Analysis of Basic IO Costs [long]
#8101062 - 04/20/07 12:58 PM
Edit post Edit Reply to this post Reply Reply to this post Quote Quick Reply Quick Reply
Part I
This post will attempt to analyze the inf costs of basic (non-set, single aspect) IOs. This is an area that I believe still needs some balance work, and I'm hoping that the devs have time to address a few concerns.
Part I of the post is this section, the introduction.
Part II is a re-post of something I wrote up in the closed beta, before I had seen the actual crafting and recipe prices as they appeared on the tables. Other than a few edits for clarity I am re-posting it verbatim. Part II is itself a long post, so be warned.
Part III is a comparison of the recommendations I made in Part II with what actually exists in-game on Test. I believe there are some ranges of basic IOs where the cost is out of line with the value of that IO when compared with training, dual-origin, or single-origin enhancements available at the same level range.
OK, let's begin. Remember the next post is a few weeks old at this point.
Guide to Market Participation
Common IO Pricing Analysis
Post Extras: Print Post Remind Me! Notify Moderator
Protea
Legend~Market Watch
Reged: 09/25/04
Posts: 1969
Loc: in ur auction stealing ur l00t
Re: Analysis of Basic IO Costs [long] [Re: Protea]
#8101096 - 04/20/07 01:02 PM
Edit post Edit Reply to this post Reply Reply to this post Quote Quick Reply Quick Reply
Part II
This post will analyze the current cost of the TO/DO/SO upgrade path and compare the resulting total cost of ownership with never-expiring Invention Origin enhancements. Starting from some basic design goals and the existing upgrade economy, I will derive suggested total costs for common Invention Origin enhancements.
Note that I have not looked at what common IOs actually cost; I deliberately did this calculation before looking so that I would have an unbiased outlook.
I know that others have posted on this, but the calculation is so long I thought it needed its own thread. Lemur_Lad can snowflake me all he wants
I. Groundwork
Power curves
The power curves for common IOs and TOs/DOs/SOs are quite different. While IOs gradually increase in power with level, the current system (with required upgrades every 5 levels) gives a sawtooth pattern. Power bumps up at levels 2, 7, 12, etc, then gradually decrease. For each of the three domains of this graph (the TO, DO, and SO domains) the average value of this sawtooth function is a constant.
It is possible, though hardly common practice, to bump up the average value of the sawtooth by refreshing enhancements twice or even three times per 5 level period. We will use the cost of this uncommon practice in comparing costs with IOs at some of the mid- to high-level upgrade points.
Terminology
First, for a given enhancement slot in a given power, we will call the cost of investing in enhancements for that slot over the lifetime of a character the total cost of ownership (TCO) of that slot. For TOs/DOs/SOs (hereafter referred to as TDSOs), that total cost is paid out in increments. For common IOs that survive the life of the enhancement slot, that cost is paid out in one lump sum.
Note that the inf savings, if any, of switching over to an IO system midway through the career of the character actually decrease with level, because of the costs already paid out into the TDSO system.
Note also that IOs do in some sense become obsolete; the point at which an IO is superceded by the average benefit of the TDSO system is called the crossover point, and the number of levels between the purchase of the IO and the next crossover point is called the IO's lifetime.
Guiding principles
In order to determine a fair price for common IOs, we need to establish some guiding principles, which are the mathematical expression of desired game results. They are fairly easily stated as follows:
- Common IOs need to be economically competitive with the SO upgrade path. This is something of a sine qua non; if they are not economically competitive, no one will use them when there is a way to get equivalent power with TDSOs, i.e. prior to lvl 45. The savings also have to be substantial enough to justify the additional complexity of dealing with storing, sorting, and potentially even buying recipes and salvage.
- Common IOs need to be economically feasible for characters with no inf benefactor. If the system is economically beneficial but out of reach for poor characters, due to lower TCO but a staggering up-front cost, we end up with a "rich get richer, poor get poorer" kind of system.
- Common IOs need to have a tangible and substantive cost. While perhaps not directly unbalancing, having common IO costs that are too low will neuter crafting badges and render the use of consignment for recipes moot.
You may ask whether it's possible to satisfy both the second and third constraints simultaneously. I do think this is possible, and I'll explain why in the calculations.
