[flow_default] Transcription: 02-Bag Tearing Edge Frature.json

transcriptions/02-Bag Tearing Edge Frature.json

@@ -0,0 +1,7 @@
+{
+  "audio_file": "02-Bag Tearing Edge Frature.wav",
+  "text": "Alright, welcome to lesson 4. This might be somewhat of a long lesson, so strap in. What we want to do this lesson is fracture our bag. So to begin doing this, let's drop down an edge fracture node and let's put it between a transfer node and a vellum cloth. Alright, let's visualize what we have now. So let's turn on our primitive piece output and let's click on this visualization pen. And what we have now is these 10 pieces highlighted here in color. Since we by default have 10 initial pieces, let's push this number to like 30. So we have more pieces. And as you see, now we have more pieces and as you see Now we have more pieces highlighted here So what does this mean? What what does this actually get us? Let's check very quickly in a solver So let's dive in and let's quickly bypass our grain so we only have a cloth Solving so we'd have a quicker result to look at. Alright so as you can see what we have now is a fractured bag so the cloth is tearing up from the get-go from the first frame it's tearing up and falling down. Let's get rid of this back face tinting. Personally I don't like the color if you're working in Houdini 19, there's this helpful toggle right here. If you can click on and it will get rid of this annoying blue color. It's very annoying to me. I don't know why. All right, so what we want eventually with this edge fracture node is to fracture our cloth, which is we've made this already, but we needed to hold still until the grain starts pushing down on it and stretching it just enough so it breaks eventually. So we need a way to stitch our fractured pieces and fractured edges together until the grains break it apart. So to do this, let's get back to our setup. And right after the vellum cloth, let's drop another vellum constraints node. Let's hook our two inputs in the place. And let's change the constraint type to weld points. So what this does is it stitches our edges back together. So it's like glue constraints, but for fracturing. And we can take a quick look at this if we press play we'll find that our cloth is indeed fractured but it doesn't fall off right away as it did before since we have stitched it back together but we can still see our edges right here that we made. Okay so that's a start. Let's unbi-pass our grains so we can see what what the grains does and very quickly let's go to a vellum object. Let's go to guides and Let's turn on the false color mode in the visualization to the stretch ratio Let's Turn this next stretch ratio to like.5, so it's halfway between 0 and 1 since ratios go from 0 and 1. So this is an average way to start visualizing our stretch ratio, and we'll see how we'll use this information later. So let's do a quick flipbook and see what we have right now. All right, so what we have right now is our same back simulation as before, but now we have this color gradient from blue to red. And since we put our max stretch ratio to 0.5, that means the red color is 0.5 in the stretch ratio or above, and the blue is zero in the stretch ratio, and in between is the gradient of colors. So how is this information useful to us right now? Well, ideally what we want is the grains pushing down on the bag until the bag can't take anymore and it breaks down it gives way to the grains to break the bag and escape and to do this we have a parameter in the volume constraints the weld point constraints called breaking and what we need is to turn this on and we have this little threshold right here and the threshold is controlled by the type of the stretch or the bending that's happening to your cloth object. Since we're visualizing the stretch ratio, let's change this from the stretch stress to stretch ratio and let's put 0.5 since we know from a flipbook the bag reaches 0.5 so we're expecting to see the tearing happening at the sides since the 0.5 is reached at the sides as we can see. So let's give this a go and see if this gets us what we're expecting. Alright so what we have right now is exactly as predicted. The cloth tears up from both sides and the grains escape from the sides and that's our first problem that we're gonna have to tackle because the fracturing looks like it was predetermined and it was prefractured the bag just tore up on its own and the grains came running down the sides. What we need to see is the grains pushing down on the bag and causing the fracture and to do that we first have to have a natural looking pattern, a natural looking fracture pattern for the edges. What we have right now is very computer generated looking with these sharp angles and edges. It doesn't look natural at all. And it's very regular in the way it see how does it work and if there's any way we can manipulate it to produce an irregular, more random, organic pattern. So let's right click and go down to allow editing of contents. So we're free to dive inside, click this, and we can now look under the hood of the edge fracture node. What I'm interested in in all of this mess is this node called find shortest path. If we click on it, we'll see the curves for our pattern. So if you pay attention, these are the curves and this is the resulting pattern from these curves. So this node basically generates the curves used for fracturing and it does that using two inputs. This convert line node, which is our bag converted into lines after being reduced by the same amount that we have here. So since we have typed here 40, inside the edgeFracture node it keeps only 40 polygons. If we middle click we'll see 40 polygons and these are converted to lines and these are the second input for the fine shortest path which it uses to produce these curves. And it produces these curves also based on our actual geometry. So if we find a way to manipulate the positions of points on the actual