Messages

Addendum 2;

Rather than distinct memory slots, a more granular memory capacity is also an option. Then higher tier prototypes could take more memory than lower tier regular parts.

Addendum;

To add to the previous post: the most obvious issue is that cartridges would require a lot of inventory space. If this is judged to be too harsh, it is a problem that can be easily solved. Simply replace the idea of cartridges with the idea of memory slots integrated into the core. These then take no inventory space, and otherwise function in exactly the same manner.

One thing that has been discussed to some degree in the discord is the fact that fabbing strategies currently dominate the game to a very large degree, especially on fast propulsion. In this post I'd first like to name a number of potential issues present in the way fabbing works. Then I will introduce an example of a way I believe the current fabbing system could be improved. Lastly, I'll argue why I think such a change would solve the earlier-named issues.


Issues;
1. Hacking-focused builds can fabricate an entire extended build. This method feels somewhat illegitimate, as it is considerably easier to execute than most other strategies for assembling extended-viable builds. Because of this discrepancy with other strateges, on a lot of the faster build archetypes preparing for extended through fabbing feels rather mandatory. Especially on such builds, the strength of lategame fabbing serves to invalidate scavenger-style gameplay to some extent. This is unfortunate for the people who enjoy such gameplay more.
2. There is no compromise-option; either you are fully equipped to fab, or you can hardly fab at all. If you have enough hackware to fab semi-reliably, the only real limit in the number of items you can fab is the fabricator quarantine. Hackware builds can consistently fab enough in this allotted time to get together a complete build. Scavenger-style builds that run no or little hackware can rarely even fabricate a single item. Destroying hubs is somewhat of a remedy against this, but arguably the contrast between these two extremes is still too stark.
3. Fabbing strategies disincentivize risk-reward options to some extent. Scavenger-style builds are often incentivized to take on branches and fight dangerous out-of-depth enemies for rewards. Fabbing builds don't need to rely on such rewards; they only need hackware, and their extended succes is already pretty much guaranteed. Even if branch rewards can be useful for them as well, there is never a reason to feel forced to take a large risk.
4. Fabbing builds overlap with machine hacking builds. This is a bit overly convenient, as machine hacking is one of the most consistent ways to make 0b10 floors easier in the midgame. This ease of blazing through the midgame, combined with their strong lategame, makes fabbing strategies far too consistent and trivial.
5. Setting up a fabbing-centric build is too straightforward. Fabbing builds require little to no resource-management. Their game-plan consists of a very small number of uncomplicated steps. Get hackware; hack for schematics; fabricate your build; equip the parts and win. This again leads to a lack of meaningful and interesting decisions when playing fabbing builds. (Not because of the lack of risk, which is already covered in 3., but rather because of the lack of meaningful resource management and adaptivity.)



Example proposal of altered system;
Fabbing items requires cartridges. Blank cartridges are a limited resource, and can be obtained as branch rewards, or by prevailing against similarly difficult challenges. At terminals, the player can hack for a schematic to write it onto a blank cartridge in inventory. Once a schematic is written onto a cartridge, it can not be overwritten. Cartridges need to be inserted into a fabricator to fab, and are returned after the item is fabricated (possibly losing some integrity, so use is limited?). In addition to these changes, the hacking difficulties for both schematic pulling and fabricating are drastically lowered.

These changes will have the following consequences:

1. The number of (distinct) items that can be fabbed is limited by the number of cartridges the player acquires. This means it is no longer (easily) possible to fabricate entire builds.
2. Compromise builds are valid. Since the hacking difficulty is lower, even builds with lower amounts of hackware can meaningfully supplement their build with a small number of fabbed items.
3. Incentivizes risk-reward. Cartridges require taking risks to obtain, making it more challenging to get into a position where you can fab a large number of items.
4. There is only small overlap with machine hacking. This makes fabbing builds less oppressive and omnipotent in the midgame.
5. The setup process a little more complex. There are a larger number of decisions to make, such as which items are completely necessary and which a luxury. There is a larger number of non-trivial steps involved in getting your fabbing build into a position where it can take on the lategame.


In short, I believe the current balance of fabrication strategies can be massively improved. There are obviously a myriad ways to go about implementing such a balance change in terms of mechanics. The above contains a description of what I consider the be one example of such an implementation. In replying to this, I would appreciate it if people could distinguish between disagreement with the named issues and disagreement with the proposed example of a way to implement fixes to these issues. I think discussion about the prior is more useful, as kyzrati is unlikely to be worse than me at thinking of a system that fixes these issues.

Re: Metafield

I honestly really dislike the idea of nerfing metafield, and especially the 'overloading' nerf proposed by Tone. Below I will outline the main reasons I think metafield should not be nerfed, and in the end I will discuss my qualms with this type of nerf in particular.

