AU Class
AU Class
class - AU

Complex Topology and Class-A Surface Modeling with Fusion 360

Partager ce cours
Rechercher des mots-clés dans les vidéos, les diapositives des présentations et les supports de cours :

Description

This class is for Fusion 360 users who (just occasionally!) find themselves ready to create a complex model and don't know where to start. We'll cover the basics concepts of surface and curvature continuity, and look at some examples of difficult topology and how to avoid it. And then we'll take a deep dive into the toolset for surface modeling in Fusion 360. Finally, we'll run through a complete step-by-step example of building a model from surfaces with Fusion 360, from concept to solid body. This class will benefit students who find themselves modeling challenging topology with Fusion 360, and who want to go from "hack and whack" to planned and perfect.

Principaux enseignements

  • Learn how to use the language of curvature continuity with confidence
  • Discover Fusion 360 software’s hidden surfacing tools
  • Learn how to build complex topology from individual surface patches
  • Learn how to convert surfaces into a solid model

Intervenant

  • Avatar de Paul Munford
    Paul Munford
    Paul Munford is a CAD geek, Customer Adoption Specialist for Informed Design and Autodesk Expert Elite Alumni. Based in the UK, Paul's background in manufacturing items for the construction industry gives him a foot in digital prototyping and a foot in Building Information Modeling (BIM). Paul was a speaker at Autodesk University for the first time in 2012, and he says it's the most fun anyone can have with 250 other people in the room.
Video Player is loading.
Current Time 0:00
Duration 1:04:14
Loaded: 0.26%
Stream Type LIVE
Remaining Time 1:04:14
 
1x
  • Chapters
  • descriptions off, selected
  • en (Main), selected
Transcript

PAUL MUNFORD: So thank you very much for attending this class, Complex Topology and Class-A Surface Modeling with Fusion 360. So how many people are using Fusion 360 right now? And that's pretty good, most people. How many people would say you're from some industrial design or product manufacturing background? Do you use Fusion with another product right now? Are you kind of transitioning?

AUDIENCE: NX.

PAUL MUNFORD: NX. Cool. OK. So just to remind us of why we're here. So this is what I'm going to deliver to you, just to make sure everybody signed up for the right class, right? So we're going to learn how to create curvature continuous organic 3D models with Fusion 360. But this is not a T splines class. This is traditional surfacing, so this is how you might apply those traditional surfacing techniques inside Fusion.

So in this class, I'd like to better use the language of curvature constituted with confidence, use some of Fusion 360's slightly hidden away surfacing tools, understand how to build complex topology from individual surface patches, and understand how to convert surfaces into a solid model. So that's our objectives for the next hour and a half. So just to introduced myself for those people who don't know mean.

My name is Paul Munford. I'm an application engineer with a company called Graitec. We're an Autodesk reseller based in the UK. So my job is to help our customers get the best out of their investment in alternate software. Did anybody come to my class earlier on this week, the Inventor's Seven Deadly Sins of Part modeling?

AUDIENCE: No, [INAUDIBLE].

PAUL MUNFORD: If you want to take a look back on that one, I went through how the Autodesk ShapeManager kernel affects the way your models are built, and in fact, it's the same shape manager kernel that's based inside Fusion 360, so the same rules apply. So if you've ever had trouble creating Fusion models, they kind of blow up or explode or you get some which surfaces not quite what you expected, that class might also be useful. And my colleague Inderjeet did a lab at AU London this year based on the same information for Fusion in Inventor, showing some common problems you might have and how to fix them. So I thought I'd throw those in there as a couple of extra resources.

There is a presentation for this class, you can download the presentation from the AU site. I thought I'd also mention this Pinterest account I keep. So these examples I've used today, I've modeled up from things I found interesting on Pinterest. So if you're looking for some practice models, that's a good place to look. I picked some models today that are fairly generic. They're not of something. I just thought they'd be good practices, and I've exported the datasets out of Fusion, so there's Fusion export set also on the Autodesk University website.

There's a zip folder with three files in. You can download those. Upload them to Fusion. You can take a look at how the models were built and see if that helps you out. So I'm going to begin by saying what is surface modeling. Here we go. Come on in, take a seat.

What is surface modeling? Well, surface modeling is a technique for creating more complex designs surface by surface, and it's commonly used in product design, industrial design, and well, in automotive and aerospace as well and those sort of things. Does anybody use Alias? Right, so if you use Alias, that is really class A surfacing. OK, what we're going to be doing in Fusion-- Fusion is not-- I'm not going to suggest that Fusion competes with Alias, right?

But what I will say is there are lots of people who just-- most of what they do is fairly straightforward engineering stuff and just occasionally have to model up something like this. And you're kind of, well, where do I start? What do I do? So this class is really for those people who want to advance their modeling up to better produce these sort of surface models, and you've only got access to Fusion. You either can't afford Alias, or you just don't want to spend time learning it. You don't spend enough time in the product.

So this is our objective, better work like this in Fusion. We'll start here. So this is a spline. So a spline is geometry with variable radius, variable curvature. So what's useful about splines is that we can use these splines as the basis for our freeform, organic, smooth geometry. And traditionally, when we create designs in Fusion, we're looking at prismatic shapes, so shapes based on lines and arcs and circles.

But we're going to be talking about this fellow on the end here. We call this a NURBS surface. NURBS is nonuniform rational b-splines, but you don't really need to know that. It's a surface based on spline curves. Surface to surface has variable curvature. And we can use this idea of variable curvature to create smooth interactions between our surface patches.

We'll just let these guys come in. Come in, guys, take a seat. There's a couple more seats down here.

What's useful about the standard prismatic tools is they're kind of doing a shortcut for us. These tools also create individual surfaces, but they automatically knit together for us to create solid models. When we're doing surface modeling, we're generally building one surface at a time, but we have to knit them together ourselves, or make it into a solid model by the time we're finished. So let's just talk about splines for a little bit because they are at the heart of everything we do in Fusion modeling.

So with splines, we have a number of these tangency handles, and we can use tangency handles to affect the curvature of the spline along its length. So a spline will always give us the smoothest curve between the fit points-- fit points are the two endpoints, and I've just got one fit point in the middle here. But at every point, we can adjust the curvature by adjusting the tangency handle. This is known as a Bezier curve-- this one a Bezier spline.

So the tangency handles, I think of them more as like a region. The curve will try and be tangent to this handle, and the longer we make the handle, the more effect the handle has over the curve, the more the curve will try and be tangent to that handle. So we can adjust the length, and you'll find if you put a dimension on one of these tangency handles, you can put a dimension in of one, and that's kind of neutral. That's just where it starts. That's it's basic default length. If we put a size in of two, we're saying this tendency handle is going to affect twice as much of the spline. If we put in a size 0.5 is going to affect half as much.

So the portion of the spline that's around that tangency handle is going to try and be tangent to that handle. Now we also have on Bezier curves, curvature handles. So we can set a specific radius to each one of these fit points. In addition to setting a specific radius, we can again adjust the length of the curvature handle, and the more we adjust the curvature length, the more the spline will try and remain curvature to that fit point. So we can use both of these to be very precise about the amount of curvature, amount of tangency, we have at each fit point.

Now, I'm going to be demonstrating quite a lot of how we can visualize curvature by using the Curvature Display tool, the curvature combs? Do you use curvature combs in Fusion? Anybody use curvature combs in other products? So you can find the curvature comb if you right click on any entity. You don't have to be in a sketch to do this, just right click on any sketch entity. If it has curvature, you better turn on Curvature Display. If you're inside a sketch, there's an option here called Curvature Comb, you just select the geometry, click on this little icon here, and you'll see the Curvature Comb's popup.

So we can use the curvature combs to read the curvature along the spline, and it gives us some information about the spline that we can use. So for anybody who has not used curvature combs before, these sort of turquoise lines are showing the direction of the curvature. So these turquoise splines, spines rather, not splines, spines will always be perpendicular to the curve. Yep, so they're going to be perpendicular, so 90 degrees to the curvature. And the length of the spine is indicating the amount of curvature.

So if I was to put that in a [INAUDIBLE] saying, the more curvature we have, the longer the spines going to be. And the red line at the top there, shows the changing curvature along the length of the spine. So we call this changing curvature acceleration. So we can use curvature combs quite extensively to look at the smoothness of our splines and make sure we get nice smooth inputs. Good smooth inputs will give us good smooth surfaces.

So here's just a little video of a spline being created. It's always a good idea to create splines with as few fit points as possible, so I've just gone with two here. And you can see by adjusting the tangency handles, we can adjust the curvature of the line, of the spline. And you'll notice that anytime we adjust the tangency handle, we're affecting the whole spline. The whole curve is going to change and adjust. But you can add fit points in, so if you need additional fit points, right-click and add them.

