Rimworld: Another Excellent Game with the Ultimate Flaw.

Started by Edmon, November 20, 2017, 06:24:40 PM

Previous topic - Next topic

Bozobub

Thing is, it's not an "either-or" situation; you can optimize a LOT, without disappearing up your own rectal cavity ::).  Once again, I have *zero* problems (beyond w/e environment, itself) setting up my cooking/butchering/cold storage cycles, on any difficulty, with any storyteller, without bothering with silly, excess micro.
Thanks, belgord!

Vlad0mi3r

As much as it pains me to agree with Bozobub (Joke FYI). I don't see it as an issue certainly not an "Ultimate flaw". When you are looking at 1 cook per 10 colonists plus cleaning duties for that cook at the end of the bill. I am failing to see the issue. This is not with a micro managed kitchen or even auto doors. With an increase in skills as well as auto doors you can have 1 cooking for close to 15 but nothing else. This is with basic meals of course.

Now I use cooking as the example but with correct hauling setups most work functions don't or shouldn't need min/maxing as far as movement is concerned.

Actually as the only real counter point I can think of is medicine production. I do setup stockpiles next to my drug lab with neutroamine and cloth so the pawn is only back and forth for herbal meds.
Mods I would recommend:
Mending, Fertile Fields, Smokeleaf Industries and the Giddy Up series.

The Mod you must have:
https://ludeon.com/forums/index.php?topic=40545.msg403503#msg403503

dkmoo

Getting back to the point that i think OP is trying to make - I don't think that it's a "flaw" but it's more a game-design challenge of reconciling two vastly different time scales - In-game time vs RL time. OP seems to be trying to get at achieving a solution that has the "best of both worlds", which, by the very nature of the difference, is impossible. Let me explain:

Simulation games, by design, require different areas of the game to "feel" like real-life to create the simulation "immersion" that players crave. For games where pawns are involved, one of key areas to achieve this "real life immersion" is to make the animations close to "real life time". This necessitates for things like movement speed, shooting speed, hammering speed (the ding ding ding sound when pawns are working on something) to be at a pace that looks and feels realistic. The issue however, is that the game MUST speed up other areas of the game in order to achieve an overall game progression pace that makes sense for the average player's playing time (a few hours per sitting). This is not a flaw, it's just cards that we are dealt with. Looking at the infamous "meal making" example, in game it probably takes roughly the same amount of time to gather/walk as cooking the food (a movement-time-cost-to-production-time-cost ratio of roughly 1:1). The same process in real life is probably somewhere around 1 minute to walk/gather stuff in your kitchen and 1 hour of cook time (1:60).  People probably don't realize the true vastness of the difference in these two time-scales. Even slowing down production by 4x and increase movement speed by 4x like the OP suggests truly only migitate the issue by roughly 25% (1:16 vs 1:60). This brings us to the following point: as long as the in-game movement to production ratio do not approach the RL ratio, there will always be a point in which "reducing movement cost design or lose" becomes a limitation

Taking the above point one step further, only way to solve this "ultimate flaw" is to have in game ratio be close to RL ratio, which definitionally will result in one of only two possibilities:
1) keeping pawn animation speed (movement) close to RL time. But this will also drag out production time to ridiculous lengths - meals in game will take hours of RL time to complete, parka full day to make, houses and buildings days/months to complete. Imagine how bored players will be waiting for that kind of game pace?
2) speed up in-game production time - if doing this while trying maintain RL walk/production ratio, the pawns will be zipping around so fast that for practical purposes it loses all the "simulation immersion" value of having pawns in the first place.

