Factorio Forums

I "play" factorio right from first public version and notice no improvements at research part of game, it was mindless grinding at begining, it still mindless grinding.

Entire meaning of current research "system":
To research A you must do B X times, this will cost Y seconds per B.

Every research is same, midgame researched cost 60 second per step with 100 steps, forcing player to grind a lot just to get things he want, this good for games with donate (to speedup process for real money) but not for sandbox.

Most techs about open something usefull that locked for no reason, grind a bit faster or survive zerg waves.

Due bottlenecks like steam engine, miner or furnace you cant speedup process, only option to run more processes.

It will be much more logical if something like prototype system implemented:

0) Crafting by hand without tech not allowed and does not contribute to any tech.

1) Only automated assembly lines provide any tech and can craft items without tech.
2) With each item crafted, tech points generated, if item failed, more tech points generated.
3) Tech points improve chances to craft item, if chance go over 100% - provide chance to create improved version of item, like Furnace 3 or Pistol X.
4) Not needed items can be "researched" inside lab, this provide additional tech points.
5) If tech points reach specific value advanced version of item unlocked, like Iron Furnace or Deep Miner.
6) Tech tracked per type, crafting\researching furnaces does not improve guns and guns wont improve furnaces.


Player can decide to build Steel Furnace directly without unlocking tech first, due to lack of tech, process will suffer from fail rate (that can be over 100% to indicate how far player from needed tech), this will allow player to push research in needed direction and possibly get improved version of furnace long before tech is fully opened.
Items crafter without tech named "X Prototype", with decreased stats but bonus tech points if researched at lab.

Same for anything else, such system will give player a chance to get whatever he want without mindless grinding and much more faster.

Fail rate in case of simple items shoud never reach 90%, soo player will have ability to craft anything right from begining at 10x cost.

Since over time tech will allow better versions of same items, some job to replace outdated components will keep player busy.

Discuss.

Welcome to factorio forums. This post is not personal, it just depicts my opinion on your proposal.

Here go some random thought:

It sounds OK at first, but what would be the new motivation of the player to play?
I played the game to build science pack factories to satisfy the labs and make progress by unlocking colourful squares.
Even for technology that I had little use for.

Nobody likes machines that fail, because of the timing they choose to do so, thanks to Mr. Murphy.
It also sounds like a lot of random processes. That is not allowed because of replay saving.
That means they have to be pseudorandom.. And 0.5 is an example, expressing how difficult it is to tune random processes (NOTE: This line is NOT true, because rolling a dice needs just good entropy)

If you want to go realistic, the non-working or glitchy prototype should be produced. Science needs testing and feedback.
.. transport belts that randomly get stuck, inserters that rust and slow down with time, filter inserters that pickup wrong item from time to time, rockets that explode in your face.. I don't think I really want factorio to be a science simulator..

I am not sure it wouldn't also cause some bottlenecks / parallelism issues (there would be a handful of tech counters to increment with each piece, no matter of the factory scale)

Note: I played a flash game about failing technology that improved. I actually liked it, but the rating suggests that my case was a minority.
http://armorgames.com/play/14200/randobot

I like some of the ideas. But it's difficult to read and understand.
Likes:
- Randomness. You cannot exactly say, when it is finished. (maybe some things are much more ranfdom than others?)

- speedup time for each research steps with something, that can't be researched

- tech-points idea. Hmmm. Jaaa, but not really satisfied. I like that if you do something X, which has nothing to do with Y it influences Y's speed or something... For example when cutting 100 trees you get the axe-research. I don't like tech-points in general. They aren't realistic. But I like definitly the idea, that when you craft by hand, it would not count (of course not the trees).
Hm. I think getting "badges" for achiving X opens up some really interesting things and it could be really hard to achive those badges in the best order and for the new players it gives them the next target, if they don't have an own.

- I like the idea of a first and fast basic research (faster buildup) and then the advanced versions of the items are "researched differently". For me it is clear, that it takes exponentially much material for every step (step 1 1000, step 2 10000, step 3 100000, only then there is a need to really expand to other places...). Edit: I would like that this kind of "research" doesn't take time. When you have the power of inserting 30 items per second (=12 smart inserters), you need about 30 seconds to get 1000 into it, then why not reaching level 1 immediately? When you have the power to feed 10 of those with 300 items per second why not reaching level 2 in under 40 seconds?

- I don't know if I like the idea that crafting an item improves it. But it's not unrealistic: The more you built the better you will get. I would change this in: You craft an item and then you give it into a "test-machine". There it is destroyed (completely, nothing left!) and this improves the item. The more you "test", the better it will get - see above, the exponentially need for items for every step of improvement.