So I always suggest you start with the minimum. If you don't get the curve you want, then add additional fit points in. And every time we add a fit point, we can adjust the position of that fit point, and we can adjust its tangency handles and curvature handles. So one key difference between prismatic modeling and surface modeling is we rarely fully constrain sketches in surface modeling, because they're just too many things to constrain.

With every fit point we have, we need to constrain its position, the direction of the tangency handle, the length, and if we start adding curvature handles in there, again, it will just get crazy. So in any of my engineering classes, I say always fully constrain your sketches, always fully constrain the sketches, right? So if you come from an engineering background, this is going to seem very unfamiliar, but I'm now going to say we're designers now. Just put it in, it'll be OK. We'll be OK.

You can delete fit points, and it's not going to affect the splines. So you just select them and delete them. And we can tell on Curvature Display, and as we turn on Curvature Display, we can set the comb density and scale, but as we adjust this, we'll see the curvature of the spline adjusts, so we can get some nice, smooth input curves.

So a few tips for dealing with splines. In these kind of Bezier curves, the more points we add, the more constraints we're adding, and the more difficult it is actually to get a really good smooth transition between points. So I'd always say have the minimum number of fit points you can. Don't out extra fit points in if you don't need them.

If you do add fit points in, make sure you've got a good flow, And I'll explain what I mean about flow in a moment. If we have bad splines, we can diagnose this using those curvature combs. So if we have too many fit points here, when we turn the curvature comb on, we can see we've got these inflections. Inflections are a change of direction.

And that's OK if that's we intend, but if that's not what we intend, this is going to give us some really weird surfaces. So this is not a good curve to start surfacing with. Also, maybe we've got fit points too close together, and if we look at the curvature combs there, we get some really weird stuff going on, right? So we can say probably somehow I've managed to get two fit points almost in the same space. So again, this is a bad spline to be trying to create surfaces from. It's going to give you problems.

So we're looking for something like this, nice clean splines, minimum number of fit points, really smooth acceleration from one end to the other. That's a good clean input for a surface. If we do need to have multiple fit points, we're still looking for a nice clean acceleration, a nice clean changing curvature over time. So what I mean by flow is that if I was to take a dimension between each of these fit points and plot them out in the graph, you can see we've got a nice smooth change in direction on that graph.

At the bottom here, I've just placed them as randomly as I can, put some dimensions in. You can see the graph's going to come up and down. These can be really hard to get a smooth transition here. It may look OK, but when I turn on the curvature combs, you can see the top one is definitely got a nice clean change in acceleration. The bottom one is going all over the place. So again, this will give us problems downstream if we try and build from this.

So just going back to fully constraining sketches, one tip I have is I don't fully constrain spline curves, but what I do do is I tend to build some construction geometry and kind of pin my spline to the construction geometry. And that way I can still have a parametric change in size, and my spline is not going to go crazy. So it gives me some degree of control, but I'd have to spend a long time getting everything locked down.

One thing specifically about Bezier curves is that Bezier curves will give us a completely closed loop. So that's one thing we can do with Bezier curves that we can't do with other forms of spline. And again, I tend to attach these to what I call a control frame of construction geometry, and it just means I can change the size of warp that shape from like a circular shape to more of an elliptical shape with a bit of control. Things aren't going to go too wacky or too weird.

Now, so we talked about individual curves, now let's start talking about the relationship between curves. So when we talk about curvature continuity, we're talking about relationship between curves, and this is measured using a value called G. So these two curves share no conditions whatsoever, so they've got no G. Not G0, no G. There is no G. So if we said that G is matching properties, we don't match any properties across these, so we have no continuity.

I think of that kind of like this. We may have beautiful curves, but they have no relationship, OK? This one is called G0, and now we're matching one property. The property you're matching is called connected. The terminology changes between different CAD products, right? So I'll try to use the terminology used inside Fusion, but you might have heard this mentioned as position, that kind of thing.

The two entities are sharing a point in space, that's what they're sharing, a location. And I think of that like this. So it's usually used for sharp changes in direction. From the original Tron, because I'm that old. And the characteristic for this when we look at the curvature combs is we can see a distinctive sort of V shape on the curvature comb showing that there is-- the only relationship we're sharing is a point. We're not sharing any other relationships between these two curves.

So a G1 condition, we're going to match two properties. We're matching the position of the endpoint of these two splines, and we're also matching tangency at the point where these two curves meet. So for me that one always reminds me of Michael Keaton Batman when he throws a grappling hook out the side of the car and spins around a lamppost at like 90 degrees. When we talk about G1, a G1 curve would usually be something like a fillet or radius with like a constant curvature.

But if we looked to the curvature combs here, we can see the distinctive characteristic here, is the combs, those spines of the curvature combs, are pointing in the same direction. So we know these curvature comb spines always point perpendicular to the curve, so they're pointing the same direction, we must have tangency.

There's a few more seats down here if you want to come sit down.

So let's move on, next one is G2. So G2 we're matching an extra property, we're matching connected, so they're sharing a location, we're matching tangency, so they're tangent at the point that they meet, and we're also matching curvature. So at the point where they meet, they share the same curvature. So my example here-- I wonder how many people recognize this. This is from Battle of the Planets. This is Jason's car.

When I did the research for this, I found out this car is called the G2, so that was pretty cool. You have to be quite old now to remember Battle of the Planets.

But the distinctive characteristic here is that when we turn on our curvature combs to look at these two curves, we can see that the curvature comb splines, they're of the same length. So the length of the curvature comb spine is indicating how much curvature we have at that point, and we'll make a G2 condition, we'll see they're the same length, so we're confident they have the same curvature.

Now the final one that's often talked about, particularly with those who have used Alias G3, now there is no way to create a-- specifically, there is no tool to create G3 in Fusion. You can create something that looks pretty like G3 with a bit of patience, but there's no tool to do it for you.

G3, we're matching one more property. So we're matching our two entities share the same location in space, their tangent at the point where they meet, they share the same curvature at the point where they meet, and they share the same acceleration. So acceleration, you remember, was the change in curvature over the length of the curve, so we've got equal acceleration coming in through both curves. So that example is this one. Beautiful and smooth, smooth as you can get.

And the distinctive characteristic here is looking at the red line there at the top of the curvature comb graph, that is also tangent. So we don't have any specific tool in Fusion to create G3, but if we play with our spines, and we get a nice change in acceleration across both, then we're pretty confident we're as close to G3 as we're going to get in Fusion. So with a bit of massage, we can get something that looks pretty good.

So I've demonstrated these in two dimensions, but the same idea works in three dimensions. So when we're creating blends in Fusion, we'll be looking for a G1, G2. So G1 and G2, we can create with a standard Fillet command. We can create a fillet that will be G1. In the Fillet command, we've got a G2 option, so we can create a G2 fillet. If we want something more G3, then we're going to have to hand bulit that using a loft to create a blend, and if we set G2 conditions on our inputs, and the inputs are built on nice clean spines with that kind of distinctive curvature comb, then we'll get as close as we can to G3.

So I just thought I'd put this up here. I don't know if you noticed in the recent update in Fusion, there was some discontinuity in the way they described these using different tools. So they've been through [INAUDIBLE] they describe it in the same way. So when we look at things like lofts and patches, we'll see if we get connected tangent curvature. That's the language they're using inside Fusion, so I tried to make my presentation the same. If I slip into using other words for these, and I apologize, but I probably will, you'll get what I mean.

So the question is which is best? It kind of implies that G3 is best, and that we work up the hierarchy. We should be going for G3 all the time, but I kind of wanted to dispel that. It's not a question of which is best, it's a question of which is the most appropriate, which is an aesthetic choice, right? So when we're designing something, we're thinking about do I want to have like a crease feature down here? Do I want to have a smooth blend? Do I need G3 or is G2 going to be enough?

So when we're thinking about our designs, we may be using all of these conditions as appropriate in our design. So something like a button here has got a nice sharp edge on it, so that's going to be kind of G0. G1, so we might have just softened that sharp edge off what a little filet on top, just so it's not too sharp like it's not sharp on your fingers.

There's probably going to be some G2 blends in here as well, and some of these bigger patches might be made of more than one surface, and so we're looking for a G3 relationship between those. So to all intents and purposes, they look like one curvature continuous surface, but we may have built them from separate patches. So if we're thinking about what degree of curvature continuity is appropriate, that leads us to think about our inputs.

So down the bottom there, in the bottom right, I've shown a G1 condition, like a standard filet. So this filet has the same curvature the whole way around. It's just an arc. And the input surfaces there are straight lines. So straight lines have-- depending on how pedantic you are, either no curvature or infinite curvature depending on how you think about it. But there will be no point I was trying to create a G2 or G3 relationship with those inputs. Because we can't match-- we can match tangency, as we're doing with the G1 relationship, a G2 relationship is matching curvature, and these inputs have no curvature, so we can't create the G2 relationship there.

