Can we speed up Linda?

[Spacemanspark]

 

What even happened to Yinadale anyways.

Anywhom, I do prefer ZAS, mostly because of how hilarious it is.

I doubt it's coming back anytime soon, though...

 

[davidchan]

 

What even happened to Yinadale anyways.

Anywhom, I do prefer ZAS, mostly because of how hilarious it is.

I doubt it's coming back anytime soon, though...

 

I prefer ZAS because it at least made Atmosia relevant. Now, by the time atmos is aware of a problem, it's either too late to deal with it or just not deadly enough for the crew to care. Plus I miss the days of traitor engineers flipping on their magboots and over pressurizing unoccupied areas of the ship and opening the door anytime security ran at them only to get buffeted back into a wall.

 

[Kevak]

Purrhaps we could swap in Zas for a week and take a poll?

[tigercat2000]

Kevak, not that simple. We can't just swap back and forth between ZAS and LINDA; We would have to undo months of work to enable ZAS again, which isn't an accurate representation of how it would run

[Kevak]

 

Ah.

 

Would it be plausible to alter ZAS so that it can function with the atmos piping system we have? Prolly would take a lot of work though.

 

[Wintermote]

 

Ah.

 

Would it be plausible to alter ZAS so that it can function with the atmos piping system we have? Prolly would take a lot of work though.

 

ZAS works with the current piping system. Actually, the existing atmos system, including emergency airlocks and airlock doors is entirely designed around ZAS. Changing back to ZAS likely requires editing/replacing a bunch of atmos files spread all over the place, editing everything to point to and access everything else properly and then doing a lot of doublechecking to make sure it's actually working.

 

Editing the map or adding new code shouldn't be necessary at all.

 

It's still likely a pain in the ass to get it sorted.

 

[Kevak]

Purrhaps if someone ever gets around to it, we could have it saved as an alt code?

[Wintermote]

 

I prefer ZAS because it at least made Atmosia relevant. Now, by the time atmos is aware of a problem, it's either too late to deal with it or just not deadly enough for the crew to care. Plus I miss the days of traitor engineers flipping on their magboots and over pressurizing unoccupied areas of the ship and opening the door anytime security ran at them only to get buffeted back into a wall.

 

One of the things I immediately noticed, is that with LINDA I have no reason to carry the inflatable walls anymore.

 

They used to be critical for repairing damaged areas and holes.

 

Purrhaps if someone ever gets around to it, we could have it saved as an alt code?

ZAS was actually already implemented on paradise. It was removed a few months ago I'm assuming. A rollback wouldn't be that difficult, but it means losing everything added since then, and then re-adding them.

 

Or alternatively, stripping out LINDA in the current code and replacing with ZAS and doing all the busywork to ensure it functions correctly.

 

The actual difficulty involved isn't something I know because I don't know what was done to take ZAS out and put LINDA in to begin with.

 

[GelatinousGlob]

 

Linda physics got int he way on mining the other day. First we had three miners try to pile out an airlock and got stuck by their ore boxes, but they had tools so its cool and they cut the wall out. Unfortunately, being pushed makes you lose hold of the ore box, so while two people could get out, they couldn't grab the ore boxes which also would let the other miner free. Spent a good 10 minutes trying to get the guy out from the ore boxes that trapped him. Then 5 minutes later, there was plasma walls with goodies inside. There was ad resser in the middle of the room with an item behind it so it enter the walls had to be removed, unfortunatly the torch causes the plasma walls to burn/disintegrate into free plasma. Now the server spent the next hour processing this plasma pushing out and around the mining tunnels, pushing players so they drop their ore boxes, pushing them around so monsters can kill them, causing superheated 'air', all this with like 1 kPa of plasma pushing around.

 

It still hadn't dissipated (In fucking space!) after the entire round ended.

 

No matter how much work was put into LINDA, I think it fails to do any good and whatever work it takes should be put into removing it and reinstalling or improving ZAS or a similar system. The only way LINDA really works well is if its going WAY faster than it currently is, which will use just as much, if not more, processing power than ZAS took.

 

Never did fix that airlock wall though, every time someone tried to put girders they got pushed from the minuscule amount of air left from 30 minutes ago that kept flowing.

 

Most people don't even use airlocks anymore because of linda, they just force the airlocks and let it leak because if there is a vent anywhere near the airlock, it doesn't matter. it just causes problems 30 minutes later when someone doesn't notice the walked into this oddly over or underpressurized tile in a room that is otherwise fine.

 

[Kevak]

Might be good to switch back to ZAS, I have no idea how hard that would be though.

