• Changes
  
  • The area of 400X400 tiles is explored when the game starts.
    
  
  
• Bufixes
  
  • Fixed that merging two electric network didn't merge the statistics. more
    
  • Fixed the cyclic/overlaping win/lost sound sample when the game is finished in multiplayer. more
    
  • Fixed that extra pipe covers could be drawn on top of connected pipes. more
    
  • Fixed misaligned turrets in the 4. New hope mission. more
    
  • Fixed some inconsistencies related to zero-signals in circuit networks. more
    
  • Fixed inconstency between the way inserters and logistic robots picked items from inventories. Logistic robots now prefer items at the end of the inventory and ignore inventory limit. more
    
  • Fixed LuaForce::clear_chart() would crash game. more
    
  • The report of different mods when trying to connect to multiplayer game is now scrollable when needed. more
    
  • Fixed Cargo Wagon Inserter input output inconsistency. more
    
  • Fixed the mod difference reporting when connecting to multiplayer game. more
    
  • Found workaround for issue in Visual C++ math library that was causing crashes on unpatched Windows 7. Service Pack 1 for Windows 7 is no longer required.
    
  • Fixed that placing a rotated blueprint containing a splitter was not possible in some cases more
    
  • Fixed crash after revive by player port when personal roboport is equipped more
    
  • Optimised the map bitmap refresh logic 16 times. (the freeze after loading a game or resizing window while playing big saves).
    
  • Fixed that the "you joined a paused game" message would display on all peers, rather than just the peer who just joined more
    
  • Fixed slow sprite loading on Direct3D more
    
  • Fixed the documentation of LuaGameScript::show_message_dialog more
    
  • Fixed the order of parameters of some functions in the documentation more
    
  • Fixed that the research screen would pre-select Automation even though it was disabled more
    
  • Fixed rare crash that could happen when assembling machine recipe was reset the same tick as autosave started. more
    
  • Fixed the laser/discharge defense names in the set filter dialog. more
    
  • Better error when wrong bounding box definition is given. more
    
  • Creating message dialog with non existent image doesn't crash the game anymore. more
    
  • Localisation errors will no longer stop the game, the result string will just contain the error. more
    
  • Fixed that the stone wall research was disabled in the New hope campaign level 4, so gates weren't researchable. more
    
  • The starting value of text property of textfield is properly set based on the input. more
    
  • Fixed that changing request slots of opened entity through script didn't update the ui. more
    
  • Fixed map exchange string problems with mods more
    
  • Fixed freezes on exit with recent NVidia drivers on Gentoo and Arch Linux more
    
  • Proper notification when quick bar slot can't be selected as there is no place to put the item in the cursor.
    
  • Fixed crash on Linux after running xrandr --off more
    
  • Fixed Rocket Silo GUI skipping 53%. more
    
  • Another attempt at fixing X11 copy/paste more
    
  • Better message when the server leaves a multiplayer game more
    
  • Fixed the bad values of defines.furnace_source, defines.furnace_result and added missing defines.furnace_modules. more
    
  • Fixed that modded saves could get unloadable without mods when they have entities removed in the loading transitions that were in more than one network.
    
  • Fixed that it was possible to have unlimited range with melee weapon after running out of ammo. more
    
  • Fixed that LuaPlayer::remove_item didn't remove from cursor stack when the player was using god controller. This also fixes the bug in the tight spot scenario. more
    
  • Fixed black screen after UAC poped up in main menu more
    
  
  
• Scripting
  
  • Documented extra unit group status return values. more
    
  
  
• Modding
  
  • Added action_range to mining tool prototype (with the default of 1.5).
    

You can get experimental releases by selecting the 'experimental' beta branch under Factorio's properties in Steam.

• Changes
  
  • The area of 400X400 tiles is explored when the game starts.
    
  
  
• Bufixes
  
  • Fixed that merging two electric network didn't merge the statistics. more
    
  • Fixed the cyclic/overlaping win/lost sound sample when the game is finished in multiplayer. more
    
  • Fixed that extra pipe covers could be drawn on top of connected pipes. more
    
  • Fixed misaligned turrets in the 4. New hope mission. more
    
  • Fixed some inconsistencies related to zero-signals in circuit networks. more
    
  • Fixed inconstency between the way inserters and logistic robots picked items from inventories. Logistic robots now prefer items at the end of the inventory and ignore inventory limit. more
    
  • Fixed LuaForce::clear_chart() would crash game. more
    
  • The report of different mods when trying to connect to multiplayer game is now scrollable when needed. more
    
  • Fixed Cargo Wagon Inserter input output inconsistency. more
    
  • Fixed the mod difference reporting when connecting to multiplayer game. more
    
  • Found workaround for issue in Visual C++ math library that was causing crashes on unpatched Windows 7. Service Pack 1 for Windows 7 is no longer required.
    
