Factorio Forums

blazespinnaker wrote: ↑

Sat Jan 02, 2021 3:26 am

Theoretical maximum throughput
As there is currently no hitbox for the trains, and train can clip through each other without collision, this means a busy railway can have infinite item throughput.

like I said, you haven't played it. the stations do *not* have infinite throughput, they're very limited. they're saying that a section of railway has infinite number of trains that can operate on it, which is also a stupid thing to say, the game will bottleneck you in other ways far before that, like electric network use or the ability to LOAD your trains.

Fight? Hardly. I find that a rather unfortunate thing to say that creates tension where there shouldn't be any. I find ptx0 perspective very interesting. I would have stopped long ago if I did not.

As for the style, socratic debate is a terrific method of discussion for sure. It's a natural outcome when folks hold strongly divergent views.

And comparing and contrasting different approaches to the problem of factory building video games is cool, especially in the context of future direction. Yeah, it's a little off topic, but isn't this the place for it?

But, sure, it's a lot of volume for just two people I guess. I can certainly pause my side of the conversation. Too bad though, if you ask me.

blazespinnaker wrote: ↑

Sat Jan 02, 2021 11:45 am

But, sure, it's a lot of volume for just two people I guess. I can certainly pause my side of the conversation. Too bad though, if you ask me.

Koub didn't say you had to stop, though, did he? He just said to continue it in PM, not here. Y'all filled a whole page just by yourselves, which seems to indicate that whatever you're "discussing" at this point, nobody else is interested.

Kyralessa wrote: ↑

Sat Jan 02, 2021 2:04 pm

blazespinnaker wrote: ↑

Sat Jan 02, 2021 11:45 am

But, sure, it's a lot of volume for just two people I guess. I can certainly pause my side of the conversation. Too bad though, if you ask me.

Koub didn't say you had to stop, though, did he? He just said to continue it in PM, not here. Y'all filled a whole page just by yourselves, which seems to indicate that whatever you're "discussing" at this point, nobody else is interested.

Oh, I dunno if that's necessarily true but I can see how you might reach that conclusion given the speed and volume of posts.

Anyways, whether we agree with the decision here or not, Koub has been given a role of admin by Wube. We should respect that, stop here and get back to something more central to the thread topic.

So while I obviously can't stop people from stretching this all out further and potentially getting the thread shut down - I personally won't be replying to posts further related to this adjudication.


Edit: Just ran across this old post on viewtopic.php?p=483391#p483391 Talks about optimisation and abstraction. Great thread! It sort of also suggests that integrating concepts from mods will be a fruitful source of ideas for future direction.

My impressions about Satisfactory vs Factorio.
They both good and fun to play.
But I prefer Factorio more, because here I can build a factory more different ways. It have blueprints and bots, that allow me to build much bigger factory. There are some methods of "copy" in SatisfF but they are from mods and works questionable way. Also mods for Factorio are nice and well made, making game fun to play again and again.
In SatisF it's a nice 3D view, and world exploration (because it's hand made), but base building, when it grows big, becomes too complicated, from my point of view.

I played both games, Factorio quite a few times and Satisfactory - 3 or 4 rather big bases.

I also like to mention Space Engineer - that also a game somewhat similar, since it has mass production, and circuit logic. But due to game engine limitations it does not allow to build big moveable attachments. This seriously gimps attempts to make things like "walking excavator" or "mining excavator", i.e. large scale mining machines that are attached to a main base on a planetary surface.

@Koub if Ptx0 and Blazespinnaker are both fine with it what are your opinions on just splitting the topic? It seems like it should meet the on-topic requirements of the off-topic subforum.

Jap2.0 wrote: ↑

Mon Jan 04, 2021 3:29 am

@Koub if Ptx0 and Blazespinnaker are both fine with it what are your opinions on just splitting the topic? It seems like it should meet the on-topic requirements of the off-topic subforum.

Sure, I'd be OK with that. As Kyralessa pointed out, I didn't say the discussion was not interesting, just that it hijacked a more global thread.

Yes, a thread for comparing different approaches to factory building games would be cool. I'll be observing and not participating however. Sometimes, it's a good idea to take a step back and realize it's probably a good idea to just let things go. (Great movie! =)

blazespinnaker wrote: ↑

Thu Dec 17, 2020 8:19 am

McDuff wrote: ↑

Thu Sep 17, 2020 3:01 pm

The obvious next level for Factorio would be a 3D jump.

Oh good lord. If anything, they should go reverse. Fewer graphics / animations.

Exactly what about factorio and its game play makes you think people care about graphics over mechanics?

3D isn't about graphics, it *is* about mechanics!

