Recent TEG Changes

[Kamades]

So, lets talk about the recent TEG changes.

I'm a engineer nerd, who mostly plays RPO, with a bit of RP1.  The standard maps are... we play every map on RPO, really, and so we see a lot of engines.

The recent engine changes, I am not 100% certain what they were supposed to do, but they have mostly left me frustrated, baffled, stuck in the engine compartment for 2 hours, and having to rely on the solars just to keep the lights on.

Hellburns are... strongly discouraged on RP, because, well, they kill everyone, and folks that can do a 2 hour hellburn without killing the station (on non-horizon maps) are rightly celebrated.  

The recent changes (mostly circulators) have left me... extremely frustrated.  I've spoken with Azrun fairly extensively, and reasonably understand the reasons for things changing.  

I Have spent 10-20 hours of late trying to make basic charburns work.  The circulator seems to brick in most configurations I've tried.  Hellburns work perfectly fine, but any low heat, medium pressure systems shit the bed on most maps (Horizon works, mostly, since there's an aux both before and after the engine).  I've had engines that flux constantly between 40kw and 400kw (1 flick per second), engine setups that should work just straight not making any power, and lots of time fiddling with the APC on rounds where there are no AI's and I've forgotten to set the solars to make it work.

I really, really love engineering.  But these changes have felt not fully tested, and not tested on the RP map listing.  I guess we're supposed to be the testers, but we're receving no documentation.  I've got a list of engine serial numbers, and no idea if they are supposed to indicate anything to me.  I guess we're supposed to put new kinds of fluids in, but trying to source most chems as engineering is.. tricky, when you're having to spent so much time just trying to keep the lights on.

I don't want them reverted, but I would like some testing, or some documentation.  Right now, it's been feeling like a black box, that works or does not work for reasons that do not align with my understandings of how gas pressures work and the flow of the pipes should work.  I guess what I'm asking here - how has everyone else found the recent changes to the TEG?

[Retrino]

as emily said, we are only in the initial stages of transitioning the teg. there are many things to be worked out, tested, and experimented with, as any new system needs. relearning engine behavior is definitely going to take work.

however, like you explained, having little to no information regarding how exactly these new changes are meant to work is extremely jarring. people who have been working with the engine (and have continued to) are struggling just to keep the station powered. honestly, i don’t like the idea of having A) such an unstable environment officially in the code, or B) if it is stable, not having much to go off of in terms of keeping it stable.

the behavior of the engine seems to be incomprehensible these past couple days. a set up that worked three days before would produce no power, a set up from months back would run perfectly. i’ve mentioned it already, but i’ve seen the hotloop not register any heat or pressure within it at all.

i agree with you in that they shouldn’t be reverted. i would hope, though, that any future patches that could cause such instability would be held off from merging until more personal testing is done. that, as well as more communication from azrun/the devs that had tested it before implementation.

really, i hate to see people giving up engineering entirely over this. having more understanding would benefit everyone involved, in my opinion

[Leeanei]

I think the biggest issue with the current TEG is the lack of a core, consistent burn outside of hellburns means teaching new engineering players is going to be rough. Real rough. If every round is experimentation, then the wiki isn't going to have a good consistent burn guide and mentors are going to be way less helpful beyond stick plasma in and ??.

[Kak0phony]

I think the TEG changes are heading in the right direction. The circulators were recently buffed so the cold loop shouldn't get bricked as hard when doing char burns, although I've only set up the cogmap1 TEG since the latest patch.

[Leeanei]

cog1 is the map the changes are being tested on, RP has been running into issues on all the rest.

[nefarious6th]

I've run into a lot of issues with doing basic charburns too; once I got a weird char-pipeburn setup to roll okay, but that was a few weeks ago and a "let's throw whatever at it and see what happens" build with no actual intention or thought behind it.

Training new players has been frustrating if not pointless. I was intrigued by the circulators stuff but still see a lot of cold loop bricking on charburn setups, even though it sounds like that was maybe the intent of it to fix? I've also seen what Retrino talks about, with the zero input hotloop. I have only gotten usable burns on Cog1 and Cog2, and my Cog2 burn ran away on what should've been a pretty mild charburn setup.

