MoI discussion forum
MoI discussion forum

Full Version: Any parametric updates?

Show messages:  1-15  16-35  36-55  56-75  76-95  96-113

From: Michael Gibson
24 May 2019   [#36] In reply to [#34]
Hi Marco, well the other thing I should mention is that the v4 release cycle has been pretty unusual since it has involved a low level rewrite to get to a 64-bit platform independent code base.

That is what has made some fundamental things like a native Mac build possible now, but it has also required an unusual amount of time fixing bugs and regressions.

So there actually has been less time spent on new features (other than very broad fundamental ones like being 64-bit and Mac native build) for this release than there might normally be and it's also why this release has taken a lot longer to wrap up.

Things should shift a bit for the next releases after this one, although it is also true that I purposely like to develop new features slowly so they can fit well in the existing UI rather than being just piled up on top of a big heap.

- Michael
From: mkdm
24 May 2019   [#37] In reply to [#35]
Hi "Death".

@You: "...don't buy one that isn't what you wanted and then try to force the developer to add all the stuff you think you need..."

I don't understand what's your problem.
Please, don't tell me what I can or i cannot do here on the forum.

I'm free to ask what I want to ask to Michael, new features, bug fixes, different way to do the existing things, etc...
I own a commercial license of Moi so I'm totally free to talk to Michael here on the forum in any polite way.

I think that Moi is a great software, but I also think that it has a too "slow" growing.
I think that some new modelling tools is necessary.
So I'm asking to Michael.

Stop.

Marco (mkdm)
From: mkdm
24 May 2019   [#38] In reply to [#36]
Hello Michael.

@You: "...Things should shift a bit for the next releases after this one..."

I'm confident that will be the case.

I only hope that it will happen as soon as possible :)

Thanks.
From: Michael Gibson
24 May 2019   [#39] In reply to [#38]
Hi Marco, maybe I misunderstood but I thought you wrote elsewhere that you're not doing any 3D modeling work currently? Or did that change?

- Michael
From: mkdm
24 May 2019   [#40] In reply to [#39]
Yes, you're correct.
Actually I have stopped almost all my 3D activity.

But, is it important?

I always keep my eyes on Moi because I've used it A LOT, A LOT :) and I'm waiting for a long time some new modelling features.

Simple.
I hope that Moi will offer some new modelling tools to become even more powerful.

Nothing more that this.
From: Michael Gibson
24 May 2019   [#41] In reply to [#40]
Hi Marco,

re:
> But, is it important?

Certainly - MoI is meant to be a functional tool that helps people get their work done. Knowing how it is being used helps with refining it.

- Michael
From: Death
24 May 2019   [#42] In reply to [#37]
"I don't understand what's your problem.
Please, don't tell me what I can or i cannot do here on the forum.

I'm free to ask what I want to ask to Michael, new features, bug fixes, different way to do the existing things, etc...
I own a commercial license of Moi so I'm totally free to talk to Michael here on the forum in any polite way.

I think that Moi is a great software, but I also think that it has a too "slow" growing.
I think that some new modelling tools is necessary."
---

A little sensitive today? ;-)

I didn't tell what you can or can not do in the forum. Just like you, I have an opinion on MoI and it is that it is the BEST modeler to use. You may disagree, but it doesn't mean that I have to share your opinion.

My experience with parametric modeling is that it is more of a pain in the butt than it actually helps. Way down in the tree one makes a change and easily screws up the whole model. Then its time to fix the rest of the steps one by one, without having the comfort of seeing the actual change, but only the cascade of it up the tree. I found it easier to model with MoI, things can be quickly and easily re-done if one just kept a few curves, or even from scratch is fast enough to beat the pain of parameters that get mangled up with a just single change and propagate badly through the whole change history. I can do in Moi what would take days in MAX, in just a few hours. And Booleans, well, you know that story, nothing beats MoI there.

There are, of course, reasons one might want parametric modeling for, but I personally think it can wait until other more interesting items have been added, since its strength IMO is more for "unplanned" modelling.

Now, since I also own a commercial version, I too have the right to state my opinions here in the forum, just like you. Like'em, or hate'em, that's up to you...

