QUB | Archaeology and Palaeoecology | The 14Chrono Centre

Manual for psimpoll and pscomb

Error messages

Explanation of messages

• agedepth (timescale): failed to converge on a value for t after 100 iterations Timescales are generated by an iterative process that should have converged to give an answer within a specified accuracy after 16-20 iterations. In this case, the process has not converged after 100 iterations. Please report to KDB.
• asin Program tried to obtain arc sine of a number < -1 or > +1. Probably a program error: report to KDB.
• atan Program tried to obtain arc tangent of a number < -pi / 2 or > pi / 2. Probably a program error: report to KDB.
• atan2 Program tried to obtain arc tangent of y / x < -pi or > pi. Probably a program error: report to KDB.
• bernshtein: failed to converge on a value for t after 100 iterations psimpoll message. Bernshtein polynomials are solved by convergence that should be complete within 20 iterations. This one has failed. Check the output file, and if the problem occurred with samples beyond the range of the radiocarbon age points, trying fixing the end points with age estimates from linear interpolation. It is also possible that the initial guesses psimpoll makes for t were too far out, so the way these are obtained needs to be modified, or an age-depth range given in menu H caused problems. See KDB.
• bus error UNIX error message. The program has crashed completely. Report to KDB.
• CA: residual bigger than tolerance Correspondence analysis error message. The program has failed to converge on a solution. Check all input values and transformation options.
• calloc: memory allocation error Program was unable to make the allocation of memory for an array. Check that you have given the numbers for taxa, samples, zones, radiocarbon ages, etc, correctly (these numbers determine the size of memory requested from the computer for arrays). Consider the possibility that your computer's memory is not adequate, then report to KDB.
• core dump UNIX error message. The program has crashed completely. Report to KDB.
• Data set for analyses has zero samples psimpoll message. The program produces a reduced dataset for numerical analysis by eliminating samples outside any defined section (menu H), and samples marked for omission. After doing this, there were no samples left. The most likely problem is that you requested a section of the dataset by age, but the samples are presented by depth and were not converted. Please rerun.
• Error estimates on ages must be greater than zero, cannot complete analysis. Check values in C14 file psimpoll message. The program has read a standard deviation for an age from your C14 file as 0 or negative, and this caused a problem in the calculation of an age model. If you want an `age' to be excluded from analysis, make sure it begins with a non-numeric character (i.e., (4000) not 4000). Otherwise, check the file carefully.
• exp Attempt to calculate an exponential failed because the value was too high. Probably a program error: please report to KDB.
• Fatal error detected by operating system: please report to KDB
  SIGABRT: Abnormal termination This is a serious crash, brought about by the program trying to do something that the computer could not handle. All such conditions should be trapped by the program, so please report this one to KDB immediately, even if you know of something horrible you did that might have caused it to behave this way. The computer may need rebooting to clear memory.
• Fatal error detected by operating system: please report to KDB
  SIGFPE: Floating point exception This is a serious crash, brought about by the program trying to use a number outside the range that the computer can handle (too large or divide by zero). All such conditions should be trapped by the program, so please report this one to KDB immediately, even if you know of something horrible you did that might have caused it to behave this way. The computer may need rebooting to clear memory.
• Fatal error detected by operating system: please report to KDB
  SIGILL: Invalid function image This is a serious crash, brought about by the program trying to do something that the computer could not handle. All such conditions should be trapped by the program, so please report this one to KDB immediately, even if you know of something horrible you did that might have caused it to behave this way. The computer may need rebooting to clear memory.
• Fatal error detected by operating system: please report to KDB
  SIGSEGV: Invalid access to storage This is a serious crash, brought about by the program trying to do access memory in some way that the computer could not handle. All such conditions should be trapped by the program, so please report this one to KDB immediately, even if you know of something horrible you did that might have caused it to behave this way. The computer may need rebooting to clear memory.
• Fatal error detected by operating system: please report to KDB
  SIGTERM: Termination request received This is a serious crash, brought about by the program trying to do something that the computer could not handle. All such conditions should be trapped by the program, so please report this one to KDB immediately, even if you know of something horrible you did that might have caused it to behave this way. The computer may need rebooting to clear memory.
• fclose: error in closing a file An error was reported to the program when it tried to close a file. May occur if a disc that you are writing to is full (so please check this), but probably a program error: report to KDB.
• fgetc: no character read (end of file?) An error was reported to the program when it tried to read a character from a file. Check that your input files are complete.
• fgets: no string read (end of file?) An error was reported to the program when it tried to read text from a file. Probably a program error: report to KDB.