I'll have to come back to this post, since I forgot something and I'm not sure I hit it on the head in the next part, but there's three things that have really bothered me out of this entire process.

One is that I can't reasonably teach anyone a useful engine setup, which really is sad because engineering is one of my favorite departments.

The second, and perhaps most profound effect, is that currently hellburns are the only really viable engine setup for consistent power; which leads to people using hellburns. And unfortunately, I have seen so, so many of these on RP1 in the past few weeks, and every single time the engine has blown up, and all but once we've had to evacuate on Cog1 because it destroyed cargo (and robotics, and made the escape hallway unusable). It's just frustrating that people think hellburns are the only worthwhile option now, because on RP rounds, keeping that kind of burn stable just isn't going to happen, and that usually means we sacrifice 30 or more minutes that we could've had at the end of the round because things just get too hot to feasibly recover from, so we leave early.

I think my other problem I was trying to remember was that I've had to sit on my engines the entire round when I do them. I get that engineering is a team/collaborative thing, but I think part of the problem is the nature of engineering seems so spread-out to me (construction, mining, cargo, actual engine), that it's difficult to get more than one or two people to sit on an engine all round when there's other things (breaches, cargo, nanites, motives) that need taking care of. The only rounds where sitting with the engine has been okay for me were when I was an engineering borg and planned to not do anything remotely useful beyond getting the engine running. No motives to check in on, no further obligations, none of it. And that was fine, because in my experience with these changes, it is a matter of just a few BYOND ticks between having a decent 1mW charburn going on a setup I was familiar with, and one that jumps to 10mW and destroys inner engineering or bricks a loop completely.

I have one other question, that I unfortunately didn't get to test myself because when I did go in to do some engineering practice at 60 minutes on a Cog1 shift, the thing already melted down: I assume the lubricants are just added into the circulators from beakers or whatever. I don't really have a problem with this conceptually on a majority of the maps, departmental interactions aside, because on a map like Cog1, an extra beaker of cryo spawns in in Medbay, so I figure it's easy enough to swipe that (although not every map has spares of these, and I would definitely look into spawning at least 1 spare beaker somewhere on the map for engineering).

But if beaker/container into the circulator is the case, this is a HUGE disadvantage for any engineering cyborgs since they don't spawn with reagent containers. And, unfortunately, with the way things have been so far, I really can't see engineering not having a cyborg on it for TEG setups, because so few people can set up the engine/will want to join as engineers with constant experimental builds, and once an engine goes haywire, it can become impossible for any non-silicon players to get back in and fix it without dying.

I can try to poke around at more TEG setups and get more solid numbers or whatever; this is mostly my thoughts from the past few weeks, though if I think of anything else, I might come back and add some other ideas/results. I just feel lost on all this.

[Azrun]

I am super bummed that this has been leaving people frustrated.  I’m sorry if people have been feeling pushed into avoiding engineers.  Sorry to the engineering teams that got stuck with a lemon of a TEG earlier this week.  Sorry to mentor/admin staff for any influx of questions/confusion/rage.  

Intent
When I started playing engineer everyone was doing the same thing every time.  The only variance was how long an argument about pump settings went on for (and if led to some act of aggression).  Then either people stuck around to watch the numbers for brrrrr or they went off to never return.  I wanted to give the engineers something to fiddle with and engage with IFF they want to.  This does mean some of it going to be black box but you get to figure out how it works, if you engage with it.

---

My Reality
Changes at the start of this month were just to resolve what I thought was a bug.  It didn’t matter which side inlet/outlet has more pressure just that there WAS a differential between the two and then it was going to pull some factor of that from the inlet and pass it to the outlet.  It made the cold loop work perfectly and forever.

To do that I added a fan/blower to the simulation to pressurize the gas on the inlet.  Now if there wasn’t enough pressure to overcome the outlet you would get at least a LITTLE bit. These changes went in Nov 1st.  The cost in energy for running the blower was set to science fantasy low to not punish people having to relearn the mechanics, or at least that was the original intention.

