ExperiSci Feedback Thread

[bobbahbrown]

Note: expersci has been merged, and those wishing to contribute experiments or content in general should check out our HackMD which provides some documentation and design considerations!

Hello,

We will very soon begin testmerging the experisci science changes. Please use this thread to provide general feedback regarding the new changes.

Any bugs should be reported on the PR itself.

Best wishes,
bobbah "bee" brown

[bobbahbrown]

feedback thusfar from the first testmerge:

- slime experiments should be rebalanced (im doing this now)
- we should be careful about what nodes we are assigning blocking (non-discount) experiments to as some technologies that are essential in the meta might not get researched into very late in the round
- it may be advantageous to look into having more discount experiments over required experiments, or having the discount vs required-ness of an experiment determined at server start based on population

[terranaut]

i had this thought earlier and forgot about it and now i remembered
i dont actually play of course but i hope you kept a legit reason for scientists to make bombs to avoid the meta tell of "he makin bombs he bad"
all the best,
bobbah "monki" brown

[bobbahbrown]

(from github)

Changes from today:
- I've removed a few of the blocking experiments, made some thematic changes to the matching of experiments to web items, and moved some blocking experiments to discounts instead.
- I've updated the RD console UI to better make use of space on the detail view, now it will show prerequisite technologies when you look at the details of a node immediately, and changing tabs will show those unlocked by that node.
- Rebalanced slime scanning experiments, they will now accept any slime core of a 'tier', and require fewer of them for earlier nodes. This should greatly decrease the issues from last night.
- Added detail stages to framework to show more information regarding an individual stage on an experiment

re: what terranaut said, yes there are experiments for toxins

Going to testmerge again tonight and see how these changes play out.

[bobbahbrown]

(from GitHub)

Today's updates:
- We've added the destructive scanner which can be used to perform normal scanning experiments as well as the (new) destructive experiments. Currently the experiments for cytology are destructive in nature. Thanks @MrDoomBringer for the wonderful sprites
- We've fixed the bugs with material experiments and being unable to scan some mineral walls, floors, and the operation table.
- We've added helpful tooltips to experiments which can be seen by mousing over the question mark in the top-right corner of experiments, these help to explain how to perform the experiment.
- People will 768p monitors will no longer suffer from being unable to resize the RD window and the default sizes are scaled down to fit those peoples' monitors
- We've added a flag for experiment handlers to be 'always active', meaning they will attempt to do every experiment when possible. The doppler array has been given this flag, so from roundstart should be automatically configured to record explosions.

I understand the bugs with material experiments were rather frustrating, so I'm hoping to get that fix on the testmerge ASAP.

seasons greetings,
bobbah "bee" brown

[WineAllWine]

I've only played a little of this on bagil (ew, bagil). But I really enjoyed it. It makes science feel more like a whole department working towards a common goal, rather than a bunch of random features. Also gives reasons to make things that you wouldn't normally (who would vat-grow a cockroach?)

Overall very excited to see this on Terry

[remanseptim]

t*rry player thinking he can talk shit about any other server but event hall and maybe manuel

also experisci is great, it's peak comfy.
i for one enjoy busywork in jobs as long as it isn't tedious, that's why genetics can be fun.
my only issue aside from the occasional expected bug where scans do not work is having to replace floortiles, since they ends up making your department look dumb when you replace white ones with grey.

[oranges]

i had this thought earlier and forgot about it and now i remembered
i dont actually play of course but i hope you kept a legit reason for scientists to make bombs to avoid the meta tell of "he makin bombs he bad"
all the best,
bobbah "monki" brown

The other valid use for bombs is compressing anomaly cores, which hopefully will be added to more high end items as people find the time

[Kassori]

Thats kind of a flimsy use for toxins, your first few cores of any type can be detonated with hot plasma and cold 02, that's what science normally does with the gases left over after the maxcap currently.

Though its not a massive loss, since toxins has no real variation, its just kind of a speedrunning mini-game.

