Should the 2022 cars be moved to 2023?

[Herr_Koos]

I don't even think people agree on any axioms (all else kept equal) other than "on a long enough timescale the field will converge" which seems a truism.

So what other axioms can people agree on?

Second place is first loser?

[Zynerji]

It wont be long until bi-directional ray tracing is used with machine learning to design the air wetted "shell" of the car with CFD reinforcement.

Then it will just be who builds to most powerful desktop quantum workstations...

[Just_a_fan]

Why would ray tracing be useful in designing an aerodynamic shape?

[Zynerji]

Why would ray tracing be useful in designing an aerodynamic shape?

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

[DChemTech]

It wont be long until bi-directional ray tracing is used with machine learning to design the air wetted "shell" of the car with CFD reinforcement.

Then it will just be who builds to most powerful desktop quantum workstations...

Then, still I'd rather see a competition of 'who can best utilize their fixed, equal resources at achieving this' than a competition of 'who has most bling to buy a bigger computer'...

[Tim.Wright]

Why would ray tracing be useful in designing an aerodynamic shape?

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

Sounds like it would take more time to setup this tool than it would to actually do the development yourself.

[Zynerji]

It wont be long until bi-directional ray tracing is used with machine learning to design the air wetted "shell" of the car with CFD reinforcement.

Then it will just be who builds to most powerful desktop quantum workstations...

Then, still I'd rather see a competition of 'who can best utilize their fixed, equal resources at achieving this' than a competition of 'who has most bling to buy a bigger computer'...

It's an odd balance, but since logic dictates that in a Formula, there can only be one Perfect solution. Once that's found, we are in a bespoke but spec series.

[nzjrs]

I don't even think people agree on any axioms (all else kept equal) other than "on a long enough timescale the field will converge" which seems a truism.

So what other axioms can people agree on?

Second place is first loser?

Is cost positively or negatively correlated with the spread of the field, for example.

[Zynerji]

Why would ray tracing be useful in designing an aerodynamic shape?

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

Sounds like it would take more time to setup this tool than it would to actually do the development yourself.

For the first car, maybe.

Every evolution after that is "free".

And Dade, the developer of LuxRender was talking about this in 2009... It might be way advanced already without any reporting.

[Zynerji]

Double

[nzjrs]

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

Sounds like it would take more time to setup this tool than it would to actually do the development yourself.

For the first car, maybe.

Every evolution after that is "free".

Haha that's the dream of all automation, and the entire history of automation necessitates a bloody large asterisk after that sentence.

[Zynerji]

Sounds like it would take more time to setup this tool than it would to actually do the development yourself.

For the first car, maybe.

Every evolution after that is "free".

Haha that's the dream of all automation, and the entire history of automation necessitates a bloody large asterisk after that sentence.

It's a tool. Once cost is sunk, all work done by it is already paid for, so not "free" but less than a team of humans on recurring salaries.

[nzjrs]

For the first car, maybe.

Every evolution after that is "free".

Haha that's the dream of all automation, and the entire history of automation necessitates a bloody large asterisk after that sentence.

It's a tool. Once cost is sunk, all work done by it is already paid for, so not "free" but less than a team of humans on recurring salaries.

On the one hand, congratulations for appreciating the cost cap, on the other hand, the fallacy here is that the tool will always be enough to meet the demands of the designer. That's never been the case, that will never be the case, and the fancy ML buzzword salad will do is be slightly trendier for a finite time until the next generation comes along and proposes we use the next shiny thing.

The more things change the more the stay the same.

Edit: I say this as someone who builds buzzwordy DL tools in their 9-5 btw, I just have the responsibility of having to be honest to customers about them!

[Just_a_fan]

Why would ray tracing be useful in designing an aerodynamic shape?

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

Does a bidirectional ray tracing with turbulence model exist? Ray tracing traces light rays, does it not? Clue's in the name. How does that help with airflow modelling? And why is it better than CFD? The thing that is actually designed from the start to model air flow.

[Zynerji]

Why would ray tracing be useful in designing an aerodynamic shape?

Bi-directional ray tracing with a turbulence model allows you to determine the start and end points of the Ray. Using BiRT, you can "paint" the high and low pressure spots on the car, and how you WANT the flow-fields to look, and let the algorithm define the necessary shape to manufacture it.

Does a bidirectional ray tracing with turbulence model exist? Ray tracing traces light rays, does it not? Clue's in the name. How does that help with airflow modelling? And why is it better than CFD? The thing that is actually designed from the start to model air flow.

We're ot. I'll pm you