G-3 is the change in acceleration, and there's no changing curvature, so we can't create the G3 relationship. So if we want to have G3, we need to make sure we build the correct appropriate inputs in order to get the correct appropriate output. So the next example there, G2, I've got curved inputs. These again just arcs. They have the same radius the whole way along. But with curved inputs, I can create a 2G relationship. And at the top here, G3, my inputs spline curves, so they can change in curvature.

And so because they've got a change in curvature, I can create a G3 relationship. Yes, there's a bit of planning involved to which level of continuity is appropriate and make sure inputs are appropriate for that level. So that some of the theory. That's some of the theory, so I hope you find that useful, particularly of those who've not come from like an Alias background. And I think the really useful thing about that for me is if you teach the theory, then people can pick up any product, and say, well, that's what I want to achieve. how do I do it? And start messing around with the tools.

But let's start looking at some of the tools specifically. So we're going to be looking at the Fusion Patch environment. So the Patch environment's where we'll find our surfacing tools, and specifically, the main tools we're going to look at next will be Sweeps, Lofts, and Patches. Now when I was rehearsing this presentation, I realized that I was using the word patch a lot but to mean different things, so I just thought I'd just clarify.

So patch is Fusion's Surface modeling environment. So it's an environment within Fusion which we collect our tools together for surface modeling. Patch is also a tool which we have in Fusion to create surfaces. So sometimes I'm talking about patch on creating a surface, and also a patch is just a generic modeling term that we use for a surface. So when I'm saying patch, I hope it's appropriate from the context as to which one I mean, but if you're not sure, just put your hand up and ask me to clarify.

OK, so who's used a sweep? I know you've all used sweeps, and I'll tell you why in a moment. So the thing that's characteristic of a sweep is a sweep sweeps a profile along a path. So the one rule we have for sweeps is that this profile must stay the same the whole length of the path. It is a consistent profile. Now what that means is the profile can be rotated as it travels along the path, and it can be scaled as it travels on the path, but it can't be warped, like we can with a loft.

So a loft allows us to create curvature continuity. We can't create curvature continuity with a sweep. So if we're thinking about in which order we should use the tools, sweeps generally come first. Like if we can do a sweep, we do that first because then we can come in with a loft and create a curvature continuous surface from the sweeps to one another.

Now just a little aside here. Let me go back to here. So this is why I know you've all used a sweep, because in fact, all of these are sweeps. This is an extrude, that's a sweep. This is a revolve, that's a sweep. This is a coil, that's a sweep. Under the hood, the modeling kernel that generates these shapes has exactly the same algorithm for all of these. It's only the user interface which is different, which asks us for different inputs. So for an extrude, we're sweeping a profile along a path, but the path is implied as being perpendicular to our sketch plane.

If we're building a revolve, we're still sweeping a profile along a path, but the path is implied as being a circle round our axis. And the inputs for a coil are more complicated, but it's the same idea, we're sleeping a profile along a path every time. So these are all sweeps. And the one characteristic they share is consistent profile. That's the one rule we can't break with sweeps.

So I wanted to say here, who's used sweep with guide rail? Just a couple of you nodding. For those of you who haven't, I suggest you give it a go. What sweep with guard rail does, is we have a profile and we have a path, but the guide rail will warp, not warp, but stretch, scale, the profile as it goes along the path. And in this case, we can turn scaling on, so I've got profile scaling set to sketch.

So instead of-- stretch, sorry. Instead of scaling in all directions, it's only scaling one direction. But what you can see is the result is kind of similar to what you'd expect from a loft, but the inputs are a little bit simpler. So you can use this as an alternative instead of having to set up lots of offset guidelines and rails and everything else for lofts. You might find that sweep with guard rail is a simpler option for you. So I just thought I'd throw that in there as a tip, and encourage you to try that one out.

OK, so this is a loft. So characteristically, a loft will warp a shape, it'll create a surface model between two different profiles. So we can extrapolate between different profiles. So to give an indication of what I mean by that, at the back here, we've got a square profile, at the front here, we've got a circular profile. And if we take sections through this loft, we can see we're going kind of squarcle at the front to mainly square at the back, and somewhere in the middle we've got an exact translation between those two profiles.

So one question I've been asked before is can I sweep between two profiles? So we now know the answer is no, because a sweep has to have a consistent profile. But we could loft using the centerline loft option. So who's had a go with loft with centerline? Just a few. So those of you who haven't used it, it's actually kind of a combined algorithm. When we use a centerline loft, the centerline is behaving like a sweep path.

So our profiles are being swept along the centerline. They're remaining perpendicular to the centerline, but at the same time, they're also warping. They're warping from this sort of kind of roughly square shape at one end to the roughly circular shape at the other end. So if you want to have a really smooth transition between profiles, the centerline loft is going to be the easiest way to do it. If we put too many profiles in and try and loft from profile to profile to profile, we're giving Fusion too many constraints. And I don't mean constraints like in a sketch sense, I just mean constraints as in things to achieve.

So this is a really good way-- if the sweep isn't appropriate to what you want to do, maybe try a loft, but try a loft with centerline and just let it give you the best shape it can between those profiles. So I just wanted to dig a little bit deeper into what's actually happening there. So I've got the same input geometry here, but on the left, I've got a rail loft. So I've got the two profiles on loft in between and a third line, which I'm using as a rail, and on the right, it's the same geometry but it's a centerline loft.

So I hope for what you can see there is that from the rail loft, we're just lofting directly from one profile to the next. But the loft is kind of being pulled off to one side by the rail. It's just being pushed over. With the centerline loft, the profiles are actually following the centerline. They're remaining perpendicular just like in a sweep. So I would try that out and see if that helps you out.

Now then I wanted to talk about the differences between patches and lofts. So who's used the Patch command? Cool. Most of you. Excellent. So the difference is that a patch always creates a four-sided surface. In fact, all surfaces inside Fusion are four-sided. That's just mathematically the way it calculates. But with a patch, Fusion creates a four-sided surface, and it fits it to our input geometry-- I'm sorry, trims it to our input geometry.

So in the background is a four-sided surface, but we only see the bit that's being trimmed to our boundary condition. With a loft, a loft will also create a four-sided surface, but with a loft, the four-sided surface will be fitted to inputs. So what this means is patch is a really good tool if you're trying to cover something that has not four sides. So less than four sides or more than four sides. If you have four sides, loft is probably going to be the best tool.

But if you need to create a lofted surface that has less than four sides, it's probably better if you take a hint from the Patch command, create a four-sided loft and then trim it back to some boundaries, rather than trying to match those boundaries. So maybe I better give you an example of that. So here I have a couple of inputs.

They're three-sided inputs, and I've turned on the texture mapping for these so we can see. This kind of gives you a rough idea of the direction of these curves. These grid lines, by the way, and known as [? isopoms, ?] and it's again used in the background to calculate how surfaces match. But we can see on a patch surface, the [? isopoms ?] are flowing in the same direction, so we can imagine that underneath there is a four-sided shape being calculated. It's been trimmed back to our surface.

So the patched surface, we'll be able to continue working with, and we shouldn't have too many problems. We could thicken this, we can offset it, we could extend these edges. Shouldn't be too many problems here. But with a loft, you can see those [? isopoms ?] are flowing right to the top, and they're all connecting together at the top. And this is known as a singularity. So what's happening here is that fourth edge that we've got has been collapsed down to a single point.

And you can imagine, that point all the math is happening in a zero condition. So if anybody's ever typed divide by zero into a computer, you know they don't handle that well. So this isn't wrong. You can do this, and if this is the last surface in your model, that would be absolutely fine. But if this is one of the first surfaces in your model, you might find that this loft to point, when you try and thicken it, you'll get some weird stuff going on. And you'll find you can't extend these edges because it can't kind of scale them because the top point there is 0.

So they're not very easy to work with downstream. So maybe avoid that if you can. That's our singularity right at the top. I just wanted to just drop a hint in here. For anybody who's not been able to turn on these UV diagnostics, if you right click on a body, you want to choose Texture Map Controls, and then under Texture Maps Controls, you got Display Mode, UV Diagnostic Texture.

Unfortunately, I can't see there's any way of kind of saving it, so you can toggle it on and off. So you kind of have to do this every time, but it is useful if you want to diagnose a surface. Particularly if it's coming from another CAD package, and you've brought it in and it won't behave, maybe turn these on and just see how Fusion's interpreting that surface under the hood.

So which tool should we use, patch or loft? Well, patch is usually the best tool if you've got non-four-sided shapes. So circular shape, one side. We can't do anything with that with a loft because we need at least two entities to be able to loft, so that's not going to help us at all. So that would have to be a patch. Three sides, two sides, patch. Five sides, patch. But if it's four-sided, we probably want a aloft. If we can use a loft, we should use a loft.

So when would we use a loft instead of a patch? So again, I've used the same input geometry here, but on this side, we've got a patch, and we can see the patch is kind of-- it's got this sort of dish shape, it's got sort of bowing in the middle. So as it tries to find the shortest route between our inputs, it's giving us a slightly curved surface. A loft is always going to go directly between surfaces, so a loft is going to give us something I think of as being kind of stiffer. More direct.

