Please Nerf the Effects of Asteroid Turfs to Atmos

[BurgerBB]

 

See this turf? You can climb it. It vented 90% of the main level instantly and killed several people because someone removed it, either from an extremely precise explosion or tool. I don't know what it was because it happened quickly during a firefight I was observing. Now if this was /tg/ with atmos that didn't update instantly, I wouldn't have a problem with this, but this is Aurora. Now, I looked it up, asked some educated people, and they said that given a typical C-type asteroid, it WOULDN'T be completely air tight but the amount of time it would take for it to drain a 10x10x10 room wouldn't be seconds but rather about 5 minutes.

Assuming that a tile is 1m, a hole the size of a tile would vent it to half its pressure in 10 seconds. So that is literally worse case scenario. If you have holes inside the asteroid like swiss cheese, the total size of the holes would be something of about 50 sq cm worst case scenario, which would take about a minute, or maybe 10sq cm best case scneario, which would take 100 seconds to bring it to half.

Balance wise making it drain all the air in seconds because someone jackhammered it doesn't make sense. Realism wise it doesn't make sense either. Code wise it doesn't make sense either because this was the result of another feature that optimized the performance of asteroids.

 

Just to repeat this, it makes absolutely no sense that a compressed rock tile has the same effect on the air as space does.

Edited December 12, 2019 by BurgerBB

[Kaed]

12 hours ago, BurgerBB said:

 

 

See this turf? You can climb it. It vented 90% of the main level instantly and killed several people because someone removed it, either from an extremely precise explosion or tool. I don't know what it was because it happened quickly during a firefight I was observing. Now if this was /tg/ with atmos that didn't update instantly, I wouldn't have a problem with this, but this is Aurora. Now, I looked it up, asked some educated people, and they said that given a typical C-type asteroid, it WOULDN'T be completely air tight but the amount of time it would take for it to drain a 10x10x10 room wouldn't be seconds but rather about 5 minutes.

 

Assuming that a tile is 1m, a hole the size of a tile would vent it to half its pressure in 10 seconds. So that is literally worse case scenario. If you have holes inside the asteroid like swiss cheese, the total size of the holes would be something of about 50 sq cm worst case scenario, which would take about a minute, or maybe 10sq cm best case scneario, which would take 100 seconds to bring it to half.

Balance wise making it drain all the air in seconds because someone jackhammered it doesn't make sense. Realism wise it doesn't make sense either. Code wise it doesn't make sense either because this was the result of another feature that optimized the performance of asteroids.

 

Just to repeat this, it makes absolutely no sense that a compressed rock tile has the same effect on the air as space does. 

Edit: For people who aren't aware of the situation, like burger seems to be somewhat --v

This is not a bug or mis-implemented venting mechanic, it's an oversight in the slapdash approach the dev team took months ago to fixing problems with the asteroid.  You see, the asteroid, or "Mine" area is the largest uninterrupted 'one' in the current server map, spanning multiple levels and significantly outweighing the actual space of the station itself.  Due to our atmospheric engine (which is the one most people call ZAS), this caused a lot of lag and memory issues because the moment someone caused an atmospheric change between the inside of the station and the outside, by opening an airlock or breaking a window, the entire asteroid began trying to process the difference in pressure.  This caused problems such as areas taking 10 seconds or more to notice that they should be venting (sometimes allowing people to close a breach before ZAS could make it happen) and a strange glitch where the entire Mine area was caught in an endless loop of simulating a large pressure change, making it so miners and people on the asteroid were constantly being blown off their feet (people called this phenomenon 'space wind' and it was extremely annoying).

The 'fix' for this problem was switching all rock tiles on the server from 'simulated' to 'unsimulated', which meant that the Mine just doesn't process ZAS at all.  This helped a lot with smoothing out the problems with the server's chugging atmos, but left us with a new problem - because of how unsimulated tiles interact with simulated ones, rock tiles basically act like space tiles (which are also unsimulated), and suck all air out of the area as if it was a breach.  

