lol.. nevermind.. i figured out how to get the points along a curve at a set distance.. (array/curve)
oops

a link to scripts would still be nice though.. anyone?

Hi Jeff, I'm glad that you're liking MoI so far!

> the only sluggishness (or whatever) i've experienced so far
> is during launch but once i'm drawing, no hiccups.. (haven't
> tried anything to big yet but still.. nice 1st impression)

Yeah the start-up sluggishness is because of this anti-hacking layer that it has wrapped around it. The actual full purchased version will not have that and starts up a lot more quickly.

But that particular sluggishness is just on start up and does not impact things once it is up and running.


> i'm assuming some of the tools i'll need are available as scripts
> but the site where the scripts are (petr's moi page) seems
> to be down..

I can access it ok from over here so it seems to be back up, maybe just try again:
http://kyticka.webzdarma.cz/3d/moi/


> is there a moi script repository anywhere or maybe a
> thread here which lists available scripts?

Petr's page is the repository right now - he's got a collection of nearly all the scripts that have been posted in the forum on his site, so that's the place you will want to go to find them all collected in one spot.

- Michael

Hi Jeff,

> the inferencing(?) system is great.. i feel pretty darn
> comfortable in the 3D viewport alone..

Just watch out, once you get used to the construction line mechanism you'll start to really miss it in other apps! :)

There's a longer description of how they work here:
http://moi3d.com/2.0/docs/moi_command_reference11.htm#constructionlines

There are a lot of functions available within that one mechanism - one key difference from other inferencing systems is that a construction line in MoI is defined by 2 points - the point you start it on and then the point where you release your drag. That doesn't sound like much but it actually allows for a whole lot of functions involving distance, so things like the midpoint and reflected endpoints are available on a construction line which are not possible to get on inference systems that only radiate out rays from a single origin point instead of being connections between 2 points.

- Michael

""""""Just watch out, once you get used to the construction line mechanism you'll start to really miss it in other apps! """"""""""

Understatment of the decade? or Century!

quote m.g. -> ""Just watch out, once you get used to the construction line mechanism you'll start to really miss it in other apps! ""


ha! you're not lying.. wow
genius



a question... should the copy/paste between rhino osx and moi osx be working ?
and if not, which app (or OS?) do you suppose is breaking the link..

Hi Jeff,

> a question... should the copy/paste between rhino osx and
> moi osx be working ?

No, that part is not set up to work on OSX.


> and if not, which app (or OS?) do you suppose is breaking the link..

Right now MoI does not attempt to look for Rhino OSX clipboard data, I would need to know more details about how it labels its clipboard formats and if it is actually structured just as a regular 3DM file in memory like the Windows version is.

That's something that I can look into when Rhino OSX is maybe a bit further along.

- Michael

Hi Michael!

I was amazed to discover that you finally ported Moi to mac!!! Thanks! I have been hoping for that for a couple of years now.

Now I have a problem. I have received 2 files from a client. It is .3dm files from rhino. I can open them in moi but they appear empty, there is no objects in the browser. If I open the files in Rhino the geometry shows up fine. Even if I resave the file from rhino the file still opens empty in moi.

I don't know if this is a common problem or if it is related to the mac version of Moi.

Sorry if this is already answered in another thread. Couldn't find it.

Would really appreciate if there is any good solutions to this matter.

Kind regards

Oskar

Ok, I think i solved it. The objects in the file was nested in some kind of block instances. I had to explode them in rhino before I could save it and be able to successfully open in moi. Is this the only way to solve this?

/Oskar

Hi Oskar, yes currently MoI does not read block instances from 3DM files, so you do need to explode them in Rhino like you did in order for them to be seen in MoI.

There are also various other kinds of things that Rhino can store in 3DM files as well, like mesh objects, spotlights, annotation dots, etc... - MoI does not work directly with that stuff, it only works with regular NURBS objects from 3DM files.

In the future I want to add instance handling into MoI and then at that time I will probably get something in place to deal with block instances coming from Rhino.

- Michael

I love it (MOI)! It is very easy to use. Except for the points allocation. What ensures the surfaces do not fit (gaps).
I know now how it works in order to avoid such gaps (Always build on existing geometry).

OBJ - EXPORT
At first I thought that "Moi", divide the geometry only on export by a separate calculation, making the distribution and ignoring the points themselves.
Now i often use the function "show point" to check them ;-).
Extra: Resizing the window (bad performance) also when i open the app the window is only filled half.

