SMath Studio 0.98.5990 (26 May 2016)

#1 Posted: 5/26/2016 6:54:20 PM

Andrey Ivashov

2270 likes in 3734 posts.

Group: Super Administrator

SMath Studio 0.98.5990 available

Latest changes (against SMath Studio 0.98.5983)

SS-2301: Symbol 'i' can now be used as a function argument;
Function eval() improved to allow calculation of the functions inside definition even if function's arguments are undefined;
SS-2352: Pressing Insert Matrix button when part of expression is selected fixed;
SS-2349: Line breaks incorrectly shown in Dynamic Assistance;
SS-2351: IsDefined() function fixed for infinity and infinity now exists in Autocompletion list;
SS-2348: Crash on deleting result of calculation fixed;
Support for numbers started with decimal symbol improved inside MathRegion;
Calculation logic refactoring.

Download

Desktop Windows (desktop computer with Windows OS):
SMathStudioDesktop.0_98_5990.Setup.msi (Date: 26.05.2016. File size: 2.02MB)
Desktop Linux (desktop computer with Linux OS):
SMathStudioDesktop.0_98_5990.Mono.tar.gz (Date: 26.05.2016. File size: 1.16MB) - last version of Mono distributive required!
Cloud version (all modern browsers supported):
SMath Studio in the Cloud (Date: 26.05.2016.)

5 users liked this post

Mike Kaganski 5/26/2016 7:09:00 PM, Davide Carpi 5/26/2016 7:47:00 PM, NDTM Amarasekera 5/26/2016 11:41:00 PM, Radovan Omorjan 5/27/2016 6:33:00 AM, sergio 5/27/2016 11:51:00 AM

#2 Posted: 5/26/2016 7:09:56 PM

Mike Kaganski

184 likes in 434 posts.

Group: User

I suppose it's 5990, not 5983

Thanks for great news!

С уважением, Михаил Каганский

#3 Posted: 5/26/2016 7:46:33 PM

Andrey Ivashov

2270 likes in 3734 posts.

Group: Super Administrator

Wrote
I suppose it's 5990, not 5983
Thanks for great news!

You're right

Thank you!

#4 Posted: 5/26/2016 11:22:43 PM

Mike Kaganski

184 likes in 434 posts.

Group: User

solvQerr.sm (9 KiB) downloaded 109 time(s).

С уважением, Михаил Каганский

#5 Posted: 5/27/2016 10:58:52 AM

Igor Skarine

14 likes in 97 posts.

Group: User

Hi Andrey,
It looks like maple plugin has a problem with 5990.
All my valuable symbolic calculations are broken.
For me, the last good version is 5973.
Please look what is wrong with this sample.
This time, it doesn't report x-not defined, but hangs forever ...
Thanks,
Igor
cic3-1.sm (7 KiB) downloaded 144 time(s).

#6 Posted: 5/27/2016 11:32:45 AM

Davide Carpi

1417 likes in 2873 posts.

Group: Moderator

I guess maple() function must be recompiled following this hint. Please be patient, it is a change that affects all the functions that deals with symbolic data.

If you like my plugins please consider to support the program buying a license; for personal contributions to me: paypal.me/dcprojects

#7 Posted: 6/1/2016 4:02:08 PM

Ruben Sidranski

17 likes in 463 posts.

Group: User

I found a bug in this version. Took over 5 minutes before I terminated the file. I took a snapshot with the timer running (shows 43 seconds) after Jean confirmed 5 seconds on his machine using version 5346.

Wrote

Wrote
Is this supposed to take over 5 minutes to calculate? May be SMath as this is the second sheet today that "locks-up" with the calculation tool bas and the timer running

Re-download in case something got wrong in the transmission.
XP Home, 1.8 GHz, Smath 5346 OFFicial release => 5 seconds

Jean

Inst_Viscosity Water.sm (52 KiB) downloaded 137 time(s).

Would be interesting more collabs test the timing, thanks.

#8 Posted: 6/1/2016 6:14:40 PM

Mike Kaganski

184 likes in 434 posts.

Group: User

С уважением, Михаил Каганский

#9 Posted: 6/2/2016 1:35:56 PM

Ruben Sidranski

17 likes in 463 posts.