1. Metafield makes the earlygame challenging. On flight runs, there is very little reason to take risks. If you have a halfway decent flighthack build, you can just hack for any and all schematics you need, and build the perfect extended build by the end of factory. Any parts you fab are likely better or at least as good as anything you could find by taking on the extreme challenges in the game, like killing exiles or Zheroes. This makes flight earlygame boring and trivial. You get some decent flight parts from haulers, get your hackware stack going, put on utility shielding, and just skip your way to lategame.

The existence of metafield changes this. The fact that you can make your lategame state more powerful by getting metafield, incentivizes the player to prepare to take on Zimprinter. Doing so is extremely difficult on flight. This in turn incentivizes players to put together an insanely good build by -7 or -6. This requirement makes the earlygame challenging and interesting. The parts you can get from Zheroes or Brawn, which would otherwise be outdated by the time your fabbing gets online (or at the very least, fabbing easily replaces them, and additionally doesn't require you to keep the parts in your inventory throughout the entire midgame), now have real value.


2. Metafield makes slot distribution challenging. It has been said earlier in this thread, but metafield synergizes especially well with low-prop flight builds. This synergy incentivizes the player to plan their evolutions around acquiring metafield. This induces a significant challenge. The early- and midgame become significantly harder. At this point in the game, the player doesn't have enough energy to run metafield yet, supposing they even managed to get it, and additionally there are few good mass support utils available. The resource management this entails makes the midgame a lot more fun.


3. Near-similar lategame builds can be achieved without metafield as well. If you take a 4-prop exp. thruster build, it already nearly reaches the speed cap. The energy requirement compared to 2-prop with metafield is very similar. You have to spend an extra slot, and deal with more propulsion attrition, but you also get the extra support from the additional thrusters. Four-prop slot distribution also makes the earlygame significantly easier. Even if this setup is maybe not exactly on par with 2-prop metafield, it is very close.

The real strength of metafield lies not so much in the enabling of few-prop lategame builds. Rather, it lies in the ease of transition. If you have a suboptimal, less extreme flight build, for example a 6-prop one, metafield makes the game significantly easier around -4 and -3. At this point the player has just become able to use metafield in terms of energy, instantly allowing them to boost their speed such that they can more easily deal with things like R branches. Without metafield, this would require the player to first fabricate better thrusters to replace their old ones. With metafield, you can reach the speed you need instantly, and then worry about replacing your thrusters afterwards.


4. It's okay to have powerful parts in the game. Metafield takes significant effort and skill to obtain, and again significant effort and skill to succesfully use in a build. If it is not at least as good in-slot as the tier 8* and tier 9* stuff flighthack builds fabricate with significantly more ease, then it will not see play in lategame builds. This would be a shame, because it is a really fun item.

There is something to be said against the prevalence of metafield in most flight builds we see on the discord. One could take this prevalence as an argument that metafield is too strong. I would like to present an alternative interpretation however. It is not that metafield is too strong. Rather, there are simply not many other things bored veteran flight players can greed for in the earlygame and midgame. Rather than removing the one fun part from this major section of the game, we should add other fun alternatives that conflict with it. (Imprinting bad. Fight me. )


Reaction to nerf-suggestion:
The overloading nerf puts metafield on par with triangulator and heat shielding. Cld. propulsion already exists, and the cld. aspect is bad. Having to overload propulsion is the sign of a weak build that is already doomed to die. I really can't imagine any decently skilled player regularly overloading prop for any other reason than escaping stasis traps. Since metafield would not even help with this, it would be a truly useless item. It would be a 3 mass -20 energy utility that allows you to turn your thrusters into worse versions of bad thrusters that already exist. Metafield is an item you have to build your entire slot distribution and loadout around. If its use becomes limited like this, any time you're not using it, you're limping around on a mutilated build. The very fact that this is even being considered as a nerf makes me feel like the people who have agreed to this suggestion have never played this sort of build.

End of rant.

Today I finished my first run wherein I encountered sterilisation. These runs take quite a while, and the time I can play is a bit limited, so I had hoped that one run would teach me everything I needed to know about sterilisation, and allow me to form a well-rounded opinion. Because of several misplays on my part, combined with less than stellar luck, I feel like my last run wasn't the perfect example I had hoped it would be. Nevertheless, I think it's still worthwhile to state my opinions about this new mechanic here. I would like to add the disclaimer that this opinion may be subject to change upon further experimentation.


First of all, the positive parts. The main motivation for the implementation of sterilisation, in my understanding, was to prevent infinite robot farming. In this regard the system is undeniably succesful. The fact that it stops any and all types of robot spawns puts a very strict limit on the possible number of robots that can be slain in a single map. It does this in a way that is unlikely to affect newer players' experience negatively, and it clearly communicates to the player what is happening. The player is given ample time to sort out their parts and leave. In fact, there is so much time that, on flight at least, the very existence of the heat aspect of sterilisation felt completely irrelevant.


