On Chemistry

[oranges]

Bottom post of the previous page:

also don't think I am not aware of issues with xeniobio, they just happen to be another thread

[iksyp]

Assistants can get nukie hardsuits and one pulse pistol

What
How
That sounds broken as hell

ruins

[Jegub]

Could we have examples of observed behaviour that requires addressing in order to get a handle on what specific issues there might be as stands?

Wall o' pinions:

The initial problem with having to develop a supply chain every time would be ensuring that the starter stock went into roundstart staples such as an effective cryo mix, or trying to respond to a traitor going loud early on. Any system should be able to function for skeleton crews as well as packed stations.

From the sound of it botany could be reduced back to basic materials production. While that was okay to do once in a while, a simple meditation like wiring the solars or mining the asteroid, botany pre-manipulator was a shadow of what it could be and was populated accordingly. Providing feedstock would need to be a case of keeping a handful of specialised plants in rotation so that there was enough time available to do something personally interesting and people would actually want to sign up. A wider palette of potentially useful genes and some apparatus to enable gene mutation could offset the extra production work and encourage more people to look beyond mutation smoke and explosives.

I'd like to see scientists interacting more directly with slimes as they used to, and reaping the results of their attentiveness or neglect. There's the issue of maintaining pace and flow, perhaps the range of slime behaviour could be expanded to help with that (calling specific slimes over to you?). Since I rarely spend time in xenobio myself, however, I can't say much about it.

Allowing a greater number of people access to an already limited number of workstations would swiftly result in slapfights devolving into vitrioling, even though making a functional dispenser is now facile they are still regularly taken. I've seen people say that they won't work Hydroponics if another botanist is present, mostly because the single DNA manipulator becomes a bottleneck. Affording easier access to the processes especially for MDs and other medical staff would be desirable, some centralised reactors which can queue manufacturing runs from more abundant interfaces (laptops?) could open up availability and require some finesse from those wanting to make more questionable products due to the increased chance of oversight.

Having a set of reconfigurable physical apparatus could create mapping headaches and wouldn't address availability, if the goal is to increase complexity this could be achieved by having extra processing methods within the interface. Separating the input and fabrication would make it harder to blow your arm off, sadly, perhaps incautiously made products could become unstable at the point of handling. Manufacturing more useful precursors could take place in a separate controlled environment and require personal attention to provide the primary risk factor.

I've occasionally thought as to how the chemical interface could work differently, SpaceChem and Opus Magnum are interesting approaches but the primary factor needs to be speed. An interface I could see being quick enough to allow acceptable turnaround would be akin to the modular input-process-output chains used in Jeskola Buzz and VirtualDub's audio filtering system:


Macros worry me, beyond the inability to check them after entry; they are basically there to bypass the actual method. They risk ending up with a few people using the feature to build enough generic strings that subsequent players then copy and paste, taking no interest in how they were devised or what other possibilities there could be. Then the job calcifies because no one is engaging with it directly, shades of telescience. There was justification for allowing circuit schematics to be input because of how complex they could become, and being able to store and share chemical processes between networked users could work well, but constructing them should be done in-game to ensure engagement and give some human value to the output.
The way to reduce dreary repetition used to be a space cleaner sprayer, so you could boost your amounts for anything needed in bulk. Something like a 200 unit carboy that was too large for grenades or bags could reduce the number of production runs needed, it could be fabricated early on with glass (and plastic?), or metabolisation rates and overdose thresholds could be reduced to allow smaller dosages, especially if the basic elements were to be finite.

There's the social aspect of someone making a request and then seeing someone fulfil it for them, a reminder that they are in a living system that requires cooperation and interaction to function. The physical distance between mining and science made having a means of transferring materials indirectly a suitable solution to the logistics involved. I don't know whether it would be preferable to have departmental hoppers that process resources and transfer them to a station-wide pool in a similar manner, but there are times when even a minute or two away from your plants can mean returning to a host of problems.

