/tg/Station 13

Please post here for any feedback regarding dynamic on MRP servers here.

Hi! The way I see it, there are two problems with dynamic in MRP right now:

First: Antag value: Often enough we find ourselves in rounds that have 8 traitors (one of 'em, a malf AI) , a ninja (+ the pirates or swarmers it spawns), a wizard and spiders.Or a dragon. Or a blob.
Not exactly in this combination, but you get my idea. When the captain announces anything in the means of "Contested system", most people's plan just get thrown aside since they know there won't be much RP going around.
I'm going off numbers that were told me, and I'm noy 100% sure they're right, but I don't understand how 2 lings can be worth 40 threat, and 8 TRAITORS is 20.
Dynamic rounds are the most chaotic of them all, and while *I* don't mind them, most of the people I know complain about it, since it turns the round into lrp. There is no conceivable way a 5 or 6 member sec team can deal with these threats unless they implement lethals, which most are afraid of due to bwoinks.
The values of the antags most certainly need some adjusting to MRP's playstyle.

Second: Rate of purchase: It would seem the gamemode purchases antags randomly, as long as it has threat, at a given time (afaik, 15 minutes in the round). This is not a good idea since you EITHER have a chaotic shift from the start, or after 15 minutes of calm, everything just goes to shit. I've had captain rounds where up untill 20 minutes I was bored to death, then in the same 5 minutes we had calls of ninja sightings, dragons AND a wizard. This is almost impossible to stop in a non-round ending way.

If it was possible to spread out the purchases instead of buying all at the same time, it could solve this problem of "chaotic explosion". It would also be nice if it had rules to prevent it from buying the same round-ending antag twice (like shift-start wizard and mid-round wizard again).
For wizards, in specific, there's a lot of problems that remain behind with his death. OP items like the metacomms orb, insta-death wand, yada yada, and the most agonizing of all, reveal ghosts, which just fucks with anyone trying to do a stealth antag, or even jaunt.

I'll throw in a few of my observations too that'll most likely be repeated throughout this thread, should others wish to restate them.

1. Antag values and scaling. I don't know them off the top of my head, but I'm aware something along the lines of 8~10 traitors is equal to 2/3 lings and/or heretics (which don't add much to the round despite their high cost, unless they get rolling which, unlike the mass of regular traitors, can get things off the gecko that can wreak havoc from the instance as soon as they get them. Hereticts become an apparent issue once ascended, which requires time. Lings are, well, lings. The best they do is cause mass confusion and chaos from almost the backlines. Good for adding onto the chaos, bad for their cost in numbers.) Some of the prices can defiantly be reworked, increasing how many numbers of certain antags spawn, so on and so forth.

2. Anything past contest system throws any idea of RP out the window. Its code word for "gamer up or die to the blob, random dragon that came into the kitchen, or the ninja who so happened to hijack a borg and you're the AI. Oh no". I'll also lump in the time aspect too since this'll go hand in hand with RP too. Unless you seriously don't care about the one in twenty random antags that spawned and even more so don't care if you just die because, well, dynamic moment, then any and all consideration of RP gets destroyed. It still exists, by Hell it sure does, but its not the kinda you'd "expect" on Manuel. As for time, as I've stated on the other thread, we're MRP, shouldn't it be at least expected our rounds go on longer then LRP? Assuming their average, on their regular highpop rounds, is 45~75 min. Sure, the station is going to be an absolute mess from the dragon, blob, revolutionists/cultists so and so forth, but if we edge people to the fact that maybe we can salvage this perforated metal hell we call home instead of calling the shuttle because of the twenty said antags. Although thats something else entirely at this point and I'll leave it at that.

3. Wizard. No. Lets make ghosts visible! Said the wizard. Lets ruin all the other possible antag! He didn't think. This is self explanatory, but ghosts do have the option, key words: have the option, to intentionally or unintentionally ruin the rounds for antags, no matter who it is. If this spell were just entirely disabled for dynamic I think wizard would fit (and among any other meta-gamer items that they have). Ghost spam pointing to that locker or plant, the antag hiding within is sweating profusely.

4. 100% Dynamic. Please God no. If I wanted to play 100% dynamic I would simply log onto one of the LRP servers. However, I'm for the idea of scaling dynamic with secret in proportion to players. Sure, this may make more people avoid low pop entirely, but when Manuel starts getting people on, it won't be an issue. Even now its a 1:1 ratio (rounding all the other possible antags into 50%). While on this topic, since brothers have been disabled, and if it wasn't done already, has their spawn percentage been altered to possibly give more chance to other antags?

Dynamic itself isn't too bad of an idea, its just how its executed internally, from how certain antags behave, to how the game "buys" them. One fixer that sounds good is readjusting the cost for certain antags and how many spawn.

Just gonna copy paste here.

In the light of some efforts to revamp dynamic I'll be taking the time hopefully sometime soon to throw together a dashboard where some simulations can run and people can be able to understand all these numbers and see what tweaks would do to change things. Really be able to give proper feedback, and fine turn the experience.

