Message Boards Message Boards

2
|
10254 Views
|
10 Replies
|
13 Total Likes
View groups...
Share
Share this post:

Change in version 10.1 Manipulate when opening Slider controls.

Posted 9 years ago

In Manipulate, since version 10.1, a very annoying feature appears when opening the controls with the top right + button: For example in: Manipulate[ x, {{x, 4(default)}, 0, 10} ] If you click the little + to open the controls of the slider, it is reset to the minimum value instantly and you lose the default setting!. In past versions, the default setting (4) was kept as a starting point which was very convenient. Has anyone experienced this or is it just on my OS X10.10.1? is it a bug or a wilful "improvement"!?

POSTED BY: Erik Mahieu
10 Replies
Posted 9 years ago

Solved for me with the new version10.2 in OS X. Thanks, this was very annoying... A pity one needs to pay the upgrade to get a bug like this one solved!

POSTED BY: Erik Mahieu
Posted 9 years ago

Indeed, this must be the bug: (I was unaware of this) tagged bug report on Stack Exchange

POSTED BY: Erik Mahieu

Why is it incorrect to call them "asymptotes"?

POSTED BY: Gianluca Gorni

They look like asymptotes, don't they. However, if you use Exclusions and ExclusionsStyle, you can get the real asymptotes to show up (in different colors or dashing, if you want).

Back in the dark ages, when you had to write your own graphics code, the appearance of the nearly vertical lines when plotting Tan or Cot was evidence of a bug or sloppy coding. What you do (or did) when plotting a function was to set a sampling of points, evaluate the function at the points, and draw line segments between them. It was easy to miss the fact that the line went from a vary large positive ordinate to a very large negative ordinate, and suppress the line if this was the case. The adaptive sampling in Wolfram Language is very good, but it, by itself, does not look for singularities. This was evidently the case for Mathematica 1 through 4. The bug was fixed in Mathematica 5, and then Mathematica 6 switched from Postscript-based graphics to a more powerful system.

Besides, what you want when you ask Mathematica (Wolfram Language) to plot a function is to see a locus of points for that function, not any asymptotes, etc. The vertical lines are NOT* part of the Tangent function, and should not be drawn unless you also ask for the asymptotes.*

For some functions, Mathematica will make reasonable assumptions about exclusion points -- something it does much better in Plot3D[]. What I have repeated asked the Wolfram programmers to do is to make sure that the code already in place to detect singularities works in the case of the elementary trig functions. The language already handles the more complicated analogues for Plot3D[] and many other functions with singularities in the 2D plots.

So, to answer your question: they look like asymptotes, but we did not ask for the plot of the function plus asymptotes. Wolfram language already has an option to draw asymptotes, should we want them.

I have discussed this issue with anyone from Wolfram who would listen, and it will be a topic my conversations for this year's conference. Some of the bugs have been around so long that tech support and marketing seems the think that they are features. (For example: the default behavior of Plot[Tan[x], {x,0,6}]. The incorrect behavior at the singular points existed since version 1, was fixed in version 5, and returned in Version 6, not yet fixed. Wolfram Alpha uses the same code, so a whole generation of users must thing that the function (and Cot[], Sec[] and Csc[]) has vertical lines at the singularities -- marketing incorrectly calls them asymptotes.)

I am happy that there are so many new initiatives at Wolfram. It might be better for everyone -- especially new users who may be confused by inconsistent behavior -- if 10.2 or even version 11 were limited to fixing bugs. The only exception I would make is the Cocoa (64 bit) front end for the Macintosh version. Getting rid of legacy code would probably clear up many more bugs than would be introduced. Even Windows and Linux users would benefit, since the work would expose platform independent bugs in the front end.

Posted 9 years ago

A good example of long-lived bugs appeared a few hours ago in this forum in a posting reporting a wrong result for a limiting value of a Laguerre function.. Laguerre Limit

In one response, the poster was directed to an article in StackExchange reporting that the same bug existed since version 7. And in this case, the problem is that Mathematica produces a wrong result.

I would also like to see a version assigned to clearing unresolved issues. I use Mathematica extensively for client work, where a wrong result can be costly.

POSTED BY: David Keith

Rolf Mertig,

I'm not sure if this is a matter of more beta testing, or of keeping a lower amount of bugs alive. There are too many reports, acknowledge as bugs, that are live for several releases now, and that are not related to some hard or still not known to humankind algorithmic capacity.

Some of the "Mathematica gets this Integration wrong" kind of bugs, for obvious reasons, can last for a long time (I said some...); but bugs related to such a fundamental data layout function, like Grid (you know that this weekend I was fixated with Grid...), that last for several releases, are, in my opinion, hard to accept.

Since we don't have an open bug list (which I can perfectly understand, or at least, a fully opened one), its difficult to say if these bugs were known prior to their first release. But even if not, we know that they were known prior to their second, third and fourth release, which makes me think that even if the Manipuate bug of the OP was known, it seems that it could perfectly have been approved for release (I know that it's a harsh thinking, and an unpleasant saying, but it's solely based on shown behaviour).

I really hope that this situation gets improved, or I will start attending the conferences with a T-shirt saying, "please get [put some functionality name here] fixed" (if WR would let me in with it :-)

POSTED BY: Pedro Fonseca

We don't have an open bug list, but we have this: http://mathematica.stackexchange.com/questions/tagged/bugs

POSTED BY: Gustavo Delfino

Manipulator[.2] also already shows this bug in V10.1. It looks like it is fixed in the development version already. At least this is what I think is mentioned here

I think this bug is a perfect argument that WRI should do more and better beta-testing also with external users before releasing any new version, also so-called minor releases.

POSTED BY: Rolf Mertig
Posted 9 years ago

I have noticed the same under both Windows 7 and Windows 8.1.

Yes, its annoying. And certainly it can not be considered as an improvement.

POSTED BY: Hans Milton
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract