Mysterious scale factor for MERL BRDF?

[shocker-0x15]

Hello,

I have noticed that this code uses a scale factor x2 for MERL BRDF data:

layerlab/src/layer.cpp

Line 530 in 3e5257e

data[j] = storage[ind+ch*header.res_theta_h*header.res_theta_d*header.res_phi_d] * scale[ch] * 2; /// XXX too dark?

Is there any basis for this value? or is this just competely ad-hoc?
(I certainly feel materials look too dark with raw MERL BRDF, but I'm not sure the reason.)

Best regards,