success = better chance for skill up? site calculator says no

[mewkus]

I'm not sure if there's something wrong with the eq traders calculator, but when I enter
Skill: 225
Mod: 5%
Trivial: 384
Base Skill: 420
Difficulty: Easy
AA: 3
Uses Alternate stat: Yes
(checking grandmaster combine in smithing)
I get:
Chance of skill up on Success: 8.5%, on failure: 8.93%, overall 8.72%

That's a higher skill up chance if I fail than if i succeed. While granted 0.43% isn't all that great a difference, it's something that should be the other way one would think. Or am I not entering the data correctly?

[Ngreth Thergn]

I may have a bug to look into

may not be able to check until wednessday though.

[Ngreth Thergn]

found my bug

Ajusted Skill = 236
Success chance = 49.75%
Chance of skill up on Success: 8.5%, on failure: 8.5%, overall 8.5%

[YurienStonebow]

now can you explain it all in easy terms so i can understand it? the formulaes are confusing enough without everyone being so techinal about it

[Neebat]

now can you explain it all in easy terms so i can understand it? the formulaes are confusing enough without everyone being so techinal about it

Formulas by their nature are extremely technical beasts. I tried very hard to explain them to my wife and I think I got through the success rate formula and stopped. The best way to understand the results is probably some examples, and you can run those with the site calculator.

Then again, it depends on how much you want to know.

The Success Rate is the same as always. Only your skill and mastery matters. For trivial recipes, the RAW skill matters. Geerloks don't help. Mastery has a huge impact when you're very bad, not so much when you would probably succeed anyway.

Skilling up is hard. Two conditions have to succeed. There's no indication in the game which test failed.

The first skillup check is twice as hard when you failed to make the item. It uses your Wis/Int/other stat. Some skills take more combines to skill up and that shows up in the first formula. Here's a little table that helps me thinking about the first condition:
I'll ALWAYS pass the first check if I get my wis/int over...

Tradeskill:      On success:       On failure:
Alchemy             415               815
Baking              315               615
Brewing             315               615
Fletching           400               800
Jewelcraft          415               815
Poison Making       200               400
Pottery             415               815
Smithing            200               400
Tailoring           215               415
Tinkering           215               415

Notice that it's fairly easy to skip the first test when you succeed on most skills. In fact, with a skill under 15, passing the first test means an automatic skill up, and that's just neat.

If the first test fails, the second check never happens! Since the second column of that table is mostly impossible, a failed combine is likely to prevent you from skilling up. Anything that prevents you from passing that first test is immediate bad news, so get your stats up and do combines you're likely to succeed on. Support a gnome and buy a geerlok!

The second test is only involved if you pass the first test. The second test doesn't know anything about:
* The recipe
* Your stats
* The phase of the moon
* The (trade)skill that you're trying to raise.

The ONLY thing involved in the second test is your current raw skill level. That means it makes a nice pretty curved graph which is the same for every character on every skill. It gets a lot higher beyond 250 skill. The recent changes were all done in that second test.

(The table above was calculated based on the tradeskill numbers in another thread, some of which are unconfirmed. ALCHEMY is looking VERY suspicious to me now, compared to poison making. Is it possible the number is inflated because people find it very hard to advance based on the cost of components and not actually the number of combines? Or is alchemy really that nasty?)

[Ngreth Thergn]

(The table above was calculated based on the tradeskill numbers in another thread, some of which are unconfirmed. ALCHEMY is looking VERY suspicious to me now, compared to poison making. Is it possible the number is inflated because people find it very hard to advance based on the cost of components and not actually the number of combines?)

Alchemy was given to us directly from the code by Tanker.

[Neebat]

Alchemy was given to us directly from the code by Tanker.

So it IS true then.

Wow. I don't have a shaman, but... wow.

I haven't studied alchemy enough to understand the philosophy behind it. I know alchemy is often compared to Poison Making, but it's really hard to compare, isn't it? This means alchemists face not only higher prices on combines, required foraged components they can't forage, but also twice as many combines per skillup.

Is justified because the rewards are much better for alchemists? There's a much wider market for potions than poisons. Is it, like research, supposed to be an optional skill for the class? Supposedly the advantage of researched spells is supposed to be minor/supplemental while the core defining abilities are store bought. Most of a shaman's potions do nothing for the group with the shaman present, while poison making IS a big help to a rogue in the core function of DPS. So poison is a basic, straight-forward skill, while potions are an optional, bank-draining slog?

I dunno, just trying to see if that makes sense.

[Boleslav]

Re check the formulas. It will not take twice as many combines to skillup in Alchemy than Poison Making. It will take more, but not twice as many.

The amount more will be influenced by the spread between the trivial of the recipes being attempted and the skill at which they are attempted.

Boleslav

[Lothay]

Is justified because the rewards are much better for alchemists? There's a much wider market for potions than poisons. Is it, like research, supposed to be an optional skill for the class? Supposedly the advantage of researched spells is supposed to be minor/supplemental while the core defining abilities are store bought. Most of a shaman's potions do nothing for the group with the shaman present, while poison making IS a big help to a rogue in the core function of DPS. So poison is a basic, straight-forward skill, while potions are an optional, bank-draining slog?