So if you've got large areas to cover with a big distance in between, I would say go with a loft. If you've got high curvature conditions, I'd would go with a loft because you will get a much straighter condition. So if you've got a relatively flat area to patch in, a relatively small area to patch in, use the Patch command rather than the loft.

So I just put a little video in here just to give a couple of tips on lofting. So with lofts, we can warp the shape. We have these option to change the end conditions. We don't actually have to have another surface to be continuous to to set the end condition. If we set it to Direction, we have these two controls here, length and angle, and you can think of these being the same as our tangency handles in our spline. It's doing the same job.

The more we extend the length of the distance there, the more influence the tangency is having. And the direction gives us tangency there too, so we can affect the curvature in a loft without actually having to have an input surface to work with. If we do have an input surface, then we have the option to change from connective G0, tangent G1, or smooth G2. And once again, we have this additional weighting control that we can kind of fiddle with and adjust that weighting exactly the same way as we can do with a spline curve to get something that's, in this case, G2. You can tell by the splines there, and make sure we've got a nice condition.

So the characteristic of a loft is that we can warp a loft, we can warp the surface so we can create these nice curvature continuous conditions. So I'm going to switch over here to a G2 condition, see what that gives us. Click OK, and you can see again from the curvature combs that they are of the same length, and their tangent-- not tangent to each other. They are of the same length and pointing in the same direction so that indicates G2.

So if we wanted G3 here, we would have to play with those inputs a little bit more and see if we can get a smoother flow between the two surfaces. But I guess the key takeaway there is that when we loft, we don't have to loft between surfaces. But a set continuity, we can loft to a sketch and use the direction control instead.

So a couple of tips on lofting. So this is a standard loft, and this is the loft we see on the loft icon, square to circle. And I personally think that's horrible. This is absolutely horrible, and the reason is because we've got four input entities and one entity at the other end, so we're going from four to one.

So in the background, Fusion is generating four input surfaces and is trying to sort of marry them together and blend them into one surface at the other end. So one tip is try and match the number of inputs you have at one end of the loft to a number of inputs you have at the other end of the loft, and you'll get a much smoother result. So I've built another one here. I'd like the further. I think that looks a lot better.

But what I've done that's different is I've kind of cheated. So if I turn on the sketches on this one we can see we have a square at one end and a circle at the other. On my improved one down here, I have a square with radius corners, so I've got eight entities, right? Four lines and four arcs. And at the front, I've got a circle with little flap bits. Well, actually it's the same, four lines and four arcs. It's exactly the same geometry that I've used at both ends, I've just changed the size. And that's given me a much smoother loft because Fusion is calculating from one edge to a matching edge, and it's just creating a nice smooth surface between the two.

AUDIENCE: So then you're just trying to align the expectations with the software view geometry [INAUDIBLE].

PAUL MUNFORD: Exactly. So the question was we're just trying to align what we can do, so we're not trying to force the geometry to something it can't. Absolutely. Kind of set an expectation so that you can improve your inputs and that improves your outputs. Absolutely. It's a good way of putting it. Thank you for that. So if you can, copy and paste the original geometry, and then change it, and then you know you've got the same inputs.

There's a really nice feature in Fusion 360, and in fact, all sketches are 3D. If you want to create a 3D sketch, you can just copy and paste something in place and just drag it off the sketch plane, and it becomes a 3D object. So you can have exactly the same inputs, copy them across, and then start to change the inputs the other end, and you know you've got good conditions.

One other tip here, so I've just taken this cube and I wanted to create a blend on the cube that's not just a standard radius, but it's a little bit more kind of as close to G3 as I can get it to get a really nice smooth blend on the corner here. But the tip here is to get the nice smooth corner, I really want to use a loft because that's going to give me nice conditions. One thing with the Patch command in Fusion is you can only set one level of curvature continuity for everything unfortunately. I'd like to see that improved.

Where lofts, we can set curvature continuity on each individual input. So they're much more flexible from that point of view. But the tip here is to avoid that condition where we're going up to a singularity when going back to a single point. What I've done is made the radius down here just slightly bigger than the radius here. Just a fraction. And that gives me a fourth edge.

And the fourth edge can be really, really tiny as long as it's bigger than zero. And the loft will work much better, and can set curvature continuity at the top there. So if you can force a loft to become four-sided, it's really going to help you downstream. OK, so setting up curvature conditions.

So here's my shape here, I just want to patch in that last face. So I create a patch, and set continuity, and it's awful. Look at that. That's terrible. That's not what I expected. So why is it so bad? And it's probably because I've gone, let's have G2 everywhere. That's got to be the best, right? Because G2 is better than G1. But actually, I've set too many conditions. I've asked too much, and it's trying to be tangent and curvature continuous with edges that are flat.

So that's why we've got a rubbish surface. So if we looked at this a little bit more, we would see on this edge here, we've got a complete change in direction. So changing direction would be G0, so we don't need to have any curvature continuity set there. We can just leave it with the free condition and let Fusion work out the best for us. This input geometry down at the bottom, now we do have curvature here. This is actually curving, this input geometry, so we might think a G1 tangent? But if we go G1 tangent, we're trying to be tangent with it, so it's going to try and flow like this, but in fact, we want a change of direction.

So again, the correct input down here would also be G0, just position, and we can let Fusion work out the best one for us. Across this one, now we definitely have curvature so, Paul, tell me we can be tangent to that one because it's curved? Well, we could pick tangent. It would probably work, but if we looked at it this way, if we took a section through any of those curves, it's flat, so we don't really need tangent condition there. We can just leave free condition again. It's going to work just fine.

And then up at the top, we definitely have a curved surface, so we definitely want to set at least tangency here, if not match the curvature. But if I match the curvature, I still get something that looks a little bit weird. And the reason is because I've got a curve coming in at the top here with a variable radius, and then down the side here, I've got a constant radius. So again, I've over constrained it. Over here it's trying to be tangent and curvature continuous in the middle, but then at the edge it's matching that constant radius.

So in fact, I just need tangency at the top, and tangency is all I need to get a good surface. So if I see the conditions there from my Profile 1, Profile 2, connected is absolutely fine. At the top, I've got tangency, at the bottom, connected again. So actually I didn't need to set too many high levels of curvature continuity on that kind of patch. So I hope by explaining how these things work, if you look at these continuity controls here in lofts and patches, you'll better make better decisions about how I should use those and what result I'm going to get. And if you ever pick one of these and it doesn't work, hopefully, now you'll understand--

The question was will the Fusion Team add G3 or a CV spline, and yeah, I'd really like them too. So if you'd like to see that happen, the best thing you can do is go onto the forum. There's an idea station for Fusion, you can put your ideas forward, and then tell all your friends go vote for my idea. And if it goes up, then they put it on the list. If we get enough votes.

AUDIENCE: [INAUDIBLE].

PAUL MUNFORD: Ah, see, so it's coming. You do have in Fusion the other kind of curve, conic curve. So you've got conics, which behave a little bit like a CV spline, but you can only have one point. But you can adjust the curvature, with a CV spline, you can't. All right, buddy, back to work. So settle back down again. Hope that's refreshed everybody a little bit. It's kind of dim in here, so it's kind of relaxed.

So this is one of the examples I've prepared for the class. So this is a really cool illustration I found by this guy called Yo Kobayashi, and I modeled that up in Fusion just to see what I could get. So that's my version over there on the right. So I wanted to give a couple of tips here about actually building stuff in Fusion. One tip I always gave with any of my models is if you have specific design parameters to meet, create some user parameters first and reference them as you work.

So usually we can at least say the height, width, and depth of our object, and then if we come along and we want to make just an aesthetic change, want to tweak the length a little bit, tweak the depth a little bit, we can just change our parameters and watch everything update. I don't always get all the parameters in the first hit. Sometimes I'm working away, and I think, you know what? I think I want to add in another parameter for a specific radius I want to use across multiple features. Or maybe there's a thickness or something I want to add in, so maybe I'll come back in and add a few more in as I work. But if I can, I always start with putting some parameters in.

And then here's an example here of a more complex, one of these control cages of construction geometry. And then I've pinned my arcs and splines to it. So when I change these dimensions, the kind of datums move, and everything updates without going too wacky. But I haven't had to place too many dimensions in it to fully constrain everything.

And one more tip, you'll notice there's no blend curves in my sketch. So this is something that we think of as working to theoretical sharps. So if you can imagine you want to put a radius, or you want to put a blend in, I'll try and work to the sharp point, the point where those surfaces would meet for as long as I can before I put the blend in. And there's a couple of reasons for doing that.

