Hi Michael.

Maybe you could try some sort of partnership with some company that is specialized in CAD kernel ?

I don't know. I admit my ignorance about this field.

The main point I want to say is that IMHO in 2017 I think it's very difficult to support a big project like Moi when you have to do ALL things on your own.

Especially today where the expectations of the average users of Nurbs CAD are very high.

Many features that were considered optional some years ago, by now are considered mandatory, and developing all that stuff on your own
I suppose it's too much, even for a very talented developer like you :)

What do you think about these thoughts ?

Regards.

Hi Marco, yes I would like to license another geometry kernel to help with filleting, as I wrote earlier in this thread. There are a lot of issues around that though, one big one is how picky the other kernel is in accepting geometry. Some kernels work best with geometry that was created only entirely inside their own system. But I don't want to replace everything in MoI all at once, I want to be able to translate geometry into the other kernel without it rejecting things like closed surfaces that are more than 0.01 degrees off of being smooth, do some operations on it, and then translate it back again.

There is no way to know if a particular kernel will be a good fit for these particular requirements without spending quite a bit of time testing and evaluating it and things that involve a lot of time and work are difficult to schedule. I was planning on undertaking this for v4 but the kernel that I want to try next is unfortunately not ready to go yet so it's not likely to happen for v4.

- Michael

Thanks Michael for clarification.

@You : "There is no way to know if a particular kernel will be a good fit for these particular requirements without spending quite a bit of time testing and evaluating it and things that involve a lot of time and work are difficult to schedule. I was planning on undertaking this for v4 but the kernel that I want to try next is unfortunately not ready to go yet so it's not likely to happen for v4."

I understand your point of view and I'm not in a position to talk about these technical stuff, because as I wrote some times ago, this is not my developing field :)

And I can imagine that many other things come into play when talking about "if or not" look for an external partnership.

It's up to you.

I just only wanted to say that IMHO it could be better also for you and for the whole Moi project, to not continue to work all alone
on this project.

I think that this strategy could bring you much more benefits in the near future rather than continue to develop all the things on your own.

But, this is only a very personal thought and I'm confident that you will find a good way both for you and for Moi project.

Regards.

I was also looking a bit what Spaceclaim can do since lately i have seen few more people were mentioning it and using it.
Some really crazy ability which completely blurs the lines between cad and poly modeling since there is just so many
controls in Spaceclaim which alows you direct transformations of selections as poly modeling but with all the benefits of CAD modeling.

To go back to the topic (Fillet improvements) > you can just check this from Spaceclam and see how that fillet can not only go over the edge but over the
entire hole! > https://youtu.be/95vOhN2dTp0?t=1m9s

If you have time, please to check the entire video, so many things i guess Moi3D will never get or maybe somewhere in distant future.


Which kernel are they using to be able to do such things?

Hi klaudio.

Well...I think that the only real problem here is that a single standard single license (not educational) of Spaceclaim costs about 15 times more than Moi.

I think that we don't need to say anything more :)

Bye.

Marco (mkdm)

...Just a final thought about missing features in Moi.

Well, until Moi will be provided with a new cad kernel, I think that one good solution can be :
1) install a free version of Fusion360
2) export to it any model that Moi can't fillet or shell or offset
3) bring it back to Moi.

I don't see any other way by now.

Bye.

Marco (mkdm)

Seems remember that you can also use a free Space Claim Light : DesignSpark for filleting complex object! ;)

http://moi3d.com/forum/index.php?webtag=MOI&msg=6186.1

Pilou, believe me.

I played with DesignSpark for a couple of months and I can say that it's smoke and mirrors :)

It's simply miles and miles from Spaceclaim.

Two very different products.

p.s. I think that by now, talking about free software to be coupled to Moi, nothing can beat Fusion360.

Still a young product but it's growing fast.

<< It's simply miles and miles from Spaceclaim.
Sure but sufficient for repare Moi's complex fillets :)

Fusion 360 is always in Updating (monthly) so each time 20 minutes or more for the new version up to date! :)
(in any way a very good product very versatile)

Hi klaudio,

> Which kernel are they using to be able to do such things?