Up until now everything I've done for this was with an Excel sheet simulation that no one really asked for, and dynamic for us has been a bit limited in scope. Now I actually have enough reason to make things a bit easier.

Misdoubtful wrote:Just gonna copy paste here.

In the light of some efforts to revamp dynamic I'll be taking the time hopefully sometime soon to throw together a dashboard where some simulations can run and people can be able to understand all these numbers and see what tweaks would do to change things. Really be able to give proper feedback, and fine turn the experience.

Up until now everything I've done for this was with an Excel sheet simulation that no one really asked for, and dynamic for us has been a bit limited in scope. Now I actually have enough reason to make things a bit easier.

Is this using the Jupyter Notebooks I sent you? If so, those are based on Dynamic 2021 and not the current configurations.

Jaredfogle wrote:

Misdoubtful wrote:Just gonna copy paste here.

In the light of some efforts to revamp dynamic I'll be taking the time hopefully sometime soon to throw together a dashboard where some simulations can run and people can be able to understand all these numbers and see what tweaks would do to change things. Really be able to give proper feedback, and fine turn the experience.

Up until now everything I've done for this was with an Excel sheet simulation that no one really asked for, and dynamic for us has been a bit limited in scope. Now I actually have enough reason to make things a bit easier.

Is this using the Jupyter Notebooks I sent you? If so, those are based on Dynamic 2021 and not the current configurations.

The idea is something that will be as open ended as possible for whatever changes someone can think of, so it shouldn't really matter on which. That said it may be a while before I get around to doing it

The biggest issue I have with dynamic is the antagsplosion. Very often on Manuel I notice a lot of antags seemingly injected at the same time. Security is calmly dealing with a few traitors and all of a sudden there are spiders, a dragon, and a ninja. I would like to see the time between injections of antags to be raised. I personally think the threat levels are fine as they are on Manuel. Lings is as people said, not large enough of an actual threat to warrant their threat number. I agree with wizards having their summon ghosts taken away on dynamic, but I think having two wizards is fine.

See you guys in 3 months.

Random event antags should not be spawned by dynamic. Space Dragon, Blob, Space Ninja, Nightmare, Abductors and Alien Infestation should all cost 101 just like Swarmers. Our random event system has handled their spawning just fine. We don't need two sources of them.

Certain antags should be added to midround rolls, Changelings, Heretics, Cultists, and Revolutionaries are all lacking there, and Changelings should also be added to the latejoin rolls.

The midround roll should also happen later as Dynamic very often just blows its load early in the round, but that doesn't seem to be an option on the config from what I can see.

Traitor scaling is also really fucky at times as some rounds it will spawn 1 traitor for every 5 normal crew, and others it will just spawn 1 traitor for every 10 normal crew. A midground would be great, but I don't know the config values to achieve this.

Owegno wrote: Traitor scaling is also really fucky at times as some rounds it will spawn 1 traitor for every 5 normal crew, and others it will just spawn 1 traitor for every 10 normal crew. A midground would be great, but I don't know the config values to achieve this.

This depends on a few things.

One is the scaling. A ruleset can be picked (traitors) and scaled up, which will create more traitors.

Another is the population. Currently, dynamic works off a weird system where the amount of antags that can be picked is based on population, but only in sets of 6, and up to a certain limit (54, which is not even close to what high pop rounds can be!). This means that at, say, 29 people, you will have low numbers, but once you hit 30 everything jumps. I talked to Skoglol about this, and we think it would be best for part of Dynamic 2021's change set to include making these configurable linear equations instead, meaning they would correctly scale with population, rather than in steps.

Owegno wrote: The midround roll should also happen later as Dynamic very often just blows its load early in the round, but that doesn't seem to be an option on the config from what I can see.

Dynamic 2021 makes these changes!

As for midround traitors (and the extension of this to changelings, heretics, etc)--Skoglol has given me reason to suspect that this would be problematic as it would encourage people to gear up for the possibility of getting one of these roles. This is something I'm going to be observing during playtests on both LRP and MRP.

Generally I'd like to see major round ending threats like cult, revolution and nuke ops to be extremely unlikely to spawn before the hour mark or so, to give the round a proper escalation of threats as time goes on and hopefully bring the round to a natural conclusion instead of having to call the shuttle out of boredom

Mickyan wrote:Generally I'd like to see major round ending threats like cult, revolution and nuke ops to be extremely unlikely to spawn before the hour mark or so, to give the round a proper escalation of threats as time goes on and hopefully bring the round to a natural conclusion instead of having to call the shuttle out of boredom

By the hour mark a lot of those modes are easy to stomp due to the progress the station has made with stuff like mechs, research, a fully established medical, guns, etc

Then they'll have to work harder or blame all of the other station threats for doing a bad job

NoxVS wrote: By the hour mark a lot of those modes are easy to stomp due to the progress the station has made with stuff like mechs, research, a fully established medical, guns, etc

This. The arms race is a major motivator for the revs/cult to work quickly and contribute to the round instead of sitting on their thumbs as sleeper agents for an hour.