[flow_default] Transcription for week08 04 chest improvements.wav

week08 04 chest improvements.json

@@ -0,0 +1,12 @@
+{
+  "file": "week08 04 chest improvements.wav",
+  "transcription": {
+    "audio_file": "week08%2004%20chest%20improvements.wav",
+    "text": "A quick add-on here to the chest, which at the moment is not working too nicely. If we are twisting our chest control here, the IK1, and we twist it to the side, it's working quite okay. But if we're moving it back and forth, we can see that the chests are not really doing the right thing here, or the breasts here, they should actually go down more. And if we're looking at it without the geo, the we would have to move this backwards to get the curvature here moving properly. And if we turn this on and move it back, we can actually see you can now the spine or the chest here is doing the right thing, breast are doing the right thing if we move this middle controller back. So we could either do something where this maybe goes back automatically if we rotate this or so. The other thing that we can maybe try to do is parent the breast joint somewhere else. So at the moment we have it quite low on the spine We could probably try to parent it to this spine joint eight here instead See if that helps at all. It will probably help a little, but I don't think enough They're still staying fairly fairly stiff So what we could also try to do is another method We could try to instead parent those guys completely to the chest control now. They're probably too high. So now they follow 100% with whatever the chest control here is doing, which also doesn't look too great. Maybe a little bit better. Maybe. I'm still not too happy with it. Let's try something else. Let's try to undo here, parent them back to where they were before. So spine 8. Or spine 7 rather. Okay, that's what we had them earlier. And instead what I want to try to do is if I move this down, and by the way sideways is kind of like same issue, even more so I think. Moving it to the side makes the chest here kind of weird. It makes it really weird because I think this one should go up. It should really rotate more with this. And if we're looking at what the spine surface here is doing, it's kind of clear why this is happening. Same thing as before. If we take this and move it, then the spine is here doing the right thing. Or the chest here, breasts are doing the right thing. But the moment it's not. So what we could try to do instead is we could try to tweak the weighting here a little bit. If this is going 45, you can see that the curvature here is almost nothing. It almost doesn't change. But if we come in here and we change the weight skin weighting on the surface, the moment we only have these top points to follow 100%, if we paint this a little bit differently, let's try painting these also 100 and you can immediately see that the curvature is getting stronger. Also if we add a little bit more weights to that one. So let's paint, let's add with a value of 0.1. To this one you can immediately see we're getting somewhere here with this. The curvature is much stronger now. Perhaps something like that. Maybe not enough, let's go all the way and paint it 100%. Okay. Now I think it is working better than before. It's still not 100%. Now let's try parenting them onto this upper joint again. Let's see what that gives us. Okay, I think this is looking quite good. This is, I think, looking quite good now. So I'm actually happy with that. And twisting is still working. The problem now that we've changed the weighting is probably that this middle one does not have as much influence anymore. It's probably going to look a little bit strange. You can see it does not really have a lot of pull anymore, because all this is now weighted 100% or almost 100% to this hip or chest control, chest joint, so we don't have as much as a smooth fall off anymore. But perhaps we don't even need that middle control so much, especially now that it's always interpolating. I think it should work fairly well.",
+    "language": "en",
+    "duration": 313.91,
+    "timestamp": "2025-11-28T05:29:25.727560"
+  },
+  "timestamp": "2025-11-28T05:29:25.733054",
+  "processing_time_seconds": 71.77143883705139
+}