Multiple Matter Fabricators - Productivity Affected by Desynchronization?

[lbap]

I am using the Mindcrack Pack.

I have two UU-Matter factories, each of which has four individual matter fabricator systems... which allows the machines to run with a permanent, fully-charged internal battery... with an permanently filled inventory of scrapboxes [give or take a second], at full speed for the pack [about +1% a second]. Each factory has 64 ultimate hybrid solar panels to supply the matter fabricators with energy, 16 for each machine.

I was hoping that one factory would generate about 2 UU-Matter a minute (+x seconds, due to the 1% a second out of 100 progress bar), so I was planning to make another 2 factories (16 matter fabricators total) to generate enough UU-Matter to produce a stack of iridium ore every hour (-4/+10 minutes).

I did a mini-test, where I let two factories run for an hour (each starting at a different percent), to find the average production rate of UU matter. I started them manually, meaning that the 3.4 UU-Matter/minute is a little off, but I want to know if the average would be closer to 4 if the machines were all started with the same progress percentage and stooped at the same time. Also, if the machines were not in synch, would the average per minute be closer to 4 the longer the test? If so, how long would the test be to have an average closer to 4 UU-Matter/min?

Again, I plan to make another 2 factories, which means I want the average to be about 8 UU-Matter/min (give or take x seconds or so).

 

[Omicron]

Not quite sure what you're asking here, but in regards to estimating/extrapolating output...

- If all factories are identical, you need to test only one of them because the results are automatically valid for all others as well.
- Next, to figure out how long it needs to run to produce a useful estimate of average hourly production, you need to determine if the production rate varies over time. For example, if you use solar power, then the production rate will be variable due to the day/night cycle, especially if players sometimes sleep through the night and sometimes not. However, if you have a buffer battery that never drains fully and gets recharged during the day, and is sized big enough to also make it through rainy days (i.e. you should be able to run two full Minecraft day/night cycles on battery power alone and then recover relatively swiftly), then you have a system which is completely non-variable.
- If you have a non-variable system, then all you need to do is run one factory for exactly one hour. Use a stopwatch and note starting and finishing progress percentages as well as UU-matter produced. Calculate the difference between starting and finishing, divide by 100, and then add 1 for every unit of UU-matter. That result is automatically your average per hour, because it never varies. It is also automatically the average per hour for every individual factory, since all factories are identical.
- If you system has a variable production rate, you'll want to run it longer. The smallest useful sample size is generally considered to be three, because if you use only one or two, then an outlier result will dominate the final figure and render it useless. So give it three to five hours, ideally again with a stopwatch. You can split this by turning it off at defined times if you need to stop playing and turn it on again right when you log in the next time. This will technically affect the result, but if you do it well and sparingly, the effect should be smaller than the error of measurement you're going to have in this kind of experiment anyway, so you can safely ignore it. While your test is running, play normally. This is important, because the way you are playing (especially with respect to day/night cycles) will affect the variable production rate of the factory, and so you will not get an accurate result for normal play if you don't test with normal play.

 

[lbap]

I forgot to mention that the factories reside in an age of 24/7 sunlight.

I knew the first test was off, but I did it anyways to get an idea of the production rate. After I did a more thorough, hour-long test, where all the matter fabs started at 0%, I got 2 stacks of UU-Matter, +12. Two factories, again starting at 0%, gave me the expect total product of 4 stacks, +24. This is a higher product rate than before, where all the matter fabs started at different percentages.

Thank you for your help