Full Version: Feature request "instance" in moi

Hello

Maybe instance would be a great feature for Moi.

Instance for those who don't konw is when you create a base objet and add instance in the scene, if base objet is modified instance are automatically modified.

It's just an idea for V3 feature but I suppose you already have plenty of ideas for the future.

PP

---

Hi PP, yes instances are something that I do want to add in the future. I'm not sure yet when it will happen though.

There are a couple of different ways that other programs handle instances.

One way that is particularly interesting is SketchUp's method which it calls "components". The thing that is interesting about it is that you can have an instance which is not just a single object all by itself but is actually a piece that is cut or merged with another object like a window cut into a building.

So one thing I've thought some about is whether instances in MoI should have that kind of system where they can be a sub-part of a large object instead of only a copied full object.

However, with MoI it may be pretty difficult to set that up, in SketchUp objects are only made up of lines and planes and that helps makes it kind of easier to have that set up there.

But there is still a lot of stuff to try and think about in this area, it will probably take a fair chunk of planning time before working on this area, that's why I'm not really sure about the schedule for it.

- Michael

---

I'd have it set up like a grouping system...

So you might have a house group.
And in that might eb a bunch of stuff, including windows - which are a group themselves (of the window frame and glass, etc.)

So if you change one window, they'd all change.

And if you changed one house, they'd all change.

Basically have a library with groups in it.

---

Hi NightCabbage, actually your example of a bunch of houses kind of brings up another issue, which is how much does an instance system really belong in a modeling-focused application.

When you mentioned having a bunch of houses and changing one would change them all, I guess I would wonder - would you really be using a modeling-focused system like MoI to arrange a whole bunch of houses like that? Making a kind of big scene with a lot of repeated stuff in it is kind of something that goes along more with a rendering or animation type program instead of MoI.

One thing that I do worry about is trying to jam too many sort of semi-unrelated functions all into one single interface, it is hard to do that and also maintain simplicity at the same time.

- Michael

---

This is true :)

However I guess if you had the ability to have instances of windows in a house, then you'd also have the ability to have instances of houses in a street :)

---

...and streets in the town
...and twons in the country
...and contries in the world
...and worlds in the sky
...and skies in the univers
...and univers in... ;)

---

Yes instances are really important for me and im sure most people. I think mention of streets of houses has thrown things a bit, but an individual building is possible with moi and in that there will always be a need for instances. I don't think they even need to be too 'intellegent' just a co-ordinate transfer system where moi just remembers the co-ordinates and rotation of the clones and just copies and replaces the previous version, obviously this would work on a group and I would think require an 'update instances' button so moi isn't trying to calculate everything in real time. Thats just how i see them working - of course there is the question of how they will respond as a boolean opperator etc. so history will play a big part too! Hope they are high on the priority list!

Cheers
Rob

---

On the contrary.

Instances are not "sooo" important for all user with only one user name here. ;)

Moi3D/ Michael should not lose the focus and become an unwieldy CAD system with hundrets of features,
because many of them only good for marketing/advertisement.

Currently, Moi is the fastest and "simplest to-learn" 3D Mod, unlike other heavy 3D CAD systems...
... and I know what I'm talking about.

---

Instances are a very important feature to have at some point.
I also want to see moi stay streamlined and straight forward, but there are many parts of the modeling process that would benefit from having instances.

---

Yes, for example a simple nut repeated many times could slow the performance...
Maybe instances could avoid this.

Marc

---

"Instances are not "sooo" important for all user with only one user name here. ;)"

Im not sure i understand what you mean by the 'one user name' Ralf?

Who said that having instances would make moi an unweildy CAD system with 100s of features? Its a staple industry time (and possibly RAM saving, depending on implimentation) concept nothing too complicated for the user to get their head around which i think is the point your trying to make about it being 'unwieldy'. If Moi is to increase its popularity and therefore sales for Michael then I would think that instances would be a feature many (but not saying all) would look for if they are to fully embrace it as a main modelling app. The nuts example is a great one Marc. Moi is indeed a great modeller and very simple and fast but that doesn't mean expansion and development should be stifled because there are concerns there are too many features. Michael is great at optimising usability so im confident if instances are included then he will do a great user friendly job.

Cheers

---

+1 on instance feature.

Leonard

---

The number one missing feature, according to me. +1

---

FWIW, I requested these features as well, back when V4 was a gleam in Michael's eye :-)

HERE

and HERE

In fact, I really wanted to use V4 with a non-destructive workflow for quick iterative designs.
As an Industrial Designer who more and more wanted to ideate within MoI, it just became too hard. For the most part, MoI is a program that can easily document a finished concept design-- and I've used it for concept work as well, but needed SketchUp or other modeler to be able to work out the proportions, filleting, and other.

