old success rate function

[Wiggume]

does anyone remember the one that was on the old boards? I am taking up tailoring again on a new char and am thinking about trying some acrylia cloaks at skill 135. If I remember correctly I started making them around 125 skill on my first char with 30% success but I want to figure it out again with the formula.

it was something like

success % = skill*1.15 - (trivial -70) I think, is that right?

that formula gives a success of 45% at skill 135 with a 5% mod. The new formula on FAQ is saying 53%.

[Itek]

that's the formula for < 68 skill (or trivial...)

generally it's (skill - (target trivial * 0.75) + 51.5) with a minimum 5 and maximum 95 percent success

it's in the FAQ somewhere I think... about there being 2 formulas

[Lothay]

The answer you seek is in the http://mboards.eqtraders.com/phpBB2/viewforum.php?f=7Tradeskills FAQ FORUM</a>.

~Lothay

[Wiggume]

Yes I read the new FAQ on success rates. I think my first post might have contained too much information.

does anyone remember the success rate formula that was on the old boards? It was something like success % = skill*1.15 - (trivial -70). I just want to make sure I didnt leave something out. Neither of the formulas on the new FAQ is even remotely similar to the 1 formula that was on the old boards.

[Itek]

math, the other science

% = 51.5 + skill - (0.75 * trivial) .... presume 192 skill and 226 trivial

% = 51.5 + 192 - 169.5 = 74

or

% = skill*1.15 - (trivial -70)

% = (192 * 1.15) - (226 - 70) = 64.8

There is a very minor difference in the two formulas. The one in the FAQ seems to be more accurate in the majority of trials.

You might call the FAQ formula "the long run average" formula and the one you have "the conservative formula" ...

For example my wife did HMP (halas meat pies) and got about the expectation for skill up rate and about the expectation for success... a little low but it was a small run (19 attempts) so it was actually closer to the "conservative formula" predictions ... but the difference was 1 success on a run of 19, hardly significant.

So, yes, we understood the question. Yes the FAQ has the "right" formula as far as we have been able to find. But your old formula is pretty close, and not "invalid" by any means.

[Wiggume]

lol I surrender. All I wanted was a 'yes that was the old formula' or 'no you forgot this blah blah blah'. But thanks Itek, your post had some nice info in it. I duno if a 10% difference in success between 2 formulas is insignificant tho.. anyways I guess I remembered the old formula correctly, thx.

[Boleslav]

That's because we aren't answering the question you asked, but instead answering the question we THOUGHT you meant, lol.

Yes, your recollection of the old formula is correct.

Boleslav Forgehammer
Paladin of Brell in his 61st Campaign
E'ci - Sacred Destiny
Blacksmithing - 250 (old school)

[Itek]

Admittedly 10 percent seems like a lot.

And if you were doing 1000 attempts at the same recipe at the same trivial (meaning: you wanted to do 1000 meat pies as a 192 baker, haven't read the patch message for 6/11 yet) the difference between 648 and 740 successes is a great deal.

But if you are doing 2 attempts on, say, Velium Blue Diamond jewelry... The known "streakyness" of the RNG will make itself known.

These are the "long run" expectation formulas.

In the long run you would expect to see the "new formula" with it's higher success.

In the short run you might want to be "conservative" and expect the lower formulas prediction.

Say you need to make leather padding for smithing skill run. You might want to get enough pelts saved up so that even if you get "lower expectation" you have enough to continue the skill run.

Of course your mileage will vary. I just figure with allowing for the "fudge factor" of the RNG sometimes being kind and other times not so much... 2 success in a stack of 20 isn't such a huge thing.