• First character of two-character code on line 4 of input must be one of %, A, B, C, G, L, M, or X. See documentation for details. The character in your file is 'c' psimpoll message. Check line 4 of your main input file, and ensure that the first character on the line matches one on this list. The character read by the program is indicated at the end of the message. And check that the file name you entered was the main input file you intended.
• fopen: failed to open file An error was reported to the program when it tried to open a file. Usually through mistyping the name of the main input file, but might occur if two files had the same name (from menu E).
• fscanf: unexpected end of file An error was reported to the program when it tried to read from a file. Probably a program error: report to KDB.
• getbuf: buffer is overfull. Check that individual input items do not exceed allowed number of characters. This mesage is preceded by more detail of what the program was trying to read when the problem occurred, what it actually did read, and, where possible, suggestion of what needs to be corrected. This information should help to locate the problem, but note that the 'real' problem may lie earlier in the file than the input line that was the immediate source of the difficulty. getbuf is a function that handles the reading of many input items. Its internal buffer is 160 characters long, which is ample for all reasonable and correctly formatted input. The `individual input items' referred to may be values or labels, which may occupy lines by themselves or be parts of lines with other values. Check your input files carefully to make sure that you haven't omitted spaces (to separate numbers) or commas (to separate labels), and that the values given for the number of taxa, number of samples, and the two-character code on main input line 4 are correct for the data given. See the output file psimpoll.LOG for the contents of the buffer, which may help work out where the problem occurred, along with information on the screen at the time.
• getbuf: unexpected end of file. The end of a file was reached unexpectedly. This messge should have been preceeded by message with details of the file and line number concerned. Check that the file has all the input items expected.
• getc: reading error (end of file?) An error was reported to the program when it tried to read from a file. Probably a program error: report to KDB.
• indsplit: more than the allowed maximum of 20 splits found. Please report to KDB psimpoll message. Your data set has generated more than 20 splits in one of the taxa. This is more than anticipated would ever occur. Report to KDB and ask for the program to be modified.
• Input box is greater than 63000 characters long pscomb message. There is a limit of 63000 characters for the size of an input box, and one of those in your PostScript input file exceeds this. Please report to KDB, and this will encourage him to provide a greater limit.
• int conversion: floating point number cannot be represented as integer A floating point number was too large to be converted to an integer. Probably a program error: please report to KDB.
• lfit (gaussj): singular matrix, cannot complete analysis psimpoll message. One of the matrices generated during construction of an age model by weighted least-squares cannot be analysed. This may be a data error, or just bad luck. Check the data, and if that doesn't work, try using singular value decomposition (menu Ld.4).
• linint_s: two depths are identical or reversed psimpoll message. Problem occurred while trying to obtain linearly interpolated ages. Check depth values in C14 file, and ensure that the ages given for the top sample (defaults to $0±50$), and basal sample (no default) are reasonable (see menu L).
• loess: extrapolation beyond date limits is not allowed. Define an age for top and basal samples in menu L This technique for age-depth modelling cannot extrapolate. Check and, if necessary define, ages for the top and bottom sample (which then get treated as if they were dates) so that there will be no extrapolated vales.
• log10 Program tried to obtain base 10 logarithm of 0 or a number < 0. Probably a program error: report to KDB.
• log Program tried to obtain base e logarithm of 0 or a number < 0. Probably a program error: report to KDB.
• Memory allocation error. Cannot load COMMAND, system halted DOS error message, probably occurring after one of the program messages. The computer will need to be rebooted before you can do anything else. Probably a program error: report to KDB.
• Number of samples is zero or negative The number of samples specified on line 3 of the main input file (or elsewhere in a Tilia file) has been given as zero or negative. This needs to be corrected.
• Number of sum taxa is greater than 10 The number of taxa included in a summary diagram is limited to 10, which was expected to be enough under all circumstances. If you really want more than 10, and can convince me that this will work when plotted, I will make the necessary change.
• Number of taxa is zero or negative The number of taxa specified on line 2 of the main input file (or elsewhere in a Tilia file) has been given as zero or negative. This needs to be corrected.
• pow Program tried to obtain x to the power of y with x negative and y not an integer, or an overflow occurred (number too large for the computer to handle). Probably a program error: report to KDB.
• ran4_s: this function is somewhat non-portable Output on your computer appears unreliable Use one of the other random number generators The `pseudo-DES' random number generator (ran4_s) is not completely portable. It works fine on Macintosh, IBM PCs, and the UNIX systems that I have tried, but the version of psimpoll you are using has been compiled on some other system. As the message says, use another generator. This one may be unreliable on your system.
• Rate of change analysis: the dissimilarity measure in use has exceeded its theoretical maximum upper limit psimpoll message. Some of the dissimilarity measures used in rate of change analysis have a theoretical upper bound, and this will never be exceeded with correct input and correct programming. There has been a failure in one of these areas. Check that you are using only proportional data as input. Then report to KDB.
