10

Referring to the method I proposed for Problem with ParametricPlot Mathematica 10.0.2 throws an enigmatic error message:

Plot[Sinc[x], {x, 0, 5}, Method -> {MaxBend -> 1}]

MaxBend::deprec: MaxBend->1 is deprecated and will not be supported in future versions of the Wolfram Language. Use Method->{MaxBend->1} instead.

I don't know if this message is entirely a mistake or if it is really trying to tell me that there is a different form I should be using now. Is this same message generated by 10.1.0? Is there a known different form for MaxBend?

Community
  • 1
Mr.Wizard
  • 271,378
  • 34
  • 587
  • 1,371
  • 8
    It's probably not a bug, but a bug – Dr. belisarius Apr 15 '15 at 16:23
  • 1
    I see it in 10.1 on Win8.1. – dr.blochwave Apr 15 '15 at 16:45
  • Does it possibly mean "MaxBend" as a string? (Obviously it will not give the message if so, so you will have to verify that it really has the intended effect. A string does not work in 9.0.1.) – Oleksandr R. Apr 15 '15 at 23:24
  • @Oleksandr Good idea! However Method -> {"MaxBend" -> 1} throws the same error. :-/ – Mr.Wizard Apr 16 '15 at 01:39
  • 1
    Does the alternative Method -> {Refinement -> {ControlValue -> (* radian angle *) }} still work, tho? – J. M.'s missing motivation May 13 '15 at 00:08
  • @J. M. Yes, and without an error message. However MaxBend also works so I am just turning Off the message. – Mr.Wizard May 13 '15 at 00:10
  • 1
    Personally, I've always assumed it's a typo, that MaxBend is in fact deprecated, and the message was supposed to point the user to Method -> {Refinement -> {ControlValue -> (*radian angle*) }}. I was hoping someone from WRI would comment on your question, though, to clarify. Have you reported it to WRI? – Michael E2 Jul 05 '15 at 18:45
  • 1
    @MichaelE2 Nuts. I misremembered and misread my own comment above; I thought that form also gave an error, but since it does not I think your explanation is certainly plausible if not more. I shall remove the bugs tag now. edit: No, I did not report this to WRI. There are many more important things that need work so I feel it would be counterproductive. – Mr.Wizard Jul 05 '15 at 18:50
  • @Karsten That form does not appear to work. It throws no errors but it also doesn't have any effect. – Mr.Wizard Jul 05 '15 at 18:53
  • 2
    It still works without any message for ContourPlot as used in this answer. – Karsten7 Jul 05 '15 at 19:00
  • @Karsten Confirmed. Thanks. – Mr.Wizard Jul 05 '15 at 19:02
  • Certainly reporting it is not a priority -- I haven't done it either, and I've been aware of the message for some time. Just wondering, I guess, considering MaxBend has apparently been deprecated or at least missing from the docs since V6. – Michael E2 Jul 05 '15 at 23:44
  • A similar, useless message can be generated with the option Method -> {PlotDivision -> 2}. – Karsten7 Jul 08 '15 at 21:12
  • 1
    Autocompletion even suggest MaxBend as a Method screenshot. – Karsten7 Jul 11 '15 at 10:24

1 Answers1

2

Earlier invalid assertion redacted.

Some spelunking reveals that the innermost (accessible) plot function called is:

Visualization`Core`Plot

And using this function directly also produces the message:

Visualization`Core`Plot[Sinc[x], {x, 0, 10}, Method -> {"MaxBend" -> 1}]

MaxBend::deprec: MaxBend->7 is deprecated and will not be supported in future versions of the Wolfram Language. Use Method->{MaxBend->7} instead.

I have failed in my attempts to trace the origin of the problem further but I may try again later.

Mr.Wizard
  • 271,378
  • 34
  • 587
  • 1,371
  • 1
    Note that tr2 = With[{gr = 1/GoldenRatio}, Trace[{AspectRatio -> gr, Axes -> True, AxesOrigin -> {0, 0}, Method -> {"MaxBend" -> 1}, PlotRange -> {{0, 10}, {-0.21723357397428536, 0.999999999999993}}, PlotRangeClipping -> True, PlotRangePadding -> {Scaled[0.02], Scaled[0.02`]}}]]; tr2 === Query[35, 1][tr]returnsTrue. How do you deduce that"MaxBend"` is being duplicated? It seems to me that the multiple instances could be the result of the evaluation chain. I note that some things are not evaluated to very bottom of the expression tree, so I might be wrong. Definitely confused. – Michael E2 Jul 10 '15 at 00:29
  • @MichaelE2 Facepalm -- no, I think you're quite right and I'm crazy. Trying too hard! – Mr.Wizard Jul 10 '15 at 00:31