Friday Facts #275 - 0.17 Science changes

[dstensnes]

Dear developers. Everyone knows that science requires coffee. Maybe the science labs should need some coffee either piped in, or provided as an item in early game? No scientific progress has been made without coffee since the fire, because noone likes cold coffee, i'm quite certain....

[Rythe]

Both because this theming is not the only priority and because I don’t think this is breaking much the weak consistency in this regard, and there seems to be a substantial number of people who feel ok with this, I believe it fine.

Finally. A decent response - thank you. But that simply circles us back to "So I don't see Wube changing it. Rails tick the right boxes per them, and that's as far as it goes." like I figured this out a while ago or something.

Still, a decent response merits a decent suggestion. Rails and train networks are pre-Production packs enough that putting them in as a way to try and boost their automation is vain both for being an extra late incentive and people don't much care about what the item does otherwise when feeding them into recipes for something else. So something like the arithmetic combinator from the circuit network would be more interesting, more appropriate and could really use the boost in player awareness.

That just leaves the lingering question of why theme matters to the point where I spent as much time as I did last go around.

Bad poets have little care or ability for rhyme and structure; StarCraft and Command & Conquer are legendary where Total Annihilation isn't because two melded refined mechanics with refined immersion while one tried to get by on superior mechanics and little else.

[posila]

I agree rail doesn't fit the theme for the same reasons, but I just think of it as a "novelty" and then I don't care very much. The beacon would be the ideal representation of production science, but it's unavailable, and I don't like the recipe.

The primary intention behind design of production science recipe has always been to create a challenge that feels somewhat different and fresh compared to previous 5 science pack recipes. That is why the recipe requires large volume of an item. Previously the item was copper wire, but feedback we got was that copper wires are used in large volumes in circuit manufacturing already, so using them in the science pack recipe makes the recipe feel rather dull. That is why we replaced them with rails. Another major concerns are total raw resource costs, another important thing to us is science recipes "guiding" player into automation of important things. Logistic science pack is made of inserters and belts to force player to automate those items, if it fits theme that is nice to have, but not priority.

[steinio]

I agree rail doesn't fit the theme for the same reasons, but I just think of it as a "novelty" and then I don't care very much. The beacon would be the ideal representation of production science, but it's unavailable, and I don't like the recipe.

The primary intention behind design of production science recipe has always been to create a challenge that feels somewhat different and fresh compared to previous 5 science pack recipes. That is why the recipe requires large volume of an item. Previously the item was copper wire, but feedback we got was that copper wires are used in large volumes in circuit manufacturing already, so using them in the science pack recipe makes the recipe feel rather dull. That is why we replaced them with rails. Another major concerns are total raw resource costs, another important thing to us is science recipes "guiding" player into automation of important things. Logistic science pack is made of inserters and belts to force player to automate those items, if it fits theme that is nice to have, but not priority.

Do it as you like - it never was wrong how you decided to make your game so far.

Only because some critics are the loudest doesn't mean they are the majority.

[Rythe]

Thinking about this more, I hadn't fully considered your angle with the 'things we want players to automate', even if that's a bit vague on grenades. There's also the part where rails use iron and stone exclusively. The third problem is naming the science packs - Automation, Logistic and Production being a bit vague in distinction. Fourth would be the theme you're happy to discard when it's inconvenient. Fifth is about making the Production pack recipe distinct.

So Red as Automation, Green as Logistic, Purple as Production w/ Refined Concrete ingredient (was a good suggestion by someone)
Or Red as Manufacturing, Green as Logistic, Purple as Automation w/ Combinator ingredient (maybe costing 2-4 cables, 4 circuits)

Would be decent blends of pros/cons of each consideration.

Something I realized is you locked yourselves into the idea that 'high item flow' is how you're going to make the Production pack recipe distinct from the rest when there's other options. And really, swapping cables for rails doesn't change the underlying observation that circuit card production creates the throughput puzzle with cables plenty enough on its own so adding it to a science pack production scheme (via cables or rails) is kind of redundant in the larger scheme of recipes.