Other games are designed in one of the two options mentioned above and they don't have this problem (think EVE online for 1, and SimCity 4 for 2). However for RW and many other games in between, a true solution to this "ultimate flaw" simply does not exist. The best that these games can do is to balance other aspects of the game to create the best game-playing experience. RW does this masterfully - ie, it gives movements "more bang for the buck" (ie, hauling enough material for 40-50 feet, or 10-15 tiles, of walls in one trip), or, it reduces the movement cost impact by having "super intelligent" animal haulers that knows exactly where and when to haul materials without intervention (by comparison I can't even get my dog to fetch me a beer)

TLTR: the "ultimate flaw" that the OP laid out is a "flaw" only insomuch as trying to fit a square block into a circular hole is a flaw. It's simply an attempt to reconcile two vastly different time scales that is definitionally impossible to achieve within the confines of most simulation game designs such as RW.

Edmon

Firstly dkmoo, I'd like to thank you for your write-up. You understand exactly what the issue is, which is great, but as a game designer I don't agree with your conclusion.

Quote from: dkmoo on November 21, 2017, 07:37:30 PM
However for RW and many other games in between, a true solution to this "ultimate flaw" simply does not exist. The best that these games can do is to balance other aspects of the game to create the best game-playing experience.

That's why it's called the Ultimate Flaw. There is no "true" or "simple" solution to the problem. We're not using the word lightly, it's a flaw built right into the heart of the nature of the game. One that requires extremely carefully balanced design to compensate for.

The basis of what you say is correct, it cannot be truly reconciled without time phases (Games with a build phase and a combat phase, but this isn't that type of game either). However, it can be balanced to feel right and to be strategically interesting.

Quote from: dkmoo on November 21, 2017, 07:37:30 PM
Even slowing down production by 4x and increase movement speed by 4x like the OP suggests truly only migitate  the issue by roughly 25% (1:16 vs 1:60). This brings us to the following point: as long as the in-game movement to production ratio do not approach the RL ratio, there will always be a point in which "reducing movement cost design or lose" becomes a limitation.
Right, but 1:16 is a hell of a lot better than 1:60. Plus, now all we need to do is make the bonuses for having a really nice room, with tool kits and various other items in it, that doesn't have things in it that should not be there, be approximately 8% of total productivity for it to be worth making an interesting room. Over a movement minimized one.

Though you will always get gains from minimizing movement, the bonus of doing that should not be so all consuming like it is in Rimworld. Pawns can waste 70%+ of the day just walking around doing nothing productive, if you aren't power gaming your designs. Which means organic building of bases to be somewhat like a real-life design is strategically awful. Which shouldn't be the case.

The balance isn't right and leeway in having a movement inefficient base design is too little and too punishing.

I disagree with your premise that Rimworld has got the balance right, let alone "masterfully". Movement is the primary consideration of everything you make and it really shouldn't be.

Soupy Delicious

oooh, I quite understand this ultimate flaw you speak of.  I think he's right!

Weyrling

Quote from: Edmon on November 22, 2017, 05:50:28 AM
That's why it's called the Ultimate Flaw. There is no "true" or "simple" solution to the problem. We're not using the word lightly, it's a flaw built right into the heart of the nature of the game. One that requires extremely carefully balanced design to compensate for.
While you may have a point regarding the actual implementation of movement/combat/production speed ratios, I disagree with the premise that failing to achieve a literally impossible solution counts as a flaw.

dkmoo

Ok, i think we all want the same thing but are hung up on the semantics of the definition of "flaw". So for the sake of progress lets just all agree that it's at least a game design challenge that needs to be addressed - either by 1) overhauling the current pace/feel of RW to achieve closer to RL movement to work ratios or 2) make the gap in the in-game to RL ratio have less of a game impact by implement design mechanisms in OTHER areas of the game to balance out the anomalies created by the time scale and ratio gap.

Personally, I prefer to leave the pace as is b/c 1) I don't want to make production time longer - i'm already playing on 3x and don't want game progress to drag on longer (which, if we want to make any meaningful impact to address the "challenge", will need to at least drag out production pace somewhat materially). 2) I don't want to make walking speed/animation faster - playing on 3x already feels like i'm watching a movie on permanent fast-forward, and 4x dev mode like a time-lapsed video.