When I first heard of issues maintaining power on RP servers I checked the maps and was able to get the following powers with 2 furnaces:

• Atlas 1MW
  
• Clarion 360kW
  
• Destiny  300kW
  
• Horizon 500kW (layout of this map confused me and was achieve 989kW)
  
• Cog1 - Very inconsistent as pressure was equalizing too quickly
  
• Cog2 600kW
  

These times were after 15-30minutes, with some tweaks I was able to get even more power from them.  Just because I got those outputs it is an unfair expectation that someone working through this on their own, without having stared at the code, be able to do the same without a decent amount of trial and error which turns into a LOT of real world time.  Temps are going to slowly drop from furnaces and thus the power is going to dip as well over time. Once pressure equalizes you are kind of stuck having to pull gas from the system and then feed it back in or have temps rise somehow.  It sounded like this was generating too much of a spinning plates exercise on RP as you had to ensure pressure didn't end up equalizing.   So RP plate spinning, poor results on Cog 1, and lower barrier to entry  were why I boosted the minimum amount of pressure the circulators would generate.  This was then costed by having the circulator use more power to generate pressure (moving from what I thought was more science fantasy and more into science fiction with an ideal fan).  Sweet I get the costing I want and it should make things easier.  That change went in on the 13th.

After staring at my commit and learning that I can’t even spell “variant” correctly my most recent changes went in.  This was to provide the groundwork for future TEGs!  It was all coming together! You could put fluids in the circulators! You can [REDACTED] and repair the circulators!  There are TEG’s that favor the [REDACTED], just like real life TEGs!  For about 17sh hours there was a bug, with aforementioned TEGs, that could cause the round to start with a non-functional TEG, fortunately Kamades reached out to me after beating his head against a brick wall for an entire RP round.  This got fixed shortly thereafter and now there are some TEGs with some minor power-output tweaks.

---

Mechanics:

• You want to maintain a loop of gas flow on either end.  Gas is going to flow from high pressure to low pressure so depending on the map you may have to do something to adjust that flow by either heating it up.  Depending on the map the cold loop is going to be an issue because cooling the gas compresses making it essentially go backwards (but thankfully there are valves that stop that).
  • If AUX ports are available on the outlet side of the TEG you can improve the pressure differential but filling up a gas canister there.
    
  • Freezers can be used to heat up the gas a bit to improve pressure differential.
    
  • The fan/blower should keep you covered.
    
• Circulator isn’t moving!
  • Ensure the APC has power and someone isn’t ruining your day.
    
  • If circulators are running but the TEG isn’t generator power they will drain the cell in the APC.
    
  • Ensure SMES has sufficient power output to charge APC so this doesn’t happen in the first place.  If the pressure differential continues to rise so too will the output of the SMES.
    
• Lube?
  • Use fluids with high viscosity!  Talk to a chemist near you! Bring them a fuel tank or something so you don’t use their precious resources.
    
  • Lubricant dictates how well the circulator functions which then is a +/- to that loops contribution to the TEG.
    
  • Char burns are the best time to play with lube! You should be able to achieve a steady power output and see the impact on adding the substance has on the system.
    
• But I don’t want to screw it up?
  • The TEG is shipped with sufficient lubricant to be equivalent to pre-patch.  That is your baseline.
    
  • Wrench the circulator with the maintenance panel open to open the drain valve.  Be sure to close it when you are done.
    

In practice most stations don’t pull much power, and the excess is just gravy.  Excess power feeds PTL (which is currently costed for Hellburns) or the nightmare of a hotwired station.  Some stations only have one SMES for the TEG so that is even less power as you only have one battery to fill.  Setting up the solars means you should have the freedom to bugger up the TEG ... and worst case vent the gas and start over.  Yell at miners to get you more rocks as needed and use that gas extractor!

Ultimately if you can’t get any power out of the engine there is a problem.  @ me in questions on discord or send me a DM, and if I’m available I’ll try to give some guidance.  I would much rather people have their own “Aha!” moment with some helpful hints or letting you know what is possible, but if you are lost and/or frustrated let me know and I’ll try to point out the problem... or I'll see about implementing a fix.

