Imported Mesh Models

[steina81]

I am finding the 10,000 facet limitation to be rather restricting. We can simplify the model and remove facets but the model becomes a 3D pixelation.

Is there a way we can bring the model in and then divide along a plane?  So maybe bring in a 10,000 plus model but divide it into four sections to convert to solid.

Allan Steinkuhl

[James.Youmatz · Accepted solution]

Hi @steina81,

Soon that workflow will be implemented into Fusion 360 with a mesh editing environment as well as an increase to the 10k facet limit! Check out the video here!

Since it's not here yet, have you checked out Autodesk Mesh Mixer? It's a free mesh editing software that I personally use before throwing my mesh into Fusion. I use it to clean up my mesh as well as what you were describing with sectioning off my mesh. If not, check it out here!

Thanks,

[Anonymous]

Do you know when soon will be?  The video you link to is a year old.

Dan

[steina81]

Dan,
I, like you, am anxiously awaiting.
One of the tech guys said it is coming soon.
They currently can convert 10k facets to TSpline but the stuff I am playing with, that is an extremely small piece of the total model.  The tech guy said this 10k limitation will be raised.
I am subscribing to Meshmixer.  It appears to have some benefit.  I have not learned it yet.
Allan Steinkuhl

[Anonymous]

The last expectation created was sometime this summer.
I am waiting for it as well, as every thing I do involves multiple meshes.

[Anonymous]

My workflow uses Rhino and Zbrush.  My files contain 100,000's or 1,000,000's of shells.  There is no was to get to <10,000.

Rhino is a great engineering tool and Zbrush is great at modeling.  They are both sloooowly creating functionality that crosses the two platforms but they both lack any real printing or machining capabilites. 

Anyhow, one platform would be great.

Dan

[colin.smith]

There will be a mesh preview in Fusion 360 by the end of July 2016.  The mesh workspace will have functionality based on the Mesh Mixer technology.  

This will allow you to import a mesh body and perform operations like scale, move, rotate, reduce/remesh, cut, repair, etc.  

This is the basic work flow we want to support in Fusion. 

The section in the center of the flow chart doesn't represent the all of the mesh editing. 

Not all the parts are there now but we will be adding functionality over time.

One aspect that isn't obvious from this flow chart adding the ability to handle large/dense mesh bodies and be able to convert Triangles mesh to Quad mesh and from Quad mesh to TSplines.

We are working on that also. 

Hope that helps to clarify our direction. 

Colin

[PhilProcarioJr]

@colin.smith

"One aspect that isn't obvious from this flow chart adding the ability to handle large/dense mesh bodies and be able to convert Triangles mesh to Quad mesh and from Quad mesh to TSplines."

Curious....does this mean there are plans to make high face count T-Splines responsive?

[innovationforge]

You might want to look into TSplines for Rhino. There isn't a mesh element restriction there (though there is certainly a practical limit based on system power). You may also find Rhino to be more flexible (in the short-term) to section your mesh as necessary.

Here is a project I did last year. It started life as a 3DS Max triangle model (for a visualization) and ended up as a NURBS for all kinds of fun shenanigans. My process was 3DS Max, Quadify script to retopologize, export OBJ to Rhino, use TSplines to convert to NURBS, kick back to Inventor for down range tasks. The body alone is 570k+ triangles...

[colin.smith]

@PhilProcarioJr, first I should say that we are constantly working on performance improvements.  I think you will see some improvements in TSplines performance in the middle of this year. 

My comment that you quoted was directed more towards the self-imposed face count limit in the current Mesh Quad to Tspine conversion that exists.

[steina81]

I am greatly looking forward to upgrades to meshes.

My experience using Brand X was long and tortuous.

I have a grandson's first birthday coming and I am looking forward to cleaning the rocking horse head up in Fusion.

Allan Steinkuhl

Sent from my iPhone

[steina81]

I am greatly looking forward to upgrades to meshes.

My experience using Brand X was long and tortuous.

I have a grandson's first birthday coming and I am looking forward to cleaning the rocking horse head up in Fusion.

Allan Steinkuhl

Sent from my iPhone

[steina81]

I am greatly looking forward to upgrades to meshes.

My experience using Brand X was long and tortuous.

I have a grandson's first birthday coming and I am looking forward to cleaning the rocking horse head up in Fusion.

