How do I edit the "Boss Loot" under "Loot" file?

[choi]

How do I edit the "Boss Loot" under "Loot" file? 

Please tell me how to edit "Boss Loot" to return specific items, I am not familiar with this so it would be helpful if you could provide detailed explanation.

[Boidster]

Can you be more specific about what you are trying to do? Are you talking about adjusting the loot that comes from the dropped bags from Demolishers? (These are the only zombies which currently have the 'boss' loot drop)

[BFT2020]

First, the lootgroup we think you are talking about, it has 4 groups in it that the game will reference to.  If you notice count="all", that means the game will go to each group.  So for groupZpackBoss2, the game will roll 1 item.  For groupZpackBoss03, there is only a 50% chance of an item pulled from that group (as determined by loot_prob_template="med".  If it said count="1" instead of "all", then only one of the 4 loot groups would be chosen based on probabilities.

 

<lootgroup name="groupZpackBoss" count="all">
    <item group="groupAmmoLarge" count="2"/>
    <item group="groupZpackBoss01" count="2"/>
    <item group="groupZpackBoss02" count="1"/>
    <item group="groupZpackBoss03" loot_prob_template="med" force_prob="true"/>
</lootgroup>

 

To go further, lets continue with groupZpackBoss2 group.  Here is that lootgroup below:

 

<lootgroup name="groupZpackBoss02">
    <item group="groupApparelClothes" loot_prob_template="low"/>
    <item group="groupApparelHazmat" loot_prob_template="medHigh"/>
    <item group="booksAllScaled" loot_prob_template="medHigh"/>
    <item group="groupToolsTiered" loot_prob_template="medHigh"/>
    <item group="groupArmorScaled" loot_prob_template="medHigh"/>
    <item group="groupCraftingRare" loot_prob_template="med"/>
    <item name="armorMiningHelmet" mods="modArmorHelmetLight" mod_chance="1" loot_prob_template="veryLow"/>
    <item name="armorSwatHelmet" mods="modArmorHelmetLight" mod_chance=".2" loot_prob_template="low"/>
</lootgroup>

 

The game is going to pick one item from that group and the odds are based on the prob templates listed.  Lets say the game picked groupCraftingRare.

 

lootgroup name="groupCraftingRare" count="1">
    <item name="resourceForgedIron" count ="1,5"/>
    <item name="resourceForgedSteel" count ="1,5"/>
    <item name="resourceMilitaryFiber" count ="1,5"/>
</lootgroup>

 

The game is now going to pick one of those items (note count="1").  Because there are no probabilities listed, you have an equal chance to get Forged Iron, Forged Steel, or Forged Fiber.  Then it will randomize how much you get based on the count.

 

If you want something to be guaranteed, you can enter this inside the groupZpackBoss lootgroup:

 

<item name="toolBeaker" count="1"  loot_prob_template="guaranteed"/>

 

Then when the bag drops and the person loots it, they will be guaranteed to get a beaker in the loot bag every time.

 

If you want to keep randomization, you can also change the loot groups.  For example, groupZpackBoss02 can be changed to

 

<lootgroup name="groupZpackBoss02">
    <item name="toolBeaker" loot_prob_template="med"/>
    <item name="armorMiningHelmet" mods="modArmorHelmetLight" mod_chance="1" loot_prob_template="veryLow"/>
    <item name="armorSwatHelmet" mods="modArmorHelmetLight" mod_chance=".2" loot_prob_template="low"/>
</lootgroup>

 

Now only 1 of those 3 items has the possibility of being pulled.

 

TFP have created baseline lootgroups which are then referenced by other lootgroups.  This allows them to reduce coding and just reference back to basic lootgroups for containers.  This is something you and I can easily do ourselves if we want to change loot drops.

 

That is a simple explanation, but as Boidster mentioned in his reply, further help will require more details from you.

[Boidster]

2 hours ago, BFT2020 said:

If you want something to be guaranteed, you can enter this inside the groupZpackBoss lootgroup:

 

<item name="toolBeaker" count="1"  loot_prob_template="guaranteed"/>

 

I need to go test this, but:

 

1) I think, for the specific example of the groupZpackBoss group, since it has count="all", you would not need to use loot_prob_template at all. Just put <item name="toolBeaker" count="1" /> in there and it will guarantee a beaker.

 

2) That said, putting the loot_prob_template="guaranteed" wouldn't hurt anything; unless there is a force_prob="true" on an item in a count="all" group, every entry is considered to have a 100% probability of being chosen anyway, as you pointed out.

 

3) I think that in more generic lootgroups which do not use count="all", that loot_prob_template="guaranteed" does not acutally mean guaranteed. I think you'd need to add the force_prob="true" in there, but even then I'm not sure how it will react. Consider:

 

<lootgroup name="foobar" count="2">
  <item name="foo" loot_prob_template="med" />
  <item name="bar" loot_prob_template="guaranteed" />
</lootgroup>

 

That ought to resolve to this (my claim here is that there isn't anything special about the "guaranteed" probability template - it just maps to 1.0):

 

<lootgroup name="foobar" count="2">
  <item name="foo" prob="0.5" />
  <item name="bar" prob="1.0" />
</lootgroup>

 

And I think that means a 33% chance for a "foo" and a 67% chance for a "bar", which would be checked twice because of count="2". If we add force_prob="true" in there:

 

<lootgroup name="foobar" count="2">
  <item name="foo" loot_prob_template="med" />
  <item name="bar" loot_prob_template="guaranteed" force_prob="true" />
</lootgroup>

 

Then...it gets weird, I think? This is really what I need to test. I believe the force_prob flag means "do not add this up with the other probabilities, just use the raw probability as is". Which would mean, I think that both "foo" and "bar" would have 100% chance of returning. Because "foo" is the only item which is to be added up & divided out and 0.5/0.5 = 1.0 and of course "bar" has a forced 1.0 probability.

 

The point I'm getting at - my theory, which I need to test - is that in spite of its name, "guaranteed", the loot_prob_template doesn't actually guarantee anything except in particular situations where force_prob="true" is used. Where I might be going wrong is if the code is actually testing for the word "guaranteed" and then just monkeying with the calculations behind the scenes. That is, it doesn't treat it like any of the other prob templates. Could be.

 

Test results coming soon.

 

[BFT2020]

30 minutes ago, Boidster said:

*snip* lots of good information  *snip*

 

Test results coming soon.

 

 

Yes, I agree.  The guarantee is probably not necessary if "all" is the value for count, but I am doing everything as "theory" while at work and can't test anything out until I get home and have some free time.  Plus, I offered to do some data collecting on weapons to narrow down damage calculations (already got my special loot file that gives me only compound bows no matter what I loot 😉  ) 

 

I wonder if you could ever put so many conditions in the loot tables (and so many sub loot groups) that your computer would just give up and crash to desktop 

(or better yet, just troll people and not put any comments in it so they get lost trying to figure out where it is pulling from and how).  😁

[Boidster]

11 minutes ago, BFT2020 said:

I am doing everything as "theory" while at work and can't test anything out until I get home and have some free time.

 

Of course it would be absolutely wrong to be "working from home" and check in on the forums from time to time. Ha ha ha *tugs at suddenly-too-tight collar*