One reason is because it's going to make a more stable model in the first place. It's going to be easy for me to change those blends later, because I have them as a feature like I did an update, and it's much easier for me to remove those blends if I don't need them, and I still have good sketch geometry underneath. So I encourage you not to put blend curves in sketches, or even in our models, we can intersect our surfaces and overbill them, which I'll give an example at later, rather than putting them in the sketches. It'll make the sketches too complicated. It'll make them difficult to work with.

So the kind of workflow here, I've created my profile. The first thing I've done is a loft. And my loft doesn't have any of those blend curves in it. It's just working to a theoretical sharps, and then I sliced that loft through, patched in the additional face. I've then blended that sharp edge out, and then I stitched the whole thing together and put in my chamfers and fillets.

So those edge conditions, edge consuming conditions, I always try and build last. They are the most likely to go wonky if you change parametrically. Because if I wanted to do a parametric change, maybe I'll just move the timeline back, do my parametric change, and bring it forward again and rebuild all those last minute fillets and chamfers towards the end of the model. And that just gives me a lot more stability.

So here's an example I wanted to use. So again, this is just a cool example I found from my Pinterest account. But this is a company called MTECH, made something called a power tree, and so I wanted to talk here about patch layouts. So how do we decide kind of to create these four-sided lofts? How do we work out what we're going to do?

So I wanted to give an example of that, plus also, this is an example of something that has a branch in it. So we got a couple of things being brought together, and this is kind of like a typical surfacing problem. So if you know how to do this, then you're kind of set. You will come across this at some point. So I'm going to just scribble over this now, so I'll have to sit down for a little bit.

I'm going to use the marker tool here. So the first thing I'm going to say is whenever I'm looking at a project I'm modeling up, I'm beginning to assess it in terms of things like what does it have that can help me? Well, this has symmetry between the front and the back. So if I was to say I've got kind of a centerline down here, so I only need to model half of this, and then I can mirror it around at the end. So that's going to inform the way I'm going to build this.

I'd also be looking for things like what parameters should I create? So I probably want to have like an overall height in there. And I'm probably going to have an overall width, and an overall depth. So those are things that I can put in and I can control with parameters right in the beginning. I'm also going to think about location. So where do I want the center point to be? Where do I want that centerpoint, the origin, to end up on my finished model? Again, that's going to form the way I'm going to start sketching.

So those are things I always think, about whether it's a prismatic model or whether it's a surface model. I'll just change color here. So the next one is can I build any portion of this surface using a sweep? So do I have anything that has a consistent profile? So in this case, I could say, well, this portion down here kind of roughly around here, has the same section the whole way down. So I could build that little bit with a sweep. I'll just indicate that over here as well.

And right up at the top, it's the same section through to about there, so I can build those top two bits with a sweep too. So those are going to be my primary surfaces. The surfaces I do first. And then I can start creating secondary surfaces, surfaces that are going to use those to set their continuity. So I'd probably look at something like this, and say how do I fit a four-sided patch there?

Well, I'd probably just take a line straight down here, and then I can create a loft from this bottom edge to this top edge, maybe even using this side down here as a guide rail, and that's going to give me a four-sided patch up here. And I can do the same thing on this side, four-sided patch right there. Same thing on this side, four-sided patch right there. And that's going to give me some nice input geometry, and I just then need to fill in the middle. So if I just sketch these in over here. Same idea. Same idea.

Now, one thing we could do is we could say if I take a centerline down that surface there and a centerline down that surface there, we could create these edges so they matched the centerlines. But the reason I haven't done that, I just added this little edge in at the top, one, to avoid singularities, and two, so I can set continuity condition across these little extra bits here. It's going to give me a smoother input.

So when we're thinking about this, we think about these main surfaces. These are what we call primary surfaces. So these are the ones we build first. So maybe I'll just mark those up one. These are all number one surfaces. I'll change color again. These ones that we create off the primary surfaces are our secondary surfaces. And in fact, this endcap down at the bottom here would be a secondary surface as well. We can't build that without the first primary surface in place.

And then finally, the patch in the middle would be our tertiary surface. So these are the surfaces we have to build third that will bring this entire thing together. So when we're doing a surface in project, I'd always encourage you to sit down for 10 minutes before you start, and just do one of these sketches and try to work out what your patch layout should be. Spend a few minutes thinking about it.

My old boss used to say, if you're going to spend even a day working on something, it's worth 10 minutes thinking about it before you start. You will save yourself time. And again, this isn't T spline. This is not organic push pull stuff. We're going for more precision surfaces here, so you'll save yourself some time in the long run if you spend a bit of time thinking about it. So let's have a little look inside Fusion.

I've got the power tree open here. So this is the version of it I created, so just to give you a look about how that might work out. If I take the timeline back down to the beginning here. So I am quite methodical in the way that I work. So right at the top of the tree, I have something called my front layout sketch. And my layout sketch will have here my main datums, those main points that I want to control with parameters, so I can make sure this flexes.

So one tip is if you do want it to be parametric, periodically just change some parameters and make sure things are updating correctly without having too many problems. The last thing you want is like Friday afternoon the boss comes in and says, just make small change, and you make this small change and your model explodes, right? Now you're not going home early. So take a little time just to check that your model is flexing the way you expect it to.

And I don't want my first layout sketch to be too complicated. So I've put sketch lines in there that I'm going to use in multiple positions in my design, and then I created another sketch over the top of that for additional features, because I don't want to put too much in one sketch and have it too cluttered or too complicated. So I'll just reference that in. Now again, these two sketches are right at the beginning of the tree, and I'm quite methodical about that.

I don't always build them in this order, but I manipulate the tree to get them in the right order. So I try and get all my sketches at the beginning of the tree if I can, so I can find them easily, and I can adjust things in a sketch as easily. And then I'll start by creating datums, so any three dimensional planes that I want. I will build from-- just turn these on. I can't talk and press buttons at the same time. I will build using that geometry.

So these work planes are planes where I know I'm going to put a sketch on later for those extrudes for those swept surfaces, and I'll build them. So I've got them working from the same sketch. When I manipulate the sketch, these are going to move. And then I can start building in some features, so when I get first. So I've got that first thing down the bottom. If I move it along a little bit, you can see I've got the two at the top being swept down, and I started building in my lofts. And as I work my way along, you'll see that's the last loft in there, and I've got it patched in.

So you can start to see how that comes together, and then eventually we'll turn that into a 3D solid. Yes, sir.

AUDIENCE: [INAUDIBLE].

PAUL MUNFORD: I'll tell you what, if I move it back to the end. So the question was, can we put a lighter shade on it so we can see the edges? So I've just turn on edge display there, so you can kind of see how that patch lays out, but this is-- in that Fit folder, there's an export copy of this so you can take it away, open it up, and have a look at how I worked on that.

Zebra, I'm going to demonstrate that in a little while. So you'll see those. OK, let's jump back into the presentation. OK, so the next thing, so we've thought about the continuity we'd like to have, we've planned our inputs accordingly, we've done a little sketch for a patch layout, so we've thought about how we're going to do that. We started building our surfaces, and at some point, we're going to have to turn these into solids.

So we've got two options here, the power tree option is pretty much smooth the whole way round, and most of our edges meet. So I've used this option here called Stitch, and stitch will just stitch all these surfaces together, and as long as we completely enclose a volume, so as long as there's no gaps, it will return it to us as a solid. So now we have a solid we can use downstream for manufacturing, we can do tooling, whatever we need to do. The other option over here is called a Boundary Fill, so the nice thing about a boundary fill is it allows us to do something called overbuilding.

So you can see these surfaces I've got on here. Instead of trying to get them to meet perfectly and get them all trimmed up and match perfectly, I've just made sure they definitely overlap each other, and there's no ambiguity about that. Made them really big. And then we can use a boundary fill just to kind of flood fill the resulting hollow in the middle. And also with boundary fill we can use work planes to kind of cap off the boundaries.

So in this case, if we do have something which is symmetry, we can flood fill using the work planes and centerline, and later on we can mirror that over. So as you're building again think about am I going to overbuild and flood fill this later, or am I going to build something that's very precise with all edges that match and then stitch? And I'll be quite honest with you, if it's a complicated model, you might use both. So you might have to build some surfaces, stitch them together, and do some more work and then-- yeah.

So a couple of things that go along with that. One is reverse normal. So have you noticed here that surfaces in Fusion have two faces? Who's noticed that? About half. This yellow face is indicating the back of the face. So this is pointing in the wrong direction. The front face will be whatever your default material is at the moment. So I've got the default still set. That's why it's this sort of shiny gray color.

So if I try and stitch this, it's probably going to fail. So before I stitch it, I need to use the Reverse Normal button. And normal just means again sort of perpendicular to the face. So normal is the face direction. So by reversing normal, it will flip this face around and then we can stitch it properly. You'll notice in lofts, you can loft from input one to input two, and if the loft comes out facing the wrong way, you can actually switch the inputs around, and it will flip the face around the other way.

