Skip to main content

Report Bugs

If relevant please include the following:

A description of the bug.

How to reproduce the bug.

The exported .sfmd save file. If the bug prevents you from exporting, it can be found in the game’s install directory as save.sfmd.

If you think the bug is related to any settings, include the settings.json file from the game’s install directory.

A screen shot of the bug if appropriate.

A copy and paste of any error message given.

The app’s version number.

********************************************

PLEASE ALWAYS INCLUDE THE EXPORTED .SFMD FILE IF YOUR BUG IN ANY WAY RELATES TO A SPECIFIC SETUP OF MACHINES. THE .SFMD IS USUALLY FAR MORE USEFUL THAN A SCREENSHOT UNLESS IT IS SPECIFICALLY A GRAPHICS GLITCH.

********************************************

If you want to make a suggestion instead of reporting a bug, please use

https://satisfactorymodelersuggestions.featurebase.app

Linux broken fractional scaling

Hey, I found a bug (maybe not 100% sure). If you use a 1440p Monitor on Linux (Fedora 44 + gnome) with fractional scaling (133%) so everything is the roughly the same size as an 1080p Monitor, Satisfactory Modeler does not scale the UI and you end up with a very tiny UI. However I found a few ways around it: My first way but the most resource hungry Proton Experimental Launch Options: gamescope -w 1920 -h 1080 -W 2560 -H 1440 -F fsr -S fit -f -r 144 -- %command% Where: -w 1920 -h 1080 -> Your upscaled resolution -W 2560 -H 1440 -> the resolution the game runs at -> should be your monitor resolution so upscaled text is not blurry -F fsr -S fit -> upscaling mode and upscaler filter -f -> fullscreen -r 144 -> refresh rate Linux native Launch Options: _JAVA_OPTIONS="-Dsun.java2d.uiScale=200%" %command% This uses just a java awt or swing (not sure and idc tbh) argument. 200% -> adjust until you are happy Some JVM Stuff: While debugging this I found: https://docs.oracle.com/en/java/javase/25/troubleshoot/java-2d-properties.html#GUID-2DC38ED7-7F26-4489-B3B5-C3B0B31E2786 Where it is stated: Linux (X11): Scaling is applied only if the desktop scale is 2 or greater, rounded down. This maybe because Java Swing/AWT is NOT running wayland native and fractional scaling is a wayland only feature and xwayland (compat. layer between x and wayland) does not support fractional scaling. However _JAVA_OPTIONS="-Dsun.java2d.uiScale=133%" and I do not really know why. So I am not sure why fractional scaling does not work out of the box. Maybe it is Java or maybe it is Satisfactory Modeler so I have maybe found a bug or maybe this is more like a feature request or a java limitation.

Timon Klinkert22 days ago
1

Full solver hangs when removing an unused Crystal Oscillator recipe node

I've encountered an issue where the Full solver will calculate successfully only if I leave an apparently unused Crystal Oscillator recipe node connected. The unused node appears to have zero production and doesn't contribute to the factory's calculated output. However, disconnecting either of its inputs or deleting the node causes Full to hang indefinitely. Steps to reproduce Open the attached working.sfmd model. Locate the two Crystal Oscillator recipe nodes (Standard and Insulated). Nuclear Planner>Not Raw Observe that one recipe is producing approximately 48.32 items/min, while the other has its output limited to 0. Disconnect either input from the inactive recipe node, or delete that node entirely. Expected behavior The Full solver should calculate successfully regardless of whether an unused recipe node is present. Removing a node with zero production should not prevent the remaining production network from being solved. Actual behavior With both Crystal Oscillator recipes present and connected, Full completes. Disconnecting either input from the apparently inactive recipe causes Full to hang. Deleting the inactive recipe also causes Full to hang. Restoring the node and both input connections allows Full to calculate again. Additional observations The issue appeared while changing from the standard Crystal Oscillator recipe to the Insulated Crystal Oscillator alternate. The model contains several legitimate production feedback loops. Quartz Purification supplies Quartz Crystals and Silica, with the Silica used for aluminum production to reduce Bauxite consumption. Earlier attempts to disconnect or rearrange parts of the silica network also caused Full to hang. I have found a workaround: keeping the inactive recipe node with both inputs connected. I suspect the extra node or its connections are affecting the Full solver's optimization behavior, even though the node appears to have no production. Attachments Working.sfmd model with the inactive recipe connected Screenshot showing the two recipe nodes and their calculated production

RoadieGamer22 days ago
Rejected

Making an outpost a "Blueprint" Breaks some functionality of priority Splitters

I was making a blueprint for Diluted Packaged Fuel, and I wanted the split between residual plastic and residual rubber to be uneven, but it would force the two to have equal inputs no matter what I did, until by chance I undid the thing making it a “Blueprint“ so it was only an Outpost instead. What…?