[WineAllWine]

is this still on the cards? I thought it would be coming after the November code freeze but haven't seen anything about it in ages

[bobbahbrown]

is this still on the cards? I thought it would be coming after the November code freeze but haven't seen anything about it in ages

it should be yes

between me floyd and jared its got a little ways more to go

slightly busy,
bobbah 'bee' brown

[cacogen]

slightly beesy ahaha

[vkalls]

I really like this. It's come a long way from when I first saw it on Sybil. I like that there are only some techs with experiment locks, there even could be a few more in my opinion. I also like the discount experiments. A little more variety would be nice, as xenobio and toxins aren't always something people will do. Maybe some options that might be more difficult but less boring, like scanning something from lavaland for biotech. Or scanning dead mobs for weapons tech. I also really like the unique challenges like the gaming one. Overall, a very good improvement to RnD and far above the current system.

[bobbahbrown]

I really like this. It's come a long way from when I first saw it on Sybil. I like that there are only some techs with experiment locks, there even could be a few more in my opinion. I also like the discount experiments. A little more variety would be nice, as xenobio and toxins aren't always something people will do. Maybe some options that might be more difficult but less boring, like scanning something from lavaland for biotech. Or scanning dead mobs for weapons tech. I also really like the unique challenges like the gaming one. Overall, a very good improvement to RnD and far above the current system.

hello!

glad to hear positive feedback, and as for variety you may have seen we just had a big ole' discussion about this in the coding general channel on discord. this PR is more intended to bring the framework to our codebase, as well as a small variety of sample experiments for coders to use for inspiration and a starting point for implementing their own experiments.

we hope for the variety to improve greatly over the short term following the PR's merge to the codebase, at which point all contributors will be able to propose their own experiments.

thanks for posting,
bobbah 'bee' brown

[bobbahbrown]

experisci was merged today, and those who wish to contribute experiments or content in general should check out the HackMD documentation which helps to jumpstart any thought in this area.

[cacogen]

build x meat furniture

build x mineral walls

blow up 2 welding fuel canisters

have had someone do xenobiology that round

have had someone do cytology that round

have had someone do nanites that round

scan 3 vomit decals

I liked research better when nodes were only gated by point regeneration. I'd have improved it by allowing players to choose their path through the tree ahead of time. Nodes would unlock by themselves as points built up. Research wouldn't involve standing at the console waiting.

Destructive Analysing sucked because nobody did it until people made guides. Then it amounted to following the meta strategy round after round. ExperiSci is similar in the fetch quest tedium and lack of engagement. But instead of relying on opaque values the game tells you what you need. It feels like jumping through hoops rather than unlocking things organically.

I respect the effort involved and the attempt to make research interesting and dynamic. I like the TGUI on the scanners, it makes the feature feel new and interesting. I do find the amount of information visually overwhelming, though. I wish the coder had broken it up into category submenus. Like the UI for downloading modular computer programs.

[Qustinnus]

build x meat furniture

build x mineral walls

blow up 2 welding fuel canisters

have had someone do xenobiology that round

have had someone do cytology that round

have had someone do nanites that round

scan 3 vomit decals

I liked research better when nodes were only gated by point regeneration. I'd have improved it by allowing players to choose their path through the tree ahead of time. Nodes would unlock by themselves as points built up. Research wouldn't involve standing at the console waiting.

Destructive Analysing sucked because nobody did it until people made guides. Then it amounted to following the meta strategy round after round. ExperiSci is similar in the fetch quest tedium and lack of engagement. But instead of relying on opaque values the game tells you what you need. It feels like jumping through hoops rather than unlocking things organically.

I respect the effort involved and the attempt to make research interesting and dynamic. I like the TGUI on the scanners, it makes the feature feel new and interesting. I do find the amount of information visually overwhelming, though. I wish the coder had broken it up into category submenus. Like the UI for downloading modular computer programs.

theres a big difference between "get this specific list of items that you can find at roundstart / can print" and "get this list of objects that you would otherwise already try to get (or atleast would be part of your job to get". Especially with the factor being that you have different targets each round.

[bobbahbrown]

snip

I liked research better when nodes were only gated by point regeneration. I'd have improved it by allowing players to choose their path through the tree ahead of time. Nodes would unlock by themselves as points built up. Research wouldn't involve standing at the console waiting.

Destructive Analysing sucked because nobody did it until people made guides. Then it amounted to following the meta strategy round after round. ExperiSci is similar in the fetch quest tedium and lack of engagement. But instead of relying on opaque values the game tells you what you need. It feels like jumping through hoops rather than unlocking things organically.

