What happened to Factorio updates?

[neoc]

There is a definitive answer to it, and it's when the game meets their developpers standard of what should be a finished game.
When it's in that state :
1) either the players' standards are met too, in which case they will praise the devs for the good job at delivering a finished product.
2) or the devs' standards were lower than the players' expectations, in which case the devs would be flamed for delivering such a poor quality product.

If I were a dev, I'd prefer the 1 : having higher standards than most of my customers, so that when I say "my game is finished", most will say "Indeed, and it's a gem".

Those are some nice ideas, but that's not how software development works. If you're a software developer who still believes somehow in 'perfection', you're at the very beginning of your career. Things are way too complicated to ever get something perfect here. Have you played Factorio? It's the same here, you'll never have a perfect factory (unless it's only 3 assemblers and a yellow belt).

[Oktokolo]

They either have way to little manpower or aren't allocating it with release in mind.

Or they are currently rewriting their graphics engine and don't want to give us a version wich crashes a lot.

I NEVER had a game crash not related to a mod. NEVER.

Of course not. They do not release unstable code. Instead they take their time and make it work before giving it to you.

The graphics engine is separate from other code. If you make large changes like that you at least at first try not to change anything that's not needed to be changed (for example by using an abstraction layer). That means they can switch between the old and the new engine relatively easily.

They also rewrite their widget toolkit wich probably uses the graphics engine and is used by every GUI in the game at the same time. Also they seem to dump a graphics abstraction layer they are not satisfied with while they are at it (wich is perfect timing).
The NPE and campaign are dependent on the new widget toolkit wich they already use internally. The new tech tree and research queue uses the new widget toolkit. The fluid refactoring is pre-dev. The new terrain generation and map editor are in development.
At least one QoL-Mod (upgrade planner) has gone vanilla and might be production-ready now but probably also uses the new widget toolkit.
They might be able to release the new laser beams - if they did not optimize them by using some new feature of their new and shiny graphics engine (would not surprise me).

All in all it seems the team is currently really busy. If you want new stuff now, try using existing or making new mods.

[neoc]

They also rewrite their widget toolkit wich probably uses the graphics engine and is used by every GUI in the game at the same time. Also they seem to dump a graphics abstraction layer they are not satisfied with while they are at it (wich is perfect timing).
The NPE and campaign are dependent on the new widget toolkit wich they already use internally. The new tech tree and research queue uses the new widget toolkit. The fluid refactoring is pre-dev. The new terrain generation and map editor are in development.
At least one QoL-Mod (upgrade planner) has gone vanilla and might be production-ready now but probably also uses the new widget toolkit.
They might be able to release the new laser beams - if they did not optimize them by using some new feature of their new and shiny graphics engine (would not surprise me).

All in all it seems the team is currently really busy. If you want new stuff now, try using existing or making new mods.

I think you got me wrong, I'm not the guy requesting new features. I just think with a game which is pretty much complete and stable for 2 years now they should finally release. All changes since then were minor compared to the base game, especially if you have the newbie single player in mind - and that one is the audience for release.

[Koub]

Of course not, but different people have different quality standards. I'm the kind of guy with very high standards. I'm rarely satisfied with my own performance. The result is that people who interact with me see me as a high quality deliverer, and I admit I like that. I wouldn't like to be seen as "the guy that delivers average level stuff", whatever I deliver (when I cook, at work, when I used to raid, ...).

There has been quite a set of features added during the last two years, so the game can't be considered as feature complete since 2016. And there are imperfections I can accept from an "in dev" game I wouldn't accept from a finished game. And the imperfections are ... all the things the devs are currently working on. Better GUI, better map gen, better balance (in tech tree, in combat, ...), adding minor QoL features players have massively adopted via mods, ...

[Oktokolo]

I just think with a game which is pretty much complete and stable for 2 years now they should finally release. All changes since then were minor compared to the base game, especially if you have the newbie single player in mind - and that one is the audience for release.

If you have the newbie single player in mind, releasing before the new tutorial is finished looks like a bad idea. The old campaign is just wrong for the average noob. Difficulty is rather random with some nasty spikes. Popups are annoying as hell and can't be brought back once dismissed. There are a lot of boring steps...
As they already said: 0.17 will be the first beta (the current version is still called an alpha). If they find no more bugs, they will probably just tweak the balancing a bit more to get PvP folks happy and go for the first release candidate, maybe even declaring it the final release after some additional testing time.

Yes, they could have rushed it like a AAA studio and released in 2016. But i would miss all the changes since then. And it would not have been the well-polished release they are obviously aiming for.
As a player with a thousand hours in the game i think it is great that they do the final polish and make the game even more superb than it already is.

[neoc]

If you have the newbie single player in mind, releasing before the new tutorial is finished looks like a bad idea. The old campaign is just wrong for the average noob. Difficulty is rather random with some nasty spikes. Popups are annoying as hell and can't be brought back once dismissed. There are a lot of boring steps...
As they already said: 0.17 will be the first beta (the current version is still called an alpha). If they find no more bugs, they will probably just tweak the balancing a bit more to get PvP folks happy and go for the first release candidate, maybe even declaring it the final release after some additional testing time.

Yes, they could have rushed it like a AAA studio and released in 2016. But i would miss all the changes since then. And it would not have been the well-polished release they are obviously aiming for.
As a player with a thousand hours in the game i think it is great that they do the final polish and make the game even more superb than it already is.

