Tweet
I also remember their being a patch message/message board post about the skillup rate being doubled for post-200 skills. If that is true, then it is not taken into account in this formula (the dev may have left it out, or it may be buried elsewhere in the spaghetti code, or it may just not exist). When I have more spare time, I'll dig around and see if I can find it.
-Baradur
EDIT: Finally, here it is. From the March 6, 2001 patch message (roughly 6 months after the tradeskill cap was raised to 250):
"- Doubled the chance to gain skill points in skills that are above 200."
Since this was specifically mentioned as a *doubling* of a skillup chance, and only post-200, it has to be something more than the 190-cutoff. Which would also explain the "hell levels" in the high 180s and 190s that are so often talked about. From 190-200 your skillup chace would bottom out (at 5% or less depending on stats and skill difficulty), and then after 200 it would go back up to 10% (where it was at 180).
Incidentally, it should be quite possible to differentiate between easy, medium, and hard tradeskills by skilling up to 100 on a mule with 215 int/wis (or 200 for a skill with an alternate stat) (200 gives the largest percentage gaps for skillups between each different difficulty level). Yes, testing the alternate stat property should be possible, too, but it would be much harder and take many more samples. With the difficulty test, you only need enough data to differentiate between 100% (50%), 67% (33%), and 50% (25%) skillup rates (parenthetical values are on fails), whereas, for a hard skill, the -15 int/wis would only cut your skillup chance by 4%, and that's going to take a *lot* more tests to see with any real certainty.
Skilling to 100 on a single mule with 215 int/wis (or 200 int/wis/dex/str, as appropriate) and recording the number of succeses and number of fails at each level to get a skillup should be enough data to peg the skill's difficulty with reasonable certainty, after some analysis of the data. I'll try to do such tests myself sometime soon, but my raid schedule is pretty hectic these days, so it may take a while.
-Baradur
EDIT: Finally, here it is. From the March 6, 2001 patch message (roughly 6 months after the tradeskill cap was raised to 250):
"- Doubled the chance to gain skill points in skills that are above 200."
Since this was specifically mentioned as a *doubling* of a skillup chance, and only post-200, it has to be something more than the 190-cutoff. Which would also explain the "hell levels" in the high 180s and 190s that are so often talked about. From 190-200 your skillup chace would bottom out (at 5% or less depending on stats and skill difficulty), and then after 200 it would go back up to 10% (where it was at 180).
Incidentally, it should be quite possible to differentiate between easy, medium, and hard tradeskills by skilling up to 100 on a mule with 215 int/wis (or 200 for a skill with an alternate stat) (200 gives the largest percentage gaps for skillups between each different difficulty level). Yes, testing the alternate stat property should be possible, too, but it would be much harder and take many more samples. With the difficulty test, you only need enough data to differentiate between 100% (50%), 67% (33%), and 50% (25%) skillup rates (parenthetical values are on fails), whereas, for a hard skill, the -15 int/wis would only cut your skillup chance by 4%, and that's going to take a *lot* more tests to see with any real certainty.
Skilling to 100 on a single mule with 215 int/wis (or 200 int/wis/dex/str, as appropriate) and recording the number of succeses and number of fails at each level to get a skillup should be enough data to peg the skill's difficulty with reasonable certainty, after some analysis of the data. I'll try to do such tests myself sometime soon, but my raid schedule is pretty hectic these days, so it may take a while.
Ngreth Thergn
I was just commenting to some other players about the attention the tradeskill community got from the Dev team.
Comment