So that creates an interesting question on how to make the Production pack different than the rest, which was the stated goal, and I think that Refined Concrete is better than rails in that regard (water). Combinator has a rather mundane recipe any way I slice it, but the combinator as itself is a distinct and different ingredient from anything else so novelty becomes about using something from the circuit network set of items rather than the usual.

Rocket Silos and Nuke Reactors could be changed to use Refined Concrete, so mass production of Refined Concrete for a science recipe could pay into having a production chain to build the large amounts of concrete those two items require one way or another. This seems like a very good idea in general.

[Mike5000]

Something I realized is you locked yourselves into the idea that 'high item flow' is how you're going to make the Production pack recipe distinct from the rest when there's other options. And really, swapping cables for rails doesn't change the underlying observation that circuit card production creates the throughput puzzle with cables plenty enough on its own so adding it to a science pack production scheme (via cables or rails) is kind of redundant in the larger scheme of recipes.

So that creates an interesting question on how to make the Production pack different than the rest, which was the stated goal, and I think that Refined Concrete is better than rails in that regard (water). Combinator has a rather mundane recipe any way I slice it, but the combinator as itself is a distinct and different ingredient from anything else so novelty becomes about using something from the circuit network set of items rather than the usual.

As recycling-by-disassembly is not included in vanilla a secondary goal of science recipes should be to consume some of the junk scattered across hundreds of storage chests that one tends to accumulate.

By this metric, concrete is somewhat preferable to combinators, although one could make an argument for a sink for stone furnaces instead.

The theme or pattern with the science packs is to put in ingredients appropriate to the domain of research in order to pretend that your science labs are doing something like engineering developments of tech with the odd bit of pure discovery on the side.

...

This form of a theme is nice to have and yes it does happen in many cases which is good as it does “make sense” or invoke this kind od immersion you speak about. But by no means it is a hard rule or a design goal we have set to obey, and it does not always happen. Green science is far from just logistics, engines in blue are only for a tank and solid fuel is barely built upon by advanced oil processing. I guess you could find more weak connections, but I’d say that rails going to train braking force or logistics 3 is similarly weak. In final non-intermediate products this is obviously a step worse, but there is not an infinite amount or intermediates we would have and would fit in this context.

Both because this theming is not the only priority and because I don’t think this is breaking much the weak consistency in this regard, and there seems to be a substantial number of people who feel ok with this, I believe it fine.

Since we all want Factorio to be the best it can be, and since remaining true to theme is better than not, and since using science as a sink for junk is better than not, doing otherwise is not "fine" absent a compelling reason.

Red and Green Science are naturally tiered - Manufacturing and Engineering. The remaining finite sciences should be themed - Production (including Modules), Trains, Bots, Nuclear - to better support different play styles. One should be able to focus on Train research before Bot research or vice versa.

[Leuf]

In our world, space exploration began as an offshoot of military technology. We only managed to devote the resources for this because it was easy to swap out a nuclear warhead for a satellite, that would be really handy for aiming the rest of the nuclear warheads that we didn't convert for this. And we aren't surrounded by giant bugs that are constantly trying to kill us.

So taking military science completely out of the loop for the rocket is just really bizarre. It really gets to the heart of what is the purpose of building the rocket in the first place? I assume that in the campaign biters are going to be there all the time. It's kind of backwards to have the core of the game built around what is best for the freeplay and accommodating people who want to turn biters off rather than what is best for the campaign. People who are playing the freeplay with biters turned off are just building stuff for the sake of building stuff. It doesn't matter what the name of the science pack is or what ingredients are in it as long as it provides challenges for whatever goal the player sets for themselves. It's inherent to the idea of the freeplay that I have broken the reasons for things to be the way they are and am substituting my own reasons.

[gaelyte]

In our world, space exploration began as an offshoot of military technology. We only managed to devote the resources for this because it was easy to swap out a nuclear warhead for a satellite, that would be really handy for aiming the rest of the nuclear warheads that we didn't convert for this. And we aren't surrounded by giant bugs that are constantly trying to kill us.

