pilF

At Cineversity i received a demonstration from Dr Sassi of the mirroring proces in more detail, there's a screencapture for download. You cab fin dit here https://www.cineversity.com/forums/viewthread/5543/#22269

Thank you very much for your generous offer @bentraje i just sent you a PM I'm seriously contemplating a switch to Maya now. Philip

So my troubles with the C4D body and head rig have not ended yet. PSD is working fine now and mirroring them is working too. The combination of a head rig and a body rig is causing a lag in the mesh now. I've watched Can Erdumans explanation on cineversity building the head rig and combining it with another rig over and over. I managed to do that and bind the head with a transform constraint to the body. But moving the body deforms a part of the face. The controllers lag behind the bodyrig thus deforming part of the face. I tried setting different priorities for the constraint, checked normalisations of the weights and even made the rigs editable. Nothing worked. The face deforms. Moving the playhead one frame does synchronise the mesh again. Any suggestion on how to solve this are welcome. Made a screenrecording of it too. Philip

Crazy detail that image. The cube is missing a towel. Always bring a towel.

It was down for an hour o so, all is working well now... phew

That's odd Cineversity site and forum is generating a 503 error

I posted the quetion at Cineversity too, got an answer from Dr Sassi. For the record: "My understanding is – that the joints that drive the points for PSD must be in the Driver list. With that, the resulting Joint-pose is part of the recorded PSD result. With two joints producing pose one and then in pose two, that would result in the need for a definition of each pose, as the angle for the Shoulder changes." I now have a PSD correction on the elbow with the shoulder and the forearm set as the driver and (very important) the PSD interpolation mode set to Global Strength in stead of Per Joint. "This mode will apply a deformation percentage based on the global delta variations of all the joints. All their delta values are averaged as a common strength value" the manual says.

Dear Core 4D , I thought I knew how PSD morphs work. Clearly not. I created a PSD morph for an elbow. It works fine until I rotate the shoulder. I then get an extra deformation on the elbow joint and i do not understand why. Can anyone point me in the right direction? Philip

Found a better explanation by CGDreams, i susspect something is off with the symmetrie of my mesh:

Hi, I'm trying to mirror Left to Right a correctional PSD morph to a hip bone being rotated 85 deg. Tried to follow along with CGdreams explanation here and here but i end up with a mess. Does this mirror feature actually work? Anyone with experience on this any advice i would greatly appreciate. I'd love to not sculpt another correction. Cheers Philip

Cool result @Björn Ewers the squishing effect works. Is this the method @Fritz described in 2023?

I'd be curious to see the result @Björn Ewers

So you have a picture from a floor where you would like to place a 2d illustration? Here's what I would do: camera calibration to place the camera correctly in relation to the floor and scale the scene. Get the exif info from the photograph, camera mode/sensor sizel, lens FOV Undistort the image (straight lines make for better calibration) Load the image in C4D through the camera calibration tag Place X, Y, Z axis or a plane and an origin point Estimate the length of one line to scale the scene (you did measure things in the scene right?) Solve the camera Place a plane on the floor and combine it with the image texture of the illustration Adjust light and render I do find this whole process a bit fiddly, it is never a perfect solve. Lenses focal point not being exactly what they state especially the zoom versions. See also this usefull explanation:

Congratulations I'd like to add !

In my defence I was young and naive believing it would end the nagging advertisements ... The trust i had in companies, it was bliss

winzip but only because i paid for it long ago andfinaly managed to shut the application up, involving a small edit in the registry and some fiddeling with startup prefs.

@HappyPolygonIt's an illustration of the things neural networks and in future general artificial intelligence can do better then humans. The water represents the level machine learning has reached. Playing Go, driving cars, etc. Dale-2 is another competence raising the water level.

Ah the dreaded valley of human skills being flooded by AI. There's this interesting illustration on this concept made by a human named Hans Moravec. Max Tegmark discusses it in his book life 3.0. An good read if you are intersted in the subject, not very reassuring though. In the Netherlands there is this play partly written by machine learning. It produced crazy plottswists (inspiration) but needed a human to give it a context. Just like this image generator. So like @clarence says the machine generates new combinations humans get to make sense out of it. For now. Because, why should there be a limit to intelligence? In future their will be machines smarter then humans. And we need guidelines for the ai to behave, like parents teaching kids. Only by then we will be the children.

With all future cards utilizing this new PCIe Gen 5 power connection the future of the next gen nvidia cards looks expensive in terms of power usage. This 16 pin connection can use up to 600 Watt ... glp

One development i find particulary interesting is this instant NeRF. A neural rendering model that learns a high-resolution 3D scene in seconds — and can render images of that scene in a few milliseconds. Source: https://blogs.nvidia.com/blog/2022/03/25/instant-nerf-research-3d-ai/ And a demo, i like the disco music that goes with it too

Quite convincing except for the teeth

Hey that looks promising ...

Tried importing this bed in Modo 10.2v2:

A promesing tech preview by Insydium, Xparticles fluid dynamics on GPU more then 4 times faster then its CPU brother.