Experimenting to test forge boon bias

General
1

The aim:

To see if gums give category boons equal weight, for example, special increases the chance of special boons, on a hammer that would be aoe and magnet. We assume these are weighted equally, because if not it seems like it would be an unfair set up, but run over a x amount of attempts, does it prove to be so?

And does the type of tool affect the weighting of the boon chance?

The set up:

A single diamond hammer, a boon compound that adds 1000 boon points, special gum, character and forge with maxed stats.

The method:

Add 2 special gums then activate boon compound. Then use deconstruction resin 3, which should give back 75% of mats this will also be interesting to test).

Repeat 100 times.

This could be potentially repeated multiple batches of 100 (depending on how much time/enthusiasm/material loss I have)

Then repeat process with a spanner.

5 Likes
2

You will for sure to this on the test Server i hope^^

3

Test server seems easier on RNG. Be a good idea to do this on Live- I would be willing to provide the Decon 3 for this test. @fidach

1 Like
4

I don’t have access, so live it is!

2 Likes
5

I was forging last night on hammers and shovels. On three different hammers I had the magnetic come up three times in a row, use the boon removal, it came up, used the removal, it came back up, used it again, it came back up. Got another good, don’t remember what, had one slot left for a boon, got mag again, just to see what happened, I removed it, got mag again, deconstruct. Of my boons that I had used, out of compound boon 3 I used 8 and I got only 3 back.

Similar results with all three hammers and shovels, and I always would get back much less of the boon 3 than I used, used 5 got 1 back, used 7 got 2 back.

Frustrating, It is Rigged ! ! ! lol

2 Likes
6

Something has changed with the forging RNG % in the last few months.

1 Like
7

HA! And here i was thinking i imagine Things!
3 times decon 3 and no mats, 4 times magnetic in a row…yes, something has changed!

1 Like
8
Running results at half way through first batch (after 50)

AOE: 22
Mag: 22
Other: 6

2 Likes
9

This is probably a no…but I wonder if RNG changes when forging more then 1 item. I usually forge things in batch’s of 3.

1 Like
10

I have certainly felt that way as well.

Though the risk of material loss is just too high to do this kind of experiment with 5/6 tools!

1 Like
11
results after 100 attepmts

AOE: 37
Mag: 45
Other: 18

My take from this, if the results are approximately accurate of the percentage chances, is that there is far too high a chance of hitting a non-special boon after adding 2 special gums. You could interpret that the chances of hitting AOE and mag are the same, within the variance tolerance, but that chance of hitting a non-special is about what I would expect from a single gum, 2 gums should half that.

My other biggest take is that the boons tend to hit the same in runs.

It would go:

Boon
1,1,1,2,2,2,2,2,1,1,1,1
Rarely
1,1,1,1,2,1,1,1,1
Never
1,1,1,2,1,2,2,2,2

My observations show (not recorded) you are much more likely to hit the same boon on your next roll. Which does not match the the actual results (if you just take the percentage of 1v2v3, three being non-special, you should be more likely to hit a different boon)

2 Likes
12

Thanks for testing this. So from what i understand, if you get boon and decon it then further attempts it appears your more likely to get that same boon again multiple times in a row (May be reading this wrong)? Also that using 2 gums is not giving you a better chance of targeting a boon specific to that gum?

13

This actually seems pretty logical. If we do this with one gum and then again with 3 gums, we might be able to work what the supposed percentage chance would be.

14

This is what seems to be happening.

I don’t know about this, we would have to try with 1 gum and 3 gums to evaluate that. If we say with 2 gum there is a 40/40/20, maybe with one gum it would be 35/35/30, and 3 gums 45/45/10. Or maybe it would stay 40/40/20.

Would need a bigger sample set to know.

15

At a players view being a forger, does that seem broken to you?

16

Well, if that pattern continued, and it was 45/45/10, then that actually doesn’t seem too horrible. For some vigor you can get a lot of control of variability, especially when adding in other recipes too.

17

but if you get magnet six times in a row, that seems like something is off to me. Right?

18

Well, that can just be rng, or it might be influenced on what the previous boon was.

Having it be 40/40/20 isn’t horrible, especially if you use the boon transmute. That’d make it nearly 80/20, but it may be potentially higher based if transmute favors same boon type first.

That’s the unfortunate part of rng. We’d need a higher sample size to see if it’s truly 40/40/20 or 45/35/20. Based on previous tests done by our guild that, I think it is pretty close to 40/40/20.

19

If a fair coin was tossed 6 times in a row, would HTHTHT be a more likely sequence of events, or HHHHHH?

2 Likes
20

Math…
image

1 Like