Xeroxiraptor: Boffins to print 3D robot dinosaur Dino-loving boffins in the US have embarked on their very own Jurassic Park-esque experiment to bring the actions of Earth's favourite prehistoric lizards to life. 3D printer creates dino bones for robots The researchers, from Philadelphia's Drexel University, are using 3D printing to create dino-bones and then attaching …
Lets see about combining the plots of those:
One is a film about a theme park where the attractions get out of control and start killing people, based on a book by Michael Crichton. The other is a film about a theme park where the attractions get out of control and start killing people, based on a book by Michael Crichton.
Shouldn't be too hard...
“Technology in paleontology hasn't changed in about 150 years,”
Actually, that's total crap. The biggest advance in paleontology was about a decade or so ago, when they first started doing computer modeling of bone and muscle structure to do, well, basically, what they're planning to do here, only in a computer model. That breakthrough was responsible, amongst other things, for the radical change in how diplodocus and other long-necked long-tailed were understood to work, and how dinos in general were understood to move. The seminal Walking With Dinosaurs was based on this work. This new 3d printing thing is entirely deriviative.
...but if you're just trying to model movement, why bother printing them? Can't you just 3D scan the bones and then build and move a virtual model? Or am I missing the point?
Pirate because I wonder whether they'll put the models on the Pirate Bay so you could print your own dino.
I'm sorry, I THINK that what both this and the previous post just said is "Why do I need to do the experiment, when we've already modelled it on the computer"
Welcome back, ancient Greek dude. You natural philosophers zoned out there for a while after the Enlightenment in the 18th Century. With your four elements and humours 'n' stuff derived from pure reason as to how the universe must work. How right you were. Messy experiments are for children, not people who already understand the universe....
Cutting the sarcasm, the reason is that the one thing which never fails to jar when watching CGI is just how unnatural animal movements are. And crucially how it has NOT IMPROVED in ten years. Not one iota. There will be howls of protest, but sorry, no, the emperor has no clothes, and CGI movement is bad. The articulation is fine, but any idiot can see that the muscles and tendons are simply not modelled - there is never any tension or gravity.
The conclusion is that this stuff is HARD, not that animators are stupid, but there is a major gap here - CGI is a minimum of 20 years away from solving this. I personally don't know what the problem is. It is "apparently" clear that this is THE big money problem in CGI, and that the amount of processing for a FE model of a decent muscle is way, way less than radiosity for hair, or polygon count for a landscape, and yet we are where we are. Visibly and embarassingly so.
Sorry, Mr. voice, but IMHO you post conflates "Technically impossible to do" with "Too expensive for a semi-crappy CGI documentary". The scientists have been using computer mechanical simulations of dinosaurs for a long time. The computer models can be made as complex as required, including things like effects of muscle pressure on nearby muscles, balance and weight of the 'pieces', mechanical resistance to traction... . Those are things you'll have trouble putting in a mechanical model.
Anyway I'm sure there are also a few areas in which having also mechanical models would help the researchers's work.
I wish, but not correct.
The computer models *CAN* be made as complex as required, but they are WRONG. I will say that again. There is something wrong with the models. A "crappy" CGI documentary, I don't know, but the bottom line was that Jurassic Park had as much money thrown at it as God, and but is still clearly *WRONG*. Very, very good for CGI, and astonishing for animals that haven't lived for 65 million years.
But if you want to see muscle pressure on nearby muscles, go and look at a Bernini or a Michelangelo (male). That's what muscles look like. Ain't no computer model that comes CLOSE.
"There is something wrong with the models"
Could you please be more specific?
And, just to clarify my point:
What we see in CGI -i.e. WWD- is an artwork. Paleontologists define the bone models, mark the places where the articulations go and give some other parameters. The CGI artists try to make a dinosaur inside those parameters. A paleontologist checks the results and gives them -hopefully- an OK.
What the 'complex computer models' are for is for testing whether a given arrangement of muscles is 'functional' or not, and whether it supports the kind of locomotion attributed to that species, or if that attributed locomotion is possible or makes sense at all.
"...go and look at a Bernini or a Michelangelo..."
Bernini and Michelangelo had a big advantage over paleontologists studying dinosaurs. They had living models. If we had living dinosaurs nobody would be bothering at all with this 'computer model' crap. Observation and an autopsy now and then would suffice.
... but kinda wondering how a scaled down model can give an accurate representation for research?
Surely the very mass of these massive creatures had a huge impact on the dynamics of their movements?
Bone density? - Surely that would work hand in hand with how the sinew and muscle worked alongside the bone structure?
The truth is, a great deal of the research is always going to be based on conjecture - nature is constantly amazing us and for all we know, T-Rex may have hopped around like a kangaroo. Unlikely, but possible.
Scaled down physics just doesn't work - we know this - so although it's a cool idea, I can't see it giving any serious insight into the mechanics of dinosaurs.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