So try that too, because that's easier than doing a reverse normal. But with patches often, it's reverse normal. Now the other command you might like to know about here is Unstitch. So unstitch is the opposite of stitch. And the reason this one is useful, so one way to start off with surface modeling is actually to start with your prismatic shapes, just start in a solid modeling environment, do some solid models, and then what we do is we delete faces we don't need and then patch them back in again.

So again, this is the example I showed you earlier of saying I want to have a nice sort of G3, as close to G3 as I can get blend on the corner of this cube, so I'll start off just by creating a solid cube. That saves me from having to stitch all this stuff together because it's a shortcut. It's already stitched. Chamfer off the top edge, and then I want to delete it. But Fusion has this really cool tool that if you delete a face, it will automatically retrim all the other faces back to get rid of it.

So in this case, you'd have to unstitch to unstitch all these surfaces, then you can delete the faces you don't want, then you can build the faces back in, and then you can stitch it all back together again. So if you come from Inventor or other tools that don't rematch the faces when you delete them, you might like to know about that one.

Now, I wanted to talk about a couple of ways we can then assess our surfaces. So we've created good clean input geometry, we've created surfaces, we've turned those surfaces into a solid, now how do we check that we have good flow? Or if we have issues, how do we check where the issues are? So one thing you'll find when you're surface modeling-- one thing I find certainly, I'm often turning it from shaded to shaded with edges.

So shaded with edges just shows my patch layout so I know where everything is. But if I make the material like a nice shiny material, in fact, the default steel is not too bad, but I might change it to something like silver, something really reflective, turn off the edges, just by spinning the model around, you get a pretty good view as to whether you've got good continuity or not. And so one tip here is changing the shaded to shaded with edges thing every time you've got to go through that workflow is a bit of a pain.

On the Autodesk app store under Fusion 360 apps, there's a nice bloke here called Pawel, he's created a visual styles changer. It's got a couple little buttons you just click on, it just switches visual styles. He's done it for fun by the looks of it. It's completely free. So you may want to check that out, and see if that helps you work a little bit faster. So that's something you can have turned on the whole time when you're working to kind of assess the result you're getting with, and see whether you're getting a good result.

But if you are wanting to diagnose a surface and see where any issues are, then you could turn here to Curvature Map Analysis. So this time I've included a little animation here, and I'm changing the scale of the curvature map, and the curvature map is showing me the least curved areas, which in this case are colored green. And the most curved areas are going to be colored red. So this is a really good way of finding out singularities.

Like if you've got something that just looks wrong, you turn this on, and you set the scale, and you get a red spot where you didn't expect one, you've got an issue with that surface. You can go in and work out what you need to do to correct it. So I say this is a pretty cool tool for diagnosing issues. Actually, we use this tool quite a lot for diagnosing issues when we do imports from other CAD software as well. If we've got a surface that won't thicken or won't behave, we'll use this to diagnose and see if we can do something with it.

But the other one you were [INAUDIBLE] time is zebra striping. Who was asking me about zebra striping? It was you, sir, wasn't it? There we go. So zebra striping, you just turn it on, it projects some black lines onto the surface, and as you move the surface around, you can look at the continuity. And for anybody who's not familiar with zebra striping, the way you read those if it's G0, we'll get a definite discontinuity between the striping. So we can be absolutely certain that it's just a sharp edge.

If we looked at G1, you would see there's a definite-- the zebra striping does meet but there's a definite change of direction again. It's a very sharp change of direction. As we get smoother, you can see that although the black lines meet, they come away at a nice smooth angle, and that lets us know again we can set G2. We'd expect seams like that. We can't set G3, but we can do our best to get as damn close as we can. And if we've got nice zebra striping, we're as close to G3 as we're going to get in Fusion.

The final way we can assess surfaces in Fusion is with Curvature Comb Analysis. I'm really sad about this tool because it doesn't do what I want it to do. This tool will only let us do curvature combs on edges. Well, most of my edges have splines in them already, and I can set curvature combs in the sketch. So this is only useful if I'm assessing an edge that doesn't have a spline attached already.

Generally speaking, in most other CAD tools, you select the surface and you put a range of curvature combs across the whole surface, but we don't have that in Fusion right now. So unfortunately, if we want to assess the surface, we're either going to have to split the surface to create an edge or put a sketch through it and assess the sketch instead. So it's a little bit more work right now. So I'm hoping that's something we can improve in future.

So a couple of questions for you. Curvature continuity in the process of matching what between surfaces? Who wants to have a stab at that one. Who's still awake? Come on. Sir.

AUDIENCE: So first the tangency.

PAUL MUNFORD: Yep.

AUDIENCE: And then [INAUDIBLE].

PAUL MUNFORD: That's it.

AUDIENCE: [INAUDIBLE].

PAUL MUNFORD: Perfect. Yeah, so it's the process of matching those properties between the surfaces, position, tangency, curvature, and acceleration. So we're matching extra properties each time. Which four prop-- I just gave it away. Which four properties can we match between surfaces? Oh, I just told you. Connected, tangent, curvature, acceleration, but it's hierarchical. We're matching more properties each time.

So three commands that are used to convert surfaces to solids. A little bit if a trick question here, because I forgot to mention one of them. So name me two commands, and then see if anybody can pick up the third one.

AUDIENCE: Stitch and boundary fill.

PAUL MUNFORD: Stitch, boundary fill-- can we think of one more?

AUDIENCE: Thicken?

PAUL MUNFORD: Thicken. Absolutely. Yeah, so thicken, if we're doing surface in modeling, particularly for something like sheet metal, we'll typically use thicken for turning it into a solid. The nice thing about thicken is it keeps all edges perpendicular, which is one of the conditions of sheet metal. We have to have perpendicular edges otherwise it won't flatten. So boundary fill, stitch, and thicken. Well done, everybody.

Three tools that are used to inspect the quality of surfaces? Who wants to suggest some tools?

AUDIENCE: Curvature combs.

PAUL MUNFORD: Zebra striping, coverage combs.

AUDIENCE: And the curvature analysis.

PAUL MUNFORD: And the curvature analysis. Well done. Curvature map, zebra, and comb analysis. Brilliant. So in summary, if you're going to do surfacing in Fusion, consider curvature continuity. What do I want to achieve? Let's think about that before I start. Plan your patch layout. So before you start, sit down, do a little sketch, see if you can work it out before you start.

Build clean, graceful, perfectly curved splines as inputs. The better the inputs you have, the better the result you're going to have. Use curvature combs to check your splines as you go along. Make sure your splines are good. Work to theoretical sharp edges. So work to theoretical sharp edges whenever you can, and put the blends in later when you know the model's there.

If you can't work to theoretical sharps, overbuild, so make surfaces that definitely intersect. Then either use boundary fill or trim them back and stitch. And don't over constrain blends. So don't be using G2 if you don't really need it. Think about that. Use fill or stitch to turn surface into solids, and then use curvature or zebra analysis to inspect surfaces just to see whether you've got a good clean surface, whether you've got any problems.

And do flex your parametric designs as you go along. Don't wait make the last minute and then flex them and find that everything breaks. So who's learned something? Most of you, that's good. Who's learned something they can take away and definitely use, right? My key thing is your confident enough to go away and try this, because if you don't try, you won't remember it. Brilliant. Thank you very much, everybody. So do you have any questions for me? I'll start over here and I'll come back to you, sir. Yes, sir?

AUDIENCE: [INAUDIBLE]. Nice G3 surface and trim it, the resulting edge [INAUDIBLE]. Earlier you were saying [INAUDIBLE].

PAUL MUNFORD: So the question is when you trim a surface, does it affect the quality of the surface? In a parametric [INAUDIBLE] like this one, no. Because the underlying surface was still there in the history. We just removed a bit of it. So the only downside is if you put a split in there to assess it, well then you'd have to delete the split, and if you want to assess again later, you have to put the split back in again. So it's a bit of a problem because parametric CAD remembers everything you've done. So it's a little bit clunky. Thank you for the question. Yes, sir?

AUDIENCE: Can you directly [INAUDIBLE] the isoparametric curves and true parametrization of the surfaces [INAUDIBLE]

PAUL MUNFORD: So the question is can you directly see the parametrization of the surfaces or do you have to use a proxy? No, you have to use a proxy. That's the only way you can view it. Does anybody else have a question for me? OK, well, thank you very much. Enjoy the rest of your day.

Please remember to fill out your surveys. I know it's a bit of a pain, because they don't seem to work very well, but please have a go. We really appreciate it.

______
icon-svg-close-thick

Préférences en matière de cookies

Votre confidentialité est primordiale pour Autodesk et nous mettons tout en oeuvre pour vous garantir la meilleure expérience possible. Pour nous aider à personnaliser les informations que vous recevrez et à créer des applications, nous collectons des données sur votre utilisation de ce site.

Nous autorisez-vous à collecter et à utiliser vos données ?

En savoir plus sur les services tiers que nous utilisons et sur notre Déclaration de confidentialité.

