Repository navigation
Richards growth curve (and compiler warnings) #775
Description
This issue is redundant with issue #778.
Note from full [tpl created yesterday (July 21, 2026) ] I made edits to:
- lose all of the compiler warnings; and
- prevent the Richards growth curve option from giving nan's (See Steve Barbeaux's comment from the other day). The final full ss3.tpl file is here which should compile and run.
Apologies but I don't know how to best push to the code.
Metadata
Metadata
Assignees
Type
Projects
- Status
- Status
Activity
Rick-Methot-NOAA commentedon Jul 22, 2026
Thhanks for helping to look into this. I'll run a compare to find your change and then continue my investigation, which narrowed the problem down to the growth in the first age beyond the Amin.
iantaylor-NOAA commentedon Jul 22, 2026
Hi @jimianelli,
The first link says I need more permissions but I compiled the current main branch which I think is good enough and used that to compare to your modified code in the second link.
I'm seeing 3 types of changes:
sprintf():If Rick can incorporate your changes to growth or implement some different fix for that issue, then I can work on mapping the first two types of changes over as part of a separate pull request, which I think will make code review easier rather than having lots of independent changes to all the TPL files.
jimianelli commentedon Jul 22, 2026
wow cool that you got there and sorry, I changed permissions just now. great idea to place in separate PRs because one is lower priority than the other (for Steve anyway). Thanks for looking into this so quickly! sorry I didn't follow better channels
Rick-Methot-NOAA commentedon Jul 22, 2026
I was already making a change to fix the growth issue. Agree with Ian that all else should be addressed separately.
Jim's code basically reverts to what was there before:
case 2: // Richards
{
Ave_Size(styr, 1, g)(0, first_grow_age(g)) = Lmin(gp);
VBK_temp2 = VBK_temp * VBK_seas(0);
for (a = first_grow_age(g); a <= nages; a++)
{
temp = LinfR + (LminR - LinfR) * mfexp(VBK_temp2 * (real_age(g, 1, a) - AFIX));
Ave_Size(styr, 1, g, a) = pow(temp, inv_Richards);
} // done ageloop
break;
Jim - what compilier is giving you those additional warnings?
Rick-Methot-NOAA commentedon Jul 22, 2026
I confirm that the problem is fixed. I think Jim and I independently found the problem at about the same time (unless Jim found it yesterday and waited to tell us). Thanks again Jim for all the other code suggestions.
jimianelli commentedon Jul 23, 2026
gcc, Apple clang version 21.00 (clang-2100.1.1.101)
but they've been around for a while so I don't think it's a compiler version issue so much. also mostly quite minor.