Artisan Archive

Currently, it does absolutely nothing. This may be changed in the (near?) future. If you check out the thread entitled "This week's Mail *draft* 8-21-03" thread (the last one pinned at the top of this forum), you can see it's on the docket for review by the Devs.

I'll repost this from an earlier note I put up....

As a crafter, I have been curious about what the impact of crafting tool use and quality (experimented or not) has on experimentation, so I did a little study.

I made four weapon crafting tools with the same material. Two were not experimented on (basic) and two were fully experimented (I was able to get them to 84%). My character is a novice weaponsmith and had Firearms 1 and Techniques 1 at the time I did this study. I then took one of the basic and experimented tools and did a marathon grind of about 250 projectile rifle barrels using them.

That left me with four specialty crafting tools A) basic new; B) basic used; C) experimented new; D) experimented used. I then made 28 blaster power handlers (7 with each tool). All power handlers were made with identical resources. I checked each power handler before the experimentation stage to insure they had the same starting stats (3-Min Dam; 5-Max Dam; Attack Speed-0.11). Experimentation was then run keeping track of level of success, final minimum damage; final maximum damage; and impact on attack speed. Here's the results:

Success Level Min Dam Max Dam A.S.
Basic Used 3.00 3.14 4.71 -0.1257
Exper Used 3.00 3.43 4.14 -0.1200
Basic New 2.00 3.71 7.43 -0.1814
Exper New 3.29 2.86 6.57-0.1857

All Used 3.00 3.29 4.43 -0.1229
All New 2.643.29 7.00 -0.1836

All Basic 2.46 3.43 6.07 -0.1536
All Exper 3.15 3.145.36 -0.1529

A new specialty crafting tool improved max damage (p=0.02) and attack speed (p=0.04), as compared to a used one, which was statistically using a student t-test even with this small a sample size. Interestingly, using a unexperimented tool versus an experimented tool had no statistical impact on the final product. The p-values on final minimum damage, final maximum damage and attack speed were greater than 0.5. The p-value on success type was 0.17.

Again, my sample size is small. Guess my conclusion is keep those tools fresh! As for whether an experimented tool has in impact, the jury is still out...