Above all any changes would need to preserve responsivity. When a blob mutates and you have a dozen people coming in with severe burns there isn't really time to wrangle soxhlets in order to gear up for the revised situation. If the station just got several big holes punched in it and there are no surviving engineers some timely metal foam could make it livable or at least escapable. If no one else is available to subdue a rogue AI you can be hard pushed to get some thermite and metal foam to their chamber before detonation as things stand. When a roboticist needs to rejuvenate a brain it has to happen before the inhabitant decides to log off and do something else instead. To function within the context of a mutually satisfying game pace there is a need for the barrier to production to be swiftly surmountable, and to an extent this defines the upper bounds of its complexity.

Finally, there's the question of how such sweeping changes to various mechanics could be implemented. Merging 'trek'chem and goonchem which was comparatively minor got complicated by the balance issues that cropped up and there were instances such as when cryo chems couldn't be made because a precursor had been removed. A year on I would be asked by previously accomplished chemists how the revised system was working out, because they hadn't felt motivated to get back into it. This would be altering three departments simultaneously, so it would require extensive play-testing for bugs and unforeseen consequences by volunteers before it went near the actual servers in order to avoid alienating people with frequent revisions.

[Vile Beggar]

well the lord has spoken

[Shadowflame909]

Can't disagree with that thinking.

[Zarniwoop]

Can't disagree with that thinking.

I'm sure the folks it should matter to will find a way to blow it off.

[oranges]

Appreciate the detailed comments Jegub

Having a set of reconfigurable physical apparatus could create mapping headaches and wouldn't address availability, if the goal is to increase complexity this could be achieved by having extra processing methods within the interface. Separating the input and fabrication would make it harder to blow your arm off, sadly, perhaps incautiously made products could become unstable at the point of handling. Manufacturing more useful precursors could take place in a separate controlled environment and require personal attention to provide the primary risk factor.

I think you're imagining apparatus much bigger than I actually intended, these would be two/three square table scale sets, the interfaces you mentioned are not far off the mark for the kind of chaining systems I'm thinking about, I'm just saying they should have real on station physical "item" equivalents, not just exist soley inside an abstract tgui interface.

I am not sure what you mean by availability, there's already only one or two chem dispensers, so it's not sufficiently different from that approach.

In terms of basic supply, simple chems should be available in chemistry by default, and or be sourceable from cargo, chems that lead to higher tier reagents should be found in botany/xenobio, and require processing to be made useful. Argument can even be made for allowing higher tier chems to also be purchaseable from cargo, as long as the cost was significant enough that the botany/xenobio routes would always be preferable and nothing should be purchaseable that can be readily made into useable reagent effects without processing through the production chains.

I agree that low pop round can present scaling issues, but I'd rather solve these by giving chemists access to botany and xenobio on low pop than by giving up on the idea of more engaging chemistry, issues around managing plants and xeno without constant attention are perhaps open questions.

Macros worry me, beyond the inability to check them after entry; they are basically there to bypass the actual method. They risk ending up with a few people using the feature to build enough generic strings that subsequent players then copy and paste, taking no interest in how they were devised or what other possibilities there could be. Then the job calcifies because no one is engaging with it directly, shades of telescience. There was justification for allowing circuit schematics to be input because of how complex they could become, and being able to store and share chemical processes between networked users could work well, but constructing them should be done in-game to ensure engagement and give some human value to the output.

Agree, any kind of copypaste gameplay shoudl be strongly avoided, I regret macros, they were not something I should have been keen on.

There's the social aspect of someone making a request and then seeing someone fulfil it for them, a reminder that they are in a living system that requires cooperation and interaction to function. The physical distance between mining and science made having a means of transferring materials indirectly a suitable solution to the logistics involved. I don't know whether it would be preferable to have departmental hoppers that process resources and transfer them to a station-wide pool in a similar manner, but there are times when even a minute or two away from your plants can mean returning to a host of problems.

I appreciate this, but surely the chemist would come to botany or xenobio to request and then get reagent sources

Above all any changes would need to preserve responsivity. When a blob mutates and you have a dozen people coming in with severe burns there isn't really time to wrangle soxhlets in order to gear up for the revised situation. If the station just got several big holes punched in it and there are no surviving engineers some timely metal foam could make it livable or at least escapable. If no one else is available to subdue a rogue AI you can be hard pushed to get some thermite and metal foam to their chamber before detonation as things stand. When a roboticist needs to rejuvenate a brain it has to happen before the inhabitant decides to log off and do something else instead. To function within the context of a mutually satisfying game pace there is a need for the barrier to production to be swiftly surmountable, and to an extent this defines the upper bounds of its complexity.

