1.14 Wolf Struggle Tests

[Timber Wolf]

(Note: Here are links to the previous test -  .393 and .426)

The test scenario: A 100% stalker character wearing no clothes walking toward a wolf.  Measuring condition loss for 20 struggles.  Used a macro to automate 8.3 mouse clicks per second.  This time I used some different character states, always using a hatchet.  I was carrying 50 kg for the overburdened tests and 6 pieces of fresh venison (3 scent bars) for the stinky tests.

I was a bit surprised by the results.  I thought being exhausted would have an adverse effect.  These results seem to show there is little to no difference between these different states.

Edited September 18, 2017 by Timber Wolf

[Ruruwawa]

Awesome!  I have a question about the data.  Are the numbers in the table remaining condition (n), or condition lost (100-n)?

[Timber Wolf]

1 minute ago, Ruruwawa said:

Are the numbers in the table remaining condition (n), or condition lost (100-n)?

They are condition lost (100-n).  But in the case of the sprained ankle (96-n), and exhausted (98-n).

[Ruruwawa]

@Timber Wolf thanks, just doublechecking.  The SD for the exhausted and overburdened states seem to suggest this states are slightly protective, and that just can't be right lol.  You conclusion, that these states have little to no effect, seems to be right.  Good work, and surprising too.

[Timber Wolf]

1 minute ago, Ruruwawa said:

The SD for the exhausted and overburdened states seem to suggest this states are slightly protective, and that just can't be right lol.

No, that doesn't seem likely!   My guess is that with more tests all of the numbers would even out.

[Ruruwawa]

2 minutes ago, Timber Wolf said:

My guess is that with more tests all of the numbers would even out.

My guess too.  I don't think there's enough evidence of a difference to make more testing worthwhile, though.

[Timber Wolf]

6 minutes ago, Ruruwawa said:

I don't think there's enough evidence of a difference to make more testing worthwhile, though.

Probably not, but I might run some more tests in a state of exhaustion.  I really thought that would make a difference.  And since the sprained ankle tests had a much lower standard deviation, I kind of want to see what happens if I have several injuries.  And now that makes me wonder what difference it might make if I am dealing with intestinal parasites.  Hmmmm.  I guess I have more tests to run.

[Ruruwawa]

1 hour ago, Timber Wolf said:

Probably not, but I might run some more tests in a state of exhaustion.  I really thought that would make a difference.  And since the sprained ankle tests had a much lower standard deviation, I kind of want to see what happens if I have several injuries.  And now that makes me wonder what difference it might make if I am dealing with intestinal parasites.  Hmmmm.  I guess I have more tests to run.

Test assbite while you're at it?  

[MarrowStone]

@Timber Wolf, the exhaustion numbers confused me, I always believed that was the most important factor.

Have you tried with a character with a fully depleted sprint bar instead of exhausted? I've no clue how youd do that consistently though.

Edited September 19, 2017 by MarrowStone

[Timber Wolf]

1 hour ago, MarrowStone said:

@Timber Wolf, the echaustion numbers confused me, I always believed that was the most important factor.

I was also surprised by this and am going to test it again.
 

1 hour ago, MarrowStone said:

Have you tried with a character with a fully depleted sprint bar instead of exhausted? I've no clue how youd do that consistently though.

I did run a depleted sprint bar test back in .393, but it didn't seem to make a difference.  In order to perform the tests, I would spring until the meter ran out and then immediately walk up to a wolf.

On the other hand, in that same test group I did fully exhausted tests and it definitely did seem to make a difference.  Although back then, I was only running 10 of every test, so the numbers might have been a little off.

I'll also add that when I was running these tests while exhausted, it was clear that the fight meter was filling up much more slowly than it normally does.  Never-the-less, the condition I lost wasn't really any different than when in the other states.

[TROY]

I'm concerned about the use of the mouse click macro. While doing so may help standardize the tests, 8.3 clicks per second seems like considerably more than an average player would be able to do without a macro

Edited September 19, 2017 by TROY

[Timber Wolf]

1 minute ago, TROY said:

I'm concerned about the use of the mouse click macro. While doing so may help standardize the tests, 8.3 clicks per second seems like considerably more than an average player would be able to do without a macro

That may well be true, but it's not more than I am able to do.  And since that's the speed I've been using since I started these tests, I need to keep it the same in order to be able to compare the tests through the different versions.

Furthermore, I really don't think it matters what speed I use as long as it's always consistent.  I'm not looking to find out exactly how much condition is lost in a particular situation, as opposed to finding out how much of a difference there is in condition loss between the different situations.