Allan Steinkuhl

Sent from my iPhone

[Anonymous]

Really exciting!

a couple of questions: 

 - I'm importing bodies today coming from meshmixer -> memento -> Fusion. Where it says 'import Mesh', will there be integration with meshmixer or will I have to D/L to U/L? 

 - In the workflow it says 'reduce/remesh', and then 'convert triangle to quad'. In other answers you have indicated that there will be no limit or a much higher limit on the number of quad faces. Will there be a requirement to reduce/remesh before converting, or will it be simply subject to power of your machine?

 - Will the conversion be done locally or offloadable to cloud? If I have a GPU, will it speed the process?

  - in the diagram you show a path to 3d Print (additive) but not a path to CAM (subtractive). Can you clarify if there are any restrictions or requirements for using the mesh bodies to generate toolpaths for subtractive? For instance, I assume you will need the mesh to be quads and converted to T_Splines or BREPS? Can you confirm the valid paths? Also, curious why CAM is omitted but 3d Print included?

Ok, that is like 4 questions. I got really excited. 

Thank you for sharing this news. 

we are only grilling you as we cannot wait to get started using it...:-)

[PhilProcarioJr]

@colin.smith

Thanks for the info Colin, I hope major improvements are being made to performance of T-Splines because currently although you can convert 10,000 polygons to T-SPlines you can't really work with that many because performance is terrible. I really want to be excited about the tools coming out this summer unfortunately I haven't seen anything that has me excited. Only because what is coming out this summer is nothing I would use much. The core T-Spline tools are where my interest lies. Whenever you can please share with us the teams plans for the core T-Spline tools.

Thanks Colin.

Phil

[colin.smith]

Hi @PhilProcarioJr,

The conversion limit is not addressed yet but it is on our list.

In the first preview it will be mesh editing tools we brought in form Mesh mixer.

As for T-Splines, we have nothing in the short term with regards to tools, however we will be working on improving the T-Spline performance.

Hope that helps. 

Colin

[PhilProcarioJr]

@colin.smith

Thanks for the info Colin.

So it seems the real use for T-Splines at least for the foreseeable future is going to continue to be a conversion tool in regards for advanced organic forms. I really wish there was more on the board in terms of tools for T-Splines...I really feel like this will become a lot of missed opportunities for Fusion in the near Future. Other CAD apps are starting to take a real interest in melding the Sub-D modeling world and CAD. I hope you guys don't fall behind that curve.

Just my 2 cents.... 

[Anonymous]

@colin.smith, when you say 'The conversion limit is not addressed yet but it is on our list.', 

can you help us understand what that means a bit more?

practically, if you look at the workflow as it stands today

I would find a model on meshmixer I want to work with,

edit it to include just the faces I want, 

then run memento to effectively de-res it into quads

and then oull that into fusion.

IOW, I am waiting as long as possible to pull into fusuion so I can edit at hi res and eliminate as many low value facets as possible before I oull it in,

to maximize the use of the limited 10k resoluton available to me.

Is what you are saying that I will be able to do all of that inside fusion, but the 10k limit will still be in place, at least for the short to medium term? 

so this is really a tools consolidation for now, but will not afford an increase in available resolution?

[colin.smith]

Hi @Anonymous,

With the introduction of the Mesh workspace we want to get to a point where we support the following workflow:

Tri Mesh > Edit Mesh (optional) > Convert to Quad Mesh > Convert Quad Mesh to TSpline > Edit TSpline (optional) > Convert TSpline to BREP

The conversion limit I was referring to is the current limit that is imposed on converting a quad mesh to a TSpline. Fusion is limited to 10K faces on the quad mesh.

Hope that helps, 

Colin

[Anonymous]

---

@colin.smith wrote:

Hi @Anonymous,

With the introduction of the Mesh workspace we want to get to a point where we support the following workflow:

Tri Mesh > Edit Mesh (optional) > Convert to Quad Mesh > Convert Quad Mesh to TSpline > Edit TSpline (optional) > Convert TSpline to BREP

The conversion limit I was referring to is the current limit that is imposed on converting a quad mesh to a TSpline. Fusion is limited to 10K faces on the quad mesh.

Hope that helps, 

Colin

---

It does help.

I can understand why you are progressing in this way.

my feedback would be that my personal top priority would be the easing of the 10k quad face limit when cinverting to T-Splines, as it imposes severe resolution limits on the output I can generate into a CAM path.

Unless you extend the CAM path generation so I can do it directly from a mesh, which I believe is today not an option, the current development path will provide for convenience but not address my biggest concern.