Friday Facts #376 - Research and Technology

[XenoCyber]

First I want to thank the devs for all the hard work, its not said enough what a great game Factorio is and how spoiled we are as players by the devs hard work.

I like the idea of triggers for science but if the idea is thematic, i.e. why would you know how to build an oil refinery before you saw oil. Then you can use the same logic for pump jacks. why would you know what a pump jack is before oil. A solution i think would work is that to unlock the oil stuff you need to hand mine oil first (like you mine iron in the beginning).

On a separate note from speed runners and people who watch speedrunners. We hate it when they forget to split and then we dont know what time is what. Some games such as plateup have livesplit integration something similar in Factorio would be awesome.

[TheoMarque]

As last part of FFF, Now i play with lab research productivity mod. Just 2% per level. I am on 13 level that need 768k all non military science packs. I Just stooped spending tons of SP for just next 2% more productivity research, because cost a time and UPS.

Productivity for items can break a ratios, the most fun part for me, i suggest levels by just 25% not 10% to avoid numbers that cannot be divided by 2 and forced to math on dividing by 3 or in some situations by 7.

Other thing is capacity of belts, inserters, rails. On large bases throughput seems to be too slow and push to absurd belt balances that just only consume UPS. Throughput is just glass ceiling for factorio that cannot just optimize by spam more outposts, more trains, more inserters, more other things linearly. Awful, because is lack of creativity and pushes me to just copy & paste parts of factory. Transport lines and systems need to improve.

[DudebroPyro]

Darn, nothing addressing quality being used to increase the light range of lamps. I was really hoping for that

[Molay]

Some people raise the problem that you unlock all the buildings for say oil processing as you go "mine" oil, so then you're not at the place where you produce the buildings to actually process it. Bit of an annoyance for people that are familiar with the game, indeed.

May I recommend to take a bit more inspiritation from SE and enable setting recipes via the map view as a baseline thing (could limit it to the current surface if it's a balance issue otherwise). That way a veteran can set up his mall to produce the things he knows he's going to unlock, and as he plops down the pumpjacks, he can quickly go to map view, set the recipes, and as he comes back home has the machines needed to build out the processing plant. It's a huge convenience how you can set recipes remotely in SE, and conditions etc. I really really hope that gets integrated, whenever I play something not-SE I really miss that feature!

I think this would go a long way in avoiding the minor annoyances with the discovery system, but also generally improve convenience and thus "fun". And as I said, if for balance reasons you can only use it on your current surface, that seems quite reasonable. Could maybe make it usable cross-surface later once you have established some uplink-satellite-whatever thingy that lets you interface with machines on the other planet in that same remote way? So that's a cool new toy to unlock later too.

[spiral_power]

I love end-game.
As I wrote in one comment to the previous FFF, the 20,000 hour limit is too short, so please increase the number of digits more. Make it at least longer than the human lifespan.

By the way, the problem with the current vanilla end-game is that the resource depletion is fast, even if infinite research is proceeded and mining efficiency is improved. I believe that the ideal would be to eventually make (almost) no maintenance of mining stations necessary.
In that aspect, Bob and SE have done well.

[adam_bise]

Just plop down some pumpjacks, power them up, and let them fill up their output pipe. You would have done it anyway.

[svalorzen]

Given the videos, it seems the recycler can output items of a higher quality than the input. If so, I'm not a huge fan of this idea. Conceptually, it'd make much more sense if a quality-enhanced recycler could improve the chance of extracting high-quality modules from high-quality items, without generating them out of thin air.

[jodokus31]

One question comes to my mind, when thinking about the trigger techs:
Do I have to select it beforehand? And/or can it trigger, while researching something else? IMO, it should be totally automatic, without selecting anything, and finish, while researching something else.

Maybe it's also time to hide further research a bit, esp. for new planets, because it could contain spoilers, when digging through the tech tree?


Regarding the quality concept. I think, it can be a nice mid/endgame resource sink, which has to compete with research. It's a bit similar to producing modules vs. research. And now the productivity researches are also competing with other researches like mining prop, artillery range.

I'm not sure, if it has to be 5 tiers of quality and if it is really useful to build factories with intermediate stage tiers.

I also miss the opportunity to buff belts with quality. IMO it would be enough, if you don't have different speeds at each stage, but at least at some stages.
And blue belt only, would also be enough. I would imagine something like:
- normal 45 items per sec.
- uncommon 60 (double red belt)
- rare 90 (double blue belt)

[ManaUser]

Regarding quality names (yes I'm still stuck on that), I'd just like something fairly simple and neutral, like: Standard, Good, Fine, Exceptional, Perfect

I've noticed a lot of the player suggestions have had a negative term (such as "rough") for the lowest, but that seems misleading.

On the research queue: Yay!

