Release of EvoluteTools T.MAP *beta*

  1. I’m not sure - as I said the Rhino instance whited over and said “Not Responding” for a while.
  2. Yes
  3. Approx 50k - I tried to reproduce the example of the sphere above.

Many thanks for letting us know. We will work on making it more user friendly. Were the results up to your expectations?

When starting from a default sphere, the results in the poles is not good. I guess this is unavoidable due to singularity nature of the control points. Other tests are satisfactory.

I did notice that the mapping is not continuous over a seam, for example when doing a Torus (indenpendent of Curvature or Boundary type)

Hi Menno,

if you use the settings proposed by Alex in his previous post to mesh your torus there shouldn’t be a seam to begin with. As Alex mentioned, this will be done automatically in the background in the next release:

Best,
Mathias

Perfect that works, thanks for pointing out the instructions :smile:

Sorry for hijacking this thread but Olivier how did you do these objects? I assume not in rhino, or am I wrong (Tspline, Grasshopper?). And just started testing T.Map, and I really like it, thanks for handing it out for free, until launch!
Regards
Per

Why have you included utterly no instructions or sample files, just some texture images? I have no idea what curves I’m supposed to select or create to select. I can push keys and work a mouse as good as anybody, but I do need to know which keys to actually push and if you don’t actually explain it, which keys to push will remain a mystery since “ParametrizationFromMesh failed” is not useful feedback. And what is “mixed integer parametrization” anyway? The descriptions claim I input a mesh and with one click I get some sort of quad output. But the actual commands want a second mesh or some sort of curves as input as well.

Nik, many thanks for your feedback. I agree that the user interface can be improved. Instructions can be found in the help file: Please F1 after starting etTmap to open the help file for EvoluteTools T.MAP, or go to

Help -> Plug-Ins -> EvoluteTools T.MAP

In short, T.MAP either aligns the mesh with curvature directions (no need to select anything else) or user-specified curve directions (in this case you will be asked to select the curves you want to align to).
You can download an example file here. We are looking forward to your feedback. In case you have trouble remeshing with T.MAP, please post your example here or send it to support@evolute.at.

A final beta version of T.MAP has just been released. Find out about it here, including a lengthy description of the plug-in features and download links: http://blog.evolute.at/?p=685
We are still working on final improvements to the interface, smoothing the results and using sharp features for alignments, these will all be present in the commercial release.
As always, we are happy to receive any type of feedback!

Sorry for late reply, I have used Tspline to make those shapes.
Regards.

An update to EvoluteTools T.MAP beta is available for download, including

  • mesh extraction: several fixes in boundary treatment
  • added option None for directions

Hi,

I have installed the lastest T.MAP beta (Loading EvoluteTools T.MAP 2.5.15 (151122_EVOLUTETOOLS_PRO_2_5)…)
and since then my rhino5 (SR12) will not fully shut down anymore. This means that rhino seems to be closed, but in the taskmanager I can still find a rhino process consuming 200 MB of RAM.
When I start another rhino this is very sluggish and only killing the processes will help.

Furthermore the toolbar will never save where it had been relocated, so I have to do this over and over again.

No workaround needed, just to let you know for further investigation.

Hi, many thanks for letting us know about these problems. We aim to fix both in the next beta release, next week. Does the shutting down problem happen every time you use T.MAP? Did you find T.MAP useful in general?

Yes it happens every time I am shutting down rhino. Only condition is to
start the T.MAP no need to do anything else on rhino.

T.MAP looks promising, but seems to be a little limited if we want boundaries like limits on beam length and evened out distribution of lengths etc.
I guess that it will somehow link into the workflow of the evolutetools, isn’t it?

Many thanks for your feedback.

Please let me know whether I understood correctly:

  • You would like to be able to specify a maximum edge length.
  • Distribution of edge lengths: T.MAP basically distributes edge lengths as equally as possible. What would you like to be able to influence?

Certainly T.MAP can nicely be combined with EvoluteTools LITE / PRO, but it’s meant to work as a standalone product.

Again, many thanks for your feedback, pleased don’t hold back if you have more.

PS: The toolbar problem has been fixed in the meanwhile and will be part of next week’s release. We are still looking into the shutdown problem.

Yes or even better some boundaries.
Assume a glass grid shell, for the width of the panels you should not use values >2.5 m (fabrication issues), whereas you do not want it to drop smaller than 0.5 m for most parts of the structure.

As said previously one direction should be <2.5 m but the other one could easily be up to 6 m. Therefore a ratio and/or boundaries would be nice to have.
Of course these boundaries can only be soft bounds for not limiting the solver too much, but knowing about a soft bound it would still bring you closer to the desired mesh.

Here is another question on the planarity of the quad mesh.
I tried with your testsample (fertility.3dm) but always got the message "Mesh consists of several connected components… _weld …) . Not being able to fix it on your test model I came up with a torus to testdrive T.MAP.
While choosing DirectionNoBoundaries to None/Curvature resulted in the obviously plane quad mesh. Trying the Curves option after drawing a partial helix on the torus surface in combination with a small number of levelsets resulted in heavily warped quads which is understandable,

The question is whether it would be a good idea to have a planarization boundary, so that you have a chance to weigh the importance of planarization.

By the way, will you come up with a floating licence on T.MAP as well?
We are thinking about buying the evolute PRO and T.MAP but only if we can have floating licences for all software parts.

Hi,

This is possible and we could implement it pretty quickly, but it is limited to certain types of patterns. If you have no singularities, you can of course scale distances for the two directions independently. If there are singularities you can typically only make the pattern denser without destroying the mesh structure, and whether you can do this independently for the two directions or only for both at once depends on the type of the singularities (let me know if you want us to elaborate on this).
Anyways, we think it is a great suggestion and we’ll definitely work on it once we get the plugin to the stable state.

Exactly. Now there’s several options one could use to enforce these bounds once the parametrization is done.

If one of the bounds is violated almost everywhere you might need to change the ideal value and parametrize again (there’s even ways to do this without recomputing the whole thing in many cases). This iteration could also be done automatically until the bounds are met, we will think about this.

If you like the output but have problems in some specific areas, you could extract a mesh and use some local subdivision methods followed by a relaxation, both of which are in EvoluteTools PRO (or even LITE).

Enforcing or even optimizing for planarity directly is not something that the T.MAP algorithm is designed for. However, as you observed yourself, if you set the pattern to follow the mesh curvature you will generally get the best results in that respect. Such a T.MAP output mesh would then be ideally suited to optimize for planarity using EvoluteTools PRO. There, you can easily observe the planarity measure and optimize until a given bound is met.

Yes, this option should be visible in our web shop within the next few hours.

We would of course be very happy if you’d decide to use our tools! Please let us know if there’s anything else we can help you with.

Many thanks for pointing us to the problem with the example fertility.3dm downloadable from our website. It has been fixed in the meanwhile, please download it again. Moreover we added a link to a publicly available benchmark model database, which is useful for experimenting with T.MAP:

Fertility example
Benchmark models (obj files, 500mb)

We have identified this problem and solved it. An update to T.MAP will be released within a week. Registered beta users will automatically receive an email containing a download link.

We are currently preparing a series of video tutorials about T.MAP. The first episode is already recorded, watch it here:

Following episodes will cover boundary treatment and general good practice about the plug-in.