Latest Posts

Topic: Measuring working time of productionsites

hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-01-23, 10:40

the effect of my change is that it combines the cycle based statisitcs with time based so a succesful cycle is always weighted with one third (we could change this to a quarter by setting stat_duration_ to 4 as well) and skipping/ failing is weighted higher for buildings with short working cycles and lower for long working cycles. this results in buildings with shorter Productioncycles rising and falling quicker in percentage. And this was the side effect I wanted to have together with the 100% fix.
Anyhow this was just a trial and if I can't convince you about the advantages even for AI (Ai in theory reacts quicker on economic changes) I'll do it in another way.


Top Quote
Tibor

Topic Opener
Joined: 2009-03-23, 23:24
Posts: 1377
Ranking
One Elder of Players
Location: Slovakia
Posted at: 2020-01-23, 11:16

Maybe we just split these statistics - one "pure time based" and new one "enhanced time based"


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-01-23, 12:39

Tibor wrote:

Maybe we just split these statistics - one "pure time based" and new one "enhanced time based"

That wouldn't make sense, as I believe the benefit is not worth the effort and cost of ressources. Rather then that I'll keep it like it is and propose to reduce the time base a bit from 10 to 7 minutes perhaps.


Top Quote
Tibor

Topic Opener
Joined: 2009-03-23, 23:24
Posts: 1377
Ranking
One Elder of Players
Location: Slovakia
Posted at: 2020-01-23, 12:45

hessenfarmer wrote:

Tibor wrote:

Maybe we just split these statistics - one "pure time based" and new one "enhanced time based"

That wouldn't make sense, as I believe the benefit is not worth the effort and cost of ressources. Rather then that I'll keep it like it is and propose to reduce the time base a bit from 10 to 7 minutes perhaps.

Yes, tweaking that 10 minutes sounds better to me. If you (and more peoples) believe that statistics should be changing (reacting) faster than now, it is an answer. 10 minutes was just arbitrary value so changing it by few minutes would be fine for me...


Top Quote
stonerl
Avatar
Joined: 2018-07-30, 00:03
Posts: 327
Ranking
Tribe Member
Posted at: 2020-01-23, 12:56

@hessenfarmer would you be so kind and open a PR on GitHub? So, we can have Windows builds and it is easier to find it.


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-01-23, 12:58

stonerl wrote:

@hessenfarmer would you be so kind and open a PR on GitHub? So, we can have Windows builds and it is easier to find it.

ok, I can do this although I'll change it anywaay due to Tibors concern


Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-01-23, 13:00

Tibor wrote:

Yes, tweaking that 10 minutes sounds better to me. If you (and more peoples) believe that statistics should be changing (reacting) faster than now, it is an answer. 10 minutes was just arbitrary value so changing it by few minutes would be fine for me...

I'll do it this way then, but I didn't get your concern about the other solution. Maybe you could explain to me what is it about exactly.


Top Quote
Tibor

Topic Opener
Joined: 2009-03-23, 23:24
Posts: 1377
Ranking
One Elder of Players
Location: Slovakia
Posted at: 2020-01-23, 21:21

My concer would be like: "if it is not broken do not fix it".

The statistics was created to be used by AI. From AI's point of view there is no reason to tweak it, so I would prefer not change it at all. But I admit that changing 10 to 12 or 8 will not make big difference, but you propose much bigger changes. And this might not be an end of it.

I understand your reasons and am not against tweaking user-visible statistics, but then two separate numbers would be best for all. And still an improvement over old attempt-based statistics.....

Edited: 2020-01-23, 21:22

Top Quote
hessenfarmer
Avatar
Joined: 2014-12-11, 23:16
Posts: 2646
Ranking
One Elder of Players
Location: Bavaria
Posted at: 2020-01-24, 07:08

@ Tibor:
Ok understood. I have started to do it in the other way yesterday. will upload it soon. I just experimented a bit with excel before and found that a time base of 5 Minutes will let us cross the 90% mark after 11 minutes. Would this be ok for you?


Top Quote
Tibor

Topic Opener
Joined: 2009-03-23, 23:24
Posts: 1377
Ranking
One Elder of Players
Location: Slovakia
Posted at: 2020-01-24, 08:00

hessenfarmer wrote:

@ Tibor:
Ok understood. I have started to do it in the other way yesterday. will upload it soon. I just experimented a bit with excel before and found that a time base of 5 Minutes will let us cross the 90% mark after 11 minutes. Would this be ok for you?

This sound interesting, I had never make such calculations. Can I interpret this that it takes 10 minutes to go from 10% to 90% and vice versa? Providing that site is fully working / idle over that 10 minutes...

This would be OK for me.

But I still dont like manipulation with production time duration...


Top Quote