March 21st, 2023 Livestream
Q&A: How fast belts can cause floating-point precision issue?
This question has 3 related videos
- November 12th, 2024 Livestream Q&A: With the issues with Mk.6 Belts, would it have been better for Mk.3 Miners have two outputs? https://www.youtube.com/watch?v=oVaqY1NgJGQ
- September 5th, 2023 Livestream Q&A: How does Satisfactory deal with margin-of-errors in production? https://www.youtube.com/watch?v=95-N_VVnGXg
- October 18th, 2022 Livestream Q&A: Are you actually going forward with the rebalance so there aren't precision issues? https://www.youtube.com/watch?v=sec9STlUvB0
Suggested Wiki Reference Code
<ref name="yt-irkPXvgOdtk">[https://youtu.be/irkPXvgOdtk YouTube - March 21st, 2023 Livestream - Q&A: How fast belts can cause floating-point precision issue?]</ref>
Topics
Transcript
how fast this is a question regarding the the floating Point position issue with belts it sounds like calculating out of positional spline can be solved without Precision lost so this is a simulation process where error accumulates so yes that is correct, that has been solved internally, I believe the issue we have currently is that rendering of the simulation is not accurate
[Music]
, yeah and I think there's been two issues there's been one issue where belts were actually incorrectly like calculating or something I don't know what the actual fix was but there was an issue where when you were dividing belts you would lose out on on of a position you would actually not lose out of precision actually it's a different issue there's a different issue between the the pipes and this belts have never really had a Precision issue the rendering of belts have had that however pipes have a floating Point Precision issue and that one is harder to go around because the calculations for fluid dynamics is a lot more complex so that one is more difficult to solve but I think we want to solve that, and I think there is a way to get around solving that we just don't know when we will get around to it foreign pipes work completely differently