- unlocking. This goes hand in hand with the "test-machine": You cannot research some things, until some other items reach the right level. I think this is a very interesting game-element, when you make a scenario with 2 players, one can only research and the other only test.

- chances to craft item: Hm. I don't know. I suggested in the other thread different types of factories. I see this as one very interesting type of factory/furnace, which may craft unresearched items by chance, but this costs much raw-material. I would give it a try, maybe it's a good idea.

- build stuff without unlocking and then it fails. This is like in reality!! But I really unsure about this. I like randomness, but I see this for a mod. This remembers me to Railroad Tycoon Deluxe, where the cars etc. get's older with the time. I personally don't like that too much. I would prefer, that there are the robots, which replace the stuff, because I don't like the micromanagement. But there are others with other preferences.

Good ideas. Keep going!

RawCode wrote: Due bottlenecks like steam engine, miner or furnace you cant speedup process, only option to run more processes.

Well this can be considered as part of the strategy (how scalable the factory is), the player has to make lot of decisions, like: "Make it smaller and kick the research fast, or start it scalable and save time in the future?" etc etc.

MF- wrote: It sounds OK at first, but what would be the new motivation of the player to play?
I played the game to build science pack factories to satisfy the labs and make progress by unlocking colourful squares.
Even for technology that I had little use for.

I have to agree with this.
Tt might not be obvious, but it is really not easy to make the motivation for the player to spend and expand the production, in transport tycoon, when you have huge company with millions of $$$ you are not really motivated to upgrade your routes or expand, you have unlimited money anyway (Well I was making huge mountain in the center of the map as goal, but it is quite silly).
The main motivation for expensive research and fight is to burn resources and motivate player in higher production.

MF- wrote: It also sounds like a lot of random processes. That is not allowed because of replay saving.

This is not accurate, we use random generator in the game for lot of stuff (fish movements, wind direction changing, some parts of map generator, and much more), the whole trick is, that the generator is deterministic, and the state of the generator is part of the savegame.

MF- wrote: Note: I played a flash game about failing technology that improved. I actually liked it, but the rating suggests that my case was a minority.
http://armorgames.com/play/14200/randobot

Nice little game, I couldn't help myself, I had to finish it

ssilk wrote: - speedup time for each research steps with something, that can't be researched
There should be (and is planned) other aspects changing the speed of machines.

ssilk wrote: - tech-points idea. Hmmm. Jaaa, but not really satisfied. I like that if you do something X, which has nothing to do with Y it influences Y's speed or something... For example when cutting 100 trees you get the axe-research. I don't like tech-points in general. They aren't realistic. But I like definitly the idea, that when you craft by hand, it would not count (of course not the trees).

This is the ethernal Oblivion style versus Baldurs gate style leveling system argument
I'm much closer to the baldurs gate style (you choose what you upgrade), I even have xp mod for oblivion, that changes it
But I understand both.

ssilk wrote: Hm. I think getting "badges" for achiving X opens up some really interesting things and it could be really hard to achive those badges in the best order and for the new players it gives them the next target, if they don't have an own.

- I like the idea of a first and fast basic research (faster buildup) and then the advanced versions of the items are "researched differently". For me it is clear, that it takes exponentially much material for every step (step 1 1000, step 2 10000, step 3 100000, only then there is a need to really expand to other places...). Edit: I would like that this kind of "research" doesn't take time. When you have the power of inserting 30 items per second (=12 smart inserters), you need about 30 seconds to get 1000 into it, then why not reaching level 1 immediately? When you have the power to feed 10 of those with 300 items per second why not reaching level 2 in under 40 seconds?

I love this idea!
Combination of achievements and upgrades per game.
Let's say, you smelt 1000 ore, now you can smelt ore 10% faster (or unlock research for it, or unlock upgrade for furnace).
Hey, you built 1000 circuits? What about upgrade that makes it cost just 2 wires instead of 3?

ssilk wrote: - I don't know if I like the idea that crafting an item improves it. But it's not unrealistic: The more you built the better you will get. I would change this in: You craft an item and then you give it into a "test-machine". There it is destroyed (completely, nothing left!) and this improves the item. The more you "test", the better it will get - see above, the exponentially need for items for every step of improvement.

