-
Notifications
You must be signed in to change notification settings - Fork 8
/
TODO
125 lines (93 loc) · 3.72 KB
/
TODO
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
TODO LIST:
==========
Features to be added in release 1.1:
TESTING/DEBUGGING PHASE:
{
* Dark energy equation of state parameters:
31082010 DONE. TESTING REQUIRED.
* Redshift space corrections.
07092011 DONE
Only differences in the way the redshift space correction is
calculated.
* Problems with Qk calculation. Spline points may not be enough
to capture the details. The cpp code currently reproduces the
fortran code!
* Check for memory leaks
* Test the cosmic emulator power spectrum with the Smith et al.
Check their paper too. According to Heitmann et al. 2010, it can
cause problems in getting unbiased estimates. So need to change
it.
* Three things to be checked:
a. Convergence issues for Q(k)
-- An idea to deal with this is under investigation now!
Integrate out to k*rzetamax first and then the fourier integral
out to infty. This removes the convergence problem but is 3
times slower. The final results are same nevertheless, so we go
for speed.
b. Spline number of points for Q(k,rmin)
-- This is unclear
c. Redshift space corrections
-- Test with mock catalogs
d. Q(k) for large scales
-- This is unclear, the reason for the non-convergent integral
is because we pass an increasing function to qawf!?
}
TESTED against the Fortran version of the code
{
* Support for halo exclusion, radial dependence of the bias.
04092011
* setrzeta() DONE
04092011
* Change Rvir to R200. DONE
04092011
* The spline for xi_L is only the redshift zero one, remove all other
apparatus
* Addition of dnlDdln1pz routine to cosmology
07092011 DONE: Perhaps Frank's code needs to reduce the interval
for the derivative. It is 0.01 now, but 0.0001 seems better.
* Avoid calls to get totNc and totNs, as these are entirely unnecessary
07092011 DONE
*xigg and xigd looks ok at various redshifts! May be some sign
issues on large scales, however its mostly very close to zero
there.
* Separate the Wp and ESD calculation in to separate parts, to avoid
unnecessary computations.
07092011 DONE
* Interface to get the wp and ESD correct - do it with wrapper
functions
07092011 DONE
}
NEXT TODO (In order of priority):
* Data in binary format for ease of reproducibility
Interface with CAMB, neutrino masses and so on!!!
* Support for newer Zehavi et al. dataset : Check out the Zehavi et al.
support branch
* Massive neutrinoes, support for reading Pk lin output by CAMB:
Checkout the massive neutrino branch. Neutrino branch requires halo
exclusion addition
* Departure from non-flat cosmologies :
Delta_crit(z) needs to be changed
* Cosmology dependence of c-M relation :
Currently only through F and
K parameters.
* Cored profile for the Nsat(M) relation :
Equations ready
* Parameter to introduce super or sub-Poisson behaviour for Nsat(M) relation
* OpenMP parallelisation
Identifying the changes to be made for version 1.1
DONE:
=====
src/cosmology.cpp : Main changes will go here. Constructors will be changed and non-flat cosmologies should be allowed.
src/powerspectrum.cpp: Smith power spectrum not defined for w0 != -1 cosmologies, or non-flat cosmologies.
src/hod.cpp: Changepar routine requires more parameters now, getnewpar, print routines changed, getOmk added.
src/mcmc.cpp: New parameters added, parameter sequencing changed to better reflect degeneracies, Additional prior added on the curvature data.
src/dat/sampleconfig: Changed parameter orders to reflect changes in the code.
Add possibility to analyse for flat cosmologies.
No changes here
=================
src/main.cpp
src/data.cpp
src/gauleg.cpp
src/grid.cpp
src/haloes.cpp
src/fisher.cpp