Jump Shot Settings

[Borch]

Based on my prior analysis, I think a 70% reduction would be appropriate. Less than that (80 or 90 percent) seems to not really do much.

[Borch]

Looking through the logs today, it appears that actually a 70, 80, or 100 percent difference in jump shooting reduction makes no difference, they are all about the same, so there is no reason to change it from 100 percent. Basically, the only thing that would make a difference is just shutting off the jump shot, but that won't fly.

[10001110101]

It seems our hurdle is that of two issues being solved by one solution. It could be that's just outside what's feasible given the game and our resources. I've wanted the jump shot issue dead, buried, and forgotten since the beast first reared it's head probably a decade ago now.... Having said that I do want the game to be enjoyable for all and I've racked my brain trying to come up with solutions that compromise with everyone. Because there are two issues even communicating them has been difficult. Even for those that seem to understand there are multiple issues involved etc. I've sat here several nights now looking through the extreme setup trying to find anything that might be helpful. There are settings that limit how long someone can be in a certain stance etc. (player control) and I would have sworn that I've played servers that had similar restrictions on jumps. For instance if you jump once you go full height but subsequent jumps each get less and less effective. But I cannot find these settings if they exist in the mod. If there are settings that can impose limitations on successive jumps I would think we might be able to use that (maybe combined with other settings) to at least try something. But again I came up blank. I don't know if it's worth approaching Patmansan at this point or not. His site is still up but it seems to be archived/mostly dormant.


Borch, I appreciate that you attempted to bring data to the table to try and make this an informed decision for all of us. I don't know if it's my own cognitive limitations or the limitations of the data but I just had a hard time figuring out how to apply it given (I assume) it included all weapon data (thus lumping the two issues) as well as the unexpected results that seem to run in the opposite direction expected at 90% (again I might be misunderstanding). There's so much in the combo of the game/mod that seemingly ends up with unintended results/consequences it has to be frustrating to do much of anything with it. Again, thank you for trying to make it manageable and accessible.


I realize the data seems to show little change at 80%. While I don't dispute that in any way there were demonstrable effects to rifle jump shots as Happie and I tested them. Does that play out in live play? I can't say for certain but I don't see why it wouldn't. I would expect that in the data rifle jump shots are a fairly small percentage of the total (admittedly a guess) but if our goal is to reduce the impact of this aspect of the game maybe it's worth doing even if it doesn't solve everything? I can't believe I'm the one breathing life into this but here I am.... I really do want things to be enjoyable for everyone. I'll try to clear my thoughts up again tomorrow and maybe add more then. I've got at least one more thought but it touches on everything in the game so I'm hesitant to bring it forward given how change frustrates many of us... me more than most.

[Borch]

The amount of damage done to a target is the multiplication of three (four) factors, afaik:
the 'BASE DAMAGE' of the weapon
the 'LOCATION multipiier
the reduction due to jumping
and not considered, the reduction due to range

So for a weapon, the base damage that results in death can be calculated. 100 hit kills a target, so:
100=base_damage*multiplier*reduction
so for a specific weapon, the amount of reduction needed to make it hit 100 exactly is:

reduction=100/base damage/location_multiplier

any result greater than 1 does not kill, ever on a single shot.
Note that all these calculations up to this time, and data from the server log files has been for a single kill.

weapons.pdf

spreadsheet showing changeover values for jump reduction

(87.73 KiB) Downloaded 801 times

[TheBigBobert]

You would need total jump shots divided by jump shots tanked to get the right % you're looking for.

While shots fired is a large number, the number of players is a small sample size on any given day. And, most likely, the number of people utilizing jump shots frequently.

Therefore, you need to look at who was actually playing each day to see if the notorious jump shooters were playing.

For example, many of the days with low jump shots tanked were days I didn't play

In combination with the advice above, if the data exists for individual players, that could be more useful for determining the effectiveness of jump shots in terms of total shots taken, as players like me always play with the same weapon and our abilities don't vary too much day by day.

(This doesn't take into account corrective actions taken when handicaps are put in place. First day of changes a lot of different things didn't work and it was tough for everyone to play. After the initial corrections it was tough but it forced me to pay more attention and be even more accurate. By the time we got up to 100% again it felt like it was 200% -- like training with a weighted vest!)

[Borch]

Indeed, looking at all of the 440,000 some-odd shots that were in the logs would result in a small number of shots.

Note that the PDF is the analysis of how the game works, not including any log information. If you have a hit of 45 and a multiplier of 2.5, that would result in 112.5 damage, which is over 100, which would cause a kill.

But so if you wanted to find out for a weapon with a 45 hit that had a 2.5 multiplier's cross-over point, where a reduction would cause a death exactly, you calculate the amount that would reduce 112.5 to 100.

112.5x=100
x=88.8%
So any reduction over 88.8 % would result in a tank, not considering other factors, such as range. So a reduction of 85% would result in a kill, 89% would result in a tank. The pdf shows the cross-over points.

Note that the server had been running at a reduction for a few weeks before the meeting (I believe at 60%, I don't recall exactly, though the logs seem to indicate that) then the week following the meeting, it was kicked up to 70%, then up to 80%, etc., so there should not have been any "getting used to" the jump shooting settings if you had played at all before the meeting.

So when analyzing the logs, I looked at the weapon hit location times the base_damage of the weapon to see what the normal hit was. If this value was reduced by the amount that it was set to for that week, then I took that as a jump shot.

Doing that, the actual numbers of tanks due to the jump-shot reduction was pretty small. I've not looked at it for a while, though, so I do not remember the exact amounts, nor do I wish to put any more effort into analyzing it.

"You would need total jump shots divided by jump shots tanked to get the right % you're looking for."

Note that a jump shot is not logged if it is a miss. Candidate jump shots were determined by:
those where the actual damage amount = the base_damage * hit_location_factor * reduction_for_that_week
But since this calculation matches that first discussed above, it doesn't really add much information.