Here are a few pictures

Hi blankidea, I'm glad that you are liking MoI and those are some great results!

> What ensures the surfaces do not fit (gaps).
> I know now how it works in order to avoid such gaps
> (Always build on existing geometry).

You're talking about OBJ export here too, right?

If you've created 2 surfaces that are sitting right next to one another, make sure that you use the Edit > Join command to join them together so that they share a common joined edge between them.

MoI does extra work on joined edges to make a "watertight" mesh that has the same vertex structure along the joined common edges so that there are no holes in the mesh. If you have separate surfaces that were created individually and just happen to be sitting next to one another then they get meshed separately and the vertices may be different where they touch. So make sure they are joined together to avoid that.

Note though that the meshes for each surface can have different UV Quad structures, the underlying quads for the mesh come from the UV space of each surface, see these previous posts for some more information on vertex alignment:

http://moi3d.com/forum/index.php?webtag=MOI&msg=1536.30
http://moi3d.com/forum/index.php?webtag=MOI&msg=2102.45


> Extra: Resizing the window (bad performance)

Yup, this is one of the last remaining areas I have left to work on.


> also when i open the app the window is only filled half.

I think this one may be solved by some fixes that will be coming for the next beta but I'm not entirely sure. Please let me know if you run into it again when the next beta is out.

I'm not quite sure yet when the next OSX beta will be out, sometime pretty soon but I'll be out of town for a week so if I can't get it out tomorrow it won't be until sometime the week after next.

Thanks,
- Michael

Your are great :-) looking forward for the next beta! Did you plan a early bird sale (pre sale)

Thanks for your answer and your tip! I still need to learn, but its easy to get result. In fact the motivation is higher then in other 3d programs because you don't have to think in and like a quad :-). So MoI is perfect for technical stuff... and to get the shape you want faster. So in my case I use MoI to create hardware pieces and import them into cinema 4d for further modifcations.

I have a other question: I was trying different shapes and test how fast they can be merged (Union)
If I "bool" the rips of a tire piece by piece its much faster then selecting all together to "bool" them.

Or did I make something wrong?
PS: Sorry for my english

Hi blankidea,

> looking forward for the next beta! Did you plan a early
> bird sale (pre sale)

Right now I don't have any plans for that, it's just more simple to only have it for sale when the full release is all ready.


re: technical stuff - yup, that's definitely where MoI is strongest in! For stuff like organic shapes or character models those are usually better done in a poly modeler, but mechanical type shapes where a lot of the design can be driven by 2D curves and boolean operations can be really convenient and quick to do in MoI.


> I have a other question: I was trying different shapes and
> test how fast they can be merged (Union)
> If I "bool" the rips of a tire piece by piece its much faster
> then selecting all together to "bool" them.

This can sometimes happen with boolean union depending on the order of how all the objects actually end up getting processed. If it ends up trying to boolean together many of the pieces that are not actually touching each other that basically makes for a more complex multi-volume intermediate result which is more complex.

In order to try and counteract that boolean union tries to sort the objects and process the biggest one first, but I think in your case here that is getting confused because the underlying surfaces on your tread pieces is fairly large.

The "underlying surface" is the full surface that is underneath all the trim curves, trim curves mark areas of surfaces as cut away regions or holes. If your original surface that you trimmed or booleaned was fairly large then those underlying surfaces can be large too and that may be throwing off the size sorting mechanism.

You can see this "large underlying surface" if you take one of your tread objects and then use Edit > Separate on it to break it into individual surfaces and then turn on surface control points with Edit > Show pts.

You can shrink an object's underlying surfaces down to be right to the size of its trim boundaries by using the ShrinkTrimmedSrf command (either hit tab and then type in ShrinkTrimmedSrf and then press Enter or set it on a shortcut key).

So for your case here try running ShrinkTrimmedSrf on the tread objects - does that then perform better than before?

- Michael

> Right now I don't have any plans for that, it's just more simple to only have it for sale when the full release is all ready.

Too bad really ;-( but I understand you

> ShrinkTrimmedSrf
Thanks again for your tip. Works great a few second and the boolean operation was done. Even if I don't know what the operation does...
because I saw no difference on the object itself (show points was on), but again it works great. I bind the command now to a hot key.

Sorry to bother you! I have to read the manual

Hi blankidea,

> Thanks again for your tip. Works great a few second and
> the boolean operation was done. Even if I don't know
> what the operation does...