This is our goonstation to share, so lets keep it awesome.  Thanks for the feedback and please keep it coming.


most of my testing has been on Atlas and Cog2 as those two are the most convenient maps for silly reasons

[Kamades]

[*]Use fluids with high viscosity!  Talk to a chemist near you! Bring them a fuel tank or something so you don’t use their precious resources.

How is viscosity defined\measured in game? I've not seen a way to get a readout of that statistic, but I'm not the biggest sci nerd out there.

As well, if you want to replace the circ lubricant, you're going to need *400* units. Of things that are, presumably, not basic chem dispenser liquids. Experimentation on this system from the Engineering side is going to require quite a bit of assistance from Science, to the tune of perhaps as long as a quarter of their shift if they are making unfamiliar reagents? As well, there is zero indication to me from the serial numbers as to what is different from one to the other - does it change what lubricant it wants? What it's ideal gasmix is? The best temperature ranges? "CIRC-FX95922L" indicates nothing of use to me, just that there is now a serial number.

When I first heard of issues maintaining power on RP servers I checked the maps and was able to get the following powers with 2 furnaces:
SNIP
These times were after 15-30minutes, with some tweaks I was able to get even more power from them.
SNIP
Temps are going to slowly drop from furnaces and thus the power is going to dip as well over time.

For myself at least, and I think a few others, the goal and pride of engineering is keeping consistent, SAFE power for 2 hours. Hell, sometimes 3-6 hours for the (admittedly rare) shifts where that happens. On every station - including the ones with sorta crap layouts (Looking at you, Destiny\Clarion\Atlas!). What I've seen with recent changes lines up with nefarious6th - a lot more hell burns, a lot more exploded engines, and several shuttle calls due to these problems, cutting the entire station's round short.

On a more personal note - I am also uncertain as to if you wish to class these engineering changes as Secret or not. If they are secret, or some subsections are, that's going to make the engine a lot harder to teach through the Mentor system. If they are not secret, a design manifesto available for people to read if they wish to, so they can understand what, exactly, has changed and how to correct the issues might be really useful.

[Azrun]

There isn’t currently a way to my knowledge that surfaces the specific viscosity of a fluid.  The most observable way currently is in how a spilled liquid spreads out when poured/spilled.  Having a chems have a rheometer and/or engineering a viscometer might be interesting but it seems like something that would only be used until everything gets cataloged.

Empty lubrication system is going to reduce effectiveness of the gas transfer the most -40%, putting terrible solids in is going to reduce it to -30%, and bad fluids -10% but that is all worst case.  Otherwise things can provide up to a boost of 10% and with some specific items providing more.  The amount of benefit is proportional to the ratio of fluids in the system. 20 units of space lube is going to be the same as 400, assuming that is the only thing in there.  Once you start mixing things it is then going to be proportional, which makes it a more difficult problem to solve.  I thought I had added a low lubricant penalty but that will have to be a future PR with some kind of check oil light…

Circulators start half full, with a max of 400 units.  This is really to cover future cases where the lubricant can leak so you can walk away and mostly ignore the problem. I liked the idea of someone coming back to the engine room wondering why the TEG bled out and there was slippery/flammable/toxic/caustic fluids everywhere.

All lubricants function the same across the board.  The only exceptions I have planned are some TEG’s that are intended for event/admeme/pray/antag/shenanigans.  And those are intended to be quite apparent in regards to what they do/don’t want.

Haven’t touched anything gas mix related…. Yet.  Send all your hellburning love/hate to https://forum.ss13.co/showthread.php?tid=15447

Best temperature ranges are intended to be left to the playerbase to engage with if they want to or they can spoil themselves looking at the code.  Otherwise I would expect them to see a +/- 10-15% power output.  There is only a chance you will see one of these per round and the behavior is going to be consistant across circulators and TEG.

