High stat = more skillup NOT successes

[Ngreth Thergn]

Sorry for another sticky thread but I want to squash rumors.

One of the Devs (tanker) managed to see and understand the formula for skillups while working on the new UI. (He ***MAY*** even be willing to post it if he is allowed) He was constantly interrupted while he was talking about it though, so what he was saying got confused by folks.

Due to the formulas... there is a greater chance of a skillup on a COMBINE SUCCESS than a failure. This can be AS HIGH as (but not necessarily always) 100% higher chance of a skillup on a Combine Success as a Combine Failure.
The HIGHEST of int/wis (or str/dex where they matter) has a "mitigating" effect on the skillup chance DIFFERENCE between a combine failure and a combine success.

As an arbitrary example (by me)

Let’s say you have a BASE 20% chance to get a skillup on a combine failure, and a 40% chance to get a skillup on a combine success...

With a High Stat (highest int/wis or alternate) you will still have a 40% chance for a skillup on a combine success... but may have say a 30 % chance of a skillup on a combine failure. (remember ARBITRARY numbers here... no quoting this as promises or exact…)

Because of the interruptions his language got confused... and when he said "success" he meant "successful skillup" not combine success.

So he is only talking SKILLUPS!!!!!

A side effect of a better chance of a skillup on a combine success, means you can skillup faster when you try items closer to your trivial, because you have a higher combine chance when the trivial is close than when it is far away. I.E. more combine successes = faster skillup. The high stat means you have a higher chance of a skillup on a combine failure... but never as high as the chance of skillup on a combine success. But it will overall increase your chance of skillups since people do tend to fail from time to time.

Again he was only talking SKILLUPS when we talk about the benefit of the high stat

[dmuerte]

This is new?

Ive noticed this for a long time, this is why i went up in jc by slowly changing gems to the next harder one and not just starting out with plat bars

Its nice to have them come out and actually admit it though, thanks for the info

Hey wait a minute does this mean that if you use a gearlock you get more skillups because you have a better chance of success?

[Taushar]

Originally posted by dmuerte
Hey wait a minute does this mean that if you use a gearlock you get more skillups because you have a better chance of success?

I think the answer is depends on. From what I understand from the Fan Faire is that if your stats, like wis/int/dex/str, is low, you may be better off with using geerlock since the skill up difference between success and failure can be large. However, if you are at higher stats, it should not matter and you may actually benefit from higher stats than geerlock successes. This also differs from tradeskill to tradeskill, for example the skill up rate for brewing will be different than smithing given all the rest being equal (like stats).

Taushar

[Ooby]

Ok, quick agreement

Tinker is at 198
Making elemental bow cams trivial at 282, i get less skill ups, than on geerloks which are trivial around 215.

It is true.

[arturio emerys]

This makes sense to me, it is most noticable in Jewelcraft when you closely follow the trivial progression. Going that way, you suceed perhaps 90% of all combines and you progress faster.

Art

[Masamune]

How about...

the "rarest" skill - begging?

What stat does it rely on and how are skillups controlled on it? It is INT or WIS based, or is it CHA based (or INVERSLY so? Low CHA = higher skillups)

Asking this because I have been "stuck" at 193 begging for, well... years. Being an enchanter, my CHA and INT are pretty well maxxed, of course but I have been unable to get a SINGLE skillup out of hundreds and hundreds of attempts, both on NPCs and on pets.

Actually, this started in the 170's and I spend my remaining training points to get it as high as it is. Since doing so over a year ago, I've tried extensively to skillup but have yet to ever succeed in getting a point.

Since begging IS used in the BIC quest, this might be a pertinant question to bring up now (even though >150 skill isn't required).

[Thicket Tundrabog]

It sure is great to get this finally sorted out.

I recall the general consensus (although not unanimous) was that skillups were higher the closer you got to the trivial. We had our suspicions but we weren't sure why this was the case.

There was a lot of disagreement on whether you got more skillups on successes, failures or it didn't matter.

Thanks for the summary.

Thicket

[Bregalad]

At very low skill (below 60), making recipes that triv not far above your skill, geerlok gives a tiny advantage, since its positive effect on success chance outweighs negative on 2nd check. For example, at 50 making triv 60 with geerlok skillup chance is better. At 50 making triv 100, having a geerlok lowers skillup chance.

As you go higher, negative effect of geerlok ramps up, peaking at 181. (at 181, on easy skill, making a triv 200 recipe at 265 stat, your chance to skillup without geerlok is almost twice higher than with it). From 181 to 190, geerlok negative tapers off, and at 190, geerlok is only positive.

I made an Excel spreadsheet, which can easily show skillup chance at any given point.

B1: raw skill goes here
B2: recipe trivial goes here
B3: skill hardship goes here (i.e 4 for smithing)
B4: Mod % (5 for geerlok)
Success % chance without skill mod:
B5: =MAX(MIN((B1 - (B2 * 0.75) + 51.5),95),5)
Success % chance with skill mod:
B6: =MAX(MIN(((B1 * (1+(B4/100))) - (B2 * 0.75) + 51.5),95),5)
B7: Adjusted stat. (highest of int/wis/alt, -15 where applicable)
Skillup % chance without skill mod:
B8: =MIN(1,(B7*10)/(B3*10*(200-B5))) * (200 - MIN(190, B1))/2
Skillup % chance with skill mod:
B9: =MIN(1,(B7*10)/(B3*10*(200-B6))) * (200 - MIN(190, B1 * (1+(B4/100))))/2

[Adainae]

I'm going to test this today. The other day I did 400 combines of unfired star ruby stiens for 13 skill-ups (217-230) So today I'm going to try the red ceramic bands (trivial at 242) and see if I get more skill-ups per number of combines.

Looking forward to it working

EDIT: 240 combines for 230-242. *cry* all that money spent on star ruby stiens!!

[Kraggie]

Geerlok's do not affect the 2nd check. The number used there is your unmodified skill. Check the other threads on the skill-up formula, etc and you'll see where this has been confirmed.

Skilling-up by using recipes with trivials close to your current unmodified skill does seem to give you an advantage in the Skillup per success/failure department, so stick with that

[Bregalad]

If that is indeed the case, then yes, skill mod item should be used throughout.

B1: raw skill goes here
B2: recipe trivial goes here
B3: skill hardship goes here (i.e 4 for smithing)
B4: Mod % (5 for geerlok)
Success % chance without skill mod:
B5: =MAX(MIN((B1 - (B2 * 0.75) + 51.5),95),5)
Success % chance with skill mod:
B6: =MAX(MIN(((B1 * (1+(B4/100))) - (B2 * 0.75) + 51.5),95),5)
B7: Adjusted stat. (highest of int/wis/alt, -15 where applicable)
Skillup % chance without skill mod:
B8: =MIN(1,(B7*10)/(B3*10*(200-B5))) * (200 - MIN(190, B1))/2
Skillup % chance with skill mod:
B9: =MIN(1,(B7*10)/(B3*10*(200-B6))) * (200 - MIN(190, B1 * (1+(B4/100))))/2

Adding a new formula then:
*** Skillup % chance with skill mod, assuming 2nd check not affected:
B10: =MIN(1,(B7*10)/(B3*10*(200-B6))) * (200 - MIN(190, B1))/2

[Luddaen]

Ok, I dont have excel, but I have a C++ Compiler and know some basic C++. What would be the C++ code to create a program to calc this automaticaly?

[Verdandi]

We have some more current discussions of the various formulas floating around now, so unstickying this.