Champion system stars: diminishing returns

Valencer

I think it might be interesting to discuss the champion system stars and their diminishing returns. So here's 2 examples:

The Tower: Magician
Reduces the Magicka cost of spells by x%

# of points - total cost reduction - increase for point

1. 1.0% 1.0%
2. 1.5% 0.5%
3. 1.9% 0.4%
4. 2.3% 0.4%
5. 2.6% 0.3%
6. 2.9% 0.3%
7. 3.2% 0.3%
8. 3.5% 0.3%
9. 3.7% 0.2%
10. 4.0% 0.3%
11. 4.2% 0.2%
12. 4.4% 0.2%
13. 4.7% 0.3%
14. 4.9% 0.2%
15. 5.1% 0.2%
16. 5.3% 0.2%
17. 5.5% 0.2%
18. 5.7% 0.2%
19. 5.9% 0.2%
20. 6.0% 0.1%
21. 6.2% 0.2%
22. 6.4% 0.2%
23. 6.6% 0.2%
24. 6.7% 0.1%

The Lady: Elemental Defender
Reduces Flame, Frost, and Shock damage dealt to you by x%

# of points - total damage reduction - increase for point

1. 1.0% 1.0%
2. 1.6% 0.6%
3. 2.2% 0.6%
4. 2.6% 0.4%
5. 3.1% 0.5%
6. 3.5% 0.4%
7. 3.9% 0.4%
8. 4.3% 0.4%
9. 4.6% 0.3%
10. 5.0% 0.4%
11. 5.3% 0.3%
12. 5.7% 0.4%
13. 6.0% 0.3%
14. 6.3% 0.3%
15. 6.6% 0.3%
16. 6.9% 0.3%
17. 7.2% 0.3%
18. 7.5% 0.3%
19. 7.8% 0.3%
20. 8.1% 0.3%
21. 8.4% 0.3%
22. 8.7% 0.3%
23. 8.9% 0.2%
24. 9.2% 0.3%
25. 9.5% 0.3%

Questions:

• If 0.1% is the lowest the system can go, does this mean all the points from 25 to 100 in the Magician star give 0.1%?
• EDIT: Just checked another star with lower values and some points into this star actually don't give an increase after a while (for instance: every other point does not increase the value). Is that intended?

Suggestions:

• Give us a template character with at least 100 CPs so we can fully test a star

Any other thoughts on this? Does it seem like the diminishing returns are too high or too low?

I'll post some more passives later. Writing them down takes a minute or 2.

Terminus

Actually, I was just thinking about this a few days ago in regards to effective end-game testing.

For example, what if one of the passives that requires 75 points is completely overpowered or wonky? Based on the fact that it takes 400k exp to get just one point, aka 1200k exp for each attribute, it will be a major pain in the ass to test those without help from Zenimax.

Valencer

Good point. Every constellation has 4 passives and the last one is unlocked for 120 points invested into the constellation. I guess we should get at least 120 points then.

Berinima

I actually would like to have a template with all 1200 Points because I would like to know how the resource pool scales (you get some percentage on your health/magicka/stamina for every point you spend) - I feel this is the main source for the power gap that might occur. Because as we all know skills scale with your pool and this is outragously stupid in the first place.

Valencer

Yeah, that's one thing I noticed too.

0 points into The Warrior: 12495 health
24 points into The Warrior: 13177 health

0 points into The Thief: 7958 stamina
23 points into The Thief: 8425 stamina

0 points into The Mage: 17973 magicka
23 points into The Mage: 19028 magicka

Seems to scale off your full max stat after applying gear and attribute points (not base) and makes quite a difference. And this is just 23-24 points into each category.

EDIT: The stat boosts seem to have diminishing returns too:
0 points into The Warrior: 12495 health
12 points into The Warrior: 12956 health (+461)
24 points into The Warrior: 13177 health (+682)

Also added another star to the OP.

zward887_ESO

Clearly this system needs to display at least the second decimal, possibly even the third decimal point.

madangrypally

Here is my current spread sheet.

https://docs.google.com/spreadsheets/d/1toa6WFzSwn21JZ6IL6jOZKSj6pXZCVzBy_xHBP0KGDs/edit#gid=900954785

If you notice in the bottom there is a 1-100 point spent table. If you want to see what you get at certain points just put them in the edit section.

Rev Rielle

If anyone is curious, for the values in the opening post:

Reduces the Magicka cost of spells by x%:
y = 1x0.604 (R² = 0.99)

Reduces Flame, Frost, and Shock damage dealt to you by x%:
y = 1x0.6981 (R² = 0.99)

Both are obviously power functions (unfortunately I can't get the superscript BBcode working here), x being the points invested, with y the resulting percentage reduction.