Check out here for some more general information on how "underlying surfaces" work with NURBS objects:

http://moi3d.com/faq#Q:_Why_does_show_points_work_for_some_objects_but_not_others.3F


But it might be better if boolean union used some different method to get the bounding box rather than one that included the full underlying surface size - the reason why it uses that one right now is because it's very quick to get it.

- Michael

Hi Michael Gibson,
By the way I forgot to tell you that CMD+V does not work on the input fields.
I use the script to get the "length of a curve" and try to paste the value into the input field.
Only the right mouse button work, but you also loose the size because you change it when you come back with the mouse
into the drawing viewport.

Also I install a script called (redefine Curve) which I have to install directly into the application (by right click on the app icon "show content").
Normally you use the Library folder for the user but this one is not working.

Thanks for your overwhelming support :-)

Hi blankidea,

> By the way I forgot to tell you that CMD+V does not work on
> the input fields.

For the moment use the Ctrl key instead with Ctrl+V and it should work, but that is something that I will fix up for the next beta to make Cmd+V work for text pasting like you were expecting.


> Also I install a script called (redefine Curve) which I have to install
> directly into the application (by right click on the app icon "show content").
> Normally you use the Library folder for the user but this one is not working.

This part probably won't change for the 2.5 release - it's the same with the Windows version where you install scripts into the application's commands sub-folder and not into a centralized location like the Library folder. I do want to improve this in the future to work like that but I do not expect for that to make it into the 2.5 release though.

So to install a plug-in command into OSX MoI now you will need to right-click on the app icon and then pick "Show package contents", then go to drive_c/moi/commands - that's the place where you will need to copy plug-in command to in order for MoI to recognize them.

In the future I want to overhaul the whole plug-in installation process so that there will be some kind of UI that handles it for you and you won't need to drill into specific folders but that's how it works currently.

Thanks for the additional bug reports!

- Michael

Hi Michael,

first of all a BIG THANK YOU for providing this OSX-Beta! Unfortunately I was kept very busy the last 2 months preventing me to test drive and give comments earlier... but finally...
It runs quite well even on my old MacBook with the problematic GMA X3100 graphics chip. A bit slow for more complex projects, but this is really more than I thought with this "old 2D-card".
Apart from the already mentioned slow resizing and incomplete filling on initial application's opening (see screenshot) I only found one additional thing: the icons for the new deform tools don't show up. But taking a closer look I guess you just haven't cooked them up yet... :)

I really enjoy playing with MOI and look forward to running the final release on some newer Mac hardware....

Again thanks for this great program!
Chyman

Hi Chyman, thanks for testing it!


> It runs quite well even on my old MacBook with the
> problematic GMA X3100 graphics chip.

That's great news! But yes this card just does not fundamentally have the same amount of GPU horsepower as nVidia or ATI cards, so you may only be able to get to a medium level of complexity with it. But you should actually be able to get quite a lot of projects done with it, especially if you only deal with different parts of the project at a time instead of everything all in one single big model.

Also if things start to bog down on a more detailed model, try adjusting the "Mesh angle" settings under Options > View as described here:
http://moi3d.com/forum/index.php?webtag=MOI&msg=4643.2

Setting the "Mesh angle" to a rougher angle like 25 degrees and unchecking "Add detail to inflections" will make for a somewhat rougher looking display but also lighten up the density of the display meshes and that can help boost display performance by quite a bit.


> Apart from the already mentioned slow resizing

Yeah this is one of the last areas that I have remaining to work on before releasing a new beta.


> and incomplete filling on initial application's opening (see screenshot)

This one I hope will be solved by some updates and bug fixes to the Wineskin system that will also be included in the next beta, but I'm not entirely sure. Please let me know if you see it again with the next beta when it is out.


> I only found one additional thing: the icons for the new deform
> tools don't show up.

Yup, that's normal for the time being - those tools are new for v3 and still need a bit of attention to be finished up. Finishing that area will probably be the last step before rolling out a final 2.5 OSX version, it won't be quite finished still for the next beta.

But I'm glad that MoI OSX is able to run on those Intel graphics cards now, I wasn't sure at first if that was going to be possible. But also someone else has tested it on the oldest GMA 950 and it works there too now, so MoI OSX should work on all Intel Macs I think, including the very first ones.

Thanks,
- Michael

Thank you for your comments and the "Mesh angle" hint (that is a good performance parameter).
Indeed your OSX support turned out quite complete after all, well done!

Hope I can come up with some nice output soon...
Chyman