  • Fixed that placing a rotated blueprint containing a splitter was not possible in some cases more
    
  • Fixed crash after revive by player port when personal roboport is equipped more
    
  • Optimised the map bitmap refresh logic 16 times. (the freeze after loading a game or resizing window while playing big saves).
    
  • Fixed that the "you joined a paused game" message would display on all peers, rather than just the peer who just joined more
    
  • Fixed slow sprite loading on Direct3D more
    
  • Fixed the documentation of LuaGameScript::show_message_dialog more
    
  • Fixed the order of parameters of some functions in the documentation more
    
  • Fixed that the research screen would pre-select Automation even though it was disabled more
    
  • Fixed rare crash that could happen when assembling machine recipe was reset the same tick as autosave started. more
    
  • Fixed the laser/discharge defense names in the set filter dialog. more
    
  • Better error when wrong bounding box definition is given. more
    
  • Creating message dialog with non existent image doesn't crash the game anymore. more
    
  • Localisation errors will no longer stop the game, the result string will just contain the error. more
    
  • Fixed that the stone wall research was disabled in the New hope campaign level 4, so gates weren't researchable. more
    
  • The starting value of text property of textfield is properly set based on the input. more
    
  • Fixed that changing request slots of opened entity through script didn't update the ui. more
    
  • Fixed map exchange string problems with mods more
    
  • Fixed freezes on exit with recent NVidia drivers on Gentoo and Arch Linux more
    
  • Proper notification when quick bar slot can't be selected as there is no place to put the item in the cursor.
    
  • Fixed crash on Linux after running xrandr --off more
    
  • Fixed Rocket Silo GUI skipping 53%. more
    
  • Another attempt at fixing X11 copy/paste more
    
  • Better message when the server leaves a multiplayer game more
    
  • Fixed the bad values of defines.furnace_source, defines.furnace_result and added missing defines.furnace_modules. more
    
  • Fixed that modded saves could get unloadable without mods when they have entities removed in the loading transitions that were in more than one network.
    
  • Fixed that it was possible to have unlimited range with melee weapon after running out of ammo. more
    
  • Fixed that LuaPlayer::remove_item didn't remove from cursor stack when the player was using god controller. This also fixes the bug in the tight spot scenario. more
    
  • Fixed black screen after UAC poped up in main menu more
    
  
  
• Scripting
  
  • Documented extra unit group status return values. more
    
  
  
• Modding
  
  • Added action_range to mining tool prototype (with the default of 1.5).
    

You can get experimental releases by selecting the 'experimental' beta branch under Factorio's properties in Steam.

Hello,

Your friday facts today are written by the community manager Klonan. He's been with us here in the office for nearly 2 months now, working hard marketing the steam release and responding to all the media inquiries sent to our email. He's going to write a little about the behind the scenes of the steam release.


So after the steam release, we had a rather large increase in traffic to all our servers, as we had hoped. Shown below are a couple of graphs illustrating the data usage for the forums. As you can see, after release we had a large influx of players onto the forum, and a lot of new users looking to download mods for their new purchase. We have also seen this increase in activity in other ways, with an evident increase in user posts and new users signing up



You can also see from the reddit stats (many thanks for head moderator Zirr for providing them) that the steam release caused a big surge of interest in our game, with a lot of new unique users rushing in, and a rather large inflix of views, which has sustained over the last 3 weeks. It's also easy to see this when you check it out for yourself, where, compared to a few months ago, the level of activity and interaction is much higher.



This traffic hasn't come as a big surprise to us, as we suspected that the steam release would cause a lot of attention pointed towards us. The sales numbers have been quite kind to us, and while we can't release specific numbers, we can make a short comparison. In the last 3 weeks selling the game on steam, we have sold more copies than in the lifetime of our website. This is an incredible number for us, and we owe it to our community to thank them for their support over the years.