Cookies indispensables au fonctionnement de notre site et à la fourniture de nos services

Ces cookies nous permettent d'enregistrer vos préférences ou vos informations de connexion, de répondre à vos demandes ou de traiter les achats de votre panier.

Cookies visant à améliorer votre expérience grâce à l'affichage de contenu pertinent

Ces cookies nous permettent d'améliorer les fonctionnalités et la personnalisation. Ils sont configurés par Autodesk ou par des fournisseurs tiers auxquels nous faisons appel afin de vous proposer une expérience et des informations adaptées à vos besoins.Si vous n'autorisez pas l'utilisation de ces cookies, certains ou l'ensemble des services risquent d'être indisponibles.

Personnalisation des publicités à des fins de ciblage

Ces cookies collectent des données vous concernant, en fonction de vos activités et de vos centres d'intérêt, afin d'afficher des publicités pertinentes et d'en évaluer l'utilité. Grâce aux données collectées, ces publicités seront davantage ciblées. Si vous n'autorisez pas les cookies, elles seront moins pertinentes.

icon-svg-close-thick

SERVICES TIERS

Découvrez les services tiers que nous utilisons pour chaque catégorie et la façon dont nous exploitons les données que nous collectons sur vous en ligne.

icon-svg-hide-thick

icon-svg-show-thick

Cookies indispensables au fonctionnement de notre site et à la fourniture de nos services

Qualtrics
Nous faisons appel à Qualtrics pour que vous puissiez nous faire part de vos commentaires par le biais d'enquêtes et de formulaires en ligne. Vous êtes susceptible d'être sélectionné au hasard pour participer à une enquête. Vous pouvez aussi nous faire directement part de vos commentaires. Nous collectons des données afin de connaître les actions que vous avez effectuées avant de participer à une enquête. Cela nous aide à résoudre les éventuels problèmes. Politique de confidentialité de Qualtrics
Akamai mPulse
Nous faisons appel à Akamai mPulse pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Akamai mPulse
Digital River
Nous faisons appel à Digital River pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Digital River
Dynatrace
Nous faisons appel à Dynatrace pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Dynatrace
Khoros
Nous faisons appel à Khoros pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Khoros
Launch Darkly
Nous faisons appel à Launch Darkly pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Launch Darkly
New Relic
Nous faisons appel à New Relic pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de New Relic
Salesforce Live Agent
Nous faisons appel à Salesforce Live Agent pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Salesforce Live Agent
Wistia
Nous faisons appel à Wistia pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Wistia
Tealium
Nous faisons appel à Tealium pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Tealium
Upsellit
Nous faisons appel à Upsellit pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Upsellit
CJ Affiliates
Nous faisons appel à CJ Affiliates pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de CJ Affiliates
Commission Factory
Nous faisons appel à Commission Factory pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Commission Factory
Google Analytics (Strictly Necessary)
Nous faisons appel à Google Analytics (Strictly Necessary) pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Google Analytics (Strictly Necessary)
Typepad Stats
Nous faisons appel à Typepad Stats pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Typepad Stats
Geo Targetly
Geo Targetly nous permet de rediriger les visiteurs de notre site vers la page appropriée et/ou de leur proposer un contenu adapté à leur emplacement géographique. Geo Targetly se sert de l’adresse IP des visiteurs du site pour déterminer approximativement la localisation de leur appareil. Cela permet de s'assurer que les visiteurs ont accès à un contenu dans ce que nous évaluons être la bonne langue.Politique de confidentialité de Geo Targetly
SpeedCurve
Nous utilisons SpeedCurve pour contrôler et mesurer les performances de notre site Web à l’aide de mesures du temps de chargement de nos pages Web et de la réactivité des éléments successifs tels que les images, les scripts et le texte.Politique de confidentialité de SpeedCurve
Qualified
Qualified is the Autodesk Live Chat agent platform. This platform provides services to allow our customers to communicate in real-time with Autodesk support. We may collect unique ID for specific browser sessions during a chat. Qualified Privacy Policy

icon-svg-hide-thick

icon-svg-show-thick

Cookies visant à améliorer votre expérience grâce à l'affichage de contenu pertinent

Google Optimize
Nous faisons appel à Google Optimize afin de tester les nouvelles fonctionnalités de nos sites et de personnaliser votre expérience. Pour ce faire, nous collectons des données comportementales lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, d'ID Autodesk, etc. La version de nos sites peut varier en fonction des tests de fonctionnalités. Le contenu, quant à lui, peut être personnalisé en fonction de vos attributs de visiteur. Politique de confidentialité de Google Optimize
ClickTale
Nous faisons appel à ClickTale pour mieux identifier les difficultés que vous pouvez rencontrer sur nos sites. L'enregistrement des sessions nous permet de savoir comment vous interagissez envers nos sites, notamment envers les éléments de nos pages. Vos informations personnellement identifiables sont masquées et ne sont pas collectées. Politique de confidentialité de ClickTale
OneSignal
Nous faisons appel à OneSignal pour afficher des publicités numériques sur les sites pris en charge par OneSignal. Les publicités sont basées à la fois sur les données de OneSignal et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que OneSignal a collectées sur vous. Les données que nous fournissons à OneSignal nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de OneSignal
Optimizely
Nous faisons appel à Optimizely afin de tester les nouvelles fonctionnalités de nos sites et de personnaliser votre expérience. Pour ce faire, nous collectons des données comportementales lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, d'ID Autodesk, etc. La version de nos sites peut varier en fonction des tests de fonctionnalités. Le contenu, quant à lui, peut être personnalisé en fonction de vos attributs de visiteur. Politique de confidentialité de Optimizely
Amplitude
Nous faisons appel à Amplitude afin de tester les nouvelles fonctionnalités de nos sites et de personnaliser votre expérience. Pour ce faire, nous collectons des données comportementales lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, d'ID Autodesk, etc. La version de nos sites peut varier en fonction des tests de fonctionnalités. Le contenu, quant à lui, peut être personnalisé en fonction de vos attributs de visiteur. Politique de confidentialité de Amplitude
Snowplow
Nous faisons appel à Snowplow pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Snowplow
UserVoice
Nous faisons appel à UserVoice pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de UserVoice
Clearbit
Clearbit autorise les données d’enrichissement en temps réel afin de fournir une expérience personnalisée et pertinente à ses clients. Les données que nous collectons peuvent inclure les pages que vous avez consultées, les versions d’évaluation que vous avez lancées, les vidéos que vous avez visionnées, les achats que vous avez réalisés, ainsi que votre adresse IP ou l’ID de votre appareil.Politique de confidentialité de Clearbit
YouTube
YouTube est une plate-forme de partage de vidéos qui permet aux utilisateurs de visionner et de partager des vidéos qui sont intégrées à nos sites Web. YouTube fournit des indicateurs sur les performances des vidéos. Politique de confidentialité de YouTube

