Ideal production process level?

What's going on in Kapi Regnum? This section is for everything concerning the game itself

Moderator: moderators

Post Reply
Guest

Ideal production process level?

Post by Guest »

Hi, when I am upgrading my buildings, it says: Ideal production process level: x%. What does this mean?

I notice it is higher on my larger buildings of the same kind. A 40 sqm farm is 1%, a 20 sqm is 0.5%.

It increases by 1% every 40 sqm.

Thus, it will be at 100% at 4000 sqm.
Does this signify the most efficient building size? I notice that increasing the building size increases the production time, i.e. a 40 sqm building produces roughly 98% the product of 2x20 sqm buildings. If this diseconomy of scale (less efficient production) continues, how would a larger farm (with 100% production process level) be "ideal"?

I am trying to figure out the best way to expand my buildings, so help would be greatly appreciated. :shock:
Guest

Post by Guest »

Well this is theoretical based on my knowledge of Kapilands, since I'm not high enough yet to experience it here. But I'm fairly certain.

You've no doubt noticed that a building gets less efficient as it gets larger - each new employee adds less than the one before did.

Well if you try to mathematically represent this, the formula eventually hits a point of inflection. That is to say, with a single formula, eventually an employee would add zero. And after that, they would add less than zero, actually making the building produce less after expansion.

So what they did was make an "optimize" point after 400 employees (4000m). This means you follow one formula until you get to that point. Then it resets up a bit, and you trace that curve again. Overall it makes kind of an arcing stair-step pattern.

In practice, it's nothing to worry about really. But when you do get to 400 employees, you should optimize - and whether it does this automatically or you have to hit a link I don't know offhand. At that point, your production will get a bonus, and you'll start over at "Ideal Production Process Level 0%".

As a tip, I tend to recommend hiring the last 50 employees of each cycle at once if you can. (Yes that means 5+ days of solid expansion for most buildings.) This is because stepping from 3500 to 3510 feet gives a really small benefit, but stepping from 3990 to 4000 gives a huge one. So if you're going to expand, you should rush to that point.

(But no it's not so good that you should have one 5000m building and the rest back at 100m, if that's what anyone was thinking. Only the last 500m should be rushed.)
Guest

Post by Guest »

Great advice!
Guest

Post by Guest »

I've been expanding, and tho I haven't reached 4000 sqm yet, I do find the smaller buildings more productive.

and yeah, after all my buildings reach a certain size (say 200 sqm), I only upgrade two of them alternately. And the rest keep producing at the small building efficiency.

I reached 1000 sqm, the drop in efficiency is drastic, a 1000sqm lumberjack produces abt 35000 wood/day and a 200 sqm lumberjack makes 9000 wood/day. Thats an almost 10,000 wood/day inefficiency.

I remember someone saying on this forum that uneven upgrades are not good as all buildings' efficiency is affected. But I have done exactly that and find no truth in that contention whatsoever. I plan to sell off my smaller buildings gradually when I reach councillor. To make room for other production/markets.
Guest

Post by Guest »

pk wrote:I remember someone saying on this forum that uneven upgrades are not good as all buildings' efficiency is affected.
That almost sounds like what I said, except not quite.

Using your own numbers, adding another 10m on your big lumberjack would produce an extra 350 wood per day. Adding another 10m on your small lumberjack would add an extra 450 wood per day. So it's more efficient to expand the smaller ones, therefore putting everything around the same size.
Guest

Post by Guest »

Oh my, I am soooo nerdy.

For an update, I wanted to take a stab at the formula used to calculate production for a given SQM building.

I did a simple linear regression using four data points at 20, 40, 120, and 200 sqm. For your viewing pleasure, it approximates the production you will achieve MUCH better than using 20 sqm equivalents (like saying my 40 sqm will produce around double of what my 20 sqm does).

So, for those of you who know math, this formula takes the form of y=A+Bx. As I get more data points this will be more accurate, but I still have to expand my city as fast as I can, so I will not care too much...

