Practical Aspects of Mineral Thermobarometry
THERMOCALC PC versions: Known Bugs and Other Quirks
Several of the limitations and instabilities of earlier versions have now disappeared with v. 2.75 onwards. Nevertheless you should note the following. I do not know whether these comments apply to the Mac versions.
Average P-T mode
The following quirky behaviour patterns were noted with version 2.4 and reported to Roger Powell in August-September 1995. Nevertheless they were still present in my copy of version 2.7 (November 1998).
Excluded end members not reinstated
In average P-T mode, when you exclude one or more end members during a run in which you are going to go on to use other data files, the excluded end member (or rather the end member at its position in the sequence of end members) is not reinstated when you load and calculate the next data file. Not fatal, but it can be very confusing if you don't know it's going to happen, and would cause you to accept incomplete and unsatisfactory results if you didn't spot it. This bug is still present in version 3.1.
Number of successive data files
After you have read a number of data files successively into Thermocalc, it eventually cannot find the next one, usually the sixth or seventh. Although nothing unusual appears in the screen prompts, it may have truncated the actual string used to search for the filename. Sounds like a text buffer misbehaving.
Output file documentation for reaction calculations
In other calculation modes the reaction calculations are preceded by explanatory text which gives the T (and/or P) at which the calculations were performed and an explanation of the column headings a, sd(a), b, c, etc.. It is important to have this in average P-T mode too, if only because the output file does not record the P-T window used for the run.
Roger Powell's instructions for reporting bugs are as follows:
When you report bugs, we need all the information necessary so we can reproduce your problem. The best way to do this is to send RP:
- the datafile associated with the failure
- your logfile for the run that failed
- a paragraph on what you were trying to do
Send to: thermocalc@earth_sciences.unimelb.edu.au
Oxford users should consult me first, before contacting Roger Powell. Commonly the problem is one with a known workaround, or the "bug" may be an artefact of the way you are trying to approach the problem.
This page last modified 12 October 2004