> There is also a little free place just under it for a little text

That's normally where the name of the currently open file goes (like alarm.3dm), it's only blank when you initially start up before saving or loading anything.

By the way, here is a cleanup of your 2.0 radius version:



It was actually pretty quick to clean up - first select it and do Edit/Separate to break it into different individual surfaces. Then select all those surfaces that have various intersecting parts, and then do Boolean Merge.

That intersects everything and cuts everything up into smaller parts. You can then delete the excess parts, select it all and join it, and then use Construct / Planar to fill in the holes in the top plane.

Merge is turning out to be kind of a one-step simple alternative to Trim when you just want to slice everything up into pieces.


There is an option in the fillet engine to do something similar to this, a kind of boolean merge of each fillet piece instead of the regular attempt to trim surfaces by the fillet rails. But it doesn't seem quite reliable yet. But in the future I may be able to automate that so that oversize intersecting fillets like this will work.

- Michael

> Yeah I didn't think it was frivolous, it's just difficult to think of how to go about making a UI to do something like that in one step which is general purpose enough, yet not too terribly complex and hard to use...

There are lots of ways to approach it I think, each with its strengths and weaknesses. One easy way would be to use a network of curves as a 'scaffold' for positioning smaller pattern elements (essentially a multiple 'curve array' operation) and applying some construction rules to the pattern elements deposited by the curves. In the example I've attached, wavy curves deposit circles at regular intervals and intersecting circles get deleted. A different rule could be 'find midpoint of intersecting pattern elements and draw one larger pattern element around that midpoint'. Or indeed 'sweep area around pattern element (sweep radius = 10) and modify pattern element size based on volume of other pattern elements that fall inside sweep area'. And so on.

I don't there is a single solution that will work for all imaginable pattern types, but functions like this could help take the pain out of creating certain decorative or functional structural patterns, especially if the construction remains 'live' so you can go back and alter the construction parameters at will.

-d

Michael, is it possible to indicate the number of polygons created in the mesh export dialog?

It's always handy to know how many polygons you're creating at export time...

-Will

Hi Will, the polygon and point count is available in the command options area, in the upper-right corner of the main window.

I kind of stashed it off over there to try and streamline the dialog a bit.

- Michael

Boy, DUH! Sorry Michael!

I totally missed that info pane, my eyes are usually focused on the resultant topology of the of the mesh in the preview and the export dialog while tweaking the slider - I'm still learning MOI, I admit it!

At least now I know to look at that area for more than when adjusting parameters for modeling tools!

Thanks again!

-Will

a print option would be great. for architecture it´d be even useful to set scales like 1:100 and so on

Hi Stan, I definitely want to add printing in a future version, but it will be missing from the 1.0 version. For now you'll have to use a different CAD program to handle printing. Rhino works well for this since there is tight file compatability between MoI and Rhino.

- Michael

Hi Michael,
a few things that have come up for me.

In 3D could we have rotate about mouse pointer rather than 0,0 which wizzes things off screen when zoomed in.

Could we have a buttton that centres a new object on 0,0 I know we ought to do it right but I don't always.

I keep going back and forth between Construct and Transform, I notice that there is a bit of room at the bottom of that section, any chance of having them both on screen at once?

For us jewellers Grid Snap is a boon, but when you get down small (1/10mm) you have to switch it off, I know someone suggested that a 1/10mm grid appear when you zoom up large , perhaps a flagged reminder on screen that this has happened?

If I create a solid inside another but with a surface level with another , I find the smaller one hard to select, how should I do this?

regards Tim.

Hi Tim,

One way to select a smaller object inside a larger object is to
do an Edit>Copy of the larger object, then delete it. That will
give you free access to pick the smaller object.
Then you can click Paste and bring back the larger object.
Hope this helps.

Michael,

I'd like to ask if there is a way right now to script the arrow keys and up and down keys to move an object. Last night, I brought a model from MoI into Rhino so I could render it... The arrow keys were really helpful in positioning lights.
By using the arrow keys, you can nudge things along the world axes. Since it's not viewport dependent, you can keep your eyes in the perspective viewport.

This is just a quick preliminary rendering of the project.

Thanks,
Jesse

Thanks Jesse,
hadn't thought of that, sometimes these solutions are a bit devious!
regards Tim.

Hi Tim, thanks for the feedback!

> In 3D could we have rotate about mouse pointer rather than 0,0
> which wizzes things off screen when zoomed in.

This one would maybe be a bit difficult. The rotation point needs to be a 3d point in x,y,z space. Your mouse pointer is normally only a 2d location on the screen. There are different ways that can be used to calculate a 3d point from the 2d mouse cursor location, such as intersecting it with the grid or snapping on to a nearby object, but these normally happen during drawing commands where you have a lot of feedback for where that point is while you move the mouse around. For rotation this wouldn't happen, it would need to suddenly generate that 3d point without doing a bunch of tracking around of it beforehand. This would probably be somewhat confusing and hard to control.

The rotation center point doesn't have to be 0,0,0 - that's just what it starts out to be.

The easiest way to change it is to select some objects and use the "Reset" button that appears on the panel at the bottom of a viewport. This will zoom to fit the selected objects to the screen, and it will also place the rotation pivot point at the center of the selected objects.

So after a "Reset", you should be able to easily rotate around the selected objects without going off into space somewhere else.

