I would approach this from a different perspective (I don't know if it is viable, but it doesn't hurt to explore it): Looking at the factory as a graph of interacting entities, you will have tightly coupled areas (like an actual production plant with belts, splitters, inserters, production ...) tha...

you lost me this time. how did you determine what byte range to prefetch? and also, i assume, you have to cast those raw bytes into a valid object. how do you know which bytes go into what variable? is the variables sequence the same on all hardware architectures? how did you address that uncertain...

Late but huge congratulations on one million! You sooo deserve this hugely! I will join the party, looking forward to it so much. I also must say that this interview is very inspirational (in a good way). Even with just subtitles, I can really feel the passion you put into this endeavor - so glad th...

As for HW, we worry CPU flags which change how floating point operations are rounded. AFAIK the flags are global, so if any application changes them, it affects all applications on the system. Other than that results of instructions should be same on every x86-64 CPU. Rounding mode and other floati...

Even if your compilers all produce the same comparison operation, different FP hardware executing it may not return the same result, which could probably lead to desyncs in MP. It's not even unlikely, it's impossible. We're talking about x86-64, so there simply are no differences in how the hardwar...

These are two different things. Floating point math (without aggressive optimization) is both deterministic and approximate at the same time. You can very well deterministically compute something that should mathematically be 0 is not == 0. That is not wrong with fp math. Just because it is determi...

It is widely recognized that comparing two floating point numbers via == is bad. I don't think I could explain it better than this FAQ entry . I believe this applies here, despite any additional contributing factors to inaccuracy. Factorio requires determinism which means that if any of our compile...

I'm not sure why the inserter moves so oddly. Most of it's behavior code is probably there since version 0.1 so probably not even the guy who wrote it(kovarex) would know why. In any case, are you just trying to increase accuracy or are you also accepting and dealing with the eventual inaccuracies ...

The problem was a bit deeper. The vector given to the atan2 was rotated imprecisely. First, awesome to get it fixed now! So I'm curious. Is this in any way related to the odd shape that the inserter hand follows? inserters.png These are the positions a freshly placed fast inserter follows when tran...

That's funny, I arrived at the very same conclusion yesterday evening without even knowing the reddit thread by then, but wanted to write it up nicely today. I was going to take a look at it since Rseding said it would be fixed if they knew what caused it and if it was an easy fix. At first I was a ...

Thanks for the log files. This should be fixed in the next version of 0.15. Note, calling take_screenshot and then deleting the surface in the same tick means the screenshot won't be done for that surface. Great! I understand that screenshots are taken asynchronously. Is there a recommendation how ...

Attached is the simple mod to reproduce. However, oddly I also cannot reproduce it on my Linux installation. But there is something wrong with that anyway, but that is a topic for another thread. On Windows I was able to reproduce the crash very reliably.

I guess I figured out the issue in the current version: I do delete the surface immediately after taking the screenshot. Apparently the screenshot is now taken asynchronously so the referred surface no longer exists. I guess I can somehow work around the issue, and I suppose it's reasonable for Fact...

Based on the name of the mod "seedmapper" I'm wondering if the game.take_screenshot that is actually in the mod includes the option position parameter that points to the chunk being generated. That would certainly crash the game. At that zoom level taking a screenshot using the position o...

That code was just control.lua, however I can no longer reproduce the issue with this minimal mod in 0.15.11. Unfortunately my actual code still crashes in a very similar way. This is the log for 0.15.11: 0.001 2017-05-16 22:08:31; Factorio 0.15.11 (build 29782, win64, alpha) 0.001 Operating system:...

The following mod crashes script.on_event(defines.events, function(event) if event.name == defines.events.on_chunk_generated then game.take_screenshot{resolution={x=1920,y=1080}, zoom=0.08, path='test.png'} end end) Log: 0.001 2017-05-02 21:13:05; Factorio 0.15.6 (build 29362, win64, alpha) 0.001 Op...

Great job guys (both contestants and organizer)! I wanted to go for it but didn't find the time though. Turns out my idea with a long blue underground belt wouldn't have been very competitive.

A little late but I need to clarify this, the test I did is no guess. All the different memory speeds (1600, 2133, 3200 MHz) are with the same timings/latency. Only in this test http://imgur.com/2FPr3su I did try different timings to see how it affects the game speed. Unless I'm very much mistaken ...

An interesting anecdote related to memory fragmentation. During AntiElites last 100% achievement speedrun, he died the first time quite late in the game, where UPS already dropped significantly. After reloading the game due to the happy accident with biters, there was a noticeable increase in UPS. N...

Pretty much the only time this can happen is if you are memory-bound in a single thread and can get around that with multithreading. Basically, either NUMA No, that doesn't get you super-linear speedup. Two memory controllers are only twice as fast as one. or having the code be able to efficiently ...