I respect the effort involved and the attempt to make research interesting and dynamic. I like the TGUI on the scanners, it makes the feature feel new and interesting. I do find the amount of information visually overwhelming, though. I wish the coder had broken it up into category submenus. Like the UI for downloading modular computer programs.

this is valuable feedback that i appreciate, thank you for submitting it.

i think your sentiment regarding the 'fetch quest' style of current experiments is fair, and i hope that through future content PRs (which are already being worked on by different people), we can introduce much more engaging or interesting experiments. this PR was much more about introducing the framework and some 'example' experiments, so i can understand your frustration with the content introduced.

regarding the bore of standing at the RnD console, i would recommend checking out the modular tablet app made by Arcane. it's on the tablet the RD starts with, and the RD can download it to any computer after which anyone with rnd access can use it. it's an awesome way to perform research while on the go, as you can leave it on your belt or elsewhere and keep the UI open whilst you do other stuff. zxaber may be adding a modular-computer-based experi-scanner soon as well.

i greatly appreciate the calm manner in which you provided this feedback, and on the topic of the UI i will take a look and see if i can improve the 'information overload' when i get a chance, though i invite any interested individual to take a look.

please dont hesitate to post any more feedback or ideas, or even reach out to me on discord.

best,
bobbah 'bee' brown

[cacogen]

regarding the bore of standing at the RnD console, i would recommend checking out the modular tablet app made by Arcane. it's on the tablet the RD starts with, and the RD can download it to any computer after which anyone with rnd access can use it. it's an awesome way to perform research while on the go, as you can leave it on your belt or elsewhere and keep the UI open whilst you do other stuff. zxaber may be adding a modular-computer-based experi-scanner soon as well.

When I played this last night, my HoP tablet may have started with this app. Either way, I was able to research from it. Which was nice. I didn't mean to imply ExperiSci required standing at a console. I meant the old system did.

I don't know if scientists start with tablets or this app, though. I noticed last night they still don't have their own console, which feels cucked. A similar principle is the HoP and not the QM having access to the NTIRN app. If an RD isn't unlocking research they're relegated to using the console in the RD's office. This is impractical for a lot of reasons. I wish the "civilian console" had the same functionality as the RD's to begin with.

theres a big difference between "get this specific list of items that you can find at roundstart / can print" and "get this list of objects that you would otherwise already try to get (or atleast would be part of your job to get". Especially with the factor being that you have different targets each round.

Maybe the specific details of what unlocks what prove this wrong but I don't like the idea of gating that good shit that research gets other departments behind doing stuff that sucks that I don't know how to do and have never seen anyone do like cytology. Xenobiology is something I find satisfying but it takes ages and the controls are shitty so I'm not going to do it because I happen to be in science on lowpop and none of the jobs have been done and I'm trying to get a node for my department and I'm locked out of it by slime cores. Again maybe the specific details of what jobs like cytology and xenobiology unlock for other departments prove this wrong but that's how I feel anyway.

[Qustinnus]

regarding the bore of standing at the RnD console, i would recommend checking out the modular tablet app made by Arcane. it's on the tablet the RD starts with, and the RD can download it to any computer after which anyone with rnd access can use it. it's an awesome way to perform research while on the go, as you can leave it on your belt or elsewhere and keep the UI open whilst you do other stuff. zxaber may be adding a modular-computer-based experi-scanner soon as well.

When I played this last night, my HoP tablet may have started with this app. Either way, I was able to research from it. Which was nice. I didn't mean to imply ExperiSci required standing at a console. I meant the old system did.

I don't know if scientists start with tablets or this app, though. I noticed last night they still don't have their own console, which feels cucked. A similar principle is the HoP and not the QM having access to the NTIRN app. If an RD isn't unlocking research they're relegated to using the console in the RD's office. This is impractical for a lot of reasons. I wish the "civilian console" had the same functionality as the RD's to begin with.

theres a big difference between "get this specific list of items that you can find at roundstart / can print" and "get this list of objects that you would otherwise already try to get (or atleast would be part of your job to get". Especially with the factor being that you have different targets each round.