Peace, brother... '-)
From: mkdm
24 May 2019   [#43] In reply to [#41]
@You: "MoI is meant to be a functional tool "

Moi is also a very useful tool for artists and designer, not strictly related to manufacturing or engineering.

I have used a lot Moi for art purposes.

I think that this is a legitimate way of using Moi :)
A way of use that has equal dignity.

The fact that actually I'm not playing with 3D stuff doesn't mean that my requests are useless or less important.

And I'm pretty sure that you think so too.

Thanks.
From: mkdm
24 May 2019   [#44] In reply to [#42]
No problem for me.

@You: "...My experience with parametric modeling..."

I never talked about "parametric". Never.

I don't want Moi to be parametric!

I'm talking about some new modelling tool like for example "Cage deformation", "Bending and tapering" (beyond the "flow" tool), some surface continuity tool, a Zebra checker, Instanting and Grouping (long...long awaited)
Just to name a few.

No parametric.

Have a nice day.
From: Anthony (PROP_DESIGN)
24 May 2019   [#45] In reply to [#43]
I'm with you Death. I have the same experience and desires you mention. If you really need a low cost parametric modeller Atom3D is available. It can also calculate mass properties. I think it would be a good supplement to MoI. Michael is one man he can't do everything. Development time is going to be slow. Given how many CAD programs already exist, it makes sense to pick and choose what is going to make MoI better and how it fits in with the rest of the world. Sure it would be great to have one, low cost, CAD program that could do everything. But that's not realistic, especially from a one man development crew. I wish people would stop beating a dead horse. I read these posts over and over again. The same thing time and again. It's not happening' get over it. This isn't directed at any one person. Just 'all' of the people that post the same derivative thing. I use such and such and I want MoI to be like that and I want it yesterday. When is it going to stop. Probably never.
From: Michael Gibson
24 May 2019   [#46] In reply to [#44]
Hi Marco,

re:
> I'm talking about some new modelling tool like for example "Cage deformation", "Bending and tapering"
> (beyond the "flow" tool), some surface continuity tool, a Zebra checker, Instanting and Grouping
> (long...long awaited)
> Just to name a few.

These have been requested by quite a few people so it's pretty likely they will become a part of MoI at some point. But it won't happen for v4 and I cannot promise which specific version they will show up in.

If any of these are very important functions for what you need to do, there is other software out there that does all of these things like Rhino for example. So if you need these things now it would make sense for you to use other software that does what you need instead of MoI.

- Michael
From: Death
24 May 2019   [#47] In reply to [#44]
"I'm talking about some new modelling tool like for example "Cage deformation", "Bending and tapering" (beyond the "flow" tool), some surface continuity tool, a Zebra checker, Instanting and Grouping (long...long awaited)
Just to name a few."

Well, like I said, use MAX for that.

(PS: You can group by assigning the same Styles to different objects, then you select them as a group by just clicking that style. Instancing is only necessary to save calculation power and ram, both of which are not so much of an issue nowadays and doesn't help if the renderer/model format don't support it. Besides, some renderers allow instancing right within them. Normals I usually set up in the renderer, triplanar being the quick and easy way out (zebra eliminated), continuity is using Blend etc. correctly, you can taper using 1d scaling and bend using flow.)

Of course things like this will prob. be added eventually under the menu commands I am sure. But until then I got all I need so far.

You might want to get back into MoI and with a bit of re-thinking the mesh modeling flow (i.e., ditch it) you might just fall in love with it again...

There was a post here with 3 articles from a German magazine which showed how powerful and easy MoI modeling is. That guy created cool models with a minimum of curves and solids. Definitely worth reading.

One thing I learned: If I wait for my favorite tool to become what I really want it to be, I'll never use it, because there's always another feature one can wait for.
From: mkdm
25 May 2019   [#48] In reply to [#46]
Hello Michael.

@You: "...it would make sense for you to use other software..."

Ok. I didn't want to start and endless conversation about this topic here so let's stop here this conversation :)

THANKS ANYWAY FOR YOUR KIND REPLY. AS USUAL :)

But evidently I didn't make my point clearly enough or you didn't understand what I wanted to say.

No problem.

As I've said TONS AND TONS OF TIMES, I still consider Moi one of THE BEST fast and quick 3D Nurbs modeler, thanks to its super streamlined workflow.

And I'M NOT QUITTING MOI.