Therefore for lumberjacks (drumroll)
Hourly production estimate = 6.33 + (1.92 * Building SQM)

As you can see by the formula, and if you graph it out, the production for a building cannot be exactly calculated given a sqm, but it is possible that the true formula takes into account the production process level. :o

Please let me know if these figures fit with your city, and other production lines. My rank is Sir, so that may alter the formula also.

....

Addendum
Data Used (Lumberjacks at Sir)
SQM / Actual Hr prod / 20-sqm equiv / Linear regression estimate
20 / 42.21 / 42.21 / 44.73
40 / 83.59 / 84.42 / 83.13
120 / 241.31 / 253.26 / 236.73
200 / 387.56 / 422.1 / 390.33
As you can see, with large building sizes it is much more accurate than just using the 20-sqm equivalent.

:roll:
Guest

Post by Guest »

Try this:

ProductValue * SQM * 0.999525 ^(SQM - 1) * LevelMultiplier

Where:
ProductValue is is the unique number fore each product
SQL is your Building size in Square Meters
LevelMutliplier is a number that starts at 1 for Bourgeois and decreases for each level after that. I could probably get the formula for that if you like, but I lack data at the moment.
Guest

Post by Guest »

Knolllllllls!
How do I find ProductValue?
LevelMutliplier is a number that starts at 1 for Bourgeois and decreases for each level after that. I could probably get the formula for that if you like, but I lack data at the moment.
Try this:
Freeman 1,000
Bourgeois 0,882
Huckster 0,833
Sir 0,789
Merchant 0,769
Esquire 0,750
Councillor 0,732
Patrician 0,714
Provost 0,698
Baron 0,682
Earl 0,667
Margrave 0,652
Duke 0,638
Grand Duke 0,625
Prince-Bishop 0,612
Archduke 0,600

From this thread: http://kr.forenserver.de1.cc/viewtopic.php?t=459

I better get to Merchant quick, so I can save you some more money! :D
Guest

Post by Guest »

Thanks, I had a feeling it was around somewhere, but I didn't feel like searching. I didn't realize there was a drop between Freeman and Bourgeois though.

Incidentally, I can approximate that curve

(Rank# ^ -0,175)

But it's not quite the right fit, I would need to switch to a less concave formula structure.

Regardless, your question (ProductValue) is easy enough for anyone who can do linear regression! :P Take a sample reading for any product in any size building. You then have the other factors, so solve for the variable. That will give you the ProductValue mutliplier, which you can then put into that formula as a constant.
Guest

Post by Guest »

It's surprising how well your formula estimates the actual hourly production. It is accurate to the tenth at the few SQM's I sampled. The regression is only accurate to the tens place!

Initially I thought it was strange, the .999525 thing, but I wonder if that plays a part in the Ideal Production Process Level. For a thought experiment, if .999525^(SQM-1) at 4000 sqm is considered, it is 0.15. I wonder if that remaining inefficiency is drawn out over the next production process level curve, say from 4000 sqm to 8000 sqm+.

Perhaps those are questions that will matter in two months or more. We shall see.

For now, I am working on a bigmassive spreadsheet that will eventually calculate the value of time in the Kapiverse! All else will be discounted... Oh sustainable ROI per second heaven!

Maybe then, if I am feeling aggressive, I will just sell 200 coins and try to break the top 50. Booyah! :twisted:
Guest

Post by Guest »

Wu-Mart wrote:Initially I thought it was strange, the .999525 thing, but I wonder if that plays a part in the Ideal Production Process Level. For a thought experiment, if .999525^(SQM-1) at 4000 sqm is considered, it is 0.15. I wonder if that remaining inefficiency is drawn out over the next production process level curve, say from 4000 sqm to 8000 sqm+.
Very intuitive. I had discussed that a bit at one point here. (Where, for some reason, I suddenly thought HTML superscripts would work and I made a semi-blind guess at the formula that isn't too far off from the above.
Guest

Post by Guest »

Must.... reach.... 4000 sqm!

Do you know if buildings over 4000 sqm have the same upgrade time?
Post Reply