Quote:
Originally Posted by wnut
Unlimited CFD
|
I have mixed feelings about this. I haven't looked, but I assume the CFD computing limits were set quite a long time ago (when measured in computer technology years). I can imagine that the prior limits may be absurdly low today. They may have just felt that raw compute power is cheap enough that it shouldn't be something high on the list of limiting. Especially if it is billed as a cost saving device.
Quote:
Originally Posted by chillibowl
right! and with the spoils going to those who can afford the most computing power, hire the most staff, those who can investigate more development paths before settling on the best solution.
at risk of being a broken record on this but havent we been down this road before?
access to money should not be the defining characteristic of any sport imo.  |
I agree on the "money" stuff as that is my mantra which seeps into most of my posts. My gut tends to tell me that this may not trigger any type of significant staffing up of people resources, but may just allow them to iterate quicker. But there is another side of CFD that is missing...
Quote:
Originally Posted by EffectiveSprinkles
But look how much money they're saving with the testing ban!
|
My understanding is that CFD is only going to be as good as your modelling methods. And that you can crank out virtual designs all day (and night) long, but the real question is... will it perform as expected when actually on the car on track?
I suspect a large part of actual wind tunnel time (which I believe is still controlled) is used to correlate and confirm the CFD results. And even then, you really don't know if it is "good" until you get on track. So if you are doing a crappy job with your CFD and tunnel testing you will struggle on track. And you may be at a loss as to "why" it's not working as expected. We seem to routinely hear about "bad correlation" driving teams down dead end paths and that can set the tone for an entire season (or more).
So I also am a fan of increasing "on track" testing time. Iteration via CFD and being able to physically verify should help teams create workable solutions. But trying to stuff this validation phase into part of Friday free practice is not enough!
Richard