Z-levels for belts, inserters, machines, railways etc. It could be flat polygons for all it matters (although I'm sure they'd make it look good).

blazespinnaker wrote: ↑

Thu Dec 17, 2020 7:55 am

Here are some things I'd like to see -

- Smarter biters (a big brain biter that does surgical attacks on your infra that you have to outsmart to defend your bases
- An infinite science on communicating with biters which allows you eventually make peace with them as long as you stop polluting. (Say, after 64 rocket launches. Setting which allows you to decrease this, of course.)
- A polluting robotic force from your home system that invades and starts attacking the biters, but since you've come to peace with the aliens, you help defend them.
- The robotic force is extremely intelligent, builds its own bases using technology similar to what you have
- World building type technology which let's you start managing the world on a global scale.
- finite global resources. solar system / galaxy exploration so you start colonizing planets.
- meta assembler, see below

These things don't need a "Factorio 2" - they're either doable in the base game or moddable in (or have already been modded in).

The idea of a "3D" Factorio is simply because, as others have pointed out, you can't just update the 2D Factorio to suddenly have usable Z-levels. The best you can do is different surfaces. And you can do a lot with those, to be sure, so yeah, I would say a lot of expansion ideas can just be done in Factorio 1.

McDuff wrote: ↑

Mon Jan 04, 2021 6:23 pm

Z-levels for belts, inserters, machines, railways etc. It could be flat polygons for all it matters (although I'm sure they'd make it look good).

Agreed, absolutely, and I make the same point later on in the thread. My point really was that 3d, when not about mechanics (as all too frequently occurs due to the lure of eye candy), is an unpleasant opportunity cost that will reduce the rate of features.

I give an example from another game, where a certain type of standard factorio feature has no collision detection. This is typical, as physics in 3d are quite expensive.

The idea of modding in this stuff is interesting, but I think looking at mods as a silver bullet a bit overly optimistic. Lack of access to core code, no $$ incentive and bugs, forwards compatibility issues and community bifurcation are all issues with mods. That said, I'm writing a mod myself.

blazespinnaker wrote: ↑

Mon Jan 04, 2021 6:29 pm

[..] standard factorio feature has no collision detection. This is typical, as physics in 3d are quite expensive.[..]

like bots?

McDuff wrote: ↑

Mon Jan 04, 2021 6:28 pm

blazespinnaker wrote: ↑

Thu Dec 17, 2020 7:55 am

Here are some things I'd like to see -

- Smarter biters (a big brain biter that does surgical attacks on your infra that you have to outsmart to defend your bases
- An infinite science on communicating with biters which allows you eventually make peace with them as long as you stop polluting. (Say, after 64 rocket launches. Setting which allows you to decrease this, of course.)
- A polluting robotic force from your home system that invades and starts attacking the biters, but since you've come to peace with the aliens, you help defend them.
- The robotic force is extremely intelligent, builds its own bases using technology similar to what you have
- World building type technology which let's you start managing the world on a global scale.
- finite global resources. solar system / galaxy exploration so you start colonizing planets.
- meta assembler, see below

These things don't need a "Factorio 2" - they're either doable in the base game or moddable in (or have already been modded in).

The idea of a "3D" Factorio is simply because, as others have pointed out, you can't just update the 2D Factorio to suddenly have usable Z-levels. The best you can do is different surfaces. And you can do a lot with those, to be sure, so yeah, I would say a lot of expansion ideas can just be done in Factorio 1.

I might alone here.... but I actually think they could on a functional level. The bigger problem would be performance. If I understand it correctly, the engine already handles everything in chunks. So aside from having to upgrade the coordinate system and chunk processing to include a Z value.... its no that much different then having a large number of chunks to process. Most trouble will probably be changing underground belts and pipes to check for open blocks along the path; wouldn't even need to have break out into individual entities since the still have fixed ingress/egress points for product. Although... that does raise the question of how the current system handles capacity and animation of products passing through it.

starlinvf wrote: ↑

Thu Jan 07, 2021 10:53 pm

McDuff wrote: ↑

Mon Jan 04, 2021 6:28 pm

blazespinnaker wrote: ↑

Thu Dec 17, 2020 7:55 am

Here are some things I'd like to see -

- Smarter biters (a big brain biter that does surgical attacks on your infra that you have to outsmart to defend your bases
- An infinite science on communicating with biters which allows you eventually make peace with them as long as you stop polluting. (Say, after 64 rocket launches. Setting which allows you to decrease this, of course.)
- A polluting robotic force from your home system that invades and starts attacking the biters, but since you've come to peace with the aliens, you help defend them.
- The robotic force is extremely intelligent, builds its own bases using technology similar to what you have
- World building type technology which let's you start managing the world on a global scale.
- finite global resources. solar system / galaxy exploration so you start colonizing planets.
- meta assembler, see below

These things don't need a "Factorio 2" - they're either doable in the base game or moddable in (or have already been modded in).

