Atmos changes feedback thread

[actioninja]

Some recent changes to atmos that had pretty major ramifications have been merged for a solid amount of time now.

Use this thread for feedback about said changes. Criticism and complaints are allowed but pointless hostility will not be tolerated.

[actioninja]

Yes there was in fact another thread on this. No, it did not have anything useful to say.

The only useful post was by the maker of said changes (thefinalpotato), which is here for posterity.

Spoiler:

Thank you for opening a feedback thread, having a place to collect this stuff is a lot better then seeing issues pop up on the tracker about it.

You appear to have me confused for OmegaDarkPotato for... Some reason? If you're going to scree at me and attempt to assassinate my character, please do it properly.
While I don't play very often, I do observe a large number of rounds. I've noticed a lot of the issues you've mentioned here.

As I said in one of those issues, I don't particularly like the current state of things. More how firelocks are played with, and how draining they can be. I'm not going to apologize for making the atmos simulation work, but I do recognize that it's had serious side effects. There are some prs open that attempt to make them less painful, but I'm not sure they'd work out all that well. We could make things settle out slower, but that would cause more cold floods and increase the amount of lag the subsystem experiences.

You don't say this outright, but this has made atmospherics the role and atmos the concept a lot more deadly. I think this is useful, but we have some severe hickups. You mention delta station, and you're right, delta has all the map design elements that make these things worse. It's much more open and has huge interconnected rooms, hell all that's separating medbay and arrivals is one airlock and a lot of open space. It wasn't designed for the system we have, and doesn't preform well with it. I've done and will continue to do work on the system to make it less reliant on sealed off rooms, performance and gameplay wise, but it'll take a bit of time.

@PKP, you mention this exact same thing happening in the past, what are you referring to? It sounds like a bug with excited group equalization and space, but maybe it was tried in the past.

[ShibaInuLord]

my god atmos is so fast now that a vapour can opened in like departures on icebox will reach kitchen in like a second, funny and good

[Capsandi]

Is it light on performance? Yes(i think)
Is it realistic? More-so than viscous gasses
Is it fun?
Well, most of the fun derived from the system was from putting out fires with an extinguisher, and cleaning up spills of plasma as you watch the purple hue drain into the scrubbers, and literally outrunning a blazing plasmafire outside of the SM, getting into a disposals chute right as the flames grow close. Oh yea and also with atmos moving this fast, maintenance wings can get flooded with fire from a single bar of plasma.
So I guess it would be a good change if the performance increase was worth it, but simplifying the plasmaflood into a instant combustion of most departments and hallways is incredibly OP. Atmos's ability to flood was already OP(but fun), I mean they had the ability to do it shiftstart, every shift, with no work required to get the gas, but now that there's no counterplay what with fire extinguishers being totally not effective when the whole room equals out to being on fire anyways, and plasma being barely visible at the densities it spreads out to become, there needs to be a process in which the giant ass tanks of gas outside of atmos get filled. Until then, 2/10 not fun.

[skoglol]

Plasma fires have been dumb for a long time, the ratio between significance and effort is way off. Boomers will disagree Im sure, and I will go on not caring what they think.

[Horza]

Fire alarms constantly go off for frivolous reasons now, and they keep going off. I've played more than a few rounds where the hallway fire alarms constantly go off for the entire round because some part of some hallway is slightly overpressured or too warm. Atmos/engineering often can't fix it even if they want to because of reasons.

The other major thing I've noticed is that plasma floods, for some reason, aren't the slow-burning wall of fire slowly moving down hallways anymore, but rather become instant flash fires with no immediately visible trigger. Last week there was one round where a low-key plasma flood was being performed across the entire station, with low levels of plasma, maybe sub-0.5%, to the point where the plasma isn't actually visible yet, and something just caused all of it, in every room, to instantly combust. The entire station instantly and immediately turned into a raging fire, even rooms where there was no apparent source of ignition and where the temperature hadn't risen enough to cause spontaneous combustion. All of deadchat was wondering what the hell the admins were doing since it didn't seem normal. Admins were just as confused.

[skoglol]

Fire alarms constantly go off for frivolous reasons now, and they keep going off. I've played more than a few rounds where the hallway fire alarms constantly go off for the entire round because some part of some hallway is slightly overpressured or too warm. Atmos/engineering often can't fix it even if they want to because of reasons.