But it hasn't all been good news.After the steam launch, the traffic of our main server increased a fair amount, as was expected. The first 'normal' Friday facts after the steam release demonstrated the 'Loader', and in typical form included with it a 25MB gif showing it off. Normally a gif so large would be no issue, but due to the increase in site traffic, along with hotlinking the gif to steam,it led to an overwhelming level of server strain.



Over the course of 4 days, the gif was downloaded over 250,000 times, totaling a whopping 8.0 Terabytes of data usage. Unsurprisingly this incurred us a large overage fee from our server provider, and we vowed never to let this happen again. Then last week Kovarex included another gif in the Friday facts, and even though this one was smaller (only 2.5 MB), it still caused some similar issues

So now we have worked out how to address this. Moving forward we will post gifs to the website in the mp4 or webm format, and for the steam page we will host them on the steams artwork page in the community hub. Hopefull this will keep these sort of server troubles at bay for the most part.


So just prior to the steam release, we deployed the functionality on the website, and while this went off witout a hitch, is has led to some other issues. Notably we have had a string of users have their accounts accessed by someone else and their steam keys stolen. This has caused a large amount of work for myself, and caused us to switch from showing the key on the page, to sending the key via email. We have also had some fraudulent purchases of accounts from our website, people purchasing the game using stolen credit cards, so as soon as its ready, we will deploy an extra protection measure to our website. This will come in the form of requiring a steam ID login to validate a steam key request. For average users this won't cause any issue, but for any nefarious account creations it shoud cause them a lot of hassle.

Tangentially related to this, we are looking at ways of allowing steam users to sign up for website accounts and allow them to access the website DRM free builds. We had a lot of requests in for this feature, as a lot of people are looking to host servers, download for their other computers or just generally run the game without steam. This will take a while to implement, so for the meantime we have posted the headless server for free to download on the website.

Also talking about the headless one of the Factorio players actually wrote a chef cookbook that makes deploying the headless much easier. Check it out at github


So as we all know, when you first go head to head with the train system in factorio, normally the train comes out on top. The whole process from placing rail, plopping down locomotives and setting your first stations can overwhelm many players. It gets even worse when the player starts to play with many trains, trying to understand the difficult task of signalling their rails for multiple trains. So this week a pair of guys by the name ColonelWill and Xeteth posted a great tutorial on youtube, which i think will immensely help any new players trying to learn the train system.

You can check it out here.

As always, let us know what you think by posting on our forum

Hello,

Your friday facts today are written by the community manager Klonan. He's been with us here in the office for nearly 2 months now, working hard marketing the steam release and responding to all the media inquiries sent to our email. He's going to write a little about the behind the scenes of the steam release.


So after the steam release, we had a rather large increase in traffic to all our servers, as we had hoped. Shown below are a couple of graphs illustrating the data usage for the forums. As you can see, after release we had a large influx of players onto the forum, and a lot of new users looking to download mods for their new purchase. We have also seen this increase in activity in other ways, with an evident increase in user posts and new users signing up



You can also see from the reddit stats (many thanks for head moderator Zirr for providing them) that the steam release caused a big surge of interest in our game, with a lot of new unique users rushing in, and a rather large inflix of views, which has sustained over the last 3 weeks. It's also easy to see this when you check it out for yourself, where, compared to a few months ago, the level of activity and interaction is much higher.



This traffic hasn't come as a big surprise to us, as we suspected that the steam release would cause a lot of attention pointed towards us. The sales numbers have been quite kind to us, and while we can't release specific numbers, we can make a short comparison. In the last 3 weeks selling the game on steam, we have sold more copies than in the lifetime of our website. This is an incredible number for us, and we owe it to our community to thank them for their support over the years.


But it hasn't all been good news.After the steam launch, the traffic of our main server increased a fair amount, as was expected. The first 'normal' Friday facts after the steam release demonstrated the 'Loader', and in typical form included with it a 25MB gif showing it off. Normally a gif so large would be no issue, but due to the increase in site traffic, along with hotlinking the gif to steam,it led to an overwhelming level of server strain.



Over the course of 4 days, the gif was downloaded over 250,000 times, totaling a whopping 8.0 Terabytes of data usage. Unsurprisingly this incurred us a large overage fee from our server provider, and we vowed never to let this happen again. Then last week Kovarex included another gif in the Friday facts, and even though this one was smaller (only 2.5 MB), it still caused some similar issues

So now we have worked out how to address this. Moving forward we will post gifs to the website in the mp4 or webm format, and for the steam page we will host them on the steams artwork page in the community hub. Hopefull this will keep these sort of server troubles at bay for the most part.