The idea of a "3D" Factorio is simply because, as others have pointed out, you can't just update the 2D Factorio to suddenly have usable Z-levels. The best you can do is different surfaces. And you can do a lot with those, to be sure, so yeah, I would say a lot of expansion ideas can just be done in Factorio 1.

I might alone here.... but I actually think they could on a functional level. The bigger problem would be performance. If I understand it correctly, the engine already handles everything in chunks. So aside from having to upgrade the coordinate system and chunk processing to include a Z value.... its no that much different then having a large number of chunks to process. Most trouble will probably be changing underground belts and pipes to check for open blocks along the path; wouldn't even need to have break out into individual entities since the still have fixed ingress/egress points for product. Although... that does raise the question of how the current system handles capacity and animation of products passing through it.

Alright maybe technically, but it's a problem like trying to retrofit a car to fly. It might seem like more work to build a plane from scratch, but ultimately it's the better choice.

McDuff wrote: ↑

Thu Jan 07, 2021 11:18 pm

starlinvf wrote: ↑

Thu Jan 07, 2021 10:53 pm

McDuff wrote: ↑

Mon Jan 04, 2021 6:28 pm

blazespinnaker wrote: ↑

Thu Dec 17, 2020 7:55 am

Here are some things I'd like to see -

- Smarter biters (a big brain biter that does surgical attacks on your infra that you have to outsmart to defend your bases
- An infinite science on communicating with biters which allows you eventually make peace with them as long as you stop polluting. (Say, after 64 rocket launches. Setting which allows you to decrease this, of course.)
- A polluting robotic force from your home system that invades and starts attacking the biters, but since you've come to peace with the aliens, you help defend them.
- The robotic force is extremely intelligent, builds its own bases using technology similar to what you have
- World building type technology which let's you start managing the world on a global scale.
- finite global resources. solar system / galaxy exploration so you start colonizing planets.
- meta assembler, see below

These things don't need a "Factorio 2" - they're either doable in the base game or moddable in (or have already been modded in).

The idea of a "3D" Factorio is simply because, as others have pointed out, you can't just update the 2D Factorio to suddenly have usable Z-levels. The best you can do is different surfaces. And you can do a lot with those, to be sure, so yeah, I would say a lot of expansion ideas can just be done in Factorio 1.

I might alone here.... but I actually think they could on a functional level. The bigger problem would be performance. If I understand it correctly, the engine already handles everything in chunks. So aside from having to upgrade the coordinate system and chunk processing to include a Z value.... its no that much different then having a large number of chunks to process. Most trouble will probably be changing underground belts and pipes to check for open blocks along the path; wouldn't even need to have break out into individual entities since the still have fixed ingress/egress points for product. Although... that does raise the question of how the current system handles capacity and animation of products passing through it.

Alright maybe technically, but it's a problem like trying to retrofit a car to fly. It might seem like more work to build a plane from scratch, but ultimately it's the better choice.

ultimately its not that cut and dry. Is what you're changing foundational, or extensible? Because your example is confusing whats essentially an optimization problem for a foundational one. If I were to strap big enough fans, and balance thrust, I can make a car fly relatively easily. The reason you don't see this is almost entirely due to a "use case" perspective that currently values the other solutions more. Especially in how air travel is highly regulated due to safety, increasing the barrier to use it; compared to the much lower one of cars, and the incidentally much higher rate of accidents and damage. You fix the barriers to making ground and air transport comparable to use for a personal vehicle, then you are substantially multiplying the capacity of infrastructure, while making it vastly more accessible.

Hence..... cost benefit analysis is merely questioning optimization, not capability. That same model is also reflected throughout the entire electronics industry; especially due to how highly valued backward compatibility is. The first iPhone was just an iPod that could make phone calls. They didn't have to completely redesign the thing to be a phone.... they just added a radio and a UI to the touch screen, and BOOM, now it does phone stuff. And it wasn't even a good phone either. It was too bulky to be comfortable, wasn't easily intuitive to existing phone users, and was stupidly expensive as either a phone or a MP3 player. But it was "good enough".

They could had redesigned it from the ground up to be better out of the gate, but there was nowhere near enough incentive to justify the time, effort and expensive on an unproven product. And as time when on, they changed the design to add functionality and optimize existing ones.

So why rebuild the entire engine from scratch when you can still extend the functionality for an acceptable cost? If you look over the history of Minecrafts development during the Notch eras, you can see clear demarcations where they had to restructure something. But the majority of development is spent building up or upon existing existing systems. This is even reflected in its mods, like Forge solving the blockid limit without having to change how the rest of the engine works.

So it won't be a Factorio 2, more like a Factorio extended :
viewtopic.php?f=38&t=95705

Factorio 2 should focus on multiplayer. Something similar to biter battles. I have so many ideas that would be so fun