The general trend I want to aim for is longer rounds, not shorter, I appreciate that it can be unfortunate to not be able to react to *every* single situation, but good gameplay arises where tradeoffs are required to be made, and you have to think about what you actually want to acheive during a round, you can definitely retool a production line, but I don’t think it should be an instantaneous thing.

Finally, there's the question of how such sweeping changes to various mechanics could be implemented. Merging 'trek'chem and goonchem which was comparatively minor got complicated by the balance issues that cropped up and there were instances such as when cryo chems couldn't be made because a precursor had been removed. A year on I would be asked by previously accomplished chemists how the revised system was working out, because they hadn't felt motivated to get back into it. This would be altering three departments simultaneously, so it would require extensive play-testing for bugs and unforeseen consequences by volunteers before it went near the actual servers in order to avoid alienating people with frequent revisions.

This is something the project has always struggled with and in general our current approach seems to work, the playerbases resistance to any kind of change not withstanding

[oranges]

Can't disagree with that thinking.

I'm sure the folks it should matter to will find a way to blow it off.

If your only contribution to every thread is going to be that we don’t' care or take others opinions into account, I'm just going to put you on dev forum approval to save us all having to skip over your posts.

Consider it your only warning.

[MisterPerson]

Sounds like Jegub is describing a fun game, but I'm not sure that it would be fun as a repetitive, multiplayer game. Once you figure out how to make whatever chemical really fast, you basically just recreate that exact sequence over and over. Even if there's no explicit macro/copy-and-paste/whatever ability, people are still gonna just blindly copy from the wiki/whatever with no real need to experiment. So trying to make a chain system just sounds like the current boring situation with a neat UI stuck on top. Again, a good game, just one that only needs to be played once.

Since the actual chemicals themselves can't be limited, I think the focus should generally be on what chemical(s) are currently being produced at any given moment and the conflict that can arise from limiting that production speed. For example, if someone orders 10 units of chemical A and someone else orders 10 units of chemical B and both take 8 minutes to produce, you have to decide what to do. Do you make A first, B first, or perhaps you compromise and make half of A, half of B, half of A, half of B to keep both satisfied. Maybe if the dude who wants chemical A brought a little bribe to guarantee they go first, hey, that would be fun. And if science wants a ton of something but upgrades your machines, woah thanks, here you go, whatever you need friend while the poor clown and his space lube request gets backlogged.

In other words, just make chemical reactions take time to complete and you'll be halfway to making chemistry interesting. Maybe not fun, but at least interesting.

[actioninja]

Sounds like Jegub is describing a fun game, but I'm not sure that it would be fun as a repetitive, multiplayer game. Once you figure out how to make whatever chemical really fast, you basically just recreate that exact sequence over and over. Even if there's no explicit macro/copy-and-paste/whatever ability, people are still gonna just blindly copy from the wiki/whatever with no real need to experiment. So trying to make a chain system just sounds like the current boring situation with a neat UI stuck on top. Again, a good game, just one that only needs to be played once.

I think you're missing the point. If you design a system with sufficient complexity, "flawless optimization" isn't trivial. A good system would have enough points of fine tuning that truly optimizing the fun out of it would be difficult. Even then, just because you can just copy a guide and take the fun out of it, you don't have to. If people want to have fun they can, and if people want to be powergaming hyperoptimizing shitlords they can as well.

[MisterPerson]

That's possible to do, but based on what was provided and assuming it's on top of the existing chemical recipes, the optimal way to make all chemicals is fairly trivial (protip: It's what you would do in your hand, just in a machine!). If someone wants to make the chem recipes more involved, well yeah then obviously that will change things. But making the optimal strategy not obvious and trivial is much harder to do than you might think.