They're using the ACIS kernel. If Spaceclaim does what you need, you should get it, you can use MoI in combination with Spaceclaim quite well. I think it supports copy and paste between MoI using the Rhino clipboard format although I'm not sure if that's still active in the current version.

- Michael

Hi Marco,

> I just only wanted to say that IMHO it could be better also for you and for the whole Moi project,
> to not continue to work all alone on this project.
>
> I think that this strategy could bring you much more benefits in the near future rather than
> continue to develop all the things on your own.
>
> But, this is only a very personal thought and I'm confident that you will find a good way both for
> you and for Moi project.

That strategy wouldn't fit with my own personal goal of avoiding management and administrative type duties as much as possible. :)

Maybe at some point in the future though.

- Michael

Hi Michael.

@You : "That strategy wouldn't fit with my own personal goal of avoiding management and administrative type duties as much as possible. :)"

Ok Michael. Thanks for sharing your point of view and for your frankness.

I wish you all the best in your work and I stay tuned.

Bye.

"""""""""That strategy wouldn't fit with my own personal goal of avoiding management and administrative type duties as much as possible. """"""""""""""""""

Lol..... Sometimes, you just can't pay me enough money!!!!

Life is short. Onward!

And small is beautiful! ;)

Hi Michael,

>> If Spaceclaim does what you need, you should get it, you can use MoI in combination with Spaceclaim quite well. I think it supports copy and paste between MoI using the Rhino clipboard format although I'm not sure if that's still active in the current version.


Yeah but Spaceclaim is super expensive. Even so i really like Moi3D and how everything works and looks...and i gladly paid for it even if i could choose Fusion360 for example which i didn't quite like (workflow related mostly in general).

The main point here is few very important features that Moi3D should have or should be able to do. I really don't need massive amount of features, i can do a lot with just what Moi3D has but it would be great if those features are really not limiting.
What i mean by that is that Moi3D is aimed mostly for artists and "concept modeling", fast and fun, but the thing is that you are losing quite a bit of that fun when you want, for example, fillet something (hah, that fillet again..lol) and it just can't do it in certain situation and fails as soon as things get a bit more complex. Or an single edge is getting in a way and fillet fails (where other software solutions just make it across the edge no problem) so you have to go though elaborate, tedious and often more advanced techniques to get rid of that pesky edge so that fillet can work. Easy to do if you have some simple planer surface but very often you just can't delete face and use "Planer" - done.

I understand though that it's not everything easy to do, cost money and takes a lot of time especially when you are doing all the work alone! I am doing also a bit of scripting myself and i know very well know how it is like when people request to just add that "small" little thing. :)


Cheers.

One thing it can teach you is to avoid putting that "problematic edge" there in the first place. A valuable skill in nurbs modeling.

With regard to the mamoths who can just "eat through" fillets like theres no tomorrow, that can be a "bad habit" also. So in a "manufacturing arena" the gobbled up area can be ARBITRARY and even DIFFERENT with various packages. Not a good thing when trying to design precise things.

With that said, one needs to recognize the "advanced fillet" moi fails at, and the "fillet at all costs" solutions.

Hi klaudio, well I agree with everything you said, but unfortunately filleting is a particularly difficult area to improve. Right now the plan that I have to improve it involves using a library from another company and they aren't ready yet so it's not likely to happen for v4. If they take too long then I'll have to re-evaluate that plan but right now that seems to be the best way to make major progress for the longer term.

During the v4 beta process I will do a little experimentation to see if I can do a few tune-ups to the existing filleter but it's not too likely to give a really big improvement and it's pretty easy that altering how it works to improve one particular case will have side effects that will make things worse for other cases.

So improving fillets is definitely something that I want to do, but right now I'm waiting on some other people's work.

- Michael

Hi Burrman.

@You : "With that said, one needs to recognize the "advanced fillet" moi fails at, and the "fillet at all costs" solutions."

Well said!

"What i mean by that is that Moi3D is aimed mostly for artists and "concept modeling", fast and fun, but the thing is that you are losing quite a bit of that fun when you want, for example, fillet something (hah, that fillet again..lol) and it just can't do it in certain situation and fails as soon as things get a bit more complex."