icon-svg-hide-thick

icon-svg-show-thick

Personnalisation des publicités à des fins de ciblage

Adobe Analytics
Nous faisons appel à Adobe Analytics pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil et d'ID Autodesk. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Adobe Analytics
Google Analytics (Web Analytics)
Nous faisons appel à Google Analytics (Web Analytics) pour collecter des données comportementales sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces données nous permettent de mesurer les performances de nos sites et d'évaluer la qualité de votre expérience en ligne afin d'améliorer les fonctionnalités que nous proposons. Grâce à des méthodes d'analytique avancées, nous optimisons également votre expérience dans les domaines suivants : communication par e-mail, assistance client et ventes. Politique de confidentialité de Google Analytics (Web Analytics)
AdWords
Nous faisons appel à AdWords pour afficher des publicités numériques sur les sites pris en charge par AdWords. Les publicités sont basées à la fois sur les données de AdWords et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que AdWords a collectées sur vous. Les données que nous fournissons à AdWords nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de AdWords
Marketo
Nous faisons appel à Marketo pour vous envoyer des e-mails dont le contenu est ciblé. Pour ce faire, nous collectons des données concernant votre comportement en ligne et votre interaction envers les e-mails que nous envoyons. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, de taux d'ouverture des e-mails, de clics sur des liens, etc. Nous sommes susceptibles d'utiliser ces données en combinaison envers celles obtenues auprès d'autres sources pour vous offrir des expériences améliorées en matière de ventes ou de service clientèle, ainsi que du contenu pertinent basé sur un traitement analytique avancé. Politique de confidentialité de Marketo
Doubleclick
Nous faisons appel à Doubleclick pour afficher des publicités numériques sur les sites pris en charge par Doubleclick. Les publicités sont basées à la fois sur les données de Doubleclick et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Doubleclick a collectées sur vous. Les données que nous fournissons à Doubleclick nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Doubleclick
HubSpot
Nous faisons appel à HubSpot pour vous envoyer des e-mails dont le contenu est ciblé. Pour ce faire, nous collectons des données concernant votre comportement en ligne et votre interaction envers les e-mails que nous envoyons. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, de taux d'ouverture des e-mails, de clics sur des liens, etc. Politique de confidentialité de HubSpot
Twitter
Nous faisons appel à Twitter pour afficher des publicités numériques sur les sites pris en charge par Twitter. Les publicités sont basées à la fois sur les données de Twitter et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Twitter a collectées sur vous. Les données que nous fournissons à Twitter nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Twitter
Facebook
Nous faisons appel à Facebook pour afficher des publicités numériques sur les sites pris en charge par Facebook. Les publicités sont basées à la fois sur les données de Facebook et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Facebook a collectées sur vous. Les données que nous fournissons à Facebook nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Facebook
LinkedIn
Nous faisons appel à LinkedIn pour afficher des publicités numériques sur les sites pris en charge par LinkedIn. Les publicités sont basées à la fois sur les données de LinkedIn et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que LinkedIn a collectées sur vous. Les données que nous fournissons à LinkedIn nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de LinkedIn
Yahoo! Japan
Nous faisons appel à Yahoo! Japan pour afficher des publicités numériques sur les sites pris en charge par Yahoo! Japan. Les publicités sont basées à la fois sur les données de Yahoo! Japan et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Yahoo! Japan a collectées sur vous. Les données que nous fournissons à Yahoo! Japan nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Yahoo! Japan
Naver
Nous faisons appel à Naver pour afficher des publicités numériques sur les sites pris en charge par Naver. Les publicités sont basées à la fois sur les données de Naver et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Naver a collectées sur vous. Les données que nous fournissons à Naver nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Naver
Quantcast
Nous faisons appel à Quantcast pour afficher des publicités numériques sur les sites pris en charge par Quantcast. Les publicités sont basées à la fois sur les données de Quantcast et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Quantcast a collectées sur vous. Les données que nous fournissons à Quantcast nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Quantcast
Call Tracking
Nous faisons appel à Call Tracking pour fournir des numéros de téléphone personnalisés dans le cadre de nos campagnes. Vous pouvez ainsi contacter nos agents plus rapidement et nous pouvons évaluer nos performances plus précisément. Nous sommes susceptibles de collecter des données sur votre utilisation de nos sites en fonction du numéro de téléphone fourni. Politique de confidentialité de Call Tracking
Wunderkind
Nous faisons appel à Wunderkind pour afficher des publicités numériques sur les sites pris en charge par Wunderkind. Les publicités sont basées à la fois sur les données de Wunderkind et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Wunderkind a collectées sur vous. Les données que nous fournissons à Wunderkind nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Wunderkind
ADC Media
Nous faisons appel à ADC Media pour afficher des publicités numériques sur les sites pris en charge par ADC Media. Les publicités sont basées à la fois sur les données de ADC Media et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que ADC Media a collectées sur vous. Les données que nous fournissons à ADC Media nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de ADC Media
AgrantSEM
Nous faisons appel à AgrantSEM pour afficher des publicités numériques sur les sites pris en charge par AgrantSEM. Les publicités sont basées à la fois sur les données de AgrantSEM et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que AgrantSEM a collectées sur vous. Les données que nous fournissons à AgrantSEM nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de AgrantSEM
Bidtellect
Nous faisons appel à Bidtellect pour afficher des publicités numériques sur les sites pris en charge par Bidtellect. Les publicités sont basées à la fois sur les données de Bidtellect et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Bidtellect a collectées sur vous. Les données que nous fournissons à Bidtellect nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Bidtellect
Bing
Nous faisons appel à Bing pour afficher des publicités numériques sur les sites pris en charge par Bing. Les publicités sont basées à la fois sur les données de Bing et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Bing a collectées sur vous. Les données que nous fournissons à Bing nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Bing
G2Crowd
Nous faisons appel à G2Crowd pour afficher des publicités numériques sur les sites pris en charge par G2Crowd. Les publicités sont basées à la fois sur les données de G2Crowd et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que G2Crowd a collectées sur vous. Les données que nous fournissons à G2Crowd nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de G2Crowd
NMPI Display
Nous faisons appel à NMPI Display pour afficher des publicités numériques sur les sites pris en charge par NMPI Display. Les publicités sont basées à la fois sur les données de NMPI Display et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que NMPI Display a collectées sur vous. Les données que nous fournissons à NMPI Display nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de NMPI Display
VK
Nous faisons appel à VK pour afficher des publicités numériques sur les sites pris en charge par VK. Les publicités sont basées à la fois sur les données de VK et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que VK a collectées sur vous. Les données que nous fournissons à VK nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de VK
Adobe Target
Nous faisons appel à Adobe Target afin de tester les nouvelles fonctionnalités de nos sites et de personnaliser votre expérience. Pour ce faire, nous collectons des données comportementales lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP, d'ID d'appareil, d'ID Autodesk, etc. La version de nos sites peut varier en fonction des tests de fonctionnalités. Le contenu, quant à lui, peut être personnalisé en fonction de vos attributs de visiteur. Politique de confidentialité de Adobe Target
Google Analytics (Advertising)
Nous faisons appel à Google Analytics (Advertising) pour afficher des publicités numériques sur les sites pris en charge par Google Analytics (Advertising). Les publicités sont basées à la fois sur les données de Google Analytics (Advertising) et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Google Analytics (Advertising) a collectées sur vous. Les données que nous fournissons à Google Analytics (Advertising) nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Google Analytics (Advertising)
Trendkite
Nous faisons appel à Trendkite pour afficher des publicités numériques sur les sites pris en charge par Trendkite. Les publicités sont basées à la fois sur les données de Trendkite et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Trendkite a collectées sur vous. Les données que nous fournissons à Trendkite nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Trendkite
Hotjar
Nous faisons appel à Hotjar pour afficher des publicités numériques sur les sites pris en charge par Hotjar. Les publicités sont basées à la fois sur les données de Hotjar et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Hotjar a collectées sur vous. Les données que nous fournissons à Hotjar nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Hotjar
6 Sense
Nous faisons appel à 6 Sense pour afficher des publicités numériques sur les sites pris en charge par 6 Sense. Les publicités sont basées à la fois sur les données de 6 Sense et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que 6 Sense a collectées sur vous. Les données que nous fournissons à 6 Sense nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de 6 Sense
Terminus
Nous faisons appel à Terminus pour afficher des publicités numériques sur les sites pris en charge par Terminus. Les publicités sont basées à la fois sur les données de Terminus et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que Terminus a collectées sur vous. Les données que nous fournissons à Terminus nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de Terminus
StackAdapt
Nous faisons appel à StackAdapt pour afficher des publicités numériques sur les sites pris en charge par StackAdapt. Les publicités sont basées à la fois sur les données de StackAdapt et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que StackAdapt a collectées sur vous. Les données que nous fournissons à StackAdapt nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de StackAdapt
The Trade Desk
Nous faisons appel à The Trade Desk pour afficher des publicités numériques sur les sites pris en charge par The Trade Desk. Les publicités sont basées à la fois sur les données de The Trade Desk et sur les données comportementales que nous collectons lorsque vous naviguez sur nos sites. Il peut s'agir de pages visitées, de versions d'évaluation activées, de vidéos lues, d'achats, d'adresses IP ou d'ID d'appareil. Ces informations sont susceptibles d'être fusionnées envers des données que The Trade Desk a collectées sur vous. Les données que nous fournissons à The Trade Desk nous servent à personnaliser les publicités numériques afin de les rendre plus pertinentes. Politique de confidentialité de The Trade Desk
RollWorks
We use RollWorks to deploy digital advertising on sites supported by RollWorks. Ads are based on both RollWorks data and behavioral data that we collect while you’re on our sites. The data we collect may include pages you’ve visited, trials you’ve initiated, videos you’ve played, purchases you’ve made, and your IP address or device ID. This information may be combined with data that RollWorks has collected from you. We use the data that we provide to RollWorks to better customize your digital advertising experience and present you with more relevant ads. RollWorks Privacy Policy

Voulez-vous améliorer votre expérience en ligne?

Nous souhaitons vous offrir une expérience optimale. Si vous choisissez Oui pour les catégories présentées dans l'écran précédent, nous collecterons vos données et les utiliserons afin de personnaliser votre expérience et d'améliorer nos applications. Vous pouvez modifier vos paramètres à tout moment en accédant à notre Déclaration de confidentialité.

Votre expérience. Votre choix.

Nous respectons votre confidentialité. Les données que nous collectons nous aident à comprendre votre utilisation de nos produits, à identifier les informations susceptibles de vous intéresser, mais aussi à améliorer et à valoriser votre engagement envers Autodesk.

Nous autorisez-vous à collecter et à utiliser vos données afin de personnaliser votre expérience ?

Découvrez tous les avantages d'une expérience personnalisée. Vous pouvez gérer vos paramètres confidentialité pour ce site. Pour en savoir plus sur les options disponibles, consultez notre Déclaration de confidentialité.