II. Calculations
Average benefit for TDSOs
We would like to know what the average of the TDSO sawtooth is for the best-case upgrade scenarios, so that we can compare them to IOs:
Code:
Upgrading once per 5 levels: (+3, +2, +1, 0, -1): Avg TDSO level: +1
Upgrading twice per 5 levels: (+3, +3, +2, +1, 0): Avg TDSO level: +1.8
Upgrading thrice per 5 levels: (+3, +3, +3, +2, +1): Avg TDSO level: +2.4
Note that upgrading multiple times every 5 levels doesn't give quite an overall +1 benefit, since the bonus is still capped at +3; still, we can fudge factor this a bit and say that to get +2 average SOs, you'll need to upgrade twice, and to get +3 average SOs, you'll need to upgrade three times.
TCO Table for Damage TDSOs
I chose damage TDSOs because the numbers are (usually) round, and they are perhaps the most common enhancement type overall. Other types of enhancements are simple scale factors of these numbers: accuracy is slightly more, recharge slightly less, etc.
Code:
Level Type Cost Remaining
5 TO 500 284965
10 TO 1000 283965
15 DO 5985 277980
20 DO 7980 270000
25 SO 30000 240000
30 SO 36000 204000
35 SO 42000 162000
40 SO 48000 114000
45 SO 54000 60000
50 SO 60000 0
Total TCO 285465
Comparison Table for Damage IOs
Code:
Level IO % Equiv Lifetime Breakeven Cost
10 11.7 -3 DO 5 1000
15 19.2 +3 DO 10 13965
20 25.6 -3 SO 5 N/A
25 30.2 -1 SO 0 none
30 34.8 +1 SO 23 240000
35 36.7 +2 SO 18 408000
40 38.6 +3 SO 13 486000*
45 40.5 +4 SO 8 N/A
50 42.4 +5 SO 3 N/A
** Attaining this value with SOs requires double upgrades; breakeven cost is 2x
*** Attaining this value with SOs requires triple upgrades; breakeven cost is 3x
To calculate the breakeven cost, I looked at the corresponding table entry in the TCO table above. For example, after buying a level 25 damage SO, I have 240k inf left to invest in that slot. If at level 30 I buy a damage IO instead, I have saved myself 240k minus the cost of that damage IO.
I did fudge the breakeven costs a little bit. In order to get +3 DO efficacy, a character would need to upgrade 3 times every five levels; but the difference between a +1 and a +2 DO is much less than the difference between a +1 and a +2 SO, so the double- or triple-purchase upgrade plans are not very appealing. At the higher levels, however, these other purchase plans are the only real way to compare IOs to TDSOs, so I left them in place.
The breakeven cost represents the maximum anyone should be willing to pay for an IO. In other words, going above this cost would immediately violate the first guiding principle I listed above.
In order to be competitive with TDSOs, IOs need to cost significantly less than the breakeven cost. But how much less? To answer that question, I need to point out an anomaly in the comparison table:
At level 25, there is no power incentive to use IOs as opposed to SOs. A level 25 IO is actually less powerful than its TDSO equivalent. This is the only level in the entire curve where the lifetime of the IO is "zero".
There is a subtle opportunity here. Characters with inf benefactors have zero incentive to use IOs at that level; instead, the rational plan is to use SOs at 25 and switch to IOs at 30. So level 25 IOs are only relevant to poor characters. We can tweak the cost of these IOs in order to satisfy the third guiding principle above, by making level 25 IOs cost the same amount as level 25 SOs. Characters can then buy into the IO system with enhancements that never expire and pay only a small penalty in their power curve. The rich might still be able to get richer, but the poor can now get richer, too.
With all this in mind, I present:
III. Final Results
Suggested Total Costs for Damage IOs
Code:
Level IO % Equiv Lifetime Breakeven Cost Suggested Cost
10 11.7 -3 DO 5 1000 1500
15 19.2 +3 DO 10 13965 9000
20 25.6 -3 SO 5 N/A 12000
25 30.2 -1 SO 0 none 30000
30 34.8 +1 SO 23 240000 120000
35 36.7 +2 SO 18 408000 240000
40 38.6 +3 SO 13 486000 300000
45 40.5 +4 SO 8 N/A 400000
50 42.4 +5 SO 3 N/A 500000
Note that the values for 45 and 50 IOs represent an extrapolation of the table's existing trends, since there is no direct comparison in the TDSO system.