[Screemonster]

Sounds like Jegub is describing a fun game, but I'm not sure that it would be fun as a repetitive, multiplayer game. Once you figure out how to make whatever chemical really fast, you basically just recreate that exact sequence over and over. Even if there's no explicit macro/copy-and-paste/whatever ability, people are still gonna just blindly copy from the wiki/whatever with no real need to experiment. So trying to make a chain system just sounds like the current boring situation with a neat UI stuck on top. Again, a good game, just one that only needs to be played once.

I think you're missing the point. If you design a system with sufficient complexity, "flawless optimization" isn't trivial. A good system would have enough points of fine tuning that truly optimizing the fun out of it would be difficult. Even then, just because you can just copy a guide and take the fun out of it, you don't have to. If people want to have fun they can, and if people want to be powergaming hyperoptimizing shitlords they can as well.

the eternal problem with the "git gud" mindset - if you try to learn for yourself instead of using the already-solved optimised flowchart on the wiki, the powergamers will scream at you for being inefficient. If you just do the wiki to appease the shitheads that scream at you in OOC after the round for daring to not be perfect, you'll get moaned at for not really understanding it and also get fucked over next time coderbus changes something that makes the wiki inaccurate, in which case the shitheads will still scream at you for not instantly magically divining the correct solution out of the ether

[actioninja]

That's possible to do, but based on what was provided and assuming it's on top of the existing chemical recipes, the optimal way to make all chemicals is fairly trivial (protip: It's what you would do in your hand, just in a machine!). If someone wants to make the chem recipes more involved, well yeah then obviously that will change things. But making the optimal strategy not obvious and trivial is much harder to do than you might think.

That's literally what orange man is proposing here though, making chem recipes more involved.

[iamgoofball]

hey, I pr'd macros originally after ripping off the idea from goon

heres why I added it:

if a chemical is only balanced because it takes 5 minutes to click all the buttons to make the chemical, it's not balanced

it's only balanced if you're not a mega autism sperg, and we have mega autism spergs up the wazoo here at ss13

[iamgoofball]

if chemistry is reworked to:
1. have randomized effects per chemical, and randomized chemicals period
2. not have static chemicals at all effects wise, just turn everything into an effect
3. get rid of the UI carpal tunnel crunch

then removing macros makes sense

otherwise, dont bother

[MisterPerson]

That's possible to do, but based on what was provided and assuming it's on top of the existing chemical recipes, the optimal way to make all chemicals is fairly trivial (protip: It's what you would do in your hand, just in a machine!). If someone wants to make the chem recipes more involved, well yeah then obviously that will change things. But making the optimal strategy not obvious and trivial is much harder to do than you might think.

That's literally what orange man is proposing here though, making chem recipes more involved.

No details were provided, so I'm assuming the worst. I'm fun at parties.

[Capsandi]

we need to do away with the chem dispenser entirely, make all reagents gathered from xenobio and botany sources and add actual work recipes to distil and mix chemicals
i.e you have to heat things to temperatures, burn off side products etc

This guys sayin, What were all thinkin! At least what I'm thinking! The chem dispenser is just that: free chems at no cost at all(Its been a long time since power in the things have been an issue) And we could theoretically wipe them from the codebase right now! All of the chems in the dispensers can be ground from objects found throughout the station. Xeno and Botany could have chems to give to chemistry, but for xenobio, we already have slime jelly, which could be utilized in more reactions. Some of the more lackluster or non-existent effects with slimecores could also generate chemicals otherwise available or unavailable. Why keep it to those departments? Walls could rust over in maintenance and an assistant could help a fella out by scraping it into a beaker.

NSFW:

The clown could donate some of his goof juice... nope, cant do it

The SM or other engine could generate "liquid supermatter" having all sorts of uses! Where I imagine this leading to is a chemist on the brink of curing the disease which plagues the station, and having to navigate a now run-down station's many crevasses to get the ingredients. I like where your headed with this. Orange man... smart?

[somerandomguy]

I didn't know how badly we needed liquid supermatter until now
Drink it? Dust.
Touch the container without gloves? Dust.
Get hit by it? Dust.
Make smoke with it? Dust.
You get the idea