And without a basic AO shader, it's really damn hard to evaluate the form. The current renderer just does not reflect light how you would like it to. So, I'll keep using MoI for CAD type work where I need to document a design in NURBS, but have moved to Blender where I can continue working out designs using non-destructive, realtime rendered workflows.

Personally, I understand the greatness of having a passionate developer who wants to focus on ease-of-use for new users. The problem is-- users can grow out of the limited functionality a single programmer is able to facilitate-- and for Michael that includes running a business, managing a website, having the best support in the business, and I'm sure a host of other jobs. Keep in mind Michael's goal is not to compete directly with other CAD packages-- not even Rhino. His sole differentiator is ease-of-use and workflow speed.

The CON side of the equation, is feature requests take years to evaluate and implement, and the product continues to fall farther behind from a capabilities standpoint. A public roadmap showing rough timescales and feature milestones *might* go a long way. Still, Michael's a perfectionist. The cake is done when he says it's done-- so roadmaps probably don't mean much other than set and dash expectations. Sorta like Sienfeld's Soup Kitchen, though with a much kindler, gentler person in charge! :-)

---

Hi Chipp, yep that's a pretty good overview!

re:
> The problem is-- users can grow out of the limited functionality a single programmer is able to facilitate--
> <....>

I guess from my point of view I wouldn't really classify this as a "problem" exactly. If MoI helps someone get started in CAD but later on down the road as they need more specific things they need to transition to other software, how is that a problem? That person has now successfully broken through the wall and is now able to get more of their work done with computer aided tools than they were before.

If MoI helped them get to where they need, that seems like a pretty useful thing to me, even if they eventually don't need MoI anymore.

Also there are plenty of people in that situation that like how MoI works for their basic modeling and drawing and so they will continue to use MoI and many even still spend the majority of their time in MoI.

MoI is so inexpensive that it's very feasible to use it along with other software in this way.

Of course I do expect to continue to add stuff into MoI to expand its capabilities and reduce the need to use additional software as time goes on.


> The CON side of the equation, is feature requests take years to evaluate and implement,
> and the product continues to fall farther behind from a capabilities standpoint. A public
> roadmap showing rough timescales and feature milestones *might* go a long way.

A public roadmap is pretty unfeasible for me just from how I work. When ideas come up that enhance basic functions I tend to give those top priority and push other things further off. A good example of this is the multi-core enabled file reading in v4.

I like to have some general target ideas that I guess you could call a roadmap of sorts but if I were to publish it then it would get treated like a guarantee rather than suggestions for future directions.

There has definitely been a slowdown in new feature areas for a while now - that's because the v4 release was a major rewrite to get to 64-bit and also a Mac native build. That turned into about a 4 year effort to make those things happen but it's a much better foundation. Once v4 is wrapped up I should be able to return to a more normal feature development pace.

- Michael

---

Hello,

In my experience MoI is a great primary tool that can be augmented by other specialty applications and plugins. I've found MoI to be a valuable learning tool as well as a modeling tool. Because the features in MoI are very accessible I've been able to view how others model in different applications and think about strategies to approach a similar model in MoI. When I determine an operation is too time consuming or unavailable in MoI I switch into a package that will allow me to complete a step with a more optimal result in less time. I haven't found another application like MoI that allows me to so easily mimic an effective workflow done in another design program, for me that' a really big deal.

---

MoI being an accessible entry to CAD is just a side-effect of an overall non intrusive user experience that fosters on the fly concepting and flow, in my opinion. That's a much greater value than training wheels, at least for illustrators and concept artists that use 3D to get imagery. MoI's fluidity allows for a state very close to the one I get during painting, but that is just a very biased opinion of someone that started CAD with MoI and is not looking to build real things.

---

MoI - Moment of Inspiration - only these three words perfect characterise target use of this amazing piece of software. In my opinion MoI doesn't aspire to be a competitor of Solid Works, although MoI offers more than enough utilities for designers, digital artists or even for Startup companies that needs CAD software for their projects.

MoI allows quick concept design in right time - in fleeting "moment of inspiration" and this is it. IMHO MoI offers more than only quick concept design, MoI is accurate enough as well, so that project to be switched from idea into real product. It would be nice of course to have in MoI grouping, instancing, modules for parts assembling, metal sheet bending, technical documentation, work flow history with possibility to change parts, but then we talk about Solid Works :-)

Ciao
Mik

---

Seems you have all that in Fusion 360 but maybe without the pleasure to do ... ;)

---