So taking military science completely out of the loop for the rocket is just really bizarre. It really gets to the heart of what is the purpose of building the rocket in the first place? I assume that in the campaign biters are going to be there all the time. It's kind of backwards to have the core of the game built around what is best for the freeplay and accommodating people who want to turn biters off rather than what is best for the campaign. People who are playing the freeplay with biters turned off are just building stuff for the sake of building stuff. It doesn't matter what the name of the science pack is or what ingredients are in it as long as it provides challenges for whatever goal the player sets for themselves. It's inherent to the idea of the freeplay that I have broken the reasons for things to be the way they are and am substituting my own reasons.

Mylitary science is completely out of the loot because there's no way to send an intercontinental missile to the bitters. We just need to contact the others humans that are, maybe, not far away.
But if a research permited to have some nuclear bomb to send to the bitters, then it would be logical to search that before.

[theolderbeholder]

[...]because there's no way to send an intercontinental missile to the bitters. We just need [...] a research permited to have some nuclear bomb to send to the bitters, then it would be logical [...]

Uuuuhhh. I like the way you think. I really like it. Someone buy that man a cigar.

(sorry for blatantly misquoting your post)

[bobucles]

I really love the changes. The new science recipes are way better and go a long way to smooth out the tech progression. The recipes are a great way to force automation of all the right items to help guide players on their next steps. Blue science is a pretty hard wall so I appreciate the effort to introduce it in much smaller steps. My only real beef is with

and also decreased its power consumption so it is fun to use even with Personal solar panels early on, and if you invest into the Energy weapon damage, it becomes quite good later.

I feel that personal lasers are in a good spot for the late game and making them globally more powerful will only screw up late game lasers. The real problem is that PSPs in general are just way too weak:
- The difference between a PSP and pocket fusion is a 6-fold increase in energy output per tile.
- PSP costs MORE RESOURCES per kw than pocket fusion
- It was far less trouble to skip the PSP and jump straight to fusion.
- Players typically don't put enough solar energy on their suits because they need to invest far more into PSPs than everything else put together

If those aren't red flags on your tech progression then nothing is. Get rid of the 1x1 panel. Nothing else uses 1x1 equipment other than dev items. Make a shiny 2x2 art for the PSP and buff it 50% for 60kw output. That makes solar panels about 25% as good per tile as fusion, and is good enough to make low level suits that aren't 75% PSPs. Keep the resource cost the same so that PSPs cost roughly the same as every other level 1 module in the suit. Finally, change the suit dimensions to get rid of 1x1 choices entirely:
- 4x6 modular armor (6 modules)
- 6x8 power armor mk1 (12 modules)
- 10x10 mk2 (25 modules)
This is a buff to the lower tier armors across the board. Better low level suits make them a more attractive early game option that players will use longer, while not changing the already adequate late game.

[zyklame]

Nice changes to the science packs.

I'am just left with 2 Questions:

1.) Is it intentional that the space sience packs still neet mitytary to unlock?
Because you might want infinite research on bitter free maps but again still have to setup military witch is contrary to the statement in the FFF.

2.) While donwloading the postcard i found out its not a jpg as the file extension pretends, its webp.
Any reason to name it .jpg anyway? My filebrowser got cunfused by the extension and din't show a preview. (after naming it properly it did)

[Oktokolo]

Yep, armor grids need a resize (regardless of whether there will be more 1x1-sized equipment).
And personal solar is indeed too weak to be usefull for powering personal roboports (it is fine for night vision, an exosceleton or a shield though).

An idea for improving QoL considerably would be to make construction bots not require power. That would not be consistent or make any logical sense. But there would be no more cheese-picking recharing bots from around the player and powering personal roboports using personal solar could become feasible.

[Lizzy]

This has been brought up a few times, but not addressed, so I'd like to bring it up again in its own post: since there has been an effort to bring more attention to modules though science packs, is there also an effort to make them easy to use?