[Fox McCloud]

 

ZAS was actually already implemented on paradise. It was removed a few months ago I'm assuming. A rollback wouldn't be that difficult, but it means losing everything added since then, and then re-adding them.

 

 

Easily said when you're not the one who did all the coding for it.

 

It took weeks to get get LINDA up and running and PR ready, then a good month or two to get all the bugs and oddities about it sorted out and up to the same standard it was on TG.

 

"It wouldn't be that difficult"? You're literally talking about ripping out an entire system that took several months of troubleshooting to get where it is today--not to mention that the current scrubbers, air alarms, and fire alarms are also currently updated to TG's standards as well--so you'd have to rip those out and re-do those too. Oh, did I mention that gas datums, fire protection, mob temperature handling, being on fire, etc are all tied in with LINDA as well?

 

There's also game mechanics implications of it, as well---fighting fires will return to being impossible, chemistry's hotspots will function poorly, and flamethrowers will also cease to function at all.

 

It's not a simple matter, at all--it would require just as much work that went into it to revert it all and place ZAS back in---and it's still going to end up being more expensive than LINDA currently is, computationally.

 

I'm not saying LINDA is perfect--it's not--no atmos system in all of SS13 is---and I"m not saying we CAN'T consider making LINDA faster, in the future, let me make that clear.

 

[Wintermote]

 

It took weeks to get get LINDA up and running and PR ready, then a good month or two to get all the bugs and oddities about it sorted out and up to the same standard it was on TG.

 

"It wouldn't be that difficult"? You're literally talking about ripping out an entire system that took several months of troubleshooting to get where it is today--not to mention that the current scrubbers, air alarms, and fire alarms are also currently updated to TG's standards as well--so you'd have to rip those out and re-do those too. Oh, did I mention that gas datums, fire protection, mob temperature handling, being on fire, etc are all tied in with LINDA as well?

 

There's also game mechanics implications of it, as well---fighting fires will return to being impossible, chemistry's hotspots will function poorly, and flamethrowers will also cease to function at all.

 

It's not a simple matter, at all--it would require just as much work that went into it to revert it all and place ZAS back in---and it's still going to end up being more expensive than LINDA currently is, computationally.

 

I'm not saying LINDA is perfect--it's not--no atmos system in all of SS13 is---and I"m not saying we CAN'T consider making LINDA faster, in the future, let me make that clear.

 

For starters, I said rolling back would be easy (and it would be), and then in the very next paragraph explained it would be more difficult getting everything else re-added, or to take out LINDA and replace with ZAS. It's like half your post is just violently agreeing with me.

 

Second - flamethrowers worked with ZAS, I don't know if they work better or not in LINDA, but I do remember FEA flamethrowers being downright awful compared to the ZAS version.

 

The rest of the issues you named can be fixed and replaced with a modicum of effort, and should be replaced back to ZAS. It has always worked better and been more fun than FEA. The move to LINDA is just mindboggling to me considering what little it adds and how much it takes away from the game.

 

I mean, it's like you broke a window, replaced it with cellophane and duct tape, and are now complaining that putting a real window back in is going to take some time and effort. I know, I'm totally surprised too - who could've seen that coming?

 

But it doesn't matter - No code gets a revert unless a coder wants it so, regardless of how the playerbase feels about it, and when public opinion turns against you, then you go all passive aggressive and make excuses and bullshit instead of just outright telling people what you really mean to say: "Fuck you, we don't give a damn what you think of our changes." Sense, logic, reason and even public support one way or another has never mattered. What's most important are personal egos. Fucking hell even the server host doesn't like the change, but coders act like arrogant assholes and boss him around like he isn't even in charge of the place.

 

[Spacemanspark]

I don't think we need to be hostile about it.

[NTSAM]

 

I don't think we need to be hostile about it.

Very much agreed.

 

[TullyBBurnalot]

 

*grabs defusion kit*

 

Let's keep this civil, shall we? There's no need to start flinging accusations around. Keep the discussion to the topic at hand.

 

[GelatinousGlob]

 

RABBLE RABBLE RABBLE!

 

Don't need to get all crazy about this. That said, its better to revert now than months from now with even more crap added on making it harder. It sucks when shit doesn't work out as well as planned but it happens.

 

[Fox McCloud]

 

RABBLE RABBLE RABBLE!

 

Don't need to get all crazy about this. That said, its better to revert now than months from now with even more crap added on making it harder. It sucks when shit doesn't work out as well as planned but it happens.

 

 

It's already been months and months.