I (and probably others) told the devs at the time that they were implementing this that they would need to come up with a solution for it, such as making the tiles under the station itself an alternate, simulated version of rock tiles, but as with many things on this server at the time and apparently still now, they deemed quickly finished halfway measures sufficient.  Now we suffer for this oversight every time someone breaks the flooring on the station and causes a magical unsimulated vacuum rock to appear.

Edited December 12, 2019 by Kaed

[MattAtlas]

42 minutes ago, Kaed said:

The 'fix' for this problem was switching all rock tiles on the server from 'simulated' to 'unsimulated', which meant that the Mine just doesn't process ZAS at all.  This helped a lot with smoothing out the problems with the server's chugging atmos, but left us with a new problem - because of how unsimulated tiles interact with simulated ones, rock tiles basically act like space tiles (which are also unsimulated), and suck all air out of the area as if it was a breach.  

I (and probably others) told the devs at the time that they were implementing this that they would need to come up with a solution for it, such as making the tiles under the station itself an alternate, simulated version of rock tiles, but as with many things on this server at the time and apparently still now, they deemed quickly finished halfway measures sufficient.  Now we suffer for this oversight every time someone breaks the flooring on the station and causes a magical unsimulated vacuum rock to appear.

1. It's not an oversight. It's intended behaviour.

2. All tiles must have either a space turf or an asteroid turf under them.

3. When you say things like 'but as with many things on this server at the time and apparently still now, they deemed quickly finished halfway measures sufficient', despite being unaware of the continued effort it took to find the cause and a fix for space wind (a year and more, in fact), while also putting in question the decision of several developers that are far more experienced in DM than you are, it makes for a weak argument.

4. The fix is complete. The fix works. What about it is halfway done? You don't like the effect. It does not mean you get to call it "incomplete" and say baseless claims about developers in the same sentence.

I would not be opposed to the unsimulated turf/simulated turf interaction getting looked at, though it'd take some time to figure out how.

[Alberyk]

There is no real way to fix or solve this, without sacrificing the performance, besides fully changing the maps to have less z-levels or use less asteroid tiles at all.

[MattAtlas]

2 minutes ago, Alberyk said:

There is no real way to fix or solve this, without sacrificing the performance, besides fully changing the maps to have less z-levels or use less asteroid tiles at all.

Wouldn't it still mean that they'd have space tiles under them anyways? Which would have the same effect.

Only way I can think of is making tiles harder to blow up.

[Kaed]

44 minutes ago, MattAtlas said:

1. It's not an oversight. It's intended behaviour.

2. All tiles must have either a space turf or an asteroid turf under them.

3. When you say things like 'but as with many things on this server at the time and apparently still now, they deemed quickly finished halfway measures sufficient', despite being unaware of the continued effort it took to find the cause and a fix for space wind (a year and more, in fact), while also putting in question the decision of several developers that are far more experienced in DM than you are, it makes for a weak argument.

4. The fix is complete. The fix works. What about it is halfway done? You don't like the effect. It does not mean you get to call it "incomplete" and say baseless claims about developers in the same sentence.

I would not be opposed to the unsimulated turf/simulated turf interaction getting looked at, though it'd take some time to figure out how.

You're right, I'm a novice developer, and have very little idea on what was involved in the fix. But I find it hard to believe that there is absolutely nothing that can be done to mitigate this problem, and that it was 'intended' for rocks to suck out all air in the station when they are revealed by the floor being destroyed.  It feels to me more like a compromise being passed off as intended, because it's utterly nonsensical for asteroid rocks to behave this way. 

The last person who did any serious work on fixing ZAS (that I'm aware of, which isn't the same as factually correct) was Lohikar, and he left the team over a year ago. So maybe the current devs don't want to or don't have the experience to fix ZAS, or maybe it's not even fixable.  But the fact remains is that this map you (the collective you, as in the server staff, not you specifically) have all created had a serious flaw in design that you fixed by basically writing out atmospheric simulation entirely on the majority of it, and as a result, incidents like the one that spawned this thread occur now. 