But for a long long time (except for 64Bit and the latest SubDiv tool) it didn't receive a TRUE UPDATE about new modelling tools.

This is a fact.

Have a nice day a best wishes for your job and your things in general :)

Ciao.
From: James (JFH)
26 May 2019   [#49] In reply to [#20]
Michael,

It is indeed exciting that dimensions are coming to Moi3d.

These scale-independent arrows and numerals will also be beneficial to vector nodes and to the numbering node, respectively.

I have looked at your video, and see that the arrow head remains in the plain of the dimensioning; which makes sense.

Would it be possible to have a script option for arrowhead display to in alignment with the screen view, because vectors
are not tied to a plane? Does that make sense? Karsten has achieved this by making the arrowhead a 3D solid element, but your new arrowheads would be a more elegant solution IMHO.

If this is too much trouble, please disregard; I just thought it was better to ask earlier rather than after dimensioning had be finalised.

James
From: Michael Gibson
27 May 2019   [#50] In reply to [#49]
Hi James,

re:
> Would it be possible to have a script option for arrowhead display to in alignment with the
> screen view, because vectors are not tied to a plane?

I think it could be possible, I'll experiment with it and see.

The way the annotations work is there will currently be 5 different types of objects for them: Linear dimension, Radial dimension, Angular dimension, Leader, and Text.

They all have a plane associated with them, but if you make a leader with 2 points you could have its plane be any plane containing those 2 points and that would give you a line with an arrowhead on one end of it. Does that sound like it would work for that if there was an additional property to show an annotation's arrowhead flat to the screen?

- Michael
From: James (JFH)
27 May 2019   [#51] In reply to [#50]
Michael,

>> Does that sound like it would work for that if there was an additional property to show an annotation's arrowhead flat to the screen? <<

In the case of text for numbering node it would be ideal to be always displayed "flat to the screen"

However, I'm not sure that I explained myself well in regards to arrows.

Have a look at this image: https://i.ytimg.com/vi/bFzZB4zhqLo/maxresdefault.jpg
or alternatively starting at 13:01 on this video: https://vimeo.com/49043607

The arrows are not "flat to the screen" but are oriented along the vector, yet nevertheless, rotated so that arrowhead is always face on.

I hope this is clearer
James
From: Michael Gibson
27 May 2019   [#52] In reply to [#51]
Hi James, so what I meant by "flat to the screen" is having an arrowhead drawn as if it was in a 2D vector program, like it's always the same proportions just rotated only in the screen plane like these:



So all those arrowheads above all have the same area of filled pixels on the screen.

But that video you linked to makes me wonder if I'm not getting the right idea since it shows some arrowheads that appear to have different proportions and filled areas like here:



But the image you linked to looks like all the same pixel filled area like in the first image above...

- Michael

Image Attachments:
james_arrowheads1.jpg  james_arrowheads2.jpg 


From: James (JFH)
27 May 2019   [#53] In reply to [#52]
Michael,

>> But the image you linked to looks like all the same pixel filled area like in the first image above... <<

Yes that image was a poor choice because the vectors are on the same plane, and hence there is no foreshortening of vectors viewed head-on (or tail on).

Below is Karsten's existing solution for vector display. Although it works well, the arrows are not scale-independent & so an arrowhead may obscure much of the model if viewed closeup & head-on.



Your dimension arrows (as shown in your video) would be suitable except a vector is one-dimensional not tied to a plane.
Perhaps this is more difficult then first thought.

Thanks for looking into this
James

Image Attachments:
karstenVectors.jpg 


From: Mip (VINC)
27 May 2019   [#54] In reply to [#53]
Hi Michael,

To avoid possible confusion with visible lines when printing screenshots in black & white, do you think you'll be able to insert a small gap between dimension elements (extension lines, arrows) and objects?

I really like those arrows.

Michel
From: Michael Gibson
27 May 2019   [#55] In reply to [#53]
Hi James, ok I think I understand better now - you don't want a 2D type screen facing arrow, you want one with perspective foreshortening but that does not look like it is in a fixed plane.

I experimented with a couple of things for that but could not come up with a good result, the current approach is pretty focused on a planar looking type shape.

- Michael

Show messages:  1-15  16-35  36-55  56-75  76-95  96-113