Then, the negative parts. There are multiple angles to consider here, but I will stick with looking from the viewpoint of score, as I think it allows me to most concisely get across my point.

The point of the current competitive scoring system is to allow players to compare runs to one another, and to their own previous runs, and to gain an abstract measure of the quality of those runs. There are multiple ways of defining this abstract quality, but in cogmind, I would like to think that it is primarily meant to convey the strength of the player's build and overall strategy, with a specific focus on their combative abilities. Since it is difficult to measure something as abstract as 'combat strength', we take the easiest approach; the more bots you kill, the stronger you apparently were. Though this system has its disadvantages, it works surprisingly well for most run-of-the-mill runs. That is, until we start looking at the very extreme cases.

Under the new sterilisation mechanic, it can occur that a cogmind that still has ample combative strength is unable to get more score, simply because there is nothing left to use its combative strength against. This creates the dilemma that score no longer functions as an abstract indicator of combat strength, and thus loses its original meaning.

This effect disproportionally impacts some types of builds over others. Since the limit is per floor rather than overall, builds that delay their combative ability for later benefits are affected far more than builds that rely solely on combat throughout the entire game. This upsets the balance that exists between these types of builds. Because of this, one of the primary functions of score (comparing score with other players) becomes skewed, because it starts to depend heavily on playstyle.

In essence, the new sterilisation mechanics seeks to solve the issue of the lack of challenge that existed in relatively easily farming access forever, by completely removing the challenge in the first place. As the ambient heat feels completely irrelevant, I think I can safely say it does nothing to replace this challenge with another. Above all else, to me this simply feels Not Fun. One of the most rewarding aspects of cogmind is that you are always able to challenge yourself to perform better than on your last run. Sterilisation seems to discourage this. I would like to have my scores as an abstract measure to compare myself to, to gauge my own improvement at the game. And while I feel like I can always improve still, I suspect my score will not grow with this improvement anymore, because of the soft cap that is sterilisation. This is extremely frustrating.


Remarks:
I realise that my entire argument hinges on the fact that it is possible to engage sterilisation and still do extended, and then have core and parts to spare yet. While I am completely convinced that this is perfectly feasible, my current lack of having done so makes my argument less credible. I'll do my best to demonstrate my point as soon as time permits it.

Another remark is that, as will be clear from my explanation above, the negative effects of sterilisation are likely to only affect a very minor subgroup of players. In fact, I would not be very surprised to find out that I am the only one that thinks this way. I can hardly ask the mechanics of the entire game to be adjusted to suit my tastes. However, I still feel there is some value in sharing my candid opinion in a structured manner, so as to at least allow for meaningful discussion of the topic.

Signed,
Someone with apparently too much time on their hands (Ye gods, I should have stuck with discord.)

I thought a while more about the proposed digging change as a result of the above discussion. I tried to explicitly list all the points in the game where I feel required to do multi-tile digs, and realized most of them are through reinforced walls rather than dirt. The most heavily impacted place is clearly caves, but whenever I steal I tend to have to run past enemies anyway, and this isn't instantly game-losing.

I suppose I don't object too strongly to the proposed change. It does certainly simplify things. Though if it gets implemented I would strongly prefer it if the recent melee digging nerfs were reverted. And perhaps if stealth gets impacted too heavily we could increase the number of secret doors as zxc suggested.

You do make some fair points. I can see why you would be in favor of more harsh and stricter rules as a way to create more of a challenge for the player. In fact, I'm very much in favor as well of making cogmind more challenging in general. However, not every type of difficulty is equally fun and challenging in my opinion.

I find it hard to meaningfully continue the argument because I think in the end it boils down to personal preference. I want to have fun playing cogmind. In my previous comment I tried to explain why I think caveins can be un-fun, and making them more harsh would be even more un-fun. Beyond this, I feel like there is very little I can say on the issue.

I have indeed played infra arcana, and I did think it was a fun game, but I enjoyed it for different reasons than the ones that make me enjoy cogmind. While the parallel is insightful in the sense that it helps me understand your point of view, I don't really see it as an argument for the proposed change.

On a side note, I dislike this format of posting on forums. It feels a lot harder to express myself and to meaningfully engage in discussion. I would personally prefer to just use discord...

P.S: One thing that also comes to mind in this context is the difference between challenge and difficulty. It is not necessarily the case that all changes that make the game more difficult also make it more challenging. Take for example the naked core challenge run. Such a run is undeniably extremely difficult; you have far less tools to work with than in a normal game. But, on the other hand, the decision space is much smaller. At any point in the game, only a handful of options is relevant. This makes the number of decisions you have to make smaller, and each individual decision easier, if not less impactful.