Right now the only action that doesn't require going through machines one by one (or completely rebuilding them) is "deploy one type of module to all empty slots in the machine". No fast-replace like upgrade. No put just one per machine. Also lots of complicated scenarios with mixed-mode modules that I want replaced without altering their ratio. Or maybe even I do want to alter their ratio by exactly one. Or maybe I am going through a partial upgrade of modules and I want to spread the upgrades evenly as they are assembled.

Modules are complicated, sure. But maybe more than they need to be.

[featherwinglove]

Philip017 used Walluv Text.
It is very effective.
featherwinglove is in a bit of a pinch, it looks like it might cry.
featherwinglove toughed it out so you wouldn't feel sad!

...i hope that you balance this a bit better allowing even higher oil richnesses. the max setting on the richness meter at current is still too low for me imo. and once the pumps get drained it's speed modules and beacons all around to try and get the oil out.

I totally agree with this! Even current and older versions demand far too much oil vs. what the map provides relative to other resources. I can't remember if it's vanilla, but I wind up using Bergius process a lot (before it was vanilla for sure!) Unlike Philip017, I tend to tank heavy and light oil for cracking vs. making solid fuel, until I'm ready to make that crapton of solid fuel. ...although a major part of the reason I do this is because I have mods (Bob's and Charcoal Processing), which produce huge piles of solid fuel as a waste product.

don't know if i like the change to using rails in purple science. but it will produce an interesting logistics challenge, and now i'm gonna will need a ton of stone coming down the bus to feed into production science.

Yeah, that's another thing. It seems odd for an advanced research pack to use great piles of primitive stuff.

i have personal robots very early, despite not being able to power them, i mine them by hand, it's alot faster than building the entire blueprint by hand most times.

That's something I haven't heard before. Most people I've seen interested in that problem (speedrunners mostly), will drop machine efficiency for simpler, easier to build machine ratios (1:1 if it isn't too far off), and then spam long rows of machines and long inserters, run back and forth about eight times and *poof* a fully loaded red belt of green circuits is flowing from what was barren steppe only three minutes earlier.

maybe someone will make a mod where you actually have to build the rocket/shuttle/satellite in a separate assembly plants. transport the shuttle to the launcher pad, and then wait for the satelite to do its job and fly back down to the earth and land before collecting the science it discovered instead of a silo doing everything. a little more interactive than current.

I doubt it. It appears most people assume extreme time compression, i.e. a minute of playing time for us is at least several hours for our engineer character (24 hours into 7 minutes works out to 3h26m per minute.)

cant burn small poles/chests any more? wow! i still like the idea of recycling the old crap. i do use a recycling mod that takes apart the old crap and turns it into the product it previously was, although i have had friends that have cheesed this and used the reverse factory with productivity modules to produce free items,

I've only done this with stone via it and Hard Crafting's gravel pulverizing...

Yes... this was clearly Minecraft affecting our ideas.

I was a Skyblocker

[keldor]

I actually think "Rocket Science" is a far superior name to "Space Science" for the final science pack. In real life, the phrase "rocket science" is generally applied to designing and building vehicles that bring things into space, which is precisely the stated purpose of the tier in the dev blog. Wikipedia briefly defines "rocket science" thusly: "Rocket science is a primary branch of aerospace engineering." Finally, the phrase "rocket science" has a strong cultural connotation as being applied to something horrifically complex and difficult and technologically advanced, which works very nicely.

"Space science" is a more rarely used phrase, but it generally refers to the science that is done when you're already in space. Experiments done aboard the International Space Station are "space science". The observations done by probes visiting outer planets are "space science". In a way, it would be the hypothetical next tier up, happening after you're already sending satellites up. In any case, I don't think it means quite what the developers want it to.

[featherwinglove]

I have tried to explain both in the article and here earlier that the “are rails advanced enough” is not just whether the recipe and item unlock consist of early parts (technically it is green tier without oil which is early). It’s also important when do you actually want to use the item - you can start setting up rails early, but there is never too late for that, and at any point adding in the game the rail network is almost certainly allow your factory to scale like crazy (and scale = production). That’s why I think the rail is a fitting ingredient into a science pack all about production.