Group: User

Mike,
Still seems like a computational bug. Mine went 5 minutes before I closed it down

Wrote
Re-download in case something got wrong in the transmission.
XP Home, 1.8 GHz, Smath 5346 OFFicial release => 5 seconds

Jean

#10 Posted: 6/2/2016 7:37:47 PM

Mike Kaganski

184 likes in 434 posts.

Group: User

Wrote
Mike,
Still seems like a computational bug. Mine went 5 minutes before I closed it down

Yes, but is this truly SMath bug? Or some extension's? Did you try to investigate before writing "I found a bug" to the New Smath Version advertisement topic?

I have no maple/maxima on my system, could that be a thing?

С уважением, Михаил Каганский

#11 Posted: 6/2/2016 10:24:20 PM

Jean Giraud

983 likes in 6866 posts.

Group: User

Wrote
Yes, but is this truly SMath bug? Or some extension's? Did you try to investigate before writing "I found a bug" to the New Smath Version advertisement topic?

I have no maple/maxima on my system, could that be a thing?

======================= I read you Mike ===========================

1. It may be a problem with local machine/OS ... conflict with others [I doubt].
2. Maybe with the Maxima solve block: solve(μT(t)-q(t),t,2,30) => disable the solve block.
3. Maple is not invoked in this attached document => eliminate Maple
4. With the symbolic Ah ! but only Smath symbolic.
-------- Thiele continued fraction Smath symbolic down to the plot => 4.5 sec
-------- Thiele continued fraction Smath numeric down to the plot => 0.5 sec
A thiele continued fraction calculates bottom to top [24 Aops ... Arithmetic operations]
This Thiele in Smath symbolic is in the range of 8 pages wide, my count 326 Aops

The μ350(x) is a J_Fraction equivalent to the continued fraction still existing in Maple.
It reduces the Aops [in this case down from 24 to 13. It computes faster as well as reported in C.T. Fikes [IBM of years 70]. I think I still have the algorithm of J_Frac,
in case some of you developpers would be interested.

The rational fractions [typical normalised Padé style] are expandable in J_Fraction,
but here again: Smath symbolic expands in horrible mess as the continued fraction.

The Newton-Wallis method is a way to compute continued fraction top to bottom.
My brain has diluted on this, maybe Mathcad as well.

Now, whoever wants: test and report versions to versions.

Cheers, Jean

Forum Compute Test.sm (25 KiB) downloaded 95 time(s).

#12 Posted: 6/2/2016 10:39:11 PM

Ruben Sidranski

17 likes in 463 posts.

Group: User

Wrote
Yes, but is this truly SMath bug? Or some extension's? Did you try to investigate before writing "I found a bug" to the New Smath Version advertisement topic?
I have no maple/maxima on my system, could that be a thing?

That was partial answer. I just upgraded to 5997 and there was an update to the Maple V wrapper. System went 14 seconds before I terminated it. I disabled the plugin and it took .98 rev 5997 0.760 seconds to process. Andrey fixed something in the new version as I was over 14 seconds with Maple disabled before. Sorry for not being clear before.

#13 Posted: 6/3/2016 2:43:44 AM

Jean Giraud

983 likes in 6866 posts.

Group: User

Wrote
That was partial answer.I just upgraded to 5997 and there was an update to the Maple V wrapper.System went 14 seconds before I terminated it.I disabled the plugin and it took .98 rev 5997 0.760 seconds to process.Andrey fixed something in the new version as I was over 14 seconds with Maple disabled before.Sorry for not being clear before.

Are you talking about

Forum Compute Test.sm (25 KiB) downloaded 101 time(s).

This document has nothing to do with Maple. I would be damned deceived if Smath couldn't
add 2+2 by itself. It just does from Windows resources. Are you the only user of the machine ? If not: kill all the birds around. In the attached work sheet, there are two
timing. So, which one are you talking about ?

Cheers, Jean.

#14 Posted: 6/3/2016 3:17:53 AM

Jean Giraud

983 likes in 6866 posts.

Group: User

Oh ! there is another point, most important, like this: Some "modern users" have dual core, quadruple core.
That kills the compute engine [slows down]. There was a Mathsoft collab crying about Mathcad 11 takig minutes
to calculate a work sheet with his bomb dual core that mine, steam engine, computed in fraction of second.

#15 Posted: 6/3/2016 11:33:41 AM

Ruben Sidranski

17 likes in 463 posts.

Group: User