Therefore i'd focus on balancing out the other areas of the game - ie by making movements more "bang for the buck" - in theory, if we make 1 movement count as 60 movements in terms of hauling, we'd get effectively the same result as a 1:60 RL ratio with a 1:1 in game ratio. RW already does this to a degree with a somewhat outlandish hauling mechanism (see my wall construction and animal hauler example in the first response). I feel like this is enough, but OP's opinion is that we need more, which I can live with - just difference in preference. If we want to add more such mechanism to further balance out the challenge, maybe Tynan can consider the following two areas that come to mind:

1) make certain types of workstations STORE a sizable materials in them, at least 2 stash size worth, and set it to replenish (requesting haul) only with less than 50% remaining - this will guarantee that each movement into the bench has the max haulable material being delivered, instead of the single-product-ingredient per trip that's currently in game.  Implementing this for at least low-cost-high-frequency items like meals and drugs will drasticallly reduce the movement cost. Players current work around this with by having single tile high priority stashes next to the benches but that still doesn't fully resolve the issue b/c 1) meat spoil, 2) pawns still "shift" and move a little bit in between each production 3) Haulers can still deliver to the single stash inefficiently (ie, delivering only 5 meat b/c the single tile already has 70). Storing material inside the bench would solve this. Fueling, fermentation, and refining currently employ this mechanism so it probably isn't that difficult to expand it to other benches.

2) making the production mechanism less "conveyor belt driven" by making it less dependent on deliverable ingredients. . ie - making production speed more impacted by pawn conditions (speed boosts, etc, like the OPs original post). Adding additional non-deliverable ingredients - maybe power, or some new type of resource...this probably requires more creativity...

OFWG

Quote from: dkmoo on November 22, 2017, 01:46:30 PM
So for the sake of progress lets just all agree that it's at least a game design challenge that needs to be addressed

Nope, I think it's fine the way it is. Not a flaw or a challenge, and certainly not fatal.
Quote from: sadpickle on August 01, 2018, 05:03:35 PM
I like how they saw the naked guy with no food and said, "what he needs is an SMG."

Murdo

The dominance of movement on base design (and the resulting ugly, illogical, literally cut-corners builds) can be addressed in a number of ways. More could certainly be done in that regard, and it may come down to mods, but that is a treatable symptom. We can address that within the current limits of game design.

But it feels like we're talking about a failure to achieve micro/macro hyper-realism, and I would argue there has been  no such attempt made to have failed. Rimworld is in some ways an abstract microcosm, in others a micro-intensive network diagram... balanced internally (for better or worse) on the premise that it creates its own strategy and tactics.

asanbr

I remember figuring out something similar in Rollercoaster Tycoon as what you describe for theme park.

For Rimworld, however, even if it may be optimal to do as you claim, you haven't solved the game just by that.

There are all the other problems and systems you need to take care of. Survival, defense, cold snaps, manhunters, etc etc.

So I don't see the problem in terms of Rimworld.

Britnoth

QuoteIt is, by far, most optimal to make meals by having the butcher table, cold room and kitchen surface all surround the exact same chair. so the pawn does not have to move from that position to do anything or get anything. It is orders of magnitude better than any other design, not just a little better.

Sounds like a terrible setup for efficiency. Just because you have reached a point where you cannot improve your design, does not mean that an improved design does not exist.

When the global map was added, I attempted the equator -> pole -> pole -> equator challenge. To build up food for each push to the pole I would settle over the winter and mass produce meals. Two cooking stoves were enough to pump out perhaps 2000 meals in the space of 20 days.  :-*

gipothegip

I don't view having to build your bases efficiently as a flaw. That's a plus IMO, this game requires you to design things intelligently.

I will agree (but not fully) that the ability to have everything in the same room at full efficiency can be problematic. However, the room will be very dirty and ugly, making pawns unhappy (so it's not completely imbalanced).

In the case of the kitchen, storing food in there will  contaminate your meals and make your pawns ill (since it'll be dirty). Furthermore you'd have to carefully manage the storage, as you wouldn't want food to spoil, but making it cold will lower the efficiency of the workstations.