Fire alarm status is area based, but can be triggered by all the individual fire alarms in the area. If areas equalize or heat spreads faster, I could see that causing constant alarms.

[cacogen]

The newfound sensitivity of fire alarms would be less of an issue if fire doors could be opened by hand, like the ones outside the shuttle. Otherwise it just further encourages getting a crowbar at the beginning of the round.

[skoglol]

Well, yes. But remember that the temperature limit was set assuming old atmos. They can likely be tweaked a bit now.

[TheFinalPotato]

I'll have more to say here later, throwing this out for now

Is it light on performance? Yes(i think)
Is it realistic? More-so than viscous gasses
Is it fun?
Well, most of the fun derived from the system was from putting out fires with an extinguisher, and cleaning up spills of plasma as you watch the purple hue drain into the scrubbers, and literally outrunning a blazing plasmafire outside of the SM, getting into a disposals chute right as the flames grow close. Oh yea and also with atmos moving this fast, maintenance wings can get flooded with fire from a single bar of plasma.
So I guess it would be a good change if the performance increase was worth it, but simplifying the plasmaflood into a instant combustion of most departments and hallways is incredibly OP. Atmos's ability to flood was already OP(but fun), I mean they had the ability to do it shiftstart, every shift, with no work required to get the gas, but now that there's no counterplay what with fire extinguishers being totally not effective when the whole room equals out to being on fire anyways, and plasma being barely visible at the densities it spreads out to become, there needs to be a process in which the giant ass tanks of gas outside of atmos get filled. Until then, 2/10 not fun.

Yeah, this is an issue I agree. Should probably raise the initial visibility on gas overlays. I've also seen some very fucky things around firestarting, really really need to look into that. Instant station wide flame is very much not intended

Fire alarms constantly go off for frivolous reasons now, and they keep going off. I've played more than a few rounds where the hallway fire alarms constantly go off for the entire round because some part of some hallway is slightly overpressured or too warm. Atmos/engineering often can't fix it even if they want to because of reasons.

Part of this is around sealing off sources of heat, the sm not leaking heat is probably the big one here. Another aspect is fire alarms being triggered by cold, which as I mentioned on sparks pr is for other reasons. I do strongly believe that how firealarms react to heat/cold needs to get tweaked to better support large areas, and reduce fire alarm fatigue

Well, yes. But remember that the temperature limit was set assuming old atmos. They can likely be tweaked a bit now.

I've been meaning to do this, it's a bit too low right now, they're below the superconduction ceasing threshold, so the alarm can finish, the firelocks open, and a bit of cold plasma gets in and ignites the whole area again. Not good behavior

[NecromancerAnne]

Echoing the concern with the plasma, the sudden spreading of atmos effects across the whole station is very noticable with water vapour. Enough that I suspect you could abuse it to achieve old freon levels of mass station freezing.

I think this might be less of an issue with the changes and more how much of an effect even trace amounts of gas are having, given even trace plasma was feeding into a widespreading fire, and a breach on Icebox causes a lot more havoc than on other stations due to the infinite amounts of junk gas that starts flowing in.

It's definitely cool but it might be a little extreme at the moment and probably some numbers held over from the old system not really being appropriate for the current.

[gottingtonballbags]

In regards to the changes on atmos, I think that while fast and actually realistic gases are cool (since this game literally was spawned from an atmospherics sim), the fact that countering atmos instability in a room (i.e making it breathable and warm) is a relentless pain in the ass is what breaks it for me. I would say that if you want to keep the gases fast moving and the temperature dynamic, then measures to counter atmos problems should be equally as efficient as methods to solve them (i.e make space heaters more efficient and wider-range, let rooms refill faster, etc. etc.) to help reduce the "firelock station" shuttle calls I keep on seeing. It's absurd how fast the firelocks drop, too. Break one wall on meta and all of a sudden the entire main hallway is locked down (slight hyperbole).

[CDranzer]

The only thing I've really noticed is that fire alarms are more temperamental. Also, one time there was a plasma fire blazing two inches in front of my face and I didn't seem to suffer any negative consequences whatsoever until it had been burning for several minutes, so that was nice, if a little confusing. It was in a big open part of Tram though, which may be why. Temperature equalization or some such. Also, I really kind of wish we had an on-screen thermometer that always gives you a general idea of the ambient temperature. I feel like I should be able to tell the difference between 5c and 40c without pulling out my PDA, even if it's only a general idea.

