... this project is left as interpreted from insufficient information.
Done SS 6179 ... would be wise to cross check other SS version [7019].

Project Fission.sm (152 КиБ) скачан 47 раз(а).

Wrote

... this project is left as interpreted from insufficient information.
Done SS 6179 ... would be wise to cross check other SS version [7019].

Project Fission.sm (152 КиБ) скачан 47 раз(а).

Hello Jean,

Is this what you mean? Checked with version 7019, and plots do not work.

Wrote

Hello Jean,

Is this what you mean? Checked with version 7019, and plots do not work.

Thanks Radovan.
SS 7019 needs be be doctored quite seriously.
What can be so hard to digest ?
As it looks, SS 7019 does not recognize exponents < 10-15
If so: should be easy to fix ?
If it tabulates the low exponents, then the plot is guilty,
that does not make sense either ...
Maybe too early to tag SS 7019 big toe a dead patient !
Cheers ... Jean

... Is there any red ?
Copy/paste in IrfanView, Resize-Resample to enlarge the ^-
I wish the paper shows some result or graph.

SS 6179 ... exp(709.782)=1.796412028*10^308
SS 7019 ... exp(709.782)= red or something

Wrote

... Is there any red ?
Copy/paste in IrfanView, Resize-Resample to enlarge the ^-
I wish the paper shows some result or graph.

There are some results for the MeV matrix. When the numbers of rows j>11 the red is coming back.
Although there are few differences between the pictures. Some numbers are zero on the first one contrary to some small numbers on the other one.
Is there any significance of this?


Regards,
Radovan

Thanks Radovan for your reply.
YAF.NET refused to quote your message entirely ?

1. Does YAF.NET freaks the ranges to bracket format ?
2. Is it 7109 that freaks the ranges to bracket.
3. What does the bracket range format 7109 vs normal ranges 6179 ?
That kind of problem was observed recently [from recollection NDTMA ?]

What about re-code the ranges 7109 as per 6179 ?
No more from the Originator [graph, results ...]
Maybe not a public paper ... or home made for classroom ?
More puzzling: more Doctors are gone fishing ...

Cheers ... Jean

Wrote

What about re-code the ranges 7109 as per 6179 ?
No more from the Originator [graph, results ...]
Maybe not a public paper ... or home made for classroom ?
More puzzling: more Doctors are gone fishing ...

Cheers ... Jean

not related with ranges function. even rewriting as for(4)
function, equation gives the same problematic result.



I have downloaded several old versions and found the last
working version is 0.99.6824. After this version there is
none giving the expected result.

Regards

Hello Jean,

I just put 11 in the range because any number greater than 11 will produce the famous error "...above max. positive number...".
The only thing I can figure out is that the version [7109] (and some versions before) has some serious problems with float/symbolic calculations.
Therefore I can consider this as a serious bug.

Regards,
Radovan

Wrote

Hello Jean,

I just put 11 in the range because any number greater than 11 will produce the famous error "...above max. positive number...".
The only thing I can figure out is that the version [7109] (and some versions before) has some serious problems with float/symbolic calculations.
Therefore I can consider this as a serious bug.

Regards,
Radovan

Simple: tag the big toe of all SS past 6179 Dead Bodies
exp(709.782)=1.796412028*10^308 ... What do you get ?
Have a good day [night] Radovan ... Jean

Wrote

Wrote

Hello Jean,

I just put 11 in the range because any number greater than 11 will produce the famous error "...above max. positive number...".
The only thing I can figure out is that the version [7109] (and some versions before) has some serious problems with float/symbolic calculations.
Therefore I can consider this as a serious bug.

Regards,
Radovan

Simple: tag the big toe of all SS past 6179 Dead Bodies
exp(709.782)=1.796412028*10^308 ... What do you get ?
Have a good day [night] Radovan ... Jean

Thank you Jean. All the best to you.

Wrote

The only thing I can figure out is that the version [7109] (and some versions before) has some serious problems with float/symbolic calculations.
Therefore I can consider this as a serious bug.

My Kalkulator is older than my Mathcad 8 Pro. [2000]



Does it suggest some profitable code ?
Up until now, this bug hasn't been attended.
How many more undiscovered ?
Thanks Radovan for your dedication, sincerely ... Jean

... last brick in the wall.

Wrote

... last brick in the wall.

exp(709.782)=1.79641202802057*10^308
That's what it has to be checked first SS 7019
Next, wrt the long thread: if exp(x) OK
then why it scraps the Originator project vs SS 6179 that does it.
That revisit to insure the bug has not yet be incinerated.
May be it should, being so little attractive.

Wrote

Wrote

... last brick in the wall.

exp(709.782)=1.79641202802057*10^308
That's what it has to be checked first SS 7019
Next, wrt the long thread: if exp(x) OK
then why it scraps the Originator project vs SS 6179 that does it.
That revisit to insure the bug has not yet be incinerated.
May be it should, being so little attractive.

Helo Jean,

From this thread, the examples and comments it is obvious that something is fishy here regarding that the same thing is working in the previous SMath versions but not in the recent ones. There is no answer from anyone yet what's wrong.

Regards,
Radovan

Цитата

Helo Jean,

From this thread, the examples and comments it is obvious that something is fishy here regarding that the same thing is working in the previous SMath versions but not in the recent ones. There is no answer from anyone yet what's wrong.

Regards,
Radovan

I figured this is not a problem related only with Smath Studio.

https://groups.google.com/forum/#!topic/comp.soft-sys.matlab/s8NTeTEQV7I
https://forum.dlang.org/post/rhnvcskyumpflxpcstzy@forum.dlang.org
https://books.google.com.tr/books?id=OjUyDwAAQBAJ&pg=PA269#v=onepage&q&f=false

For example, you can calculate exp(709.782) via Google.
But not exp(709.783), it gives nothing.

Online maxima can not calculate also. http://maxima.cesga.es
Wolframalpha can calculate though. https://www.wolframalpha.com/input/?i=exp%28709.783%29

Maybe the previous source code was not suitable to 64-bit IEEE 754 binary arithmetic in those previous versions.
I think Andrey changed the calculation code according to IEEE and because of it infinite overflow error popup.

Regards

Let me repeat my main point about this error which I believe it is a bug. It is not about exp() function.
Please try in every software you want to calculate this simple addition of two small numbers.



SMath will issue an error (with the rather confusing message) but every other software I tried will not. The result will be either zero or 1.5e-115. That is the point.

Regards,
Radovan

Wrote

Let me repeat my main point about this error which I believe it is a bug. It is not about exp() function.
Please try in every software you want to calculate this simple addition of two small numbers.



SMath will issue an error (with the rather confusing message) but every other software I tried will not. The result will be either zero or 1.5e-115. That is the point.

Regards,
Radovan

You are right Radovan, this should be a bug. I have checked the addition on old versions.
0.99.7016 (and older versions) can calculate the mentioned additon, 0.99.7100 and after doesn't.

Let me clarify something, exp(709.783) is still not calculated on 0.99.6824 too.
But somehow the Project Fission.sm calculations can be done with 0.99.6824.

There is some serious issues over here.

Regards

Wrote

Let me clarify something, exp(709.783) is still not calculated on 0.99.6824 too.

The utmost limit 709.782 except Kalkulator and Casio
from some method switching from exp(x) to some series.
That kind of magnitude has certainly no interest.
Why 7019 is bug ?

The Free42 calculator for Android supports the value of 14,140 for the exp () function
exp (14140) = 8.3941 E+6140



Best Regards

Carlos