So just prior to the steam release, we deployed the functionality on the website, and while this went off witout a hitch, is has led to some other issues. Notably we have had a string of users have their accounts accessed by someone else and their steam keys stolen. This has caused a large amount of work for myself, and caused us to switch from showing the key on the page, to sending the key via email. We have also had some fraudulent purchases of accounts from our website, people purchasing the game using stolen credit cards, so as soon as its ready, we will deploy an extra protection measure to our website. This will come in the form of requiring a steam ID login to validate a steam key request. For average users this won't cause any issue, but for any nefarious account creations it shoud cause them a lot of hassle.

Tangentially related to this, we are looking at ways of allowing steam users to sign up for website accounts and allow them to access the website DRM free builds. We had a lot of requests in for this feature, as a lot of people are looking to host servers, download for their other computers or just generally run the game without steam. This will take a while to implement, so for the meantime we have posted the headless server for free to download on the website.

Also talking about the headless one of the Factorio players actually wrote a chef cookbook that makes deploying the headless much easier. Check it out at github


So as we all know, when you first go head to head with the train system in factorio, normally the train comes out on top. The whole process from placing rail, plopping down locomotives and setting your first stations can overwhelm many players. It gets even worse when the player starts to play with many trains, trying to understand the difficult task of signalling their rails for multiple trains. So this week a pair of guys by the name ColonelWill and Xeteth posted a great tutorial on youtube, which i think will immensely help any new players trying to learn the train system.

You can check it out here.

As always, let us know what you think by posting on our forum

Hello,

these Friday Facts are focused solely on our view of gameplay issues in Factorio, and our plans to solve these.

We played more Factorio these days as we were testing the multiplayer, and as we played I realized again, that once you reach a certain point in the base building, with big smelting setups supplied by trains, several platforms for each resource type, you have to spend more and more time by managing the mines. Once we had 6 copper mines, there were always 1-2 that had to be deconstructed, and rebuilt elsewhere. Even with blueprints for stations, for the mine, for double rail junctions etc. It becomes more and more routine work and the game stagnates. Also the feeling that the mine is a short term thing to be removed just a few hours later isn't good. This basically limits me from building a real megabase. There are 2 solutions to these problems, and I would love to combine them.

• Resource generation tweaks:
  Resource generation would be tweaked, so the richness of the mining sites would grow the further you are from the center of the map, the same way the size of enemy bases grows. This would actually add a lot of strategy to the game, as it would always be questionable, whether it is better to use the nearby mines, or whether to expand deep into enemy territory, and create a defensive position of the mine that will last for a very long time. This would also open new possibilities for difficulty settings, as higher difficulty settings could decrease the starting richness of the mines, but it wouldn't imply the need to just tinker with lot of small mines. This would just mean that you need to go deeper into enemy territory compared to easy settings, to get the same amount of resources.
• Long term mining drills
  These would be unlocked by late game research, with a high resource cost. Their advantage would be that, in the long run they would be able to squeeze much more resources from the mining site compared to the normal mining drills, mining some kind of "dirty ore". They would get X times dirty compared to the normal mining. The dirty ore would have to be additionally processed in your base, so it would add infrastructure and logistic challenges if one wants to go this way. In the long run, you could go much further with the scale of the factory, as the complexity is moved from the temporary mine to a long term facility in your base.

With these two combined, the local logistic network with robots, repair packs, walls and turrets that would be re-supplied by a special train from time to time would make sense, as keeping the mine alive would be a long term task that is worth the automation and the bottleneck will be something else other than just the speed of building new mines in the very late game.

This also brings me to an idea, that once you are building the megabase, the only late game and expensive research is the almost infinite follower robot count research, which kind of sucks as it is the only option. You can actually finish it without reaching the full potential anyway. I was thinking, to just add extra expensive tiers to most of the upgrade technologies, so there is something to research in the long term, and would be a very cheap way for us to improve the late game possibilities.

The intention is that the player views this as an optional extension of the research, and to give them and their factories some additional work to do, for a substantial bonus. But the problem arises if they don't intend on building a huge factory, but they are simply completionists, so they force themselves to work on it even when they don't enjoy it anymore.

