DeltaV-Station
Revert New Xenoarch
About the PR
Reverts https://github.com/DeltaV-Station/Delta-v/pull/4470/, bringing xenoarch back to it's pre-rework state.
Why / Balance
The general reaction I've seen to new xenoarch is mixed at best, and I'm inclined to agree. On a technical level, it feels buggy, with some triggers being wonky to obtain alongside other triggers, some nodes being entirely impossible to trigger (without artifexium), some triggers being needlessly confusing in their wording and/or mechanical activation, and some deltav-specific mechanics still missing/not properly functional. On a conceptual level, the new system is extremely predictable and extremely safe, lacking the dangerous aspects of old xenoarch. There is a literal ten second timer you can use to escape the area before it does anything, and you can see exactly what the upcoming nodes do before you trigger them.
New xenoarch was a valiant attempt at improving the system, and it does succeed in some ways, namely the UI. However, I believe it didn't hit the mark.
Technical details
If we wanted to revert this cleanly, we should have done it before the upstream merge. Unfortunately, we're now post-upstream-merge, meaning I had to just let God take the wheel when resolving the merge conflicts. A lot of this code doesn't even exist on upstream anymore, and I get the impression the original new xenoarch PR didn't care too much about preserving deltav comments on imports and whatnot, so I'm just gonna leave it as-is for the sake of my own sanity.
I "tested" this in the sense that I confirmed it ran, and confirmed that the xenoarch interactions I could think of (artifexium, crushing, just normal xenoarch stuff) are still functional and don't crash the server. Something felt weird with the analyzer picking up new artifacts, but I couldn't reproduce anything so it's probably fine. CI/CD shall judge my soul. if anybody else wants to pull this PR and test it on their own, that would be awesome.
Notably, the UnlockNodeCommand has been nuked and I might've fucked up and nuked XenoArtifactUnlockNodeCommand too, but you can just right-click activate artifacts anyways so who care
Media
Requirements
- [X] I have tested all added content and changes.
- [X] I have added media to this PR or it does not require an ingame showcase.
Breaking changes
Changelog
:cl:
- remove: The new artifact system has been reverted. Get ready to print a lot of artifact reports again.
RSI Diff Bot; head commit 1ac105edb045291a2e8feafab565c1fdd635a8f0 merging into 143ddf812987d274f5921e98811fc768cace91dc This PR makes changes to 1 or more RSIs. Here is a summary of all changes:
Resources/Textures/Objects/Specific/Xenoarchaeology/xeno_artifacts.rsi
| State | Old | New | Status |
|---|---|---|---|
| artifact-activation |  |
Removed |
![github-actions[bot] avatar](https://avatars.githubusercontent.com/in/15368?v=4 "Published by a pub.dev verified publisher"){kind=small}
I dont play epi often nowadays but how about just hiding nodes you didnt trigger yet? Or otherwise improving on the new system instead of reverting to this dated system
I dont play epi often nowadays but how about just hiding nodes you didnt trigger yet? Or otherwise improving on the new system instead of reverting to this dated system
new system is fundamentally flawed, for a number of reasons (ex. multitrigger nodes and their 10-second delay) if you remove multitriggers (and by extension remove nodes generating in parallel (and by extension make only one node active at a time)) and hide nodes you haven't triggered yet, then you basically just have a worse version of old xenoarch (albeit with a UI)
also, unrelated: requesting direction review on this one ASAP, the longer it takes to get this merged, the more likely it is that it's gonna end up in merge conflict hell
requesting direction review on this one ASAP
direction review time, 24 hours, etc. you know the drill.
Direction Approved, However we will revisit Xenoarch with next upstream merge
Direction Approved, However we will revisit Xenoarch with next upstream merge
We... we will? >.>
Could we please provide more concrete details about the problems with new xenoarch? Revisiting xenoarch later will be much easier if we have a better grasp on the backlash. Currently we've only documented generic complaint ideas, with no concrete examples to help us improve the system.
