Difference between revisions of "Talk:Invention Accuracy Recipe"
m (added sig) |
m (updated a few comments) |
||
Line 6: | Line 6: | ||
A couple of points I noticed | A couple of points I noticed | ||
1) the high levels really get crazy amounts of resale value. A level 50 whom I have been doing most of my testing with gives a distorted pciture of the whole. | 1) the high levels really get crazy amounts of resale value. A level 50 whom I have been doing most of my testing with gives a distorted pciture of the whole. | ||
− | 2) a Level 15 IO crafting cost is a | + | 2) a Level 15 IO crafting cost is a pretty good 'bang for the buck', slightly worse than a DO at only a minimal incremental cost (assuming the recipe dropped) which keeps you goin until much higher |
− | 3) Level 20 recipes costs seem out of whack. Both visually with the kick in the curve, and are more expensive than the theoretical base cost of a level 20 SO, to which it is inferior. | + | 3) Level 20 recipes (and to a lesser extent level 25) costs seem out of whack. Both visually with the kick in the curve, and are more expensive than the theoretical base cost of a level 20 SO, to which it is inferior. |
+ | 4) level 30 IO recipes offer ~90% of an SO value, for ~110% of the cost and are at 'sweet spot' in the cost/benefit. | ||
Traditional accuracy enhacers have a perfectly linear cost per level (above 10 anyway) | Traditional accuracy enhacers have a perfectly linear cost per level (above 10 anyway) | ||
Line 16: | Line 17: | ||
The best fit I got to IO recipe costs was exponential, but its not quite right. I'll try a few more complicated models when I get the chance. Knowing Crytpic it will be easy to get to in Excel, its just a matter of finding the key. | The best fit I got to IO recipe costs was exponential, but its not quite right. I'll try a few more complicated models when I get the chance. Knowing Crytpic it will be easy to get to in Excel, its just a matter of finding the key. | ||
[[User:Catwhoorg|Catwhoorg]] 05:51, 11 April 2007 (PDT) | [[User:Catwhoorg|Catwhoorg]] 05:51, 11 April 2007 (PDT) | ||
+ | [[User:Catwhoorg|Catwhoorg]] 08:26, 11 April 2007 (PDT) |
Revision as of 15:26, 11 April 2007
I plotted out the recipe costs alongside those of SO,DO and TO. Note inf is a log scale.
A couple of points I noticed 1) the high levels really get crazy amounts of resale value. A level 50 whom I have been doing most of my testing with gives a distorted pciture of the whole. 2) a Level 15 IO crafting cost is a pretty good 'bang for the buck', slightly worse than a DO at only a minimal incremental cost (assuming the recipe dropped) which keeps you goin until much higher 3) Level 20 recipes (and to a lesser extent level 25) costs seem out of whack. Both visually with the kick in the curve, and are more expensive than the theoretical base cost of a level 20 SO, to which it is inferior. 4) level 30 IO recipes offer ~90% of an SO value, for ~110% of the cost and are at 'sweet spot' in the cost/benefit.
Traditional accuracy enhacers have a perfectly linear cost per level (above 10 anyway) TO = 96*level -384 DO = 383*level +383 SO = 1152*level +1152
The best fit I got to IO recipe costs was exponential, but its not quite right. I'll try a few more complicated models when I get the chance. Knowing Crytpic it will be easy to get to in Excel, its just a matter of finding the key. Catwhoorg 05:51, 11 April 2007 (PDT) Catwhoorg 08:26, 11 April 2007 (PDT)