[cacogen]

I would say that if you want to keep the gases fast moving and the temperature dynamic, then measures to counter atmos problems should be equally as efficient as methods to solve them (i.e make space heaters more efficient and wider-range, let rooms refill faster, etc. etc.) to help reduce the "firelock station" shuttle calls I keep on seeing.

This is how I feel. I think too the firelock problem should be solved directly (because at this rate it's never going to be solved properly) by making them able to be opened by hand. It would take longer than with a crowbar. If the proper fixes were made to make them less of a common occurrence then it could be reverted.

[PKPenguin321]

might be tripping but a tiny bit of plasma seems to poison the air just a teensy bit on more or less the entire station for the rest of the round. might be a quirk of tramstation actually

[steampunch]

In regards to the changes on atmos, I think that while fast and actually realistic gases are cool (since this game literally was spawned from an atmospherics sim), the fact that countering atmos instability in a room (i.e making it breathable and warm) is a relentless pain in the ass is what breaks it for me. I would say that if you want to keep the gases fast moving and the temperature dynamic, then measures to counter atmos problems should be equally as efficient as methods to solve them (i.e make space heaters more efficient and wider-range, let rooms refill faster, etc. etc.) to help reduce the "firelock station" shuttle calls I keep on seeing. It's absurd how fast the firelocks drop, too. Break one wall on meta and all of a sudden the entire main hallway is locked down (slight hyperbole).

There definitely needs to be better counters to atmos problems than there currently are. I've died to atmos nonsense far too much in the past month I've been back on TG, and every other shift feels like it ends in one way or another because of atmos. Firelock hell is especially irritating, cause it's not really engaging, it's just annoying enough that you want it to end even if that means ending the entire round.

[Omega_DarkPotato]

breathing plasma the whole round thanks to fastmos or having everything get locked down with firelocks because of a breach is really unenjoyable.

At least it's gotten me to go and put 2 minutes of time into some anti-toxin seiver pills (2-3u per pill at 950C works wonders, by the way, against toxin damage)

[Domitius]

I feel useless trying to re-pressurize area as an atmos tech. Handling breaches is so demoralizing as there is no way to effectively heat the air back up as it cools down too quickly to do anything. I like the new system but there is no interesting way in interacting with it besides calling the shuttle.

[Intercept0r]

I feel useless trying to re-pressurize area as an atmos tech. Handling breaches is so demoralizing as there is no way to effectively heat the air back up as it cools down too quickly to do anything. I like the new system but there is no interesting way in interacting with it besides calling the shuttle.

Valuable feedback. There's a job whose purpose is to unfuck atmos and fix airlocks. You're saying that job doesn't currently have the tools to make it happen.

I don't fully understand the mechanics though - doesn't repressurizing a room flood it with gas that has the same temperatue as the canister? The few moles of atmosphere that remain in a depressurized room should not drastically lower the temperature once flooded with many more moles of room temp air, so where does the "coldness" come from?

[TheFinalPotato]

There's a lot of useful feedback here, while I'm not going to address each point directly, it really has been helpful to hear your thoughts. Thank you.

That said, I've drafted a design doc for atmospherics/atmos tech. The focus is on promoting repairing/maintaining the station, alongside the creation of tools to better maintain the station. Figured this would be a good spot to put it.

https://hackmd.io/@tgstation/ryUHjU5ZO

[cacogen]

That was a semi-interesting read but it lacked the specificity I'd have liked. Even just outlining the specific problems like space heaters being ineffectual or whatever if you're not sure how you'll fix things yet.

From what I've seen in general you seem agnostic on the whole firelock issue. As though it's a problem of a lack of knowledge on behalf of the players or something instead of an issue with how firelocks work and how fastmos affects them.

There are never enough atmos techs, either. The ones that do exist don't even know the basics and have little interest in doing their jobs. While the game is designed around these people fixing atmos fastmos is going to continue to cause problems.

[Capsandi]

I agree with much of what your saying with how reactions are used in the atmospherics tech gameplay loop. I think that many reactions are balanced around the supplies present in atmospherics, which result in a more stagnant atmosphere outside of atmospherics. The result of this is of course only those who code dive knowing anything more than co2, plasma, oxygen and nitrogen, which is tragic considering how the system could be as interwoven with the rest of gameplay as the chemistry system is, instead of the only two alterations to the atmosphere being a breach or plasma-fire.