geometry input, maybe we'll get a more interesting curve looking pattern. So the easiest way to do this is by simply dropping a mountain node so we can apply a noise to our point positions. So let's hook this between our input geometry and the find shortest path node. And immediately you'll see we have now an irregular pattern before if we go back, we'll see that a pattern still look in like squares because we still have quad connections in our box, but it's not in a regular order like this. It has a more of an irregular order to it. We can play around with element size and amplitude for a mountain to get different results. So what you have now basically is an asset that produces infinite edge fracture patterns based on the element size and the amplitude you put in your mountain. But as you can see, this alone isn't enough to produce a realistic result because of our quad connection. So we need to fix this. So let's drop down a remesh node to remesh our geometry to an irregular pattern that we can produce more natural result out of. So let's hook this before edge fraction node and let's give this more of a resolution. So let's take down our target size to.0125. I think that's more than enough. We now have a very decent high resolution geometry. So let's see how this affected our edgeFracturePattern. Okay, now we have something more natural than before even without our mountain if we bypass this even with the default looking edge fracture pattern it looks much more than before this is what we had before which is some boxes scattered on our bag and what we have now is something that you would expect in real life more or less so let's start playing around with a mountain node to give this some irregular pattern in hopes to achieve a natural look. And actually I already played around with this a lot and I made a lot of different results before starting this tutorial so I know exactly the numbers that produces the result I want but feel free to play around with the numbers and do a lot of versions and pick up your favorite one at the end. I think it's great fun and it's very, and it can be very informative to how this node works and how many different types of fracture patterns that you can produce. And you will begin to see the difference between 3D generated patterns and realistic patterns and how can you manipulate that best to serve what you need to achieve in your personal projects. So let's click this to one and for the element size I found this number to be best and for offset I offset it by 126. All right, so you see what we have now is a very irregular looking pattern. It's much more natural looking than we had before. We can play around with the remesh, we can take down the iterations number even lower, so we can have something different, something like this, which has more fracture down the bottom. So if the bag tears from the bottom, we'll have more cloth fragments to look at. Let's take the initial pieces up as well. So let's say 45 pieces. And I think that's a good point to stop and see what we have now. Oh, I almost forgot as well, since we have rematched geometry, this means point numbers and IDs are different. So we need to reselect our pinpoints. You can have a procedural more cleaner solution where you drop a box and have this box group of points act like pinpoints, but since we won't be selecting our Pinpoints again, and this is a very personal project that I want to direct every aspect of I'd like to pin the animation points manually and We won't do this again anyway, so it's not a big deal deal so I've picked my points again and this time I made sure to do it in a more precise way than before you can pick your points in any order you'd like you can have it suspended from just the angles it will get you an interesting look as well but for my taste this kind of suspension is what looks best for me right now so let's make another flipbook and see what this gets us. Alright so as we can see we have much more looking result from the areas that are being fractured right now. We don't have this quad looking fracture pattern but we still have our bag fracturing from the side as if it was already predetermined. So to fix this we'll have to work on our grains actually since we now have very low resolution grains and very few grain particles we need to add to our grains resolution more than this so we'll have more grains and it will push down more on the bag and will have a desired effect but for now let's see what happens if we lower our break threshold value and if this will get us a better result. So let's go to a vellum constraints, the world constraints and let's take this down to 0.25. And I'm gonna get rid of this full scholar mode now since we already know what our ratio is. And let's get rid of the wireframes as well. And make another flipbook. All right, now we're talking. Look at that. That looks a lot better than before. It looks much more natural, much more organic. The problems that still remain is the bag breaks very quickly before we can see it stretch and really... I want to see it really struggle with the grains before it gives way. I want the back to stretch very far until it can't take it anymore then it breaks so we can see that stretch and that force it will make it look a lot better. And the grains are very low as well as I've said we can see right now even before it breaks the bag we can see the default pattern of the grains since the sprites are too big the point scale is too big. So in the next lesson we'll push the quality and the resolution of the grains further, a lot further. We'll work on adding some variation to our sand as well. All the sand we have right now is dry sand. I like a mix of dry and clumpy or wet sand. So we can see how we can achieve this as well. I know this has been kind of a longer lesson, so thank you for making all the way through and we'll see you in lesson 5.",
+  "language": "en",
+  "confidence": null,
+  "duration": 943.24
+}