[TROY]

Carry on

[Timber Wolf]

For this next group of tests I focused on the state of exhaustion and I started up a brand new game.  @TROY's question above made me curious as to what the results would look like if I did not use the macro, so I did a round of tests banging away on my controller as I do when I am normally playing the game.  When both of these rounds of tests showed numbers similar to the exhausted tests I ran in the first group, I decided to spin up version .426 using the same game save to see if anything has changed.  Here are the results:

 

As you can see, there was no meaningful difference between these tests.  So exhaustion might have once been a factor, but it appears it has not been for a while now.

[MarrowStone]

To me it almost looks like clothing is now the only factor and clicking.

[Timber Wolf]

19 hours ago, MarrowStone said:

To me it almost looks like clothing is now the only factor and clicking.

The weapon you select is still an important factor.  I've used the hatchet for all of these tests, because my previous testing showed it to be the best option.

[Hotzn]

Excellent work, as always, @Timber Wolf!

[MarrowStone]

3 hours ago, Timber Wolf said:

The weapon you select is still an important factor.  I've used the hatchet for all of these tests, because my previous testing showed it to be the best option.

Oops, and weapon..

[WanderingPalm]

More incredibly useful information from Timber Wolf!

While the damage suffered may not be any different anymore with different exhaustion levels, it's still worth noting that being exhausted makes the struggle bar fill slower (as Timber Wolf mentioned in the test)...though why that doesn't equate to more damage being sustained, I have no idea. So it's not that exhaustion has no effect on struggles, just that it's essentially irrelevant.  I wonder why they changed it?

Actually, except the Stink test, all of these results surprised me. I wonder if seeing the durations for each struggle might shed some more light on the results? Surely fighting the wolf for longer would make you more susceptible to taking damage (regardless of what the exhaustion test showed). Or am I wrong in thinking that the duration is less variable than damage?

[Timber Wolf]

I wanted to find out the difference between the wolves in Stalker and Interloper, so I ran some more tests.  I used my standard scenario - 100% character, well rested, no clothes, 8.3 cps, and an improvised hatchet.  Here are the results:

Edited October 14, 2017 by Timber Wolf
Clarified improvised hatchet for the tests

[Timber Wolf]

On 9/30/2017 at 2:22 PM, WanderingPalm said:

Surely fighting the wolf for longer would make you more susceptible to taking damage (regardless of what the exhaustion test showed). Or am I wrong in thinking that the duration is less variable than damage?

Here's my best explanation:

Each wolf struggle will likely have multiple individual striking events, for both your character and the wolf.  The very first striking event will be the wolf's, unless you click in the very short period of time between the struggle first beginning and the wolf initially making contact with you.  

Each individual wolf striking event calculation begins by using Random Normal Distribution (the normal distribution is sometimes informally called the bell curve*) based on a predetermined condition loss value.  Then the clothing protection value is applied. *Edit: I also think the weapon choice probably has a value applied here.

Each individual character striking event calculation begins by using Random Normal Distribution based on a predetermined condition loss value for the wolf.  Then the weapon choice value is applied.

The struggle will continue until the wolf's condition falls below a particular threshold or your character dies.  Sometimes you can get in the first strike and win the struggle immediately, incurring no condition loss at all - or very little condition loss if you almost time it just right.  And I think the faster you click, the quicker you will have your next striking event. 

*

Edited October 14, 2017 by Timber Wolf

[earthy]

Interesting results, but I'm not sure if I can get behind the conclusion. In the exhausted state you seemed to experience quite a few short(<10%condition lost) struggles, which seem to heavily influence the average. I'm also inclined to think hatchet may not be the best choice for a test like this, given that shorter duration struggles would be more prone to variance influencing the conclusion. On the other hand, it's also possible given the stalker vs interloper results that stalker is simply more forgiving. Do you expect the results of Normal vs Exhausted in interloper would show the same trend? 

Nice work once again though!

[Salty Crackers]

How did you get a hatchet in Interloper? If you're using the improvised hatchet, the results will change drastically, regardless of which difficulty. I would recommend using the hammer for the Stalker v. Interloper tests.

[Timber Wolf]

Just now, Salty Crackers said:

How did you get a hatchet in Interloper? If you're using the improvised hatchet, the results will change drastically, regardless of which difficulty. I would recommend using the hammer for the Stalker v. Interloper tests.

Sorry, I should have been more specific.  I used an improvised hatchet for both the Stalker and Interloper tests.

[Timber Wolf]

So, how important is it to have protective clothing?  I ran more tests while wearing a new wolfskin coat, deerskin pants, deerskin boots, rabbit mittens, and a toque - giving me 29% protection.

An interesting side note - I scared off the wolf 18 times while getting these 50 results.  So, the scare rate was 18/68 = 26.5%