Actually, I believe at the time Alchemy was revamped (the very first version of Alchemy was severely borked) the philosophy to to prevent Alchemy from becoming "Shaman in a bottle" and eliminating the need for shaman in groups. So Alchemy was made difficult, time-consuming and expensive. And the more useful potions were really difficult/expensive to make.

I haven't actually played EQ since April of 2003 at this point, but there was a great deal of very heated debate about the topic back in the day.

[Neebat]

Re check the formulas. It will not take twice as many combines to skillup in Alchemy than Poison Making. It will take more, but not twice as many.

The amount more will be influenced by the spread between the trivial of the recipes being attempted and the skill at which they are attempted.

Boleslav

Sorry, Boleslav. I really wanted to say something like "the shaman has to do up to twice as many combines under similar conditions" but by the time I had all the disclaimers in there, the sentence was incomprehensible even by my standards.

Of course the difference between poison making and alchemy is not constant. In fact, it can actually be more than double.

So let's check the formula. Here's how I was figuring it:

I pick 200 in the base stat because that's exactly where dex stops having any impact on the rogue's odds of skilling up from a successful combine, and it's fairly easy for anyone to achieve.

At 200 dex, when successfully making a poison, the rogue has a 100% chance of passing the first check. (200 / 1*2 = 100)
At 200 wis, when successfully making a potion, the shaman has < 50% chance of passing the first check. (185 / 1*4 < 50)

I didn't bother to factor in the second formula or to run the failure case, because the impact is exactly the same from there.

To run this check through the calculator on this site, I had to pick a specific skill level and trivial. I used a skill of 10 and a trivial of 20, so that both get high odds of skilling up.

Shaman gets a 36% chance of skill up.
Rogue gets a 78% chance of skill up.

Looks like more than twice as many combines to me, but then, I'm neither a rogue nor a shaman.

[KB1]

What is borked is the vendor price on a sale of a potion. Poisonmaking > by far in that aspect, so you can't just look at the number of attempts, but also what those attempts are going to cost you.

[Boleslav]

Sorry to be so abrupt in my last post, I didn't have time to poke into it in depth, but felt certain that it would not take twice as many combines to skillup in Alchemy as in Poison Making. Here is a little deeper look into things.

From 175+ both Alchemy and Poison Making will have recipes with trivials closely enough spaced so that you should have a decent chance to succeed the combines. Let's use a fictitious probability of 80% for a successful combine.

What have done is to calculate the chance to get by Pass 1 on a fail, then on a success. Then I multiplied the chance on a success by .8 and the chance on a fail by .2 and added them up to get a more rounded view of the chance to get by Pass 1.

% Chance to clear Pass 1
Stat Value-------200-------250-------300------350-----400
Alchemy----------42--------53--------64-------75------ 87
Poison------------90-------100-------100------100-----100
Ratio-------------2.2-------1.9-------1.6-------1.3-----1.2

The row labeled Ratio is the result of dividing the Alchemy percentage into the Poison Making percentage. It should be the 'multiplier' used when figuring out how many more Alchemy combines than Poison Making combines it is likely to take to get skillups.

Now that I have crunched the numbers, I must admit it is a lot closer to double the numbers than I would have thought. If you are doing your skillups at around 250 stat-wise it could easily be double. My guess is that most people heading over 250 will probably be in at least the 300+ range though.

Still, it is a major difference unless you can pump your stats sky high. As a 67 Paladin with no stat AA except the first DoN reward I top out at 300 WIS. That puts me at a huge disadvantage compared to someone with higher stats in a 'difficulty 4' tradeskill.

Boleslav Forgehammer
Paladin of Brell in his 67th Campaign
Tunare (E'ci) – Sacred Destiny

[Neebat]

Still, it is a major difference unless you can pump your stats sky high. As a 67 Paladin with no stat AA except the first DoN reward I top out at 300 WIS. That puts me at a huge disadvantage compared to someone with higher stats in a 'difficulty 4' tradeskill.

It's cool to have confirmation. Thank you for checking my math. I definately recognize that it does get better with extremely high stats. I'm still pretty comfortable with my 200 Wis vs. 200 dex comparison. In fact, 200 dex is an easy number for rogues and 200 wis may be tough for shamans until the high end.

Shamans aren't high elves or erudites. The highest shaman in my guild still had under 200 wisdom at level 60! Admittably, he was really bad at buying gear and he's gotten better now. (I think he still has at least one +INT aug.)

If a rogue is focusing on DPS, as they should be, STR and DEX are likely to be near caps even without buffs. My wife's halfling rogue has had a dex over 200 a long time, and she still has +INT augs.

Sorry to derail this thread so badly. To the original point, the calculator seems to be working now.

[kadasi]

HMMMMMMM, all this formula stuff gives me a migraine. Think I'll just continue chalking all the skill up stuff to magic, and trusting Tunare to see that it all works out ok. LOL

[Tejaye]

Calculator

Where can i find this calculator to put those numbers into?