Samuel Herron2 months ago
3
Planned

Fails to launch from macOS 27.0

Steam just gives an error popup with description: Failed to start process for this game : OS Error 0 i tried reinstalling it with the option “Create an application shortcut”, but also then still failed to launch from the launcher with same error code showing on steam. I verified the installed files from steam version from May 14, 2026 im on a M2 macbook air

sophie2 months ago
3

Cannot overclock resource wells

There seems to be an issue with resource wells and pressurizers. The program lets me place pressurizers and extractors, but there is no way to connect them. The pressurizer can be overclocked, but this has no impact on the actual modelled output, as the extractors themselves cannot be overclocked. If I want to model a pure nitrogen gas well fully overclocked to produce 300/min, I need to place two 100% clocked pure and a normal well for 120 + 120 + 60 = 300. The same issue applies to Oil and Water wells. It seems like there is just a missing means of connecting the pressurizer and the extractor that would solve the issue, or just allow overclocking of the extractors directly.

Caleb Renshaw2 months ago
1
Rejected

Fractions still appear in several places despite all settings being set to decimal.

Numbers appear in fraction format in several places on the UI despite having set ALL the options to Decimal in settings.

nznomad2 months ago
4
Rejected

Power calculation is wrong

When making my power setup I made a excel spreadsheet to understand it easier and the “Average Net Power” seems right but the “Min Made - Max Used” is wrong. It should be around 2,154,337.525 but shows 2,569,402.693. This probably has to do with boosted power, but still the (“Average Net Power” - “Max Used”) is still way below what is shown in the screen shot. If I’m wrong with my interpretation I would like to know what is causing this. Thanks for looking into this!

Brennan Duessler2 months ago
3
Rejected

Infinite calculation time

I’ve made a small graph with a few looping mergers and splitters, which works fine, and the calculation is completed after like a minute or two, however, when I then try to take the outputs from one of the loops and put it into an assembler along with a different input, the calculation just goes on forever.

virus3 months ago
1
Rejected

standard database values for the alternate "Caterium Circuit Board"

The standard database values for the alternate "Caterium Circuit Board" recipe are completely broken.The tool calculates with 120 Caterium wire and 28 plastic, which causes massive overproduction chains when building backwards.In the actual game, the correct quantities per minute (at 100% clock speed) should be 37.5 Quickwire/Wire and 12.5 Plastic to produce 8.75 Circuit Boards. Please fix the base recipe ratio constants for this node.

Robin4 months ago
1
Rejected

Uranium nuclear power plant has wrong output

Power plant states 10/min waste. ought to be 50/min.

David Southwick4 months ago
2

Github signin not working

When you click to sign in using github you get Be careful! The redirect_uri is not associated with this application. The application might be misconfigured or could be trying to redirect you to a website you weren't expecting.

magic280z4 months ago
1

Fluid rounding incorrect

Alternate title - the evils of units that are not what they seem… Game setting - Recipe Parts Multiplier 0.5 Refinery with HOR alt recipe: In game, the 30 input oil become 15. The modeler displays 20. The cause seems to be that the config files for modeler handle fluids with an implicit multiplier of 1000. The game also handles fluid quantities as multiples of 1000 - but divides by 100 for UI display. So, naturally, the config files say the input is “-3” meaning -3,000 oil. That’s OK so far. The issue relates to the fact that the game calculates the ½ price recipe parts as 3×1000×0.5 yielding a display of 15 (but handled internally as 1500). The problem is that the modeler seems to be doing (3×0.5)*1000 with rounding on the first sub-calculation.

Arcane Tourist5 months ago
3

Failed to launch JVM

It’s even doesn't launch. Windows 11, OpenJDK Runtime Environment (build 25.0.3+11-LTS) OpenJDK 64-Bit Server VM (build 25.0.3+11-LTS, mixed mode, sharing) error: satisfactory_modeler.exe Failed to launch JVM

Alex Marvin5 months ago
2

Miners not producing expected output

All of the miners are producing ½ the output they should. In-game, an MK2 miner on a normal node produces 120/m, but in the modeler, it only produces 60/m

Ice_395 months ago
1
Rejected

Miners not producing expected output

All of the miners are producing ½ of the actual output in-game. For example, an MK2 miner on a normal node is supposed to produce 120/m, but in the software, it only produces 60/m

Ice_395 months ago
2
Rejected

25% Recipes in Blender not Rounding Up

Solid items shouldn’t ever become fractional; the game always rounds up, and guarantees a minimum of 1 input.

Michael Paddock5 months ago
2

Electrode circuit board incorrect output number

Connecting petroleum refinery to circuit board assembler cause output count updated to 0. Alt. recipe: Electrode circuit board Before connect: After connect:

Luxuria GK5 months ago

Blueprint Belt Bug

When I saved my blueprint and placed it, my belts were turned and glitched into my spliters and Constructers. Here, the belt flipped completely around and was inside the constructor. It is in the experementel Version.(1.2.2.2)

Xander vandeput6 months ago