 

LINDA, for example, was slotted in in PR #895; we're now in the #2100 range--that's a LOT of changes since LINDA got put in (with tons of changes built up around LINDA, too boot). This isn't a specific object like a type of flashlight that you can just rip out and easily place something else back in---it's a system that is the culmination of many many months of internal and external discussion, then many many more months of working on it and refining it for Paradise.

 

There's also a pipenet refactor in the...err...pipe...line....coming up that also comes from TG and also is in line with LINDA. The amount of interlocking parts that have been built up around LINDA is incredibly high; it would be tantamount to reverting our chemistry system (if not worse).

 

Again, let me make it clear. I'm not opposed to making LINDA run faster---we're just not at a point where we can do so yet, for performance reasons.

 

[Spacemanspark]

 

Would you be opposed to making it fast enough that it would be realistic to re add the ability to make people fall and fly around when the pressure in a room rapidly depressurizes?

That was the best part of ZAS.

 

[tigercat2000]

 

Would you be opposed to making it fast enough that it would be realistic to re add the ability to make people fall and fly around when the pressure in a room rapidly depressurizes?

That was the best part of ZAS.

 

LINDA will never do that. Not without turning it to process every tick, or something ridiculous like that, which is "crash the server" territory.

 

[Spacemanspark]

One can dream, at least.

[Kevak]

 

Is it plausible to hack together a system along the lines of "If a line can be drawn from a tile of above x pressure to a tile that has below x pressure (10 or lower?), objects and entities in the high pressure tiles fly towards the low pressure?

 

Might simulate depressurization.

 

[Shadeykins]

 

if activetiles <= x

{

kpapertile/activetiles = kpaav

if kpaav >= 20

{

kpapertile = 0.

}

}

 

Here's a quick and dirty C++ oriented check (Got no clue about Dreamweaver's nonsense) that would find the average KPA of tiles when there's a certain number of them active, and if the KPA is less than a certain value it would set the KPA to 0. Of course, integers aren't actually defined here (as this is just a what-if example). This method would require hooking into room definitions/dimensions (which are already set up for station blueprints), and adding some additional math or integers to calculate the dimension of the room... I'd suggest the tile # for changing to be 50%.

 

The only downside with this method is that it doesn't cause explosive depressurization, but would make a room depressurize immediately once a certain threshold is reached which would at least make depressurization a danger across a room rather than a couple tiles.

 

You'd of course have to also factor in to make sure it only ticks when tiles pressure is being reduced, otherwise you could never re-pressurize a room.

 

Seems pretty reasonable and doable to me if not a bit complicated... Personally I'm a bit confused why there's a LINDA/ZAS fight when it's perfectly reasonable, doable, and probably in the server's best interests to just rig up our own atmos system. Such a system would...

 

A) Be easier to bugcheck in

B) Suit the servers needs

C) Probably not be so damn hard to manage

D) Would not conflict with any existing content

 

Just my two cents~

 

[Tauka Usanake]

 

Seems pretty reasonable and doable to me if not a bit complicated... Personally I'm a bit confused why there's a LINDA/ZAS fight when it's perfectly reasonable, doable, and probably in the server's best interests to just rig up our own atmos system.

I think it's been stated that making such a system would be really tough and a few other projects, if not actual atmos revisions, are also in the works. I don't really like having LINDA as a system but it's something until we get back to ZAS level murder space

 

[SpaceTimeNow]

 

Do we really need to roll everything back, wouldn't it be easier to make LINDA calculate bigger areas as it would be one tile. Sort of like ZAS did it but maybe with some limitation, so it does not calculate an entire hallways as one.

 

Currently Linda seems terribly inefficient because gas has to literally travel from tile to tile which makes it so much slower resolving a atmos situation. Which probably is a performance issue in itself, Linda cannot in a resonable amount of time vent an area to space. Every frigging breach that happens will slow down the server and that will stack up over a entire round making the server slower and slower because it has to resolve all this atmos situations which it is dragging along.

 

And the crew no longer really cares about some of the remote breaches because they don't bother anyone. Which will increase the amount of unresolved breaches.

 

And the worst part of Linda is that impacting gameplay dramaticly.

 

[Fox McCloud]

 

Do we really need to roll everything back, wouldn't it be easier to make LINDA calculate bigger areas as it would be one tile. Sort of like ZAS did it but maybe with some limitation, so it does not calculate an entire hallways as one..

 

 

This isn't even remotely possible, not by a long-shot. It literally just plain doesn't function this way at all; it's designed entirely around the concept of being turf based, not based on an arbitrarily defined area.

 

I can't stress enough how literally impossible this is given how LINDA was designed, built, and how if functions.