May 31, 20206 yr https://en.m.wikipedia.org/wiki/Navier–Stokes_equations Dominique Simming since 1981 - [email protected] GHz with 16 GB of RAM and a 1080 with 8 GB VRAM running a 27" @ 2560*1440 - Windows 10 - Warthog HOTAS - MFG pedals - MSFS Standard version with Steam
May 31, 20206 yr 12 minutes ago, Dominique_K said: https://en.m.wikipedia.org/wiki/Navier–Stokes_equations Love a bit of weekend geek! Flying since the early days of 386’s! 😀 Web developer, sailor, nerd. Contributor to the G36 Improvement Project PC Specs: Intel i9-9900K on a Aorus Z390 // Pro Wifi motherboard // 32GB Corsair Vengeance Ram, 2x16, 3200mhz DDR4 // RTX3070 Gigabyte, Eagle OC 8GB // WD Blue SN550 (2400) NVMe SSD // 2x LG QHD Monitors, (27QN600) Discord Username: CaptMatto#7935
May 31, 20206 yr Author Just now, Matt Barraud said: Love a bit of weekend geek! Yeah. I wonder what he does simulating the air disturbances of the wing foil. FS20 will perhaps include a wind tunnel for third party developers 😁 ? Dominique Simming since 1981 - [email protected] GHz with 16 GB of RAM and a 1080 with 8 GB VRAM running a 27" @ 2560*1440 - Windows 10 - Warthog HOTAS - MFG pedals - MSFS Standard version with Steam
May 31, 20206 yr Ah, fluid dynamics on airfoil. I did something like this in university as a final year project. Except, I used the Lattice Boltzmann method rather than Navier-Stokes equations. Júnior Silva
May 31, 20206 yr I wonder if this is some kind of hint how serious they took the aerodynamics simulation. Which would be awesome, as even XPs' blade element theory isn't on that fundamental level of detail as an application in a flight model (afaik it was used for calculating forces on propeller blades and turned out being an efficient way of calculating a basic flight model). Edited May 31, 20206 yr by tweekz Happy with MSFS 🙂 home simming evolved
May 31, 20206 yr Author What I wonder is why he does it in Shadertoy. Maybe @n4gix @Lorby_SI @WarpD or @SteveW could enlighten us? Dominique Simming since 1981 - [email protected] GHz with 16 GB of RAM and a 1080 with 8 GB VRAM running a 27" @ 2560*1440 - Windows 10 - Warthog HOTAS - MFG pedals - MSFS Standard version with Steam
May 31, 20206 yr Nice! If MSFS will use such type of CFD coding, that would indeed be unheard of in the flight sim world, in terms of complexity. Although, I notice the vortex shedding seems to indicate an unrealistically stalled airfoil (even at low AoA). So maybe he is just having some fun and demonstrating shadertoy capabilities (eheh). In any case, a good sign that the lead flight model/physics programmer knows his stuff! ☺️ "Society has become so fake that the truth actually bothers people".
May 31, 20206 yr 1 hour ago, Murmur said: Nice! If MSFS will use such type of CFD coding, that would indeed be unheard of in the flight sim world, in terms of complexity. Although, I notice the vortex shedding seems to indicate an unrealistically stalled airfoil (even at low AoA). So maybe he is just having some fun and demonstrating shadertoy capabilities (eheh). In any case, a good sign that the lead flight model/physics programmer knows his stuff! ☺️ Yeah... probably just toying around... it was some wishful thinking by myself. 😄 But as you say, still a good sign that the lead programer is aware that flight dynamics are more than rails. Edited May 31, 20206 yr by tweekz Happy with MSFS 🙂 home simming evolved
May 31, 20206 yr Commercial Member 4 hours ago, JSilva said: Ah, fluid dynamics on airfoil. I did something like this in university as a final year project. Except, I used the Lattice Boltzmann method rather than Navier-Stokes equations. Yep, that's what NASA uses as well. This is worth a look. Dave Hodges System Specs: I9-13900KF, NVIDIA 4070TI, Quest 3, Multiple Displays, Lots of TERRIFIC friends, 3 cats, and a wonderfully stubborn wife.
May 31, 20206 yr Let see if we can get a simulation with a variant flow and some wing flex perfomance.
May 31, 20206 yr The IFR shows plenty of stutters, so don't waste GPU cycles on this level detail... Matthew S
May 31, 20206 yr 18 minutes ago, MatthewS said: The IFR shows plenty of stutters, so don't waste GPU cycles on this level detail... On the contrary. Stutters are historically caused by CPU bottleneck, so if anything moving flying model to GPU and freeing up CPU could be a good thing for stutters. So yes, please use GPU cycles for flight model. "Society has become so fake that the truth actually bothers people".
May 31, 20206 yr 1 hour ago, Murmur said: On the contrary. Stutters are historically caused by CPU bottleneck, so if anything moving flying model to GPU and freeing up CPU could be a good thing for stutters. So yes, please use GPU cycles for flight model. You might like to read what I said. No, don't waste GPU cycles on this level of detail in the flight modelling, it's unnecessary. We're only burning maybe 1 KwH of electricity on a 5 hour 747 flight in MSFS and not 50,000 kg of fuel, so there's no need for such commercial levels of flight modelling accuracy as demonstrated in the video above. We need the GPU cycles for scenery & visual effects, and not "super duper" flight models. What they showed in the aerodynamics video a few months ago looked more than sufficient, which presumably is already implemented on the GPU. Edited May 31, 20206 yr by MatthewS Matthew S
May 31, 20206 yr 1 hour ago, MatthewS said: We need the GPU cycles for scenery & visual effects, and not "super duper" flight models. What they showed in the aerodynamics video a few months ago looked more than sufficient, which presumably is already implemented on the GPU. Stop using "we". You talk for yourself only, I talk for myself only. It's subjective. I would welcome this kind of "super duper" flight model accuracy, for me present scenery and visual effects look more than sufficient. Edited May 31, 20206 yr by Murmur "Society has become so fake that the truth actually bothers people".
May 31, 20206 yr I want super duper flight model, air flows and wing flex performance rather than nice silly trees, visual effects or any scenery BS.
Archived
This topic is now archived and is closed to further replies.