Right now the serial numbers don’t mean a whole lot.  They mean as much as the serial number on your television.  This is intended to condition people to look at them and see if there are any patterns. This way if something doesn’t match the pattern it becomes apparent that this TEG has some quirk to be investigated.  If this is seen as too obtuse I can try to surface some information that is available to the CE at round start somehow.  They are not being procedurally generated such that every TEG is ever so slightly different making it impossible to figure what is what without doing the same burn for the rest of your life, the thought did cross my mind though.

I don’t understand the “consistent, SAFE power for 2 hours”.  After about 40min on Horizon the two SMES are full running off two furnaces.  Run into the engine room turn off the furnaces, card the APC, and turn off the equipment so the blowers don’t run.  If you need more power turn it all back on again.  Based on what I've read in Discord some people aren't really getting to any stable power so that is the thing I'm curious about.

It isn’t in “secrets” so as far as I know I have no power to say don’t share the info.  I want people to discover things on their own or with a fellow engineer but more importantly I want people to have fun.

For most of the month when I have gotten a chance to play it has been engineer on Goon and the round typically one of a few ways:

1. I’m the only one in engineering so I make a stupid hot chamber burn by myself choosing not to put on space suit and making sure the pipes never burst… they typically do.  But at that point I can typically pull the plug on everything so the nightmare is mostly contained.
   
2. I am asking people why they do the things they do as they set up for a Hellburn.. and then trying to repair the engine room if the engineering team hasn’t gone full toxic when things go south.
   
3. Be the guy the CE is screaming for not wearing said space suit telling me I AM GOING TO DIE while I help get the Hellburn setup.  Then run off with building materials to help repair the station.
   
4. It's a non-TEG map and I get to play!  Typically people refuse to let me feed the singularity and it is a sad day.
   

Very rarely have I gotten a chance to teach anyone how to set up a burn because either everyone leaves engineering OR I get halfway through explaining how to do a char burn before the CE has spaced the engine room and we need to go get internals on and the lesson is pretty much over.

[Studenterhue]

How is viscosity defined\measured in game? I've not seen a way to get a readout of that statistic, but I'm not the biggest sci nerd out there.

Viscosity is a property of certain chems and like Azrun said, it affects how fluids spread, how "elecflashes", like those from tesla coils, conduct through fluids and how much the fluid will slow you down. Viscosity tends to follow real-life trends somewhat e.g. milk is more viscous than water, honey is more viscous than milk, and chocolate is more viscous than honey. But some chemicals have viscosity based on how powerful/sucky they'd be as a fluid. So partially-hydrogenated space soybean oil, ipecac syrup, and VHFCS (soda engine!!) are on the end high of viscosity scale but so are LDM and napalm.

[nefarious6th]

To be clear about the hellburn thoughts: I'm not saying anything you did has changed anything about hellburns and how they work other than the frequency. My problem is people can't get charburn setups working correctly anymore and instead opt to hellburn or do combo burns.



EDIT: I have two thoughts. I had to rewrite this all because I lost my first version of it, so. I might have to come back and add-in where stuff is missing. Sorry for any flux.

1.) I forgot. Again. Consider this my placeholder text.

2.) I feel like the constant advice here and in the changelog to "just experiment" is very reminiscent of the kind of advice that I'd be given on chem and golden gas ratio stuff; which, as far as I know, both those things ARE considered secret. So hearing it here just seems to like ? unneccessarily muddle a process that shouldn't be inaccessible for people, since from what I'm understanding, no part of this is actually meant to be a secret. And then trying to teach from that is just even more difficult. I think I also have to agree with Kamades's thoughts on the serial numbers; to be a little more precise, if there is a link between X component and Y part of the serial number, I think it just needs to be stated versus a "learn it on the fly thing". I don't mean to speak for you, or your intent here, or anyone else, but I really, really think that the serial numbers need to be ? translational ? if that makes sense. In that, I think the serial numbers should be used to inform the setup, not have the setup inform the meaning of the serial. It should be about translation, not decryption, is what I guess I expect. And then I'd hope that would result in more feedback like what I think these PRs meant to evoke, where you have people come in and say "I tried X and Y and Z on this engine and got 100ZW; I wonder what would happen to output if I used lubricant Q instead of X?". I guess where I think I see my expectations not lining up with the changes so far is that I imagine engineering should be as hard as having that be a two-variable sort of equation, or three, or what have you, with something known or knowable from guidance and documentation (like a translational meaning of something in the serial number to, say, a documented bit on lubricants), but what we're being given thus far feels like everything is just totally stochastic. Nothing in that set can be known before setting up each unique engine and seeing where it goes. I know this is all very early still too, but I'd just want something more concrete to operate on because it feels like trying to hold smoke.

