I am looking both at successful smelts and at unsuccessful smelts. We need both types of data in order to figure out how alloys work. In fact, failed smelts are especially important if we want to be able to predict success vs. failure.
If you would like to contribute data to this page, please send me the following information:
- name of each mineral
- for each mineral, the number of clusters in the cart
- (optional, but desired) for each mineral, the number of stones in the cart
- the amount of brass that resulted from the smelt
Bronze
I have only collected data for three smelts; not enough to be worth examining. Please send me your data!
Brass
Success vs. Failure
While in the field, cluster ratios are the easiest value to track and use. My recommendation is to shoot for a cluster ratio of (N+1) copper to N zinc. This ratio seems to succeed much more often than not.
However, cluster ratio is not 100% reliable when trying to predict success vs. failure. If you look at the graph you will see that ratios around 1:1 or N+1:1 usually succeed, but not always.
Element ratio is also does not predict success vs. failure:
Total cluster count is also not a 100% reliable tool for predicting success vs. failure. If you look at the graph, you will see that the two failure points fall into the middle of the field in terms of total cluster count. 
Looking at this graph, I am interested by what appears to be a second, steeper line. That suggests that another factor may be involved; a factor with discrete steps rather than one with a range of values.
Derusha and I performed several experiments using a 10:10 cluster ratio of cuprite:danbaite:
| Cuprite Stones | Danbaite Stones | Yield |
| 7 | 6 | 0 |
| 7 | 8 | 0 |
| 10 | 10 | 17 |
Unfortunately, I do not have data from enough failed smelts to feel comfortable drawing hard and fast conclusions. Please send me data from any failed brass smelts you may have! (See the top of the page for information desired.)
Predicting Yield
I have looked at total cluster count, copper cluster count, zinc cluster count, cluster ratio, stone count, stone ratio, and elemental ratio. None of these seem to predict yield 100%. Cluster counts in particular (total, copper, or zinc) seem not to be determinate: some counts can either produce high or low brass yield. This is illustrated wonderfully by my graph of cluster counts vs. yield:
What is Element Ratio?
See the Element Ratio page.
Graphs
You can see all charts on one page.
Updates
2004.11.02: added data from wiki user pages Leghk, Gspusi, Derusha, Skrug. Also added data from a three smelt experiment.
Comments? Questions? Requests?
Feel free to add comments, questions, or requests here, with your name attached....
hey there see some of my smelting-experiments if you want -> Gspusi
Thanks, Gspusi! I added your data to the spreadsheet these graphs are based on.