In my view, the only difference here that warrants a revert is the fundamental difference between old and new xenoarch: old is about completing a mysterious tree by navigating using a single node, while new has a focus on activating interesting combinations of effects (durability system) with more complex triggers. This fundamental difference is where I believe we should focus discussion.
All the other problems you mentioned do not warrant a revert, in my view:
some trigger combinations are wonky ... some triggers are impossible
Can you give examples?
Looking through the triggers.yml list, I really don't see any combinations that are weird (except maybe cold and heat, which is still possible with frezon and a welding tool), and I definitely don't see any that are impossible. I dispute the notion that the system feels "buggy" - I think the community is still learning how it works.
some triggers have confusing wording or mechanics
Can you provide examples? This is an incredibly easy problem to fix. A couple triggers being misunderstood isn't a good reason to revert the whole system.
some delta-v specific mechanics are missing
We already have PRs fixing this. https://github.com/DeltaV-Station/Delta-v/pull/4486 and https://github.com/DeltaV-Station/Delta-v/pull/4705
There is a ten second timer that mitigates all danger
This is very easy to remove; I agree we should remove it so we can add back some of that danger from the old system; that would be an excellent improvement over upstream. I am out of town this weekend but can PR it on monday.
I imagine direction has discussed some of this. But documenting it concretly would help us succeed if we want to revisit this in the future.
I am personally on the fence right now; I appreciate the mystery of the old, but also the complexity of the new.
But I am mainly concerned that this revert feels premature.
In my experience in the short time this feature has been active, I've seen that players have not yet learned very basic facts about the system, for example:
- people haven't realized that the top nodes activate when you click it (I was set on fire three times one shift by colleagues that didn't know this - which was very funny actually)
- people haven't realized that you can't "go back" to lower nodes; you need to activate their full durability before moving on if you want the benefits of their effects
- people haven't realized that triggering multiple nodes will result in none of them activating (I think this is where the most confusion about the system comes from)
This shows that the community is still learning the new system. It does not show that the system is fundamentally unlearnable / bad. The feature needs more time for the community to spread knowledge about it.
I worry that we are judging the system before people have had a chance to explore what it has to offer. Frankly I think the new system has a lot of new roleplay potential, giving artis more use than just a research point battery.
how about hiding nodes you didn't trigger yet?
I think this could work if changed slightly: just hide the effects of new nodes, not their triggers. You need to know how to advance along the tree, but we could hide what effect will occur when you do advance, to add back some of that unpredictability (a positive element of the old system)!
I am personally on the fence right now; I appreciate the mystery of the old, but also the complexity of the new.
But I am mainly concerned that this revert feels premature.
In my experience in the short time this feature has been active, I've seen that players have not yet learned very basic facts about the system, for example:
* people haven't realized that the top nodes activate when you click it (I was set on fire three times one shift by colleagues that didn't know this - which was very funny actually) * people haven't realized that you can't "go back" to lower nodes; you need to activate their full durability before moving on if you want the benefits of their effects * people haven't realized that triggering multiple nodes will result in none of them activating (I think this is where the most confusion about the system comes from)This shows that the community is still learning the new system. It does not show that the system is fundamentally unlearnable / bad. The feature needs more time for the community to spread knowledge about it.
I worry that we are judging the system before people have had a chance to explore what it has to offer. Frankly I think the new system has a lot of new roleplay potential, giving artis more use than just a research point battery.
some specifics that I myself have both noticed from players and have experienced myself
-
The increase in information leads to a decrease in experimentation and danger, making proper preparation to deal with the consequences of an artifact much less impactful and necessary. The only things you really have to worry about are explosions due to well... it being an explosion. Everything else you have plenty of time to get out of the way of without needing to actually do anything.