On the research triggers, I don't totally understand what problem this is meant to solve, but I guess I'll have to see it in action.

[Panzerknacker]

Great news. I always thought that the beginning of the game was a bit overwhelming (for new players) with so many recipes available from the start. And the electronics research never did anything except unlocking other research. Also, burner tech kinda feels redundant because of how quickly you can switch to electric. I would not mind if the research for the electric versions would be rather costly.

Indeed the green science section of the game would potentially unlock complicated tech like automated trains and logic modules which I never even found time to implement in that phase considering how cheap red/green science is to progress through (especially on default difficulty). Good to read that this is being looked at.

I was really hyped by the Quality announcement btw, I think it is a very interesting addition that will allow for many different factory designs and also introducing some aspects of randomness in the game is a good thing, I liked your slot machine comparison.

About the new infinite research, one question that that comes to my mind is, how can you have inifinite productivity research because last FFF you mentioned there would be a productivity cap of 300% due to the addition of recyclers?

[BlacKcuD]

This sounds and reads amazing. This will be a day 1 buy for sure.

[DanGio]

Wow ! Excellent.

Some feedback :
- Having to research through many technologies before getting one valuable item is something I generally hate in mods, I prefer less but more powerful technologies. So I'd like to see radars and repair packs, and even lamps, under one, more expensive, "Base upgrade", or "Enhanced Field Inspection" technology.
- I wonder if the steel plate productivity research will also work in non-electric furnaces like the stone furnace.
- With quality and productivity research, it seems to me that Factorio is really entering a new era, where the bases are gonna get more organic and almost impossible to plan exactly, as factories will constantly change speed and productivity. It is a bit frightening but also exciting

[Rebmes]

I have to pitch in. I am a long time player and I also think Quality will add some excitement and depth to the game, in a way that is easy to add to existing factories, easy to get into, and indeed optional. I don't use bacons much myself for example, but in some situations I really enjoy them. Hopefully your explanation today will answer some of the questions that people had about Quality - good luck with the names, though. <- Perhaps what people are looking for are crafting terms rather than looting terms, or perhaps even factory terms of your own. After all, you're still basically writing the book with this stuff.

Thanks for having faith in your player base and holding steady, and thanks for the great example of a Quality Microfactory.

As for research, currently there's not much point to automate red science before green science, just do it at the same time and location, with the same resources. It will be interesting to see a different start, but I'm a little worried about how many hoops we'll be expected to jump through.

lastly, I like this comment above by svalorzen.

Given the videos, it seems the recycler can output items of a higher quality than the input. If so, I'm not a huge fan of this idea. Conceptually, it'd make much more sense if a quality-enhanced recycler could improve the chance of extracting high-quality modules from high-quality items, without generating them out of thin air.

[kovarex]

For the steam-power research (50 iron plates); would it show the progress then? so if you crafted 20 or so, it would be 40% filled? The top-right only shows one research technology, but now multiple can be 'researched' at once…

Yes, the technology shows the progress the same way as when you normally research.

One question comes to my mind, when thinking about the trigger techs:
Do I have to select it beforehand? And/or can it trigger, while researching something else? IMO, it should be totally automatic, without selecting anything, and finish, while researching something else.

Selecting it is optional (if you want to see the progress in the top-right corner of the game).
When there is no other research in progress, the first time you start progressing the trigger-based technology it gets automatically selected.
If there is already other research in progress, it doesn't interfere, and just quitly progresses.

On top of that, we added a notification of when any research is finished, so even when other research is in progress, you won't miss what happened completely.

Maybe it's also time to hide further research a bit, esp. for new planets, because it could contain spoilers, when digging through the tech tree?

We considered it, but decided to not do it, as the decision to keep everything open/searchable from the start still holds.
The main reason is to allow even new players to make informed decision about what do do next. And since choosing the planet is a bigger deal then choosing a technology, it is fair to let the players know what will each planet offer.

[thedoh]

From a game design point of view I don't think I like the implementation of research triggers for oil because I don't think it makes sense for the engineer to know how to make a pumpjack but not know how to use the oil. In my opinion keeping oil behind lab-based research makes more logical sense.

I have several thousands of hours (literally) of Sea Block experience under my belt. I've completed several playthroughs, so I've been exposed to the way Sea Block uses research triggers. The way Sea Block uses the concept is as a way to teach the player the fundamentals of the game (mod): The player must be able to make basic circuits, and so the first few techs direct the player there. I don't think that makes sense for oil in the base game because by that point the player should already be well established. Where I think the triggers make the most sense is as described for electronics and steam engine tech; the first foundational techs in the game.

In my first Sea Block playthrough I made an Ore Processor because it was a tech I could research. When I placed the building I couldn't set any recipe on it! "What a SCAM!," I thought. I wasted precious steel on the thing that didn't DO anything. In the base game, will a new player place a pumpjack and then not know what to do with it?

