This readme file contains status reports on the current program version. For program documentation see the TeX file(s) in the sub-directory Docs. For the list of files in the suite, see the Makefile. Version 0.0 (Sep 20, 1999) -------------------------- LO evolution works correctly. However, there are occassional bus errors on calls to MINOS. Lock status: 1) Files CTEQ_pol.f, chisq.f, chisq0.f: reserved for updating by Sourendu Gupta. Files are being updated for imposing positivity on the densities. 2) Files ape.f, CTEQ_unpol.f: reserved for updating by D. Indumathi. Files are being updated for NLO evolution in Adler-Bardeen scheme. Require file CTEQ_pol.f after release. Version 0.1 (Sep 20, 1999) -------------------------- LO evolution works correctly. Polarised densities are properly bounded at Q_0^2. There are occassional bus errors on calls to MINOS. Version 0.11 (Sep 22, 1999) --------------------------- Program crash due to MINOS calls traced to incorrect dimensioning of array "f" in files final.f and final0.f. This error is fixed. Version 0.12 (Sep 26, 1999) --------------------------- Sign error in ubar-dbar parametrisation for cteq3l densities fixed in file cteq3l.h. All I/O through stdin and stdout. Version 0.13 (Sep 27, 1999) --------------------------- Version number and chisq per DOF for each experiment now output by final.f and final0.f. The polarised gluon splitting function in ape.f corrected. Lambda, nflam, q0sq, bmass now taken from cteq3l.h instead of the input file read in inits.f. Version 0.14 (Oct 2, 1999) --------------------------- 1) Numerically optimised form of the polarised gluon splitting function now used in ape.f. 2) Data and errors now kept in a one dimensional array and a pointer array used to keep track of the experiment, Q^2 and x. This change in datacom.f makes corresponding changes in all files dependent on it. 3) The CTEQ3L parametrisation corrected in cteq3l.h. Corresponding changes made in CTEQ_unpol.f and CTEQ_pol.f. 4) The implementation of the Bjorken sum rule in CTEQ_pol.f corrected. 5) Positivity bound checks in CTEQ_pol.f corrected. Calls to bound checks moved out of chisq.f and chisq0.f to function initg1 in CTEQ_pol.f. Version 0.15 (Oct 7, 1999) -------------------------- 1) CTEQ4L densities added. This requires new file cteq4l.h for the parameters, and files CTEQ4_pol.f and CTEQ4_unpol.f for parametrisations. 2) New function added to integs.h (needed in CTEQ_pol.f). 3) The old files CTEQ_pol.f and CTEQ_unpol.f moved into CTEQ3_pol.f and CTEQ3_unpol.f respectively. Version 0.151 (Oct 9, 1999) --------------------------- 1) Small errors in initg1 in CTEQ3_pol.f and CTEQ4_pol.f fixed. 2) Bjorken in CTEQ3_pol.f and CTEQ4_pol.f modified so that the result is always well-defined. Version 0.152 (Oct 11, 1999) ---------------------------- 1) Unpolarised computation performed only once. Results stored in a new array passed through common block in file datacom.f. Corresponding changes in unpol.f, unpol0.f, final.f and final0.f. 2) Now final.f computes and prints out the polarised parton densities at a value of Q stored in q2(-1). Corresponding canges in inits.f and grids.h. Version 0.16 (Dec 27, 1999) ---------------------------- 1) Two new common blocks added in files speed.h and dsped.h (initialised by subroutine jaldi in file inits.f. 2) The parameter dependent initialisations are merged into files CTEKx.f from the two files CTEQx_pol.f and CTEQc_unpol.f. Also, the functions f1 and g1 from these two files have been moved to file strfn.f. 3) Evolution across flavour thresholds has been systematised by adding a routine thrsh to CTEQx.f, which is called by inits.f. The change to ln Q^2 is deferred until after the data is read in. Hence the format of the input file changes. 4) The program can now handle three kinds of targets (proton, netutron, deuteron). Adding neutron causes changes in strfn.f, unpol.f, unpol0.f, chisq.f, chisq0.f, final.f and final0.f. Version 0.161 (Dec 30, 1999) ? or 1.161? ---------------------------- 1) Error in the boundary value of the evolution of q_{15} is now fixed. Changes in files chisq.f, final.f and unpol.f. Version 1.161_alpha (Jan 17, 2000) ----------------------------- 1. NLO evolution added to ape.f. Matching condition changed in subroutine thrsh in file CTEQ4.f. Coefficient functions added to f1 and g1 in strfn.f. This causes a change in the parameters passed to these routines. The array klod in dsped.h is now two dimensional due to these changes. The array xld in dsped.h is initialised in routine jaldi in inits.f and used in the definitions of f1 and g1 in file strfn.f. New constants added to qcd.h. Include statements in CTEQ4.f changed to use cteq4m.h. 2. Bjorken sum rule written to NLO. This affects routine bjorken in file CTEQ4.f. 3. Definitions of q3, q8, q15 earlier had a factor of 2 error. This has been fixed by modifying initf1, initg1, bjorken and positive (in CTEQ4.f, but not yet in CTEQ3.f). Version 1.161_beta (Jan 19, 2000) --------------------------------- 1. NLO and LO polarised splitting functions put in arrays (held in apesf.h), precomputed at initialisation by a call to apeinit in inits.f, and used by pol1 and unpol1 (partly) in file ape.f. 2. Integer division error in function bjorken (in file CTEQ4.f) fixed. 3. Missed factors of cf in functions f1 and g1 (in file strfn.f) fixed. Definitions of cq in both functions also fixed. Redefined charge square factors in qcd.h and corresponding changes in strfn.f. Release Version 1.0 (Jan 20, 2000) --------------------------------- 1. Variable psna in subroutine apeini (in file inits.f) corrected. 2. Variable cq in function g1 (in file strfn.f) corrected. Related correction to variable cf1q in file qcd.h. (to be done)