Queen Of Thieves beta 0.9.1 (release candidate)

[jack1974]

Sorry, as always it's late so might have not understood completely: you are saying that it's still not as much XP as the "original" 0.8.0 ?
I'll find those debug output that escaped my axe

[myoksy]

We knew beforehand that your new formula results in lower XP gains : ) But now it's possible to estimate the rate - the highest enemy level I saw (12, in the second dungeon of the catacombs) gives about half the XP in 0.9.0 compared to the calculation used in 0.8.0.

As I said before: I don't know if advancing will feel slow reaching mid-game and actually _playing_ the game. But I am not very confident, mostly because the pacing was imo excellent until party level 20 or so in a game with normal difficulty.

I will report back when I played the current game longer.

[myoksy]

you are saying that it's still not as much XP as the "original" 0.8.0 ?

just realized that my XP prediction was wrong - incorrect order of calculations : ) so it will not be as drastic as I feared (50 % less experience around lvl 12) but still a notable difference between 0.8.0 and the current 0.9.0.

but those are technicalities and it's probably more helpful to discuss a more basic topic: I am not sure why the XP gains should generally lower. the only explanation I can think of is that you believe the game is too short (or this was reported outside of the forum). to present an opposite view: imo the length is fine, something like 20 hours per play-through (ballpark estimate) and replayability is high (four times to see all romance combinations).

only the last phase before end game was a little bit too abrupt as a result of the accelerated XP gain with high-level enemies, but this can be solved without slowing down the earlier chapters. e.g. a cut-off for typical enemies in the last chapter before the climax: int(self.cExp+(self.cExp* min([self.level, 21])/15)) - without the complex XP formula you formerly used in environment still a tad lower than before ; )

[myoksy]

a more optical thingy: PC had a hit chance of 78%, the enemy dodged the attack and the log line reads "..., damage considerably reduced! (15% melee dodge chance)". shouldn't this read "22% dodge chance"?

[jack1974]

No I think the length is fine too. To be honest I didn't had the material time (between various projects and other real-life stuff going on) to retest the game as much as I wanted after the XP change
Looking at your formula, I need to browse python docs to see what min does
But I guess that's a formula that speeds up XP gain after level 21 or so ?

a more optical thingy: PC had a hit chance of 78%, the enemy dodged the attack and the log line reads "..., damage considerably reduced! (15% melee dodge chance)". shouldn't this read "22% dodge chance"?

No because the hit chance depends also on the attackers' stats. The one in the log is the "base" target dodge chance. So in your case if you had a 7% bonus due to critical hit.
Probably is better to remove that value from the log since could be confusing and doesn't really add much (since the final "right" calculation is the one in the damage preview).

[myoksy]

But I guess that's a formula that speeds up XP gain after level 21 or so ?

exactly the other way round

min() selects the lowest value of a list. so until an enemy level of 21 the additional gain is <lvl>/15, higher level foes give the same XP as a if they are lvl 21 (to slow down leveling up in the last phase of the game)

as an example: base XP 20

lvl 1: 20 + (20 * 1/15) = 21
lvl 10: 20 + (20 * 10/15) = 33
lvl 20: 20 + (20 * 20/15) = 46
lvl 21: 20 + (20 * 21/15) = 48
lvl 35: 20 + (20 * 21/15) = 48
lvl 666: 20 + (20 * 21/15) = 48

No because the hit chance depends also on the attackers' stats. The one in the log is the "base" target dodge chance. So in your case if you had a 7% bonus due to critical hit.
Probably is better to remove that value from the log since could be confusing and doesn't really add much (since the final "right" calculation is the one in the damage preview).

thanks for the explanation and yes, removing would make sense - the value is not especially informative for mere players

[jack1974]

Thanks for the explanation. Now that I think more carefully, some time ago during early beta I lowered the XP needed to level up the latter levels, so that maybe speed up everything.
Anyway I'll just use your formula, no point in doing other tweaks now
I will do another update tomorrow, I also added a few lines in Yenna scene, if you don't pick the "pacifist" option, will explain that you obtain the letters anyway.

[myoksy]

some time ago during early beta I lowered the XP needed to level up the latter levels

the old back-and-forth game : D

Anyway I'll just use your formula, no point in doing other tweaks now

it's effectively your formula used until 0.8.0 with a small tweak

[myoksy]

I just tried it with a new normal game (first two catacomb levels, first story dungeon and some mansions)



the absolute XP values (Y axis) are not that important, it's the difference between the three calculation methods - and the one I proposed is still lower than the old method used until 0.8.0. I don't know if it's that important that the XP gains are mostly unchanged, if this should be wanted instead of "<lvl>/15" something like "<lvl>/11" is probably more exact.

for RL reasons I cannot offer my time as tester this weekend

[jack1974]

Maybe this weekend will have some time to test myself, but I think the new system (the one that will go live in the update tomorrow) should be "high enough". I don't think we necessarily need to go back to the old XP though!