While this is being contemplated, the presence of the billion mol tanks outside of atmospherics has always felt like a bandaid solution to where the station gets its gas. Of course, plasma fires are great fun for the players, and have been major forces in some of my favorite rounds, but the sourcing of this gas, even if acquired rapidly, could serve to more closely tie the techs to the rest of the departments, without disallowing those great big plasma-fires. I have a few ideas in how these gasses could be sourced.

I also agree on the wiki issue, though I'm not sure how to format reactions between different gasses in efficiency. Should there be a huge table that says "no reaction, no reaction, forms n20 at very low temperatures, no reaction"? Should each gas have its own wiki page detailing formation and reactions in which the gas is present, whether it can be found in lavaland/icemoon's atmosphere and its affect on species? I believe the latter option is preferable, if only to lower the amount of info on the one guide to atmospherics page.

[TheFinalPotato]

That was a semi-interesting read but it lacked the specificity I'd have liked. Even just outlining the specific problems like space heaters being ineffectual or whatever if you're not sure how you'll fix things yet.

I was somewhat vauge because I'm not 100% on what exactly should be fixed, this is a template for my goal, not the exact steps.

From what I've seen in general you seem agnostic on the whole firelock issue. As though it's a problem of a lack of knowledge on behalf of the players or something instead of an issue with how firelocks work and how fastmos affects them.

I wouln't say that, I have some thoughts on how to make the system a tad bit better, but that won't solve your issues with firelocks, because I'm not interested in neutering the system, just in making it controllable.

There are never enough atmos techs, either. The ones that do exist don't even know the basics and have little interest in doing their jobs. While the game is designed around these people fixing atmos fastmos is going to continue to cause problems.

It seems like you agree with me that it's partially a lack of knowledge, I don't intend to just thrust the responsibilities of fixing atmos issues on these people, I want to make the system more accessible and understandable, give them more/better tools to help solve their problems, and create a feedback loop that makes fixing these things engaging. I thought I was clear about this.

[TheFinalPotato]

I agree with much of what your saying with how reactions are used in the atmospherics tech gameplay loop. I think that many reactions are balanced around the supplies present in atmospherics, which result in a more stagnant atmosphere outside of atmospherics. The result of this is of course only those who code dive knowing anything more than co2, plasma, oxygen and nitrogen, which is tragic considering how the system could be as interwoven with the rest of gameplay as the chemistry system is, instead of the only two alterations to the atmosphere being a breach or plasma-fire.

Mood

While this is being contemplated, the presence of the billion mol tanks outside of atmospherics has always felt like a bandaid solution to where the station gets its gas. Of course, plasma fires are great fun for the players, and have been major forces in some of my favorite rounds, but the sourcing of this gas, even if acquired rapidly, could serve to more closely tie the techs to the rest of the departments, without disallowing those great big plasma-fires. I have a few ideas in how these gasses could be sourced.

We've partially discussed this on discord, but I don't think large stocks of gas are a major issue, and even if the process of sourcing them is engaging it'll just end up being a distraction from what the main gameplay loop should be.

I also agree on the wiki issue, though I'm not sure how to format reactions between different gasses in efficiency. Should there be a huge table that says "no reaction, no reaction, forms n20 at very low temperatures, no reaction"? Should each gas have its own wiki page detailing formation and reactions in which the gas is present, whether it can be found in lavaland/icemoon's atmosphere and its affect on species? I believe the latter option is preferable, if only to lower the amount of info on the one guide to atmospherics page.

I need to think on this a bit more, but perhaps.

[cacogen]

I wouln't say that, I have some thoughts on how to make the system a tad bit better, but that won't solve your issues with firelocks, because I'm not interested in neutering the system, just in making it controllable

My issue with them is how often they're triggered as a result of fastmos. The real issue is how how one breach has such widespread effects. I know that's built-in and desirable, but it's going to continue to be an issue until players are given the tools and knowledge to fix it. I know that's your point, but it seems like it's going to be awhile before it's resolved.

