Illarion.org

As some of you might know, we are working on the traders. We have found ways to determine the price of an item on a fair basis. We'd like to know what you think.

Imagine items with an attribute. This attribute ranges from 1 (lousy) to 3 (good). There are three items:

A: attribute=1
B: attribute=2
C: attribute=3

One could say, D is three times "better" than A. How would you determine the prices?

Option 1: Price=attribute
Option 2: Price=attribute^2 (squared that is)
Option 3: Price=attribute^3 (raised to the third power)

For these values:

Option 1:
Price of A=1
Price of B=2
Price of C=3

Option 2:
Price of A=1
Price of B=4
Price of C=9

Option 3:
Price of A=1
Price of B=8
Price of C=27

(note: the numbers are no prices in copper coins, just ratios)

If you need an example, take shields. A is a lousy wooden shield, B is a good metal shield and C is a divine shield of invulnerability. Please cast your vote and leave a short message why you think so. "I just feel like" is a good answer (I am not kidding). Please, do not start a discussion, it won't help at this moment.

Poll will run until saturday. German translation only upon request

Linear is not enough and cubing it is to much, squaring it is best, because it means you get a decent price from your work, but not so much you get huge ammounts of money, with very little work once your a grandmaster.

Squaring seems best to me. That way, things are within the range of affordability for new players, but not without working for it, if that makes sense.

Definitely not a friend of linear, here. I'd say cube it. For one part, A and B are not much of a difference whether you look at square or cubic. However C makes the difference. If I read that C stands basically for a magic shield, then cubed is the way to go, imho. I've always enjoyed the "low magic"-style of gameplay in Illarion, and always enjoyed people RPing out those "Ooh"s and "Aaah"s and #me's depicting astonishment when they see rare or special items. Regardless of magic qualities though, even if it's only the highest quality, I don't see why the highest quality should stand in metric proportion. When you already have a bunch of money, it's easy to make more quickly; so a square system would create a gap between A and C, killing off B, because if you can already easily access C once you could access B, why even go for B? If you're low, it's a long hard road to work up. I don't see anything wrong in that, and the cubic method would uphold this style.

Note: The prices will only hold for "mudane", craftable goods. Unique and rare items will be delt with seperatly.

Thanks for all the votes so far! I'd like to read more short opinions, with "not starting a discussion" I just ment that I just need "pro" arguments.

To help you, I want to give out another detail: It is planned that one gets 10-20% (depending on the demand) of the "real value" of an item when selling to NPCs. That means, if you want to buy an item just from the money you get from selling stuff, you have to sell the following items (I take 10%):

Option 1:
Price of A=1 -> 10 A-Items
Price of B=2 -> 10 B-Items or 20 A-Items
Price of C=3 -> 10 C-Items or 15 B-Items or 30 A-Items

Option 2:
Price of A=1 -> 10 A-Items
Price of B=4 -> 10 B-Items or 40 A-Items
Price of C=9 -> 10 C-Items or 22,5 B-Items or 90 A-Items


Option 3:
Price of A=1 -> 10 A-Items
Price of B=8 -> 10 B-Items or 80 B-Items
Price of C=27 -> 10 C-Items or 33,75 B-Items or 270 A-Items

I guess this would depend on how different the stats of normal and "uber-magic-divine-invicible-artifact"- items are. But I guess the prices you talk about are only for items which are freely available with NPCs anyway and those should only be non magic ones. For such items I think a square pattern is a good choice.

Unique or rare items will have their price dictated by the ingame market, like it has always been.

Which quality and conditionhas the items? that should have influence too,or not?

I'm sure these calculations are assuming the items all have the same base quality and condition. Hence it's irrelevant what exactly they are.

Exactly, Adano. Quality, that is a different story. All items have comparable qualities.

If anybody of you wants to propose another way of calculating prices (exponential? mixed powers like Preis=AttributeOne^3 + AttributeTwo^2 + AttributeThree ? Step functions?) send me a PM. Even though we have two complete concepts "on the desk", we could use another one for comparison and maybe for improving the current concepts.

Opinions and ideas! Bring em on!

Just some more speculation, but I believe a cubic price ladder will ultimately benefit the crafter-type characters.

I voted for cubed... linear seems too 'perfect'.. this shield is twice as good... cost twice the price...

All the numbers seem to exact, could we have a strange constant that all is multiplied with or something? or will it be taking things too far?

*laughing*.

Like an official Illa-Goods Formula =D.

f(weapon) = ......

No more like

Price = Quality(Weapon Type)(1.367) or something

This formula exists. The "strange constant" is also known. It differs for different kinds of items. However, it makes no difference if an item costs 1 quaxel and the other 20 quaxel or an item costs 40 hubis and the other one costs 800 hubis. The ratio is the same. The niveau of prices is again another topic.

I agree with this:

Hadrian_Abela wrote:linear seems too 'perfect'.. this shield is twice as good... cost twice the price...

However, I also don't think that the increase between different price groups should be so large. I think this because with exponential equations, either it will be too hard for newbies to make money, or else experienced craftsmen will become rich too easily. Would it be possible to make it somewhat linear, but instead of
option 1: 1
option 2: 2
option 3: 3

It would be something more like

option 1: 1
option 2: 3
option 3: 5

I think that with something with fixed rates like that, but not as low as the suggested linear model, would be perfect. If this isn't possible, or if you just don't want to discuss it, tell me to shut up and I'll cast my vote for linear.

Zare wrote: It would be something more like

option 1: 1
option 2: 3
option 3: 5

No need to shut up! I can use any input. You propose an offset, your model fits y=2x - 1. This is just linear like 1,2,3 even though it looks different at a first glance. The offset has only an influence if it is in the range of the final value, see below:

Attribute=10 -> 19
Attribute=20 -> 39 (~2 times 19)
Attribute=30 -> 59 (~3 times 19)

An offset would be a nice idea to guarantee a certain level of prices, I will try to incorporate it into a model.