Strange failure (a simple calculation changes some global status?)

Strange failure (a simple calculation changes some global status?) - CoolProp functions fail after a simple function usage with units - Messaggi

#1 Pubblicato: 30/03/2016 01:17:49
Mike Kaganski

Mike Kaganski

184 apprezzamenti in 434 post.

Gruppo: User

coolprop.sm (4,64 KiB) scaricato 1375 volta(e).

This file uses CoolProp wrapper plugin to get some water property.
But between two invocations, a calculation is made that changes some state, so that after it any CoolProp calculation fails.

May it be that A() somehow mangles, say, FPU state (flags)? Like exeption masking, or somesuch?
С уважением, Михаил Каганский
1 utenti hanno apprezzato questo post
Davide Carpi 01/04/2016 18:48:00
#2 Pubblicato: 31/03/2016 14:13:05
Davide Carpi

Davide Carpi

1.443 apprezzamenti in 2.936 post.

Gruppo: Moderator

Confirmed.
If you like my plugins please consider to support the program buying a license; for personal contributions to me: paypal.me/dcprojects
#3 Pubblicato: 31/03/2016 22:03:24
Jean Giraud

Jean Giraud

983 apprezzamenti in 6.866 post.

Gruppo: User

Wrote

This file uses CoolProp wrapper plugin to get some water property.
But between two invocations, a calculation is made that changes some state, so that after it any CoolProp calculation fails.

May it be that A() somehow mangles, say, FPU state (flags)? Like exeption masking, or somesuch?



I have a good idea about this plugin, having programmed WaterSteam in Visual basic
years ago. The plugin may be OK by itself but it suffers from the Smath silent
killer bug. Naturally, getting some properties uses some of the related constants.
Then, getting next properties does not renew constants previously used, and you get
any zombie result. Nothing to do untill this bug is cleared. The "clear" function
works, does not work.
If you work with WaterSteam, better get the freeware from "Katmar" [piece of gold].
Same formulation than IAPWS [quite monkey business inside IAPWS].
I programmed them in Mahcad 11, collab Fred was very happy.

Jean
1 utenti hanno apprezzato questo post
Davide Carpi 01/04/2016 18:49:00
#4 Pubblicato: 31/03/2016 22:22:27
Jean Giraud

Jean Giraud

983 apprezzamenti in 6.866 post.

Gruppo: User

Please don't reply: do this, do that
Smath does NOT do first, I don't have to do what it should do.

Forum Smath Silent Killer Bug.gif
1 utenti hanno apprezzato questo post
Davide Carpi 01/04/2016 18:48:00
#5 Pubblicato: 01/04/2016 12:32:36
Andrey Ivashov

Andrey Ivashov

2.343 apprezzamenti in 4.056 post.

Gruppo: Super Administrator

Fixed in Stable SMath Studio 0.98.5935.
Thank you.
2 utenti hanno apprezzato questo post
Davide Carpi 01/04/2016 18:48:00, Mike Kaganski 01/04/2016 19:37:00
#6 Pubblicato: 01/04/2016 18:56:42
Davide Carpi

Davide Carpi

1.443 apprezzamenti in 2.936 post.

Gruppo: Moderator

Jean, thank you for pointing out the different behavior of Clear about built-in constants. My fault, was in my knowledge and I've forget to document it.
I've updated the plugin (SS 0.97.5346 and later), now the description of Clear(...) reflects also this behavior.

2016-04-01 23_55_33-SMath Studio Desktop - [clear.sm_].png


Wrote

Fixed in Stable SMath Studio 0.98.5935.
Thank you.


Once again, thank you very much Andrey.
If you like my plugins please consider to support the program buying a license; for personal contributions to me: paypal.me/dcprojects
1 utenti hanno apprezzato questo post
Andrey Ivashov 01/04/2016 20:00:00
  • Nuovi post
  • Nessun nuovo post