[ArcaneDefence]

Finally bothering to put my bits together in one place here-

Scrubbers are simply bad at removing residual gases.
You can dump a large volume of plasma in the central primary hallway, and in a minute or two after setting the air alarm to contaminated it will be in a pseudo state of limbo. That being areas of ~.22mol plasma slowly bleeding onto the scrubbers for a far, far longer time than the initial spill took to clean up.
These clouds of plasma create obnoxious choking alerts, and leave the station with toxin build up for extended periods of time despite the actions necessary to clean it up having already been taken.

Another issue is of course the "flash flooding" where proper conditions can force gases to settle over extremely large areas.
It's piss easy to dump a canister of gas and after a few moments have it settle out to every available turf it can connect to, as I've pointed out in discord before. I don't think I need to bother explaining why that's a problem.

[pugie]

There's been a lot of controversial changes recently, all you need to do is connect to the server and look at the changelog to see. I don't mind any of these, I like a lot of them in fact... except this one.

Gonna try be a bit more clear here than I was with my saltpost on the matter in which I referred to the wrong food name player (Aldi Nutoka)

I've given these changes time to think about, I've reflected on them, and I still think they're awful and make the game drastically less enjoyable for players.

I think coders have wanted to move round times to be longer and to make PRs that flesh out jobs to encourage people to do them (fermichem, cytology, bar rp bots) so this is the awfully conspicuous outlier - fastmos causes not only gases to engulf the station in 1 or 2 ticks, a few holes in space will rapidly depressurise and freeze as close to sub-zero as possible, triggering all firelocks in an area & making the station fully uninhabitable for anyone who can't get coffee or a winter coat in time, or people who decided they wouldn't grab a crowbar shift start.

There are a lot of threats to the station & having detrimental forces isn't a bad thing, but having the entire station be depressurised is unfun & nigh insurmountable, it always warrants shuttle calls as soon as possible. This greatly limits the development of each round in which gas plays a factor in, disallows anyone to try anything cool because super fast atmos just surpasses everything.

Even those that survive aren't much better off than those that die, having to open firelocks every 10 seconds & be slowed down considerably by a hardsuit if they don't get a winter coat. Before this change, atmos changes would only noticeably effect a single area which could be fixed if one so desired, but also avoided through properly equipping (firesuit/hardsuit) then unequipping once cleared from the threat. Now the aforementioned single area is the entire station, and the singular counter is calling shuttle.

I'm sure it will be said, if anything, that this can be fixed by changing how firelocks work, but I think it goes deeper than that - I don't think fastmos holds any merit, I think it should straight up be reverted already - especially considering it's been live for a month or two now and no one seems to want to make a PR to fix the most glaring issue every single round faces now. https://github.com/tgstation/tgstation/pull/57381

[cacogen]

>no one seems to want to make a PR to fix the most glaring issue every single round faces now.
Ok to be fair I did make a PR that tried to ease the impact of firelocks a bit by making them openable by hand after five seconds. LemonInTheDark wasn't even interested in it temporarily while he does whatever he's going to do to resolve the issues that were always there, but that fastmos made matter. I don't know of any other PRs made to try to reduce them. Good post though, you did a better job of listing the various problems it causes than I have.

[Ghilker]

>no one seems to want to make a PR to fix the most glaring issue every single round faces now.
Ok to be fair I did make a PR that tried to ease the impact of firelocks a bit by making them openable by hand after five seconds. LemonInTheDark wasn't even interested in it temporarily while he does whatever he's going to do to resolve the issues that were always there, but that fastmos made matter. I don't know of any other PRs made to try to reduce them. Good post though, you did a better job of listing the various problems it causes than I have.

temporary PRs are bad

[cacogen]

Why did he say it

[ABearInTheWoods]

I made a quick low effort pr: https://github.com/tgstation/tgstation/pull/57467

But a good solution could be to de-couple the rate of process for tiles vs atoms/machines so scrubbers/heaters could still run twice a second while tile movement happens like, once a second or the old original every 2 seconds.

[TheFinalPotato]

I don't think that'd solve the issue here, we've committed to pipenets and machines marching in lockstep with environmental, and changing that will significantly change behavior.
As I understand it most of the agony from the changes aren't sourced from the subsystem ticking too often, but are instead caused by atmos "events" being more impactful, since each tick is more effective and excited groups now work again. Could be wrong about that tho.

Late to the party here I know.

[Capsandi]

Hey the depressurization/gas leak gameplay that was fun before this pr is still frustrating.