I played the campaign about two years ago and I don't remember being annoyed or bored by it. Those were the first steps to hook me up, just like a tutorial in other games. Tutorials and campaigns are not the whole game, and everybody knows that. Factorio doesn't (and obviously, hopefully will never) cater to an audience of three year old babies who's attention has to be grabbed in 3 seconds or they walk away.

The game was in a releasable state back then and it is even more so now. You either like this kind of games, or you don't, and the campaign won't prevent a significant amount of players from giving freeplay a chance.

[Koub]

Ever since I have bought the game 4.5 years ago, I think "this game is totally releasable in its state" when the last stable release for that major version is out. And whenever the next major update stable comes out, I say to myself "The devs were right, the game needed that additional step, retrospectively, the game did lack features back then. But now, it's totally releasable.
And same thing since 0.9.
I know one could argue that the devs could add features for the next 50 years and still find features to add, and that feature creep is a real thing. That's true, but at the same time, most of the features the devs announce feel needed to me. The thing is that as every one of us has different quality standards, every one of us has different opinions on what defines an acceptable release state.

[neoc]

Ever since I have bought the game 4.5 years ago, I think "this game is totally releasable in its state" when the last stable release for that major version is out. And whenever the next major update stable comes out, I say to myself "The devs were right, the game needed that additional step, retrospectively, the game did lack features back then. But now, it's totally releasable.
And same thing since 0.9.
I know one could argue that the devs could add features for the next 50 years and still find features to add, and that feature creep is a real thing. That's true, but at the same time, most of the features the devs announce feel needed to me. The thing is that as every one of us has different quality standards, every one of us has different opinions on what defines an acceptable release state.

"Releasable" has no objective definition. Your first assumption, that the game is releasable, is correct (feature-complete, bug free). Adding features doesn't retroactively change this fact. You're just bending words at this point.

[Oktokolo]

I know one could argue that the devs could add features for the next 50 years and still find features to add, and that feature creep is a real thing.

No, there definitely is no feature creep in Factorio. You just have to look at the most-downloaded QoL mods to see that there are still features missing and aspects of the gameplay that need more tweaking. After removing the tech-dependency for blueprints, they never tweaked the availability of auto-placing of entities to reflect that - hence the popularity of Nanobots.
Another popular QoL mod is Squeak Through.

Nanobots will obviously never make it into vanilla and that is fine. But i would not be surprised if the devs would add some pre-bots way to autoreplace ghosts near the player. The other one could just be assimilated into vanilla as it is - complaints would be unlikely.

[Koub]

I know one could argue that the devs could add features for the next 50 years and still find features to add, and that feature creep is a real thing.

No, there definitely is no feature creep in Factorio. You just have to look at the most-downloaded QoL mods to see that there are still features missing and aspects of the gameplay that need more tweaking. After removing the tech-dependency for blueprints, they never tweaked the availability of auto-placing of entities to reflect that - hence the popularity of Nanobots.
Another popular QoL mod is Squeak Through.

Nanobots will obviously never make it into vanilla and that is fine. But i would not be surprised if the devs would add some pre-bots way to autoreplace ghosts near the player. The other one could just be assimilated into vanilla as it is - complaints would be unlikely.

That's actually also my opinion. There are features I consider necessary that are absent from the game.

[Aeternus]

Last FFF it was mentioned the fluid system is getting an overhaul. Stuff like that needs thorough testing, 'cause if it breaks... that tends to mess up everything. I don't mind them doing the work on their own pace... the game is fun as it is, certainly functional too. Need more features or another way to play? Bobs/Angels mods got you covered, they'll basically transform the game and add an insane layer of additional complexity...

[thedarkbunny]

"Releasable" has no objective definition. Your first assumption, that the game is releasable, is correct (feature-complete, bug free). Adding features doesn't retroactively change this fact. You're just bending words at this point.

There's no purely objective definition of "releasable" in the same way there's no purely objective definition of "feature-complete." There may not even be a purely objective definition of "bug-free," given the number of bug reports closed with "it's counterintuitive and arbitrary, but the algorithm is working as designed so it's not actually a bug."

[neoc]

There's no purely objective definition of "releasable" in the same way there's no purely objective definition of "feature-complete." There may not even be a purely objective definition of "bug-free," given the number of bug reports closed with "it's counterintuitive and arbitrary, but the algorithm is working as designed so it's not actually a bug."

Factorio sold more than 1 million copies (not now, but already 1-2 years ago). I think that speaks for itself regarding to being ready for release. Those people aren't supporters in the majority, those are people who saw the game, maybe tried the demo, then decided it's worth 20 Euros.

[Oktokolo]

Factorio sold more than 1 million copies (not now, but already 1-2 years ago). I think that speaks for itself regarding to being ready for release.

Factorio entered a market without competition. It is like back then when Minecraft was young and sold milions of copies for the same reason (despite actually having some competition - wich its target group did not know about).
Wube holds the monopoly on the great factory builder genre. It certainly looks like beeing a niche genre - but with a billion of people playing worldwide, you can still sell to millions when beeing the only one serving the niche and delivering quality.

[steinio]

Don't feed the troll.

[Oktokolo]

Don't feed the troll.

Fishing is in the game for a reason...

[neoc]

Don't feed the troll.

Wow. That's all you have?