learning how to progress in the mid levels, it's tedious to work through the early levels to get to the "new" (to you) stuff.

Now that I have played 250+ hours, with several wins and high score at 40K+, I still feel that suggestion is well-founded.

[Suslik]

@Kyzrati
I can see that a lot of your points come from experience of previous iterations that I was not participating in. As in, starter level fabricators were removed to fix a problem of them being not viable in early game. But maybe instead of outright deleting them it was better to consider other options to make them viable, such as more ways to obtain shematics via Trojan()'s or maybe even carrying some schematics from previous builds? Or maybe even combine fabricator and scanalyzer into one machine?

I'm glad to know that in b8 you're focusing on early game and I'm really keen to see what it's going to feel like. If you were to re-consider any of my points though, I'd definitely want you to think again of making earlygame more varied by having more starter equipment slots and making lategame less overwhelming by reducing their progression towards the end. Say, +2 starter utility slots and +1 slot/level on last 2 levels instead of +2 slots/level.

[zxc]

'+2 starter utility slots and +1 slot/level on last 2 levels instead of +2 slots/level.'

Sounds a bit arbitrary to me, switching up the rules midway through the game.

I don't think you need to reduce the slot evolutions at the end of the game at all. However, it would be nice to have more slots to play with at the start of the game.

[Kyzrati]

Well, having two or more extra slots at the beginning ends up necessitating a change to the evolution rules by default, because it's already set such that the maximum number of slots, given a linear progression, ends up just under the interface limit.

It would also affect the balance, as in the early game would have to be made more difficult overall so that it remains challenging for experienced players, despite the fact that less skilled players won't be able to effectively use those extra slots anyway. This would further increase the spread between what players experience in the early game.

(Also note the lowest difficulty setting does already give an extra starter slot.)

Originally posted by Suslik:

@Kyzrati
As in, starter level fabricators were removed to fix a problem of them being not viable in early game. But maybe instead of outright deleting them it was better to consider other options to make them viable

But as stated I prefer to de-emphasize that part of the game--it's not about building your ideal build right away and keeping it that way, so having fabrication only be a thing later on for the occasional tweak makes more sense in that context.

[Wildfire]

The fact that you have to combat your way to midgame and then risk wasting all that progress on a build that may not work (and that you definitely don't have all the equipment for and may never find) means it's not worth playing.

I want to get better, but how can I learn the stealth play when it's only viable by the time I'm already expected to be good at it? I can't get good at it without doing it for a bit, and I can't do it until it's already midgame.

Imagine playing Halo on normal until halfway through, and then it turns into Metal Gear Solid, but you've literally never done any stealth until that moment. And when you inevitably screw up and die after a few minutes, it's another romp through half of Halo before you get another ten minutes of stealth practice.

It's simply not viable by any metric.

If the game is supposed to be about adaptation, the first half must offer the opportunity to learn the mechanics that will be crucial in the second half.

Anything else is an incomplete game.

[Kyzrati]

Are you playing on the default difficulty setting? It's set to the most challenging setting by default (and it's *very hard* xD), so if learning the basics of tactics/strategies/AI in a friendlier environment is your thing then I'd recommend changing it in the options menu.

That said, in the end different build types will have an easier or more difficulty time in various parts of the world. You'll find some places where stealth is basically not an option, and others where full combat is extremely hard to survive while a stealth hacker can come out literally unscathed. Many runs involve evolving and adapting to the kinds of challenges you want (or must!) face, so Cogmind is generally not like other games in that you find one style and stick with it throughout the entire run. This is how it's designed.

Once you're good at the first few floors, even on the hardest mode it only takes about 10 minutes to reach the mid game. (Or faster if you don't really care about building up for the mid-game--you can speed run the first few floors in just a few minutes and BAM, you're in the mid-game where it starts to open up :D)

In any case, I think the best recommendation here would be to use a lower difficulty if you're just looking for a learning/practice experience.

[Suslik]

@Kyzrati I know you're working on difficulty settings and you expect them to solve this problem of learning various playstyles by new players, but I honestly think that selecting an easy difficulty is not just something new players do. I'd never do that myself either. I think that @Wildfire does have a point that the game should introduce you to more mechanics earlier, because right now early floors have basically one mechanic: shoot your launcher at swarmers and shoot guns at everything else. Exiles do help with this a little bit, but not enough.

Not only starter equipment is too weak to allow for more playstyles early on, you don't even have enough spaces to use it anyway. Even though I do like the game, I still think that it's not the best design choice, neither from new players' perspective, nor from veterans'.

[Kyzrati]

There'll be a new blog post about the difficulty levels once they're ready. This feature should be one of the main things to come with Beta 9.

You don't really need much space to use stuff in the early game, because that section of the game doesn't really last long anyway, but the space is also necessary to leave room for more stuff (I was talking about this in the last stream, actually). There are many plans to shake up the early game, though I can't do them at this time. Prior to 1.0 would be nice, but I can't do that without more funding. (Actually this week's SITREP is going to address that.)

[Stryker]

Originally posted by Wildfire:

I want to get better, but how can I learn the stealth play when it's only viable by the time I'm already expected to be good at it? I can't get good at it without doing it for a bit, and I can't do it until it's already midgame.

This general issue is the one weakness of the game, in my opinion. It's a great game, but difficult to get the late-game transition learning experience without dedicating major hours. What I did was simply save-scum. I did not lower the difficulty, just restarted some late game areas to experiment. For me this was the difference in liking vs. loving the game. After restarting several times I learned enough to finally win the game, even back to back, without 'cheating'. It's an excellent game, don't give up on it.