LIST OF BUGS/IMPROVEMENTS, IN ORDER OF MOST RECENT FIXES FIRST.


BUG: 1 JUL  2022  PGJ

emission.f    - wrong normalization of IQUV emission coefficients, ame for all I,Q,U,V

	The original version returned IQUV values
        larger by an erroneous factor of 3.155 * (QNORM/10)	

	Consequently the intensities originally output depended on QNORM, which is wrong
	However, the relative values of IQUV are all originally correct because the calculations
	are all optically thin.

	Therefore all plasma diagnoses depending on ratios only are correct even if calculated
	with this bug.

	Thanks to S. Gibson for passing on this bug.


IMPROVEMENT:  5 Mar 2020  PGJ

atom.*, input.* in data/

	   Data for Fe XIII have been replaced with modern CC
	   collision strengths not older DW values

	   All atomic ion files now include one allowed E1 transition
	   to an appropriate excited delta-n=0 transition to mimic all
	   excitations to all higher levels using adjusted collision
	   strengths and in some cases branching ratios.



IMPROVEMENT June 6 2019 PGJ

colrd.f gammae.f, gammai.f, gammas.f

	Added a key ICOLLMM which flags if sums over M to M' collisions are
	needed.  This can speed up collision calculations by skipping loops
	in the gamma*f routines.

	CPU 
	251.449u 0.650s 4:14.06 99.2%	0+0k 0+0io 0pf+0w  New code 26 level
	1.267u 0.014s 0:01.30 97.6%	0+0k 0+0io 0pf+0w  New code  3 level
	0.668u 0.008s 0:00.69 95.6%	0+0k 0+0io 0pf+0w  no collisions

	2.457u 0.014s 0:02.49 98.7%	0+0k 0+0io 0pf+0w  Original code
	0.677u 0.008s 0:00.70 95.7%	0+0k 0+0io 0pf+0w  no collisions


BUG June 6 2019 PGJ 

gammae.f, gammai.f, gammas.f

	  Incorrect denominators (2*AA1+1) in two places have been corrected in this routines.
	  Thanks to Tom Schad for noting part of this problem.

	  Sum over multipole order K was incorrectly done from -K to K. Now the do loop
	  is DO M=0,K,2    This error led to an overestimate of collisional rates through
	  a sum over CASSUME (for strong coupling and exchange collisions only).

colrd.f gammai.f gammas.f  : not a "bug" per se but an addition if spin forbidden
          transitions are computed in large atomic models.

 	  Cases of spin-forbidden lines are now treated in the same way as M1
	  transitions by assuming that the cross-sections are dominated by strong
	  coupling and/or exchange of electrons.


EFFECTS: In some ways these two sets of changes comepnsate for one
another for calculations ehich include spin-changing or delta J > 1
collisions. For small atomic models with no such transitions the revised
code will produce smaller collision rates, by factors of two or so.
For those atoms constructed to take care of collisions via allowed
E1 transitions, these changes are exected to be small.

	  
BUG 16 Jan 2014  PGJ

gammae.f - incorrect implementation of CECOEF has been replaced
	 with a reduction in J^2_0 in the routine field_int.f

	 Note that this led to inconsistent statistical equilibrium calculations
	 when collisions were included and when CECOEF in the INPUT file was non-zero.