The other method is you can use the "Area" zoom on that bottom panel as well - that lets you pick a point and drag a zoom box around it. The point that you pick can be snapped on to any point you want, and that point will become the new rotation pivot. So if you want to focus in on the corner of a box for example, do a zoom area and snap the center point to the corner of the box and drag a rectangle around it a little bit. Now when you rotate you will pivot around that box corner.


> Could we have a buttton that centres a new object on 0,0 I know we ought
> to do it right but I don't always.

If you need to center it later on, try the Transform / Align tool. You can use 2 applications of that and switch between horizontal and vertical centering and snap the target point on to the origin and that should do it.

If you can identify a center point of the object to grab, you can also just drag it to the origin and it will snap there, make sure you have Object Snap enabled, it should be highlighted in orange on the bottom bar. Sometimes the Transform/Move command can give you a little more control than dragging since it allows you to more explicitly pick the base point.

But the best way is to just draw the object centered from the very beginning!


> I keep going back and forth between Construct and Transform, I notice that there
> is a bit of room at the bottom of that section, any chance of having them both
> on screen at once?

I want to reserve that area for some future expansion.

Which Transform commands do you find yourself using the most often? I have some ideas that I want to try for version 2.0 that should make it easier to do some types of transforms without needing to go over there.

If you are frequently using a particular transform, it might help to set up a keyboard shortcut for it. Let me know if you want some help setting this up.


> For us jewellers Grid Snap is a boon, but when you get down small
> (1/10mm) you have to switch it off, I know someone suggested that a
> 1/10mm grid appear when you zoom up large , perhaps a flagged
> reminder on screen that this has happened?

I'm kind of wary about automatic changes to the grid, even with flagged reminders. It's kind of dangerous in a certain way because normally the grid gives you a distinct sense of the scale you are working on and lets you guarantee point placement at specific values, but if it changes just in response to view manipulations, you no longer have those guarantees.

One thing that you might find useful is a keyboard shortcut that would step the grid spacing up or down on a keystroke. Paul has a script for that here: http://moi3d.com/forum/index.php?webtag=MOI&msg=509.1

Also Petr has collected some other useful scripts here: http://kyticka.webzdarma.cz/3d/moi/

Let me know if you need any help setting these scripts up.


> If I create a solid inside another but with a surface level with another , I find the
> smaller one hard to select, how should I do this?

Probably the easiest way is to hide the larger one.

Another way is to use the area select - this is the rectangle that you can use for selection when you click and drag in an empty area. If you drag starting from the right side and moving towards the left, you will get a dashed rectangle and anything that intersects the rectangle in any way will get selected. if you drag from the left going towards the right, you will get a solid rectangle and only things that are completely contained inside of it will get selected. You can use this "complete containment only" mode to capture a smaller object if you drag a rectangle that only contains it and not the larger object.

You can also sometimes use Select / Invert to help get a selection onto a difficult object as well - select everything else and then do Select / Invert and it will flip the selection.

- Michael

Hi Tim,

<If I create a solid inside another but with a surface level with another , I find the smaller one hard to select, how should I do this?>

I often work with two versions of MoI open at the same time, one for the job I'm designing in & one for "Copy & Paste" of various parts.
I can then use the "parts version" to store various pieces in their correct position, while I continue to work in the "design version".
It's also handy for those designs where you get a "second idea" while working on the first one or when you've got a lot of parts to "Boolean Difference".
In the "design version", I can then Hide or Delete those various parts knowing that I can just Copy & Paste back from the "parts version".

Hope this makes sense, Colin

Thanks aal,

these are really good suggestions, just what I need1
regards Tim.

Infinite grid in the 6 views (if it's too hard for the 3D view) (for steph) :)

> Infinite grid in the 6 views (if it's too hard for the 3D view) (for steph) :)

I guess I still don't understand why this is a problem.

If he wants the grid to be larger, why doesn't he make it larger in the grid options?

I should be able to do an infinite grid in the ortho views eventually, but since it is so easy to increase the grid to a larger size like 2000 or 4000 sections, it just hasn't been a priority yet.

Does he need it larger because he has it still set to the original default, or is he doing something where 2000 or 4000 is not large enough?

- Michael

He has habit to begin to modelize without constraint :)
For example for his "background" of "watches" start to modelize for the 0,0,0 in the defaut mode
but following inspiration arrive to the end of univers' grid because not think of the limitated size !
For architecture and urbanism!
And seems the other 3D prog has also ortho view infinite!)

Some restraint is actually helpful.

The way floating point mathematics works on computers tends to cause a loss of numeric precision when larger numbers are mixed in calculations with smaller numbers. Just simple operations like addition and division sort of get rounded to coarser values.

So it can cause problems when objects are located farther and farther away from the origin.

I'm not sure of an exact number, but things should probably generally stay within coordinates of -5,000 to +5,000 or so to get the best results.

Placing objects that are farther out in space or larger in size than that will tend to generate additional failures especially in complex commands that do a lot of calculations like booleans.

- Michael

> but following inspiration arrive to the end of univers' grid because not think of
> the limitated size !
> For architecture and urbanism!

That would be nice, but generally such unrestrained and unlimited size to objects will cause numerous calculation problems....

Beyond the calculation problems, MoI version 1.0 just isn't envisioned as a city planning tool, it is more oriented towards modeling individual objects.

- Michael