Proposals/Obsolete Proposals/Stats, Experience and Mission Count in 2.5-dev
This is a work-in-progress and specific values are likely to change. Please consult the discussion page for more details.
This is a proposal for improving soldier development for 2.5 or 2.6. It addresses the slow pace with which soldier stats increase and the way in which some improved stats are pointless or not significant enough.
Soldiers gain experience in each stat for participating in battle and performing certain actions (such as shooting an alien) relevant to that stat. The game then converts this experience into stat points which are added onto the soldier's initial stat.
The factors effecting soldier development consist of the following:
- The number of missions they participate in.
- The number and type of actions they perform in those missions.
- The experience they get for each of these actions. (See: Skills/Improvement
- The equation which converts experience into stat points.
- The soldier's initial stat points.
- The equations which use a soldier's stat points to effect actions on the battlefield (ie - how the Accuracy stat effects a soldier's actual accuracy).
Number of Missions
Currently a game takes place over the course of more than 100 missions. On playing through, I think this is excessive and plan to reduce this to about 60-70. In addition to the number of overall missions, I hope to improve the incentives for players to field more than one squad, which will spread the experience over more soldiers.
All numbers in this proposal assume that a veteran soldier is likely to see 30 missions, but may see around 40-50 in exceptional circumstances.
Conversion of Experience to Stat Points
Currently, the game converts experience points into stats with the following equation:
New Skill = Original Skill + (Experience/100)^0.6
This should be changed to the following equation:
New Skill = Original Skill + (Experience/10)^0.6
This will dramatically increase the rate of stat gain, converting fewer experience points into more stats. In addition, it will create a sharp curve so that soldiers will gain quickly in their first missions, making new recruits still worth keeping alive past the middle of the game.
All experience adjustments below assume this new equation.
Experience gain for power should be cut to 1/3 its current rate. This would keep the current growth rates (which seem appropriate) in the new experience conversion equation.
- Estimated increase ~30 missions: 13-19
- Estimated stat ~30 missions: 48-54
Experience gain for speed should be increased 3x its current rate. This stat should not increase TU too much or it would quickly become the most powerful stat.
- Estimated increase ~30 missions: 35-53
- Estimated TU increase ~30 missions: 8-10
- Estimated stat ~30 missions: 55-73
- Estimated TU ~30 missions: 38-41
Cause for experience gain
Currently this is based on walking, crouching and using weapons. This provides adverse incentives to always move your guys fully each turn. After some discussion, we've decided to use a cumulative stat growth mechanism, so that speed experience gain represents a percentage of all other experience gain.
Experience gain for accuracy should be increased 2x its current rate. This stat should not increase as quickly as the weapon skills, but if it increases too slowly it will be a drag on the improvement of any weapon skill because the two are combined when computing accuracy.
- Estimated increase ~30 missions: 21-41
- Estimated stat ~30 missions: 46-66
Utility of stat
Accuracy, along with weapon skills, doesn't improve effective accuracy enough. After some discussion, we've decided to try a new accuracy equation:
accuracy = 1 - ((accuracy_ability - 10) / 50 + (weapon_skill - 10) / 50) / 2
Experience gain for mind should be kept at its current rate. This stat is highly dependent on our psionic warfare, so it is difficult to set good rates now. The cause for experience gain is likely to change with psionic warfare. Soldiers will need to compete against alien mind stats, so we will need a large range in the Mind stat. In addition, our panic system hasn't been looked at in a long while and would also factor into this stat. Keeping the current rate, alongside the new equation, will give us a good range of values for future tweaking.
- Estimated increase ~30 missions: 39-50
- Estimated stat ~30 missions: 66-77
Cause for experience gain
This should be more closely related to fear and panic, but I haven't yet thought of a good mechanism. Ideas include getting wounded, seeing a new alien, seeing teammates killed or wounded.
Rank requirements may need to be adjusted to fit this new profile. I haven't looked into that.
Experience gain for hp should be reduced to 4/10 its current rate. I think this shouldn't grow too much because it could unbalance a lot of weapon details in the late-game.
- Estimated increase ~30 missions: 27-34
- Estimated stat ~30 missions: 127-134
Experience gain for all weapon skills should be increased by 20% in general. However, we should adjust individual rates to equalize the typical gain.
- Estimated increase ~30 missions: 34-66
- Estimated stat ~30 missions: 54-86
Note: estimates range from 3600 to 10,800 exp. Most vets will range from 3600-6000, but the estimates above take into account players' tendency to focus exp on a few vets.
Cause for experience gain
The experience for weapon stats varies a lot because some roles are cross-trained more (Assault), some are less frequently used (Close) and some get fewer points per kill. After looking at the data, my sense is that the difference in experience given for different stats is unnecessary. The two highest-experience skills (Explosives and Sniper) each given 200 exp per kill. Assault, which gives just 100 exp per kill, typically lags behind. Close, which gives 150 exp per kill, also lags behind, but that is likely because this skill is more specialized and used less often. Accuracy doesn't matter as much for Close specialists, so it's okay to lag behind.
I think all weapon skills should get 180 exp per kill, which is the original 150 + 20%.
Here is a list of items that need to be changed to implement this proposal.
- [open] Change the equation which converts experience points to stat points: New Skill = Original Skill + (Experience/10)^0.6
- [open] Change the experience-per-mission formula for strength. Reduce multiplayer to 1/3 its current rate: 50 * (weight / strength) / ENCUMBRANCE_MODIFIER
- [open] Change the cause of speed gain. It should work like HP, where it is a fraction of all XP accumulated elsewhere. It should be 0.5 points for every point earned elsewhere. (Can someone check the math on this for me? I think if it has the same equation that HP has now, before the change, it should be about 3x it's current rate. But something in my head is giving me a nagging doubt. I'm basing the assumption on the ability points improvement graph in the talk page. --H-hour 11:12, 7 August 2013 (CEST))
- Actually I think that would closer to 5x, but there would be wider variation, using the data from your spreadsheet the HP XP gained is about 2x to 7x the speed XP gained (with an exceptional 13x case), but if we'll increase the XP gain across the board... --DarkRain 02:52, 8 August 2013 (CEST)
- [open] Increase the experience gained for accuracy to 40 per enemy hit.
- [open] Increase the experience gained for accuracy per hit while using the sniper skill to 60 per enemy hit.
- [open] Change the equation which calculates accuracy.
- [open] Adjust soldier rank thresholds to ensure it works properly with new mind stat growth.
- [open] Reduce the experience gained for HP to 0.3 experience for each experience point earned elsewhere.
- [open] Set the experience gained for all weapon skills to 180.