-
Interfering nodes are frustrating and unfun to play around. This is most prevalent with the atmospheric nodes, but any node that activates under similar conditions will activate at the same time, and prevent the artifact from being completed without an excessive amount of setup. Oxygen-free environment, low temperature gas, low pressure, high pressure, any of these can potentially interfere with one another and can be incredibly difficult to set up to only meet one specific condition without a solid working knowledge of the atmos system, which is not something we ask of or require of scientist players.
-
Impossible nodes. Some nodes, due to how the stacking effects work, can be outright impossible to complete. It is possible, and in fact not entirely uncommon, to run into nodes on an artifact that have contradictory requirements that are impossible to meet within the timeframe allotted for activation. Something like requiring both high pressure and low pressure for example. These nodes can show up as early as layer two if you get unlucky with your initial nodes and can instantly dead-end an artifact. If this happens early on, or gods forbid on your first artifact, it could easily put a strangle on your early progress which can put an epistemics department behind by a good thirty minutes or more, which then leads to frustration for both epi players as well as other departments in round who are relying on them to research things quickly, such as microreactors, advanced mops, industrial ore processors, and other such items.
Thanks for your additions @AeraAuling , more details like that help us work on this.
The low pressure + high pressure combo is impossible to trigger, yeah. I think it would be a pretty straightforward Size/M PR to add a whitelist / blacklist system for node trigger combinations. I can do that on monday (I'll do it in upstream if this revert goes through)
I do agree the atmos triggers are the most difficult triggers for scientists. I think upstream maybe intended that to incentivize more interaction with engineers. We could easily turn down the probability that those triggers appear, to reduce the problem.
Regarding research output:
We could rebalance (increase) the arti research point values. If scientists feel they are not contributing as much due to nodes being harder to reach, increasing the reward value of the nodes is another way to increase research speed / make artis feel rewarding again. (an alternative to making nodes easier)
Interfering nodes are frustrating and unfun to play around.
Are there any other trigger types that often conflict / activate unintentionally? besides "low oxygen", "heat", "cold", "low pressure", "high pressure"? I don't see any that stand out to me in the list...
Personally, as a maintainer, I'd prefer modifying the current system over a giant reversion. A few things from my perspective:
-
Should this be merged, I will forgo any updates to XenoArch for obvious reasons. This makes it harder for me as a maintainer because now I still have to look at those PRs and decipher if they are implementing non-XenoArch components, and partially import them. Heavily modifying the current system is preferable to this to be honest.
-
There are no updates for the old XenoArch system. It is in its final form, unless we want to diverge and enhance it. But to be honest, I'd rather diverge from the current XenoArch system for compatibility reasons.
-
going to the new XenoArch system, then back to the old XenoArch system, I feel its only a matter of time before we just go back to the new XenoArch system as it becomes more feature-complete.
Are there any other trigger types that often conflict / activate unintentionally? besides "low oxygen", "heat", "cold", "low pressure", "high pressure"? I don't see any that stand out to me in the list...
There is oxygen free too messing with low pressure and others, unless you mean low oxygen about it.
Ok so... sorry if this sounds maldy... just trying to paint a picture of someone who does like the idea and system surrounding it and actually likes artifact research... just not it's current state with the rest.
Long Text
Whenever anything with the new system fails or doesn't work as whatever we're planning on executing, or getting a certain combination which are simply not possible roundstart, I feel like the times when I've had certain trigger nodes from the old system... namely Radiation, Plasma Gas, and Psionic Disturbance (In order from least to most annoyance).
The main problem here isn't so much that it's a bad system, on contrary I actually like it... just the amount of triggers which are simply not possible to do without relying and annoying other departments constantly (Atmos for gasses and Logi for a can NO2), has become way more in comparison to the three I mentioned earlier, especially now that you actually need to combined them in a way which makes it possible. All under the consideration that all of these triggers sometimes require the personell to have PPE which they simply don't have or usually get.