You not knowing a better solution to the problem does not mean it's 'working as intended', there is a serious flaw in the current setup.  If you don't know how to correct it, or it's even impossible to correct as you say, perhaps an alternative angle should be looked into to correct the asteroid problem.  Perhaps a map redeisgn like alb mentioned earlier.

Edited December 12, 2019 by Kaed

[Skull132]

Compromises are intended behaviour. A lot of coding is cost-benefit analysis, and at present, the CBA showed that this is the best way to go. Thus, intended behaviour. Whether it's ideal behaviour or not is a different debate. Oh and an addendum. This decision is why our ZAS is no longer lagging to fuck or taking 5 minutes to process a single broken window.

There are two solutions to this issue. One is a major rewrite of ZAS to run off of delta updates. Something that myself and Lohikar discussed at lengths and in a good amount of detail. Though the theory remains largely untested: I think Lohikar did a small prototype but he didn't get far beyond that. A technically apt coder would have to code this, though. And it's not a small undertaking.

The second solution is to do a technically worse compromise, and accept the possibility of megazones inside the station. Basically: make it so that large zones outside still utilize unsimulated turfs, but floor tiles, when removed, create simulated rock. The likelihood of a large enough to be troublesome megazone being created with this method is low: you'd have to destroy the entire station, but it's there. Further, if you want a shit compromise, then this is it from a technical side: it's not-uniform behaviour and technically a risk to server performance. Also, untested.

[BurgerBB]

There is always a fix. The solution would be to implement it so that simulated tiles are replaced by simulated asteroid turfs and non-simulated tiles are replaced by non-simulated asteroid turfs when a tile is destroyed.

Edited December 12, 2019 by BurgerBB

[BurgerBB]

Bumping this. It's off the front page and I really don't want this to be ignored. It's such a meme that this happens. An entire asteroid turf can vent a room in seconds despite this not being realistic behavior.

 

Make it so that destroyed simulated turfs are replaced by asteroid simulated turfs. Literally this is all you need to do. I made a PR for this ages ago but it went ignored after a dev said "It will be debated" but a month later I got 0 word or anything about it.

 

 

[Wigglesworth Jones]

@NewOriginalSchwann can attest to asteroid turfs killing a large amount of people (7 or so) due to an explosion destroying tiles. 

[NewOriginalSchwann]

2 minutes ago, Wigglesworth Jones said:

@NewOriginalSchwann can attest to asteroid turfs killing a large amount of people (7 or so) due to an explosion destroying tiles. 

I can. We had a round where a wizard fireballed one (1) tile, vented the medical lobby, and roughly six people died in the resulting venting/explosion (including the wizard).

The ultimate death list was:
Petra Volvalaad Head of Security
Kylee Steele Chief Engineer
Johan Ringer Warden
Marianne Stymes Security Officer
Veronika Kazlaukes Security Officer
Skkukoas Ukazozu Medical Doctor
Fatima Tadhhab Roboticist
Ka'Akaix'Voa C'thur Scientist
Daisy (Wizard)

The pressure damage was so extreme that none of us were able to get our internals on or perform any first aid before being stuck in a cycle of faint from pain -> wake up for five seconds -> faint from pain-> wake up for five seconds -> faint from pain until after about six or seven goes of that we were unable to do anything, and all died. Engineering eventually managed to patch the hole, but it was too late for any of us at that point.

[BurgerBB]

Despite that being a fucking meme, for some reason we have players and developers justifying that as intended behavior and balanced/fun gameplay.

 

It's currently the stupidest thing on the server right now and honestly I question whether or not these players have actually ever dealt with a breach like this with the new brainmed.

[Wigglesworth Jones]

It is a huge meme, yes.

 

I think making tiles harder to break will at least serve as a bandaid until a more proper solution is made.

[Faris]

I'm looking into alternatives. Namely that it stands to reason that density of walls from an x-direction should be similar to that of it from a y-direction.

Right now walls have a explosive resistance of 10 while floors have it as 1. I'm going to tweak with the floor value to find an acceptable one, or perhaps find something else that can do it. It won't nerf the venting but it'll cut down on how often it occurs.