I think this table does a good job at satisfying all constraints. IOs represent a cost savings opportunity, at the burden of increased complexity. The cost savings are substantial enough to warrant spending some money in the Consignment system acquiring recipes and salvage pieces. At the same time, they are expensive enough that crafting badges and memorized recipes have a tangible benefit.
Recipe vs. Crafting Costs
The above costs represent the total of the recipe and crafting costs. In the interest of simplicity, I would suggest splitting this equally between the two (i.e., at level 10, the recipe costs 750 inf and the act of crafting costs 750 inf).
Parting thoughts
That was quite a long post - even longer than I'd imagined. As I mentioned before, I have not actually looked at what the inf costs are in game for crafting various levels of common IO.
I'm hoping that what I see in-game will mesh with these calculations, in which case, perhaps this post will guide some of the justifications for the rates staying as they are. If not, then this is my argument for rebalancing around the numbers I give.
Peace,
- Protea
Guide to Market Participation
Common IO Pricing Analysis
Post Extras: Print Post Remind Me! Notify Moderator
Protea
Legend~Market Watch
Reged: 09/25/04
Posts: 1969
Loc: in ur auction stealing ur l00t
Re: Analysis of Basic IO Costs [long] [Re: Protea]
#8101172 - 04/20/07 01:11 PM
Edit post Edit Reply to this post Reply Reply to this post Quote Quick Reply Quick Reply
Part III
Now that I have finally had time to dive into this again (apologies, devs and other testers), let's look at the actual recipe + crafting costs for basic IOs:
Code:
Lvl Recipe Craft Total Recommended % of Rec.
10 1700 3400 5100 1500 340%
15 3550 7100 10650 9000 118%
20 16350 32700 49050 12000 408%
25 17550 35500 53050 30000 176%
30 30975 41300 72275 120000 60%
35 47025 63700 110725 240000 46%
40 74475 99300 173775 300000 57%
45 177200 177200 354400 400000 86%
50 464400 464400 928800 500000 185%
There are some interesting properties of this table.
First, I strongly believe that the level 10 and 20 IOs are grossly overpriced. There is absolutely no incentive to even craft one of these things given that the 5 levels it will take for them to become obsolete compared to TDSOs will take very little time to work through, and you'd be better served fighting and gaining XP than dealing with the salvage involved.
It's possible that badgers will flood the market with level 10 IOs they crafted, but other than that, level 10 IOs are going to be completely ignored. Why even have them if they cost this much? Level 20s have a tangible advantage in their bonuses, but they become outstripped so quickly by SOs that again I think most people will ignore them in practice.
As an aside, I was astonished that the recipe cost at 10-20 is so significantly less than the craft cost. If the devs inverted this relationship, lvl 10 and 20 IOs might still see some use by people who got lucky drops. As it stands, even the crafting cost for these items serves to marginalize them.
Second, I think there is a problem at levels 25-40. In this level range, basic IOs are still expensive enough that non-twinked characters are going to have huge difficulty justifying the expense. Yet they are so cheap at 30 compared to the current cost of twinking characters with SOs that sugar-daddy sponsors would be foolish not to use them. This is exactly the rich-get-richer, poor-get-poorer problem I was trying to avoid in my analysis: twinking not only gives you more power over the life of the character, it also saves money. I don't think twinking should be encouraged to that extent.
I would like to re-iterate my suggestion that level 25 IOs be normalized in cost to SOs. They are weaker than SOs at that level, so fully twinking min-maxers will ignore them anyway. This is the golden opportunity for Cryptic to allow people to buy into the IO system in a deliberate way that won't cripple the characters' other slotting at that level.
Last, I will fully admit that the numbers for levels 45 and 50 basic IOs were kind of pulled out of a hat. I actually have very little problem with what Cryptic chose to do here. The level 50 cost seems a bit high, but given that it can potentially allow 2-slotting instead of 3-slotting, it's hard to figure out how to map that onto an inf cost.
OK, that's it for now. Devs, you now have my opinion on record
Peace
- Protea