I can imagine, that one of the solutions could be to allow infinite technologies. You can, for example, research laser turret damage for as long as you would like, but every level is twice as expensive compared to the previous one. This solves the problem, as you can't just "finish all the research".

There are lot of changes that will make the trains more viable in 0.13 (better conditions, better rail building, possibly variant of the loader, and even the changes I mentioned in this post should help, but most of them are not planned for 0.13 as it is already quite a big release). On top of it, I'm planning to extend the cargo wagon inventory size. It has 30 slots now, and I'm considering of increasing it to 40-60. The fluid wagon is still planned as well. With all of these changes, trains might actually become more viable and practical in 0.13, which we believe can't be a bad thing.

The other part that might need some tweaks are the green science pack research costs. The problem is, that once the green science is automated, the researches are unlocked quite fast and the player doesn't have enough time to take advantage of it usually. Then he hits the wall of the blue science, the research stops and it is the time to use all the green science technology to get to blue science. The order of the green research didn't really matter so much at this point. It might help things, if the green research would be more expensive, so the transition is smoother, and the choices matter more.

The motivation for this is to provide some smoke, even when the particle based smoke is turned off in the graphical options for performance reasons. But it could actually be combined with the particle based smoke, to achieve a greater effect.



As always we would like to know what you have to say on our forum thread.

Hello,

these Friday Facts are focused solely on our view of gameplay issues in Factorio, and our plans to solve these.

We played more Factorio these days as we were testing the multiplayer, and as we played I realized again, that once you reach a certain point in the base building, with big smelting setups supplied by trains, several platforms for each resource type, you have to spend more and more time by managing the mines. Once we had 6 copper mines, there were always 1-2 that had to be deconstructed, and rebuilt elsewhere. Even with blueprints for stations, for the mine, for double rail junctions etc. It becomes more and more routine work and the game stagnates. Also the feeling that the mine is a short term thing to be removed just a few hours later isn't good. This basically limits me from building a real megabase. There are 2 solutions to these problems, and I would love to combine them.

• Resource generation tweaks:
  Resource generation would be tweaked, so the richness of the mining sites would grow the further you are from the center of the map, the same way the size of enemy bases grows. This would actually add a lot of strategy to the game, as it would always be questionable, whether it is better to use the nearby mines, or whether to expand deep into enemy territory, and create a defensive position of the mine that will last for a very long time. This would also open new possibilities for difficulty settings, as higher difficulty settings could decrease the starting richness of the mines, but it wouldn't imply the need to just tinker with lot of small mines. This would just mean that you need to go deeper into enemy territory compared to easy settings, to get the same amount of resources.
• Long term mining drills
  These would be unlocked by late game research, with a high resource cost. Their advantage would be that, in the long run they would be able to squeeze much more resources from the mining site compared to the normal mining drills, mining some kind of "dirty ore". They would get X times dirty compared to the normal mining. The dirty ore would have to be additionally processed in your base, so it would add infrastructure and logistic challenges if one wants to go this way. In the long run, you could go much further with the scale of the factory, as the complexity is moved from the temporary mine to a long term facility in your base.

With these two combined, the local logistic network with robots, repair packs, walls and turrets that would be re-supplied by a special train from time to time would make sense, as keeping the mine alive would be a long term task that is worth the automation and the bottleneck will be something else other than just the speed of building new mines in the very late game.

This also brings me to an idea, that once you are building the megabase, the only late game and expensive research is the almost infinite follower robot count research, which kind of sucks as it is the only option. You can actually finish it without reaching the full potential anyway. I was thinking, to just add extra expensive tiers to most of the upgrade technologies, so there is something to research in the long term, and would be a very cheap way for us to improve the late game possibilities.

The intention is that the player views this as an optional extension of the research, and to give them and their factories some additional work to do, for a substantial bonus. But the problem arises if they don't intend on building a huge factory, but they are simply completionists, so they force themselves to work on it even when they don't enjoy it anymore.

I can imagine, that one of the solutions could be to allow infinite technologies. You can, for example, research laser turret damage for as long as you would like, but every level is twice as expensive compared to the previous one. This solves the problem, as you can't just "finish all the research".

There are lot of changes that will make the trains more viable in 0.13 (better conditions, better rail building, possibly variant of the loader, and even the changes I mentioned in this post should help, but most of them are not planned for 0.13 as it is already quite a big release). On top of it, I'm planning to extend the cargo wagon inventory size. It has 30 slots now, and I'm considering of increasing it to 40-60. The fluid wagon is still planned as well. With all of these changes, trains might actually become more viable and practical in 0.13, which we believe can't be a bad thing.