Good idea as well.

ssilk wrote: - unlocking. This goes hand in hand with the "test-machine": You cannot research some things, until some other items reach the right level. I think this is a very interesting game-element, when you make a scenario with 2 players, one can only research and the other only test.

yes we thing the same way

ssilk wrote: - chances to craft item: Hm. I don't know. I suggested in the other thread different types of factories. I see this as one very interesting type of factory/furnace, which may craft unresearched items by chance, but this costs much raw-material. I would give it a try, maybe it's a good idea.

We need to make the probability extension to the recipe (it would be handy also for other mods, like nuclear energy etc)

ssilk wrote: - build stuff without unlocking and then it fails. This is like in reality!! But I really unsure about this. I like randomness, but I see this for a mod. This remembers me to Railroad Tycoon Deluxe, where the cars etc. get's older with the time. I personally don't like that too much. I would prefer, that there are the robots, which replace the stuff, because I don't like the micromanagement. But there are others with other preferences.

Yes, we definetly don't want the player to spend time by doing manual replacement more and more frequent (as the factory gets bigger), some variation of this idea could make sense if the player could build some kind of repair robots.
Machine would have another property, how long can they go without replacement/maintenence

instead of beeing able to craft ALL items with not much success, maybe beeing able to craft items near in the research tree.

1) Randomness part defined by random seed (read some knuth), storing seed inside replay data will allow to produce consistent results in replay, in machine independant way, soo replay will be always same if implemented properly.
Why you ignore randomness part of world generation?

2) Motivation? who need motivation in sandbox? you just doing what you want without any pressure from game, currently game push player and punish with zerg waves if player ignore pressure.
Also this "sandbox" force player to grind tech packs without any chance to evoid this mindless grinding.
Just imagine sandbox that does not allow you to build anything higher then 10cm before you build 99999 10cm objects?

3)

Nobody likes machines that fail, because of the timing they choose to do so, thanks to Mr. Murphy.

This decision up to player, if you dont want machines that fail, you just dont attempt to craft something without tech and keep crafting tech packs and dump them into labs. (this part still here, cos some techs just can's be bound to crafting something, like inserter cargo or chemisty or something else).

As for useless prototype as crafting result - only option to drop it into lab for research points, in my suggestion it done automatically and instantly - fail produce more research points.

4) Cutting 100 trees wont improve your axe, crafting 100 axes definely will improve crafting process and can allow improved or advanced versions of axe to be crafted, you can discover new more efficient method to craft item or how to craft item with less resources, but all this "bonuses" just little addition and not main target of gameplay.
Game shoud not push player, normal gameplay shoud unlock everything, breaking 3-5 iron axes (and crafting 3-5 new ones) shoud unlock steel axe, there is no motivation to grind axes, as you will get this tech without grinding.

5)

build stuff without unlocking and then it fails

Crafting process can fail, whatever is crafted is stable and working without any chance to fail...

6)

I love this idea!
Combination of achievements and upgrades per game.
Let's say, you smelt 1000 ore, now you can smelt ore 10% faster (or unlock research for it, or unlock upgrade for furnace).
Hey, you built 1000 circuits? What about upgrade that makes it cost just 2 wires instead of 3?

This called mindless grinding.
Why 1000? it can be set to 10005000 to motivate player ever better!

Normal operations shoud unlock everything without need to do something completely useless 1000 times.
Why wait for 1000 ore for 10% bonus?
I just will craft Steel Furnace with 100% speed bonus directly, right from game start, set steam engine, few assembly lines and push furnace tech, after 1000 ore i will have multiple steel furnace prototypes, few non prototype and ability to build more with 0% fail chance.

Motivation that some player need can be switched to something like missile batteries (to bombard creeper bases from safe distance), combat drones or something similar used to assault creeper bases, this a lot better then tech packs.

instead of beeing able to craft ALL items with not much success, maybe beeing able to craft items near in the research tree.

Closer to needed tech, lower fail rate.

It can be hard to craft car without tech, but if you know how to make simple inserter, it wont be hard to make long one, it can be slower or cost more resources, but it's possible.

Same with guns, cars, electric poles and everything similar, if player know how to make wheel, he can craft wheel from anything without failure chance.

Mindless production of science packs? You don't upgrade and improve your assembly lines?
It is a machine doing that (the tedious part) and I am looking after it (the fun part)

Would you prefer if each tech required 1000 unique ingredients instead of 1000 equal ingredients?
To me it sounds like a ton of work for the dev with little to no change for the player.

If you are talking about the sandbox mode of the game - yes, sure, there should be a mode with unlimited resources and everything unlocked from the beginning.

i craft nothing by hand as soon as assembly line unlocked, but crafting tech packs only plain stupid.