The only time I really do the same room thing is with drug production, and I'll admit that might be a bit cheesy. But that's really only because I didn't set up my storage / buildings properly, so it ended up in the same room instead of an adjacent supply closet. I never did it to the extent that they'd never have to move though, and my pawns still weren't happy about working in a dirty and cramped room.


I think the best solution might be to have a work speed bonus / malice depending on the condition of the room. Clean rooms with enough space should definitely be more efficient to work in, even if it has to be somewhat abstracted. This already applies to surgery, and iirc research as well.
Should I feel bad that nearly half my posts are in the off topic section?

mean

Quote from: dkmoo on November 22, 2017, 01:46:30 PM
1) make certain types of workstations STORE a sizable materials in them, at least 2 stash size worth, and set it to replenish (requesting haul) only with less than 50% remaining - this will guarantee that each movement into the bench has the max haulable material being delivered, instead of the single-product-ingredient per trip that's currently in game.  Implementing this for at least low-cost-high-frequency items like meals and drugs will drasticallly reduce the movement cost. Players current work around this with by having single tile high priority stashes next to the benches but that still doesn't fully resolve the issue b/c 1) meat spoil, 2) pawns still "shift" and move a little bit in between each production 3) Haulers can still deliver to the single stash inefficiently (ie, delivering only 5 meat b/c the single tile already has 70). Storing material inside the bench would solve this. Fueling, fermentation, and refining currently employ this mechanism so it probably isn't that difficult to expand it to other benches.

It's really bugging me, that this is not implemented in the game. I usually do 3x1 + another one at the long edge storage though, and after each refill I reduce priority back to how it's set globally in my settlement. That's a lot of hustle in a long run, but it makes a difference on higher difficulties. It's also viable to do 2x6, but pawns do glitch occasionally when it's made like that for some reason.

Renegrade

Quote from: Bozobub on November 21, 2017, 01:56:20 PM
Um...  Yeah.  Edmon, your strategy *IS NOT* actually optimal.  Other reasons have been given above, but a BIG one is the simple fact that a dirty kitchen (with a butchering table in it) will cause food poisoning quite often.

My standard kitchen is basically a dirt-factory, and yet I get almost no poisoning with a good cook.  The last code dive I saw suggests that there's a soft cap of about 1.6x the cooking skill's poisoning chance for a dirt-factory kitchen.   A 12-skill cook has a poison factor of 0.0007 (0.07%).  Times 1.6 = 0.00112 (0.112%) or about 2.24 poisonings in a 2000 meal series.  Conversely a surgery-room-clean kitchen maxes out at 0.89x bonus for 0.6 cleanliness, or (0.0007 * 0.89 = ) 0.000623 factor (1.246 poisonings per 2000).

You're much better off with a good cook than a good kitchen.  The dirt barely makes a difference at all vs. a 0-skill cook with a TWENTY PERCENT poison rate.

(0, 1, 2, 3 and 4 skill cooks give 20%, 10%, 6%, 4% and 2.5% poison chance, respectively)

Quote from: gipothegip on November 22, 2017, 06:37:37 PM
I think the best solution might be to have a work speed bonus / malice depending on the condition of the room. Clean rooms with enough space should definitely be more efficient to work in, even if it has to be somewhat abstracted. This already applies to surgery, and iirc research as well.

The game is already like this.  Pawns enjoy a significant mood bonus for being in a nice room (big size, clean, beauty, comfortable chair), and THAT translates into a noticeable bump to global work rate.  Add to that the cold work speed penalty, and you can find a wide swing between a good room and bad one.

The real problem with the food is that they only make one meal at a time.  I tried out that Feed the Colonists mod, and it vastly reduces the dependence on movement, despite only being a 4x mod (4x more meals for almost 4x more work).  If RimWorld went in that direction, then the other factors I just mentioned would matter more.  Especially if they say made 4-8x the food for 8-16x the work or somesuch.


RemingtonRyder

Mood doesn't affect global work rate any more except through inspiration.