• realloc: attempt to reallocate to a negative amount of space Failure of attempt to reallocate memory that was previously allocated because the new size of memory requested is negative. Probably a program error: report to KDB.
• realloc: memory allocation error Failure of attempt to reallocate memory that was previously allocated. Check the lengths of text items that you have included in (especially) the main input file to see if they are really excessive. In pscomb it may mean that a box from one of the input files is more than 31,000 characters long. But probably a program error: report to KDB.
• scanf: input failure An error was reported to the program when it tried to read from the keyboard. Probably a program error: report to KDB.
• Second character of two-character code on line 4 of input must be one of A, D, M, S, or T. See documentation for details. The character in your file is 'c' psimpoll message. Check line 4 of your main input file, and ensure that the second character on the line matches one on this list. The character read by the program is indicated at the end of the message. And check that the file name you entered was the main input file you intended.
• Sediment accumulation rate(s) have been calculated as negative
  This will result in age reversals, and/or negative pollen accumulation rates. Check output file (if any) for details, and reconsider age model psimpoll message. The program was asked to calculate accumulation rates, and finds that the model in use produces negative rates. The program continues, and the full gory details will be in the age-depth output files if you requested them. Note that the pollen diagram will still be plotted against age, but will have overlapping samples.
• Segmentation fault UNIX error message. The program has crashed completely. Report to KDB.
• SIGINT: interactive attention signal from user User has interrupted normal running of program.
• spline_s: two depths are identical or reversed psimpoll message. Problem occurred while trying to obtain an age model with a spline. Check depth values in C14 file, and ensure that the ages given for the top sample (defaults to 0±50), and basal sample (no default) are reasonable (see menu L).
• splint: bad xa[] input, cannot complete analysis. Check depths in C14 file psimpoll message. Spline curve-fitting is a two stage process. splint is the second stage, and it has found that two input depth values are identical. Check input from your C14 file.
• sqrt Program tried to obtain the square-root of a number < 0. Probably a program error: report to KDB.
• strcat: total string length too long A concatenated `string' of characters has become too long for the amount of memory allocated for it. Probably a program error: report to KDB.
• strftime: failed to fill array A function that formats representations of data and time for output has failed. Probably a program error: report to KDB.
• svdfit (svdcmp): no convergence in 30 iterations of its, abnormal exit psimpoll message. The program could not complete analysis of your proposed age-depth model using singular value decomposition. Check the data in your C14 file, and try using a weighted least-squares age-depth model. If this fails, report to KDB.
• svdfit (svdcmp): zu() needs augmenting with extra zero rows, abnormal exit psimpoll message. Program error, failing to trap analysis of age-depth model by singular value decomposition with more terms in the model than ages. Report to KDB.
• switch: unexpected value Program does not allow for all possible cases in a `switch' statement. This is one likely error message if your .CFG file is out of date. If deleting any .CFG file does not remove the problem, report to KDB.
• SYSTEM: The text on lines beginning like this is an error message from the computer system, and is therefore specific to that system. Note what it says, and report to KDB. Information from psimpoll or pscomb should be on the next line down, and more details may be found by looking up that message elsewhere in this section.
• TILIA format: line-length unexpectedly short. Check input file! The program recognised a file as TILIA format, but found that one of the lines with taxa names was too short. There should be at least 16 characters on these line, with a letter indicating the (sub)sum group as character 16. Taxon names begin at character 18. It is acceptable (but strange?) to have no taxon name, but the (sub)sum value must be present. Check the input file, make sure that it is indeed TILIA format, and look at the contents of the list of taxa names.
• time: calendar time not available The program attempts to obtain the calendar time from the operating system, but this appears not to be available for some (unknown) reason. Please report to KDB.
• Unexpected end-of-file reached: data set is incomplete An input file does not have as many items as it should have. If the file named immediately above this message is a .CFG file, it is probably out of date. Use the example .CFG file in to work out which features are set in your file, then delete it, and reset them from the menus on a new run of psimpoll. If the named file is not a .CFG file, check its length against the number of items that are indicated at the top (number of taxa, number of samples, number of zones, etc., as relevant).
• ungetc: error pushing character back This function is used when reading `garbage' characters from a file. The first non-garbage character is `pushed back', so that it can be read by the next input routine. There has been a problem with the `push back' process, probably a programming error. Report to KDB.
• vprintf/vfprintf: output error An error has been reported by a function that `prints' data from a variable length list of parameters. Program error: report to KDB.

Archaeology and Palaeoecology | 42 Fitzwilliam St | Belfast BT9 6AX | Northern Ireland | tel +44 28 90 97 5136