Hi, Alastair:
f2 should refer to the captial sigma in the Wilson distribution. As shown in P419 of Giacovazzo etc. "Fundamentals of Crystallography, 3rd edition", "Sigma, being a function of the atomic scattering factors, will not have a constant value in all the reciprocal space, but it will decrease for increasing values of sin theta/lambda."
f2 is not a constant with different Miller dices. Please look at the lines of (751:752) on the attached Fig2.
Did I misunderstand something?
Best,
Jiawei Wang
----- 原始邮件 -----
发件人: Alastair Fyfe <[log in to unmask]>
收件人: Jiawei Wang <[log in to unmask]>
已发送邮件: Fri, 18 Jan 2013 12:05:41 +0800 (CST)
主题: Re: [ccp4bb] Confused with pirate::Refine_HL_coeff::operator() function
Hi Jaiwei,
Kevin will be the person to answer this, but, even without understanding
the intent, the code seems reasonable. The outer declaration for
variable "f2" means it is in scope throughout. The first loop over
indices assign a value to it which it keeps after exiting the loop at
line 719. The next loop over indices, starting on line 735, references
the value of "f2" in the lik_max() statement.
Best,
Alastair
On 01/17/2013 07:02 PM, Jiawei Wang wrote:
> Hi, all:
>
> Please look at the attached figure.
>
> In the function Refine_HL_coeff::operator(), the variable f2 was declared on the line 684, and it was assigned to specific Miller indices on the line 694. Once the iteration (690-718) was terminated on the line 718, f2 could be any value.
>
> However, on the line 737, f2 was used again without pointing to any specific Miller indices.
>
> Could anybody explain the reason for that, or it's a bug?
>
> Best,
> Jiawei Wang
|