search

lesgourg / class_public

Public repository of the Cosmic Linear Anisotropy Solving System (master for the most recent version of the standard code; GW_CLASS to include Cosmic Gravitational Wave Background anisotropies; classnet branch for acceleration with neutral networks; ExoCLASS branch for exotic energy injection; class_matter branch for FFTlog)
232 stars 288 forks source link

non-linear matter power specra of CLASS and CAMB #46

Closed 0satoken closed 8 years ago

0satoken commented 9 years ago

Hi all, I have used CAMB so far and started using CLASS. I compared matter power spectra generated by both codes. The linear matter power spectra agree quite well but the problem is that non-linear matter power spectra are inconsistent with each other and the difference reaches about 10% at most. Attached figures show the result, where I adopted WMAP 9 year cosmology and no massive neutrinos. I have checked that both codes employ the same HaloFit of Takahashi et al. (2012). The version of CLASS I used is 2.4.3 and CAMB is January 2015. pk pk_nl

oscarhdz7 commented 8 years ago

Hello, I am wondering if you have made any progress on this issue? I too have seen that the linear power spectra agree but I have seen an even more dramatic difference between the CLASS and CAMB nonlinear power spectrum at nonzero redshifts. I have using CLASS 2.4.3 to get the non-linear matter power spectrum between z=0 and z=5 with the precision file pk_ref.pre and for z>3 the CLASS linear and nonlinear power spectrum are identical up to k=8 h/Mpc. CAMB does produce different nonlinear and linear power spectrum for z>3. A couple days ago I opened up the issue "halofit with z>3", regarding this problem. Regards, Oscar

0satoken commented 8 years ago

Hello, In the pull request #81 , this issue is resolved and I also have confirmed that the non-linear power spectra computed by CAMB and CLASS in the latest version are consistent with each other. I close this issue. Regards, Ken