The reason I bring this up is that I think additional difficulty should only be introduced insofar as it makes the game more challenging. This challenging aspect is part of what makes the game fun. While the proposed changes to caveins clearly make the game harder, I don't think they lead to new or more interesting decisions. I guess that's the crux of why I'm opposed to the changes.

I hope that sort of makes sense. Feel free to hit me up on discord if you want to discuss further. Writing these feels like too much of a chore though.

To respond to GJ's comments; I personally think trying to argue balance from the standpoint that cavein chance is low isn't great.

Balancing an instantly game-losing scenario by making the chance for it to occur low seems like a bad approach. If the expected return is still higher than the risk, people will take the option regardless. Then, when you win the roll, the fact that something horrible could have happened is completely irrelevant, and the risk might as well not have been there. And if you lose the roll you're back to the frustrating scenario where you lost a long run purely due to rng.

The real problem here imo stems not from the chance of low-rolling, but from the severity of the consequences. If the consequences of taking the risk are less extreme, you get a more nuanced decision.

Regarding the cases discussed in this thread, for fragile flight builds, things like getting scanned or getting a cavein are very much on the instant loss side of things. Maybe not necessarily for purposes of w0. But for extended or score runs those things are extremely terrible.

To relate back to the mechanic of digging; Suppose completely safe digging got removed (e.g. by making all dirt tiles have a chance to collapse). Then it would be way more fun to have a large probability of sustaining minor consequences from digging than it would be to have a minimal chance for catastrophic consequences. And by adjusting the probability and severity of the consequences we can still make it so that the total damage received per number of tiles dug is the same in the limit case where the player keeps on digging ad infinitum.

...I hope I'm phrasing that in a way that makes sense.

EDIT: phrasing

I didn't read through all other comments in detail yet (damn, what a wall of text ) but there's a couple of things I want to add that I think haven't been mentioned yet.

walls don't cave-in, and dirt can cave-in at any time (checked per turn and per move)

I'm very much against further nerfs that reduce the available options to dig completely without risk.
Digging is a very necessary strategy at many points in the game. To name a few instances, think of dodging behemoths in caves on a flight build, getting SHELL from the SHELL lab, digging into the Q exit prefab, etcetera. There are a decent number of instances where it feels like digging is the only appropriate response. For many of the more fragile builds, risking a cavein is risking instantly losing three or four items, which is often the same thing as instantly losing the run.

If I'm four hours into a run and my build gets squashed by a 5% chance cavein risk I couldn't reasonably have argued to not take, that's not fun.

Caveins as a mechanic should serve the purpose of disincentivizing unreasonable digging behavior. I think this goal is quite succesfully achieved already under the old system. Regardless of any flaws that the old cavein mechanics have, if they're changed I would put this as a minimal requirement: reasonable digging behavior should be completely safe, or at the very least should not have a (grantedly small) chance to disproportionally punish the player.



Thoughts on storage nerfs:
Just to add my two cents; I feel like on flight, storage is already reasonably balanced. There is a real trade-off when managing storage units. Mass for flight is super relevant. Think for example how SHELL got nerfed by ten mass, and now its only very rarely worth it to equip on flight builds, while before it was absolutely amazing. Utility slots as well are a difficult-to-manage resource. Spending extra utility slots on storage disallows you from using them for extra hackware or extra damage reduction, which both translate pretty much directly into less build-power for flight-hack and flying brick respectively.

The current system creates interesting decisions. I often find myself running flight builds that have severely reduced direct potency, for the tradeoff of being able to carry the parts I need later. This is a prime example of frontloading the difficulty to get later rewards, which is something I love about cogmind. I personally see no reason to change storage mechanics or balance when purely considering flight builds.

An argument could be made that storage needs to be nerfed because of the very extreme high-storage examples of combat builds that run amok, but I'm no expert on those so I can't really comment on their balance.

ディスコードではPimski#0183です。何か聞きたいことがあったら直接メッセージしてもいいし、コグマインドのチャンネルで聞いてもいいんです。

こちらこそよろしく！

Hey!

I'm an experienced player of cogmind. I speak some Japanese, so if you ever have a question, feel free to ask in Japanese too! I would love the opportunity to practice.

Most players, including myself, visit the discord server rather than this forum. If you ask things there you might get an answer more quickly. I'll paste a link below.


こんにちは！

コグマインドをずっと前からやってきたいわゆるベテランです。日本語も少しは話せるから、質問があったらいつでも遠慮なく聞いてください。こっちもせっかくの練習の機会ですから。（笑）

大体のプレーヤはディスコードを使っているんですから、そっちを使ってみたらもっと早く返事がもらえると思います。結構便利ですから、お勧めします。お誘いのリンクは下に張っておきました。

https://discord.gg/5EMGeHX