[0.9.2] Production stats way off

[Holy-Fire]

The production stats opened by pressing "P" seem to be way off. Hard to say what is wrong when almost nothing there is right.

e.g., when "5s" is chosen, it shows "59" on the copper plate icon, which I suppose should mean I've produced 59 copper plates in the past 5 seconds. It then shows 393.3/m which I suppose is the extrapolated average rate. But if it's 59 per 5 seconds then it should be 59*12 = 708 per minute.

In fact, 393 per minute is the correct figure. It's how about how much I'm producing with my ~23 stone furnaces. The 59 figure however seems completely made up.

The problem deepens when choosing "1m". It shows me 351 produced (which sounds about right) but then shows me a rate of 195/m. The amount is inconsistent with the rate, and the rate is inconsistent with the rate I saw for 5s (my production is steady, it doesn't change enough to warrant a difference between the timescales).

Discrepancies persist with larger timescales. Interestingly, for 10h it says I've produced 15K copper plates, and for 50h it says I produced 14K, which is less.

My guess is that the total produced in the given period is counted, but then it's multiplied by some constant factors to display the amount produced and rate, based on the chosen window; and these constants are wrong.

[Gammro]

I've noticed the same:


The production stats probably aren't constant enough to accurately measure, but the consumption rate of iron and copper ore can be called somewhat constant over the last 10 minutes. The total consumed and the consumption=rate*10 don't match up. Both are off by a factor of 1.80.

[Nirahiel]

The thing is, those values (I think it applies to power as well) are not exact values but average.
Try to look at the amount of power consumed by your lamps, at the exact moment when they turn off (and you now use precisely 0 energy) You'll get what I mean.

[FrozenOne]

Reported for quite a long time.
https://forums.factorio.com/forum/vie ... f=7&t=1914

[Gammro]

The thing is, those values (I think it applies to power as well) are not exact values but average.
Try to look at the amount of power consumed by your lamps, at the exact moment when they turn off (and you now use precisely 0 energy) You'll get what I mean.

I already thought it was an average. It's just not a good average.

I put all the values from my screenshot in a table. I put the given total in, the given rate and multiplied it by 10. Then I divided the given by the calculated. The expected value of that data set is 1.799, with a variance of 0,000698. That's pretty much constant.
The graphs in the are too different from each other to produce the same error factor.

If it'd be an accurate average. I'd say the easiest way is to take the total production in that time period and divide it by the amount of minutes in that timeframe. Then you'd at least come within 5% accuracy. Which is far more useful.

It's just that this data is currently useless, because there's no easy way to know which one is correct.

FrozenOne: We're not talking about input vs output differences(even though those are weird too). We're talking about the gap between the average rate and the total production/consumption of items.

[Holy-Fire]

Reported for quite a long time.
https://forums.factorio.com/forum/vie ... f=7&t=1914

Seen this report, but I thought the title and description weren't clear enough. Also it focused on a different problem.

The thing is, those values (I think it applies to power as well) are not exact values but average.
Try to look at the amount of power consumed by your lamps, at the exact moment when they turn off (and you now use precisely 0 energy) You'll get what I mean.

Look at the numbers. (Either in your own factory or what I quoted above.) They just don't make any sense.

[Holy-Fire]

Originally reported for 0.8.8, it's 0.9.2 now and this is still happening.

[SilverWarior]

Originally reported for 0.8.8, it's 0.9.2 now and this is still happening.

If you read through the original 0.8.8 thread about this isue you will notice that developer sad that they will look into this but they are not giving it a priority since it is not a gamebraking bug.
So please be a litle patient.

[Holy-Fire]

Originally reported for 0.8.8, it's 0.9.2 now and this is still happening.

If you read through the original 0.8.8 thread about this isue you will notice that developer sad that they will look into this but they are not giving it a priority since it is not a gamebraking bug.
So please be a litle patient.

Link?

[slpwnd]

Thanks for bumping the thread. The discrepancy between current sum (text on the picture) and average per minute (next to the progress bar) is now fixed for the 0.9.3. The problem was than during numbers normalization the numbers for average per minute got skewed by smoothing samples. There is still some small error for 50h vs. 10h (numbers for 10h are actually bit higher), but we will look into that as well.

Edit: Ok so the 50h vs. 10h actually is not a bug:) The thing is that there is a fixed amount of samples (300) for every period. Only samples that are finished (so not the current one) are used for the calculation. For 50 the sample is 10 minutes for 10h it is 2 minutes, so the difference can be up to 8 minutes. The solution would be to include the last sample as well, but this can be done only when not all the samples are taken (which in case of 50h they often aren't) otherwise the graph would be "jumping".

[FrozenOne]

Also, mind the bugs in my original thread https://forums.factorio.com/forum/vie ... f=7&t=1914
(for example on the screenshot there the colors of graph are random/not matching, and others described there...)

[slpwnd]

Also, mind the bugs in my original thread https://forums.factorio.com/forum/vie ... f=7&t=1914
(for example on the screenshot there the colors of graph are random/not matching, and others described there...)

Ok, so there was actually a fundamental bug in the production statistics calculation. The consumption and production were added up in the same variable (with different signs) for given frame. This could really skew the numbers especially for longer times (when the frame is big enough for numbers to cancel out - for 5 s interval when 1 frame is 1 tick the effect would probably be negligible). This is because of the historical reasons - the production statistics and energy statistics use the same code and using same variable for the energy statistics was an intentional optimisation. Anyway, this is now fixed for the 0.9.3.

[Holy-Fire]

Also, mind the bugs in my original thread https://forums.factorio.com/forum/vie ... f=7&t=1914
(for example on the screenshot there the colors of graph are random/not matching, and others described there...)

Ok, so there was actually a fundamental bug in the production statistics calculation. The consumption and production were added up in the same variable (with different signs) for given frame. This could really skew the numbers especially for longer times (when the frame is big enough for numbers to cancel out - for 5 s interval when 1 frame is 1 tick the effect would probably be negligible). This is because of the historical reasons - the production statistics and energy statistics use the same code and using same variable for the energy statistics was an intentional optimisation. Anyway, this is now fixed for the 0.9.3.

Thanks, but there's still a bug. Production stats for 5s interval are too high.

This is easy to see in a toy setup with only one furnace.

(There's probably some problem with the way this is sampled that causes an overestimate, which has the strongest effect in a short interval).