Even worse all within a time span of 3-5 seconds or so, hoping that the monkey you killed actually dies within the time span, whilst also the auto activation triggers in a timely matter, and we don't have a trigger which require an atmospheric setup which requires atmos time to understand, followed by wrenching it. All while hoping that an Atmosian you ask is actually capable of building the setup you need, whilst not being held back by failing to set up the TEG, and that within the first 5-10 min of shift for the first artifact before a random anomaly outshines the need of it and complete research in the next 15 min... or we explode, whatever happens first.
My personal gripes with this system would be better if...
- The Artifact Laboratories have an already linked Distro Siphon setup. (only one station has that afaik)
- Enough space so that we can put both an freezer and heater near the chamber. (Doesn't need to start with one... but dedicated ports would be nice.)
- Enough space that we can actually filter out the air from the chamber and filter selectively back in what we need...
- Honestly... a dedicated Epi Atmospheric Alert which Epi could access would help a ton
- The Handheld Artifact Scanner could maybe show the ui of the artifact we're working on and light up which it tries to trigger. 5.1. Maybe clicking on one node, Gives you text which shows what has triggered yet or not along with how much time you got left to do it?
- Psionic Disturbances work (Make it work with pulse so it's more useful)
- Artifexum can trigger all Triggers of one node instead of only one. (Also If I've seen it right, it can only be used once per node for the trigger?)
- The time for the activation afterwards isn't 10s, but maybe 3s.
- The time for the triggering extends by 3 to 5 seconds anytime we successfully trigger one of the required ones. (Which would also make else impossible reaction combination possible.
The word for the current state... unresponsive and a fear of telling a Research Intern (with zero atmos knowledge) to go link up the pipes to waste, only for 'em to accidently pipe 'em into the distro of a freezon producing artifact ( Can happen accidently if you're not watching where some pipes run).
TL;DR
- Atmos Reactions requiring an atmos setup which require space and knowledge (or a competent Atmosian)
- Artifexum seems to has lost potency
- Too short of a timer to trigger a weird combination of reactions
- Too long of a timer for a reaction to happen.
This pull request has conflicts, please resolve those before we can evaluate the pull request.
i will say, my current absolute biggest pain point with new xenoarch is the lack of glimmer multiplier. right now, artifacts are barely even worth doing compared to anomalies. https://github.com/DeltaV-Station/Delta-v/pull/4486 was poised to bring this back but it's dead in the water, I think?
i will say, my current absolute biggest pain point with new xenoarch is the lack of glimmer multiplier. right now, artifacts are barely even worth doing compared to anomalies. #4486 was poised to bring this back but it's dead in the water, I think?
Not dead in the water I completed the changes that were asked to it an never revived another review of it I believe it is now on hold do to maintainers sorting out even newer upstream merges along with the proposed changes to the new xenosystem other mentioned here. I will try again to get someone to review it
Okay glimmer has now been finally brought into the new system with #4486 being pushed, along with the proposed changes to the trigger list to prevent contradicting triggers in #4740.,I believe this revert would be a mistake to go through with. It would make much more sense to iterate and change the newer system to what we like rather than be stagnant and keep an older system just because it's what most people are more comfortable with.
This new xenoarch system in my opinion creates a bigger challenge for epi which prolongs the gameplay loop as well as preventing epi from getting all the research done extremely fast every shift like some skilled people could do in the past. This also provides a better point of entry to people who are new to xenoarch as the old system was tough to learn or teach due to the overly simple UI of the old version. Where as the new system can be almost figured out on your own for the most part due to the interactive and descriptive new UI.
The beauty of it all is that if it is a issue of balance then we can balance it. But I dont think we should change it if the big issue is just people not being used to it yet. I'm not a maintainer and have no control over it but I just wanted to share my thoughts and concerns about this change.
All of the new XenoArch updates have been merged in! I think there was 8 changes total?
For now, I'm going to close this PR. Direction will be asking the community how the changes have been working out in a week or two, and if we should revisit removing New XenoArch then. If the community still hates it or doesn't think it can be improved, we'll re-open this.