Reading this is making my head hurt, Firefox hasn't underlined anything so it isn't the spelling... The logic really is just that non-sensical, and that's from a guy who can tank a water dragon big enough to ride fitting into a red-and-white tennis ball TARDIS without it being uncomfortable in there. Burning three kilometres of train track in a science experiment is just that strange!

So I don't see Wube changing it. Rails tick the right boxes per them, and that's as far as it goes.

[featherwinglove]

I, a 2101-hour-player (and counting) say that pickaxe removal was good

Removing pickaxes saves you approximately 30 seconds in 2101 hours, costs Bob a few hundred hours extra work, and hurts thousands of other players.

Removal of pickaxes is extremely not good.

A possible good solution for those who want Factorio to be simpler would have been to default pickaxes to not wearing out in Vanilla while retaining the underlying game mechanics for use by modders.

Tbh while I like that there are a ton of different mods and modders, as a user I do not give a shit about a mod I do not play. I do not see why the devs should leave something in, because "future versions of a specific mod will be hard to update". Sure, it will be a bummer for the mod autor and users, but you can't keep everybody happy at the same time.

Casual players not caring what the truly dedicated think is par for the course and generally assumed anyway. I wish they wouldn't keep jumping in with what reduces to an explicit "We don't love you, please be quiet" like this all the time. It adds nothing to the conversation and just raises blood pressures.

[Avezo]

About PSP in early game, i think primary problem still remains - solar panels give too little power either way. Solution is to introduce some sort of burner-power source to the early armor. And while we're at it, fusion reactor should have something to do with nuclear power IMO.

Or, just make all armors vehicles with equipment grid powered through fuel.

[Jap2.0]

I support the idea of ICBMs to nuke the biters.

[featherwinglove]

I actually think "Rocket Science" is a far superior name to "Space Science" for the final science pack. In real life, the phrase "rocket science" is generally applied to designing and building vehicles that bring things into space, which is precisely the stated purpose of the tier in the dev blog. Wikipedia briefly defines "rocket science" thusly: "Rocket science is a primary branch of aerospace engineering." Finally, the phrase "rocket science" has a strong cultural connotation as being applied to something horrifically complex and difficult and technologically advanced, which works very nicely.

"Space science" is a more rarely used phrase, but it generally refers to the science that is done when you're already in space. Experiments done aboard the International Space Station are "space science". The observations done by probes visiting outer planets are "space science". In a way, it would be the hypothetical next tier up, happening after you're already sending satellites up. In any case, I don't think it means quite what the developers want it to.

I believe "space science" is uttered more often than "rocket science" these days, even if the latter is more widely spread in the vernacular. People who are interested in space stuff utter the phrase "space science" so often that it is probably outnumbers everyone drawling "it's not rocket science" after a couple of drinks at a New Year's Eve party combined. I agree there is definitely a distinction, but I don't think it means much to Factorio, at least not yet. Where modded "space" surfaces might come into play, I have space station mod for example, it then makes sense for "rocket science" to come from the rocket silo, while "space science" comes from these other surfaces, brought back to Nauvis via a downlink station which amounts to a filtered teleporter that excludes "atoms are heavy" items (I like the way Eric Raymond put it.) This restriction being put into Factorio breaks it a bit from reality in that space science comes more from spacecraft than other surfaces, while in Factorio, instantly launched spacecraft would produce only rocket science - although using better spacecraft and possibly better rockets would be able to produce more rocket science for the material used in each launch. In Factorio, the only way to produce space science is to build stuff in space and either have it watch the stars and produce science packs greenhouse-style or burn it in science pack assemblers. Probably having greenhouse-like telescopes produce observations like wood which then get processed into the science pack - one might wish to have a texture pack that makes it look a bit more space- and electronic-like because the usual inserters and assemblers processing "observations" and "data" would seem pretty strange.

But still not as strange as burning three kilometres of train track in a "production" science experiment!

Solution is to introduce some sort of burner-power source to the early armor. And while we're at it, fusion reactor should have something to do with nuclear power IMO.

I agree. I'm always using mods that solve this strangeness.