Maybe the specific details of what unlocks what prove this wrong but I don't like the idea of gating that good shit that research gets other departments behind doing stuff that sucks that I don't know how to do and have never seen anyone do like cytology. Xenobiology is something I find satisfying but it takes ages and the controls are shitty so I'm not going to do it because I happen to be in science on lowpop and none of the jobs have been done and I'm trying to get a node for my department and I'm locked out of it by slime cores. Again maybe the specific details of what jobs like cytology and xenobiology unlock for other departments prove this wrong but that's how I feel anyway.

tech nodes are a bonus not a necesity, if you cant do cytology thats your own loss.

[cacogen]

Is that true? Do you not need xenobiology or cytology to unlock things?

[JusticeGoat]

Any thoughts on having the system be dynamic based on starting pop? Maybe research requirements are lax on lowpop vs a full science team will have more to do, otherwise this just makes working the department with one or two people annoying.
Edit: Oh i looked though the doc and it seems like you are covering this.

[Qustinnus]

you shouldnt be able to research everything on lowpop in the same way that you shouldnt be able to get every item on lowpop.

[JusticeGoat]

its a 2 hours at most game, gating content because you don't have enough people is a slap in the face, just let people have fun because that is what we are here for, and if that means maybe adding a way for the system to adapt to the player count i see no downsides.

[Qustinnus]

no

This post was deleted by cacogen on Fri Feb 26, 2021 4:03 am.
Reason: not worth the drama

[JusticeGoat]

This post was deleted by JusticeGoat on Fri Feb 26, 2021 4:49 am.

[NecromancerAnne]

First off, some notable oversights or problems I've seen thus far:

A) Deconstructing a chair or table and then reconstructing it will let you effectively treat it as a new chair for the sake of scanning. I suspect this is the case with floor tiles as well. This means, with a single sheet of material, I can make and deconstruct chairs or tables to complete the material scan objective. This somewhat makes the object amount acquired arbitrary, but I don't see any actual way to make this work unless you make it use the deconstruction machine exclusively. Speaking of which...

B) You can't scan tables with the hand scanner, since it places it on top of the table, but you can use the deconstruction machine. However, the machine doesn't delete things at all it seems. So using my above method, you can endlessly scan deconstructable objects. Is this intentional? I really don't know, I got the impression it is meant to delete stuff.

C) The deconstruction machine is on a overlay layer higher than other objects. This means most objects disappear underneath it.

Most other problems I find with techweb progression appear to mostly be from holdovers from the previous progression, and also being attached to experiments not easily contributed by the jobs that benefit from that tech, which is sorta what I thought we were leading towards rather than tripling down on research being only possible from science progression.

Xenobiology has no relation to medical biotech, and yet governs all the medical biotech progress as well as robotics progress due to overlap, and a few other jobs simply get caught in colleterial purely because most of those jobs require the Biotech node. This means, if nobody is in xenobio, medical and robotics sees absolutely no progress in their equipment and can't do anything to fix that themselves even if they sought to contribute. Biotech is first for progress in quite a few tech branches. That web needs to get chopped up quite a bit, and there needs to be some contribution from robotics in this process for at least the biotech progress if we really want to keep this the domain of science, since they're the best candidates for cybernetics and medical research. Maybe it would be a good time to bring back dissection?

(Do note, at the moment scientists just kill off all the starting slimes to get the first biotech node, so progress on xenobio is usually not even being made in the first place and actively being gimped simply to keep tech progress going elsewhere, which seems counterintuitive to the intention of experiments. This assumes nobody is doing xenobiology, which isn't out of the question due to it being mostly zen gardening with slimes, which isn't to everyones tastes.)

These techweb progression restrictions aren't the case for engineering, who can do a lot to contribute to their progress only because their experiments are incredibly easy. They can even jumpstart some of the experiments thanks to their attachment to toxins via atmospherics. They are grossly favoured in being able to do most of the work themselves. Botany can contribute to acquiring their technode, but Biotech is STILL a full stop restriction to so much of the web they can't circumvent.

[Qustinnus]

I think you raise some valid points and I'll look into ways of letting medical help their own research path instead of having to rely on the powergamer job, but I disagree with the notion that any department /needs/ their tech nodes to function.

[NecromancerAnne]

'Needs' is maybe not the right word there, but it might also be more wider problems with in-job progression that the only things that progress them is less convenience items (like advanced tools) and more enabling branches of their department (most of the robotics stuff like nanites and cybernetics is locked behind tech nodes, which leaves the job extremely shallow without progress). I don't think that's a experisci issue either, so maybe not worth touching on here.