The other part that might need some tweaks are the green science pack research costs. The problem is, that once the green science is automated, the researches are unlocked quite fast and the player doesn't have enough time to take advantage of it usually. Then he hits the wall of the blue science, the research stops and it is the time to use all the green science technology to get to blue science. The order of the green research didn't really matter so much at this point. It might help things, if the green research would be more expensive, so the transition is smoother, and the choices matter more.

The motivation for this is to provide some smoke, even when the particle based smoke is turned off in the graphical options for performance reasons. But it could actually be combined with the particle based smoke, to achieve a greater effect.



As always we would like to know what you have to say on our forum thread.

• Bugfixes
  
  • Fixed the corrupted level 2 of the New hope campaign, that was producing corrupted saves. If you are already playing this level, you have to start it over, I'm sorry for the trouble.
    
  • Fixed the performance issue in the same level after train transport is researched.
    
  • Fixed that the game reported the error to be related to overlay always when it crashed.
    
  • Fixed rare crash related to splitting item in entity that is just about to be destroyed. more
    
  • Sending random packets to factorio port can still crash the server eventually, the probability will just be lower. more
    
  • Running biters over with a car or tank will make them aggressive in peaceful mode. more
    

You can get experimental releases by selecting the 'experimental' beta branch under Factorio's properties in Steam.

• Bugfixes
  
  • Fixed the corrupted level 2 of the New hope campaign, that was producing corrupted saves. If you are already playing this level, you have to start it over, I'm sorry for the trouble.
    
  • Fixed the performance issue in the same level after train transport is researched.
    
  • Fixed that the game reported the error to be related to overlay always when it crashed.
    
  • Fixed rare crash related to splitting item in entity that is just about to be destroyed. more
    
  • Sending random packets to factorio port can still crash the server eventually, the probability will just be lower. more
    
  • Running biters over with a car or tank will make them aggressive in peaceful mode. more
    

You can get experimental releases by selecting the 'experimental' beta branch under Factorio's properties in Steam.

Hello old time Factorio players and steam newcomers as well!

The steam release went pretty smooth, we received new batch of bug reports, but they were manageable and the 0.12.25 update covered most of it.
The rating is much better than we expected, the fact that we are currently the #1 in games on steam sorted by rating is something no one would even dream of. I would like to thank you all again for all the positive reviews, it feels good to read them. I plan to read a batch of them whenever I need energy to overcome some very annoying bug or technical problem. It gives me the feeling that it is worth it :)
We actually made a contest of guessing the steam ratings before the release.

The sales of Factorio went pretty well, and we probably can't talk about exact numbers due to the legal agreement with steam. I believe I can say that we sold slightly more during this week on steam, than during the whole last year on our page, which is overwhelming. This obviously makes us think about the future. First however, we need to talk about the plans for the 0.13 release, which brings us to:

The tech tree visualisation is getting closer to completion, the examples shown here are showcases of the icon organisation logic. The graphical representation (lines) and some details will be updated with the help of our graphics department, but the result will probably look close to this.

The logistics 2 technology icon is big as it is the selected one. We might add an additional way to indicate that it is the currently selected research. Requirements and continuations of the technology are both shown here.


Here you can see all the requirements of the personal roboport research.

I believe it will help players to understand the tech tree a lot, and improve readability. As the algorithm is general, it can especially help to clarify complex mod research dependencies.

I have this idea of the missing transport belt related piece for some time, so I made a prototype to see how useful is it.
The temporary name is loader and this is how it works:


It is an entity that can load or unload the contents of a container as fast as the transport belt can handle. It is 2X1, and it's much more expensive than inserters so they are still viable in many places. As it affects the core mechanics of Factorio I want to be really careful with it.
Pros:

• High throughout support for belt loading/unloading.
• Makes belts (and trains) more relevant in comparison to robots in the late game.
• Train stations could be more compact and faster to build.
• High capacity storage based on belts would be easier to build and more useful.

Cons:

• Possibly overpowered
• Another entity that has to be produced.
• Lot of the belt balancing setups will become obsolete, as the solution with loader is quite trivial.

I would like you, the Factorio players, to give us your opinions on this topic in the forum thread as usual.