In my opinion, the triggered techs should be reserved for the first couple of recipes because the design (of triggered techs) must be in service of teaching the player the foundation of the game. By the time the player gets to oil, they should already be familiar with how to build things and explore the tech tree to understand tech ordering.

[adam_bise]

I don't like the trigger techs, but I find them acceptable.

One of my favorite aspects of Factorio is the fact that you are given lots of cool tools and the freedom to do whatever you want with them. The game doesn't try to control your playstyle by restricting you from doing things that might not make a lot of sense.

In a previous playthrough, I decided to have a central water supply. It was ridiculous for sure, but it worked and provided extra challenges which was my goal. While restricting the player from making refineries without a single drop of oil in them is something I don't like (attempting to restrict my playstyle), it is acceptable because there is literally zero reason to have dry refineries.

Still, I would rather have the freedom of making a zero-sense dry refinery lol. I don't really see much of an advantage for this, other than avoiding some new player confusion.

[jodokus31]

One question comes to my mind, when thinking about the trigger techs:
Do I have to select it beforehand? And/or can it trigger, while researching something else? IMO, it should be totally automatic, without selecting anything, and finish, while researching something else.

Selecting it is optional (if you want to see the progress in the top-right corner of the game).
When there is no other research in progress, the first time you start progressing the trigger-based technology it gets automatically selected.
If there is already other research in progress, it doesn't interfere, and just quitly progresses.

On top of that, we added a notification of when any research is finished, so even when other research is in progress, you won't miss what happened completely.

Sounds great!

Maybe it's also time to hide further research a bit, esp. for new planets, because it could contain spoilers, when digging through the tech tree?

We considered it, but decided to not do it, as the decision to keep everything open/searchable from the start still holds.
The main reason is to allow even new players to make informed decision about what do do next. And since choosing the planet is a bigger deal then choosing a technology, it is fair to let the players know what will each planet offer.

OK, sounds reasonable. After all, it's not really an exploration game.

[Queuebee]

Hey Amazing Factorio Team! Wonderfully well explained FFF as always. The quality of information and the depth you go into is just top-notch. It's like unwrapping a present every time.

I have some remarks and questions written below here.

Quality System

• How does introducing randomness into crafting affect UPS?
• Is the randomness tied to the maps seed or random each time, have you thought about how this affects speedruns? (I to watch speedruns)

Research on trigger
Unlocking tech by triggers sounds great, but I wonder how you will decide on the specific the number of items/actions for each trigger.
Would triggers with a certain item/action threshold get the same effect from technology price multipliers or will there be a seperate multipliers for the triggers?

Will it be easy to see the difference between researchable and triggerable technologies (visually in the tech tree)?

Research Queue Tweaks
We already have a setting to make the technology screen pop up when research is completed, but what if we could also have a toggle to pause the game on each new research completed (on by default for new players so they don't miss it when they queued up a lot of research)?

Can't wait to play and test these new ideas.

Have a great weekend!

[nimeroni]

Now we just need an alternative for the boring re-occuring beacon build.

Why would you want an alternative to the beacon ? It have a specific niche (mass producing intermediates), which it does very well. And it will stay in its niche, since speed modules ruin quality chance.

(If anything, the new productivity will make the beacon even more valuable.)

The oil example given is a bit problematic. If you think about how this process works there's some annoying steps for the player.
1. Research pumpjacks
2. Go place pumpjacks. But then there's nothing you can do with the oil.
3. Then go back and make chem plants oil refineries.
4. Then go back to the pumpjacks and actually use the oil.

There's some distinctly unnecessary roundtripping here of the player where they have the oil and the tech, but they need to make the building but are standing in the wrong place to do so, and are making the oil but can't do anything with it.

I'd suggest that the unlockable techs should only be recipes, not buildings, so the player can pre-build everything they need, at least for buildings the player might conceivably build outside their main base.

1. Research pumpjacks
2. Go place pumpjacks
3. Go back to the base while laying down a line of pipes (and underground), and branch it to storage (note that those are already researched)
4. Make chem plants / refineries
5. Use oil

I mean, what psychopath refine oil at the outpost instead of at the base ?

Given the videos, it seems the recycler can output items of a higher quality than the input. If so, I'm not a huge fan of this idea. Conceptually, it'd make much more sense if a quality-enhanced recycler could improve the chance of extracting high-quality modules from high-quality items, without generating them out of thin air.

Conceptually, you might have high quality components that can't give everything they can because another bit of the item is not up to snuff. So you separate the good from the bad.

Game design wise, it make recycle loop slightly less bad.

[rldml]

Galdoc has a really inetesting alternative to the quality naming scheme

Optimized has to be the last quality tear. Optimized is an absolute maximum value and cannot be increased