I think the other thing I wanted to add was tangential, but important to me: What does this spell for engineering cyborgs for engine set-up and tending? Will we need to put one or more containers on their modules to they can either carry or mix lubricants that are seeming more and more essential for a successful engine, with the 40% penalty on empty currently?

[Azrun]

What has increased the hellburn frequency?  Easier and more understood at the cost of killing everyone or making an uninhabitable department?

What is making charburn setups so difficult?  What is your expectation of the setup?  My expectation is that people should be able to power the station and fully charge the SMES with some periodic interaction post setup given enough time (without the use of solars).  I would expect some action by engineering as thematically and mechanically you are providing a resource the entire station relies on. Char burns have little to no risk but consume a resource and solars also have little to no risk.  I think the frequency of needed interaction is something worth discussing.  I think understanding all of your expectations might help bridge the gap.

Since engineering borgs won’t have a reagent container...  they currently don’t have a path to engage with a sabotaged circulator to mitigate the 40% power decrease. I'd rather they had to seek out a silly carbon based life form to resolve the issue in general.  As per our side discussion an Oil Can or something similar might be a good alternative as long whatever it was wasn’t compatible with the reagent dispensers.  Seems like an interesting thematically and mechanically just not sure of the balance ramifications.  

Your points regarding the serial numbers ring true to me.  There is a lot there that honestly means nothing at the moment, and that was by design.  I'm a little torn about surfacing the specifics like on the chemistry wiki, but maybe a little footnote on the setup instructions to point you in the right direction sounds interesting. “*NOTE: This is an experimental TEG where temperatures are not to exceed 1000K.”  Then I could bury a list of different TEGs in a letter/email/etc that is findable for nice fluff.  Looking back over the current implementation there IS a lot of noise so I think I’m going to tone that down a bit so there aren’t so many moving pieces and new models won’t possibly re-use previously cataloged identifiers.

Is it only successful if it is the most optimal setup? Are lubricants already the new thing to argue over like Hellburn setups?  Were there no successful setups before circulators accepted reagents?

[pali6]

Disclaimer: I have not read all of the previous posts here, sorry. Crossposting my thoughts from Discord:

Imo there are two main groups of players when it comes to engineers (well, probably a continuous spectrum but let's focus on the extremes). The first group wants to tinker with the engine and figure out cool engine setups and stuff. The TEG changes give them many tools and options to do that which is great! The second group is people who just want to start the engine so the station is powered. The TEG changes just add unnecessary obstructions for these players for the most part.

As a consequence I think that a basich low power (char) burn should be only lightly affected by RNG and should not require chems. For more powerful and more complex TEG setups I think both the inclusion of new mechanics and RNG are good. I'm not sure how to achieve this scale properly because I barely looked at the code in the PRs but I feel like the general design should go somewhere this way. Have a realiable low power inefficient setup and then have a wide space of more complex more effiicient more powerful setups with RNG constraining you to some part of that space and you exploring that part.

[Studenterhue]

So small update, this PR has been merged, so now TEGs get a different sprite to indicate if they're a variant. I know it's likely not gonna solve all the issues here, but hopefully it should clear up a little confusion, and that was all the PR was set out to do.

Besides that, Azrun also has been trying to address some of the feedback in this other PR that's ready to go, if the merge conflict clears up. Serial numbers are simpler, there's more in-game documentation of the new mechanics, and the blower for the circulators can now be manually adjusted, among other things.

[Azrun]

The changes are in and the Wiki has been updated.