Love the meticulous path through code. Thank goodness for mathmaticians! I am truly grateful for the time you spend using mathmatics.The visual modeling is exciting to me! I want to express it with dance!
Would love to see more particles released in the upper atmosphere to be able to track the downdraft. I noice we are very uplift bias. Very cool stuff though.
I was thinking that if CM1 can do particle systems for realtime renderings then it might be possible to also render those horizontal vortices due to breakdowns by using B-splines for irregular rotational axes which the particles could circulate around in realtime.
I know this may not be possible, but how would distributed computing using something like BOINC fit into this? In addition to the supercomputer work, you could also parcel out data to run through BOINC so that you get a virtual supercomputer as well.
I think you might have the coolest toys of anybody on the planet. Hypercars and rocketships ain't got nothing on this. I think that was one of the coolest computer science lectures I've ever seen. Thank you for making this public. I hope you don't mind if I have some questions. What I'm dying to know, have you given any thought to creating guassian splats with any of your renders? You might be able to be the first to create a guassian splat of a tornado! I also wonder, is it possible to render and view the model data like you did at the end of the talk on consumer-available hardware or even just visualizing the data something you need a super computer for? And Is there any kind of simulation work or data processing that a weather/computer enthusiast could do at home? Really, thank you for taking the time to make these lectures public, that was super cool.
You can render on consumer hardware, that's mostly what I use (well I have an Nvidia Titan GPU for the big stuff). The simulations scale horribly as you increase resolution so you really need a supercomputer to do tornado modeling work. Regarding gaussian splatting, I had to look that up - I don't have software that can do that.
Dr Leigh what do You think about Charles L. Chandler's paper "The Electromagnetic Nature of Tornadic Supercell Thunderstorms". I can only imagine you both meet and discuss the subject of tornado genesis. That would be something absolutely amazing, I guess.
Also, do these tornado/supercell simulations consider the negatively charged electrostatic dust particles in the wall cloud and updraft? -- which is part of EHD-tornadogenesis.
Love the meticulous path through code. Thank goodness for mathmaticians! I am truly grateful for the time you spend using mathmatics.The visual modeling is exciting to me! I want to express it with dance!
Would love to see more particles released in the upper atmosphere to be able to track the downdraft. I noice we are very uplift bias. Very cool stuff though.
Excellent stuff as always Leigh. Thank you for sharing!
Hi Leigh, is there a user group online anywhere where one can throw beginner CM1 questions/issues at?
Congrats for your fascinating work and thanks for sharing!
I was thinking that if CM1 can do particle systems for realtime renderings then it might be possible to also render those horizontal vortices due to breakdowns by using B-splines for irregular rotational axes which the particles could circulate around in realtime.
I know this may not be possible, but how would distributed computing using something like BOINC fit into this? In addition to the supercomputer work, you could also parcel out data to run through BOINC so that you get a virtual supercomputer as well.
I think you might have the coolest toys of anybody on the planet. Hypercars and rocketships ain't got nothing on this.
I think that was one of the coolest computer science lectures I've ever seen. Thank you for making this public.
I hope you don't mind if I have some questions. What I'm dying to know, have you given any thought to creating guassian splats with any of your renders? You might be able to be the first to create a guassian splat of a tornado! I also wonder, is it possible to render and view the model data like you did at the end of the talk on consumer-available hardware or even just visualizing the data something you need a super computer for? And Is there any kind of simulation work or data processing that a weather/computer enthusiast could do at home?
Really, thank you for taking the time to make these lectures public, that was super cool.
I would kill for a single time slice of that supercell simulation 😅
you can run cm1 on consumer hardware. but you wont be reaching 30m grid size any soon
You can render on consumer hardware, that's mostly what I use (well I have an Nvidia Titan GPU for the big stuff). The simulations scale horribly as you increase resolution so you really need a supercomputer to do tornado modeling work. Regarding gaussian splatting, I had to look that up - I don't have software that can do that.
Dr Leigh what do You think about Charles L. Chandler's paper "The Electromagnetic Nature of Tornadic Supercell Thunderstorms". I can only imagine you both meet and discuss the subject of tornado genesis. That would be something absolutely amazing, I guess.
Also, do these tornado/supercell simulations consider the negatively charged electrostatic dust particles in the wall cloud and updraft? -- which is part of EHD-tornadogenesis.
Everything is cool 😎
In reality vortices aren't Phong shaded in a thunderstorm so I was wondering if CM1 can do volumetric particle irrotational vortex flow in realtime
Nope CM1 is just the model, the viz is done from saved model data files.
Yeah I picked up on that a little while after I asked. I'm really interested in how the dryline is simulated in the software.
I'm also wanting to compare QLCS tornado and convergence boundary landspout simulations using this.