F1 2019 Pre-Season Testing Thread

[Edax]

I'll try it to tell you it from my perspectiv, so all is objectiv and my own opinion:
Ferrari - Very good pace, especcially long run, short runs did look very strong aswell, both drivers very very strong
Mercedes - Long runs were very good aswell, they did a lot of sandbagging during the other testing days but HAM showed they can match Ferrari short run pace, HAM strong, Bottas was looking good but, in my opinion, not much improvement to last year.
Red Bull - was looking very good on 6 days of the 8, yesterday Gasly, again my opinion, screwed up badly for the whole team so that Verstappen only could do 29 laps today, Honda engine was very reliable and showed good pace in the speed traps
Renault - looked good aswell and for me are the 4th team on the grid, they still miss a bit of power, and will need to keep improving to stay that way, midfield teams are very close to eachother
ToroRosso - had a good pace and speed, car seems to be working very well, Albon and Kvyat are doing really well in the car, they're long run pace looked good aswell and are for me the 5th team on the grid although this will be depending on track and conditions because i guess the midfield will be even closer then last year
Alfa Romeo - looked strong aswell and can be some kind of suprise this year, strong long run and short run place, you can say they benefit from the expierence Räikkönnen has brought in, they will fight with TR and Renault.
McLaren - not really sure where they stand, the engine (renault) seems to be good, not sure if they have the aero package to be a real midfield-contender, maybe they can suprise me.

the others i can't really tell because i didn't had them really on my screen the only thing i can be sure of is ..... Willliams will be last.

Again this is my personl opinion

Thanks

Your welcome although I believe we have other users with more insight and technical knowhow, and they could tell you more precise how it stands

Not sure whether anyone really understands. Mercedes and Ferrari obvously have a very good car, as expected.

But for the rest I think it is very hard to say who is ahead of who, other than that the midfield looks more competitive than ever.

[LM10]

Why is it so important for teams to especially cover the underside of the car when it's being put on flatbeds? Besides the FW part, there isn't that much to be seen anyway, am I wrong? There are multiple shots of the rear and the diffuser going around already and other than that nothing new to be seen at this spot.

What's more, the photographers being sent out by the opponent teams take pictures, but again, what would they get to see more than we already do on the pictures on the Internet? Pretty much every detail of the cars is on here, for example.

The underfloor of a Formula 1 car is arguably one of the most critical pieces of the entire car in terms of aerodynamic performance.

That's the simplest answer to your question.

But without looking at the underside of a car you already see everything other than the middle part of the floor. Could someone explain me what’s so critical about the middle section of a floor? Because it looks relatively simple designed.

I know it's a bit of an off-topic, sorry for that. But to show what I mean, here's a pic of the floor from RB13. Isn't everything except for those channels next to the plank visible from above anyway?

The floor surely is a critical piece of the car, but which parts just seen from underside make teams love to spy on each other? The channels right next to the plank look relatively similar on all cars. Also, isn't it a matter of design philosophy/concept which should make others to hardly copy it, considering it's a critical part and should be in homogeneity with the rest of the car?

https://cdn-7.motorsport.com/images/mgl ... er-sto.jpg

[Sierra117]

The underfloor of a Formula 1 car is arguably one of the most critical pieces of the entire car in terms of aerodynamic performance.

That's the simplest answer to your question.

But without looking at the underside of a car you already see everything other than the middle part of the floor. Could someone explain me what’s so critical about the middle section of a floor? Because it looks relatively simple designed.

I know it's a bit of an off-topic, sorry for that. But to show what I mean, here's a pic of the floor from RB13. Isn't everything except for those channels next to the plank visible from above anyway?

The floor surely is a critical piece of the car, but which parts just seen from underside make teams love to spy on each other? The channels right next to the plank look relatively similar on all cars. Also, isn't it a matter of design philosophy/concept which should make others to hardly copy it, considering it's a critical part and should be in homogeneity with the rest of the car?

https://cdn-7.motorsport.com/images/mgl ... er-sto.jpg

Well from what I can discern, it's a bit of the curiosity aspect, as in, if you don't pay keep an eye on teams' floors then you may miss something if perchance there is someone that did something innovative. And there's a bit of making sure you see all sides of the cube to make sure it's a cube - by seeing all angles of the car they can build a better "knowledge base" as to why a competitor's car is performing a certain way, thereby aiding their own development by getting hints as to what may or may not work. Sometimes one finds an answer to one's problems by seeing the answer to a seemingly different problem. Hope that made some sense lol.

[Maritimer]

Would teams ever make basic CFD models of opponents cars? Seeing the floor would be important for something like that.

[PhillipM]

It's one of the few times rival teams can get clear photo's showing the actual curvature of the front of the diffuser - which is one of the most vital parts on the car aero-wise, hence it gets hidden if possible.

[Jolle]

Would teams ever make basic CFD models of opponents cars? Seeing the floor would be important for something like that.

They do some, Sauber showed a test they did on a Mercedes style intake, but only if the data can be used in some form for their own car. Their CFD time is very restricted in the regulations.

[Sieper]

I read that often, but how could FIA ever enforce that no CFD is being done (somewhere in the big companies behind the F1 teams).

[Zynerji]

Capping CFD in F1 in 2019 is borderline mental disorder.

With GPU acceleration, FPGAs, cloud computing, distributed computing, and not to mention the most dense and powerful processors ever created in most people's home computers and cell phones, there is no reason to cap the CFD.

The teams should be encouraged to build software to distribute the CFD on the Cosmos blockchain, and allow their fans to install nodes to donate the cruncher resources of their computers and phones. Then it would mean that popular teams that engage fans would get more iterations than the team that doesn't engage.

Anyways...

[PhillipM]

That would lead to a single team dominating, always.

[Big Tea]

That would lead to a single team dominating, always.

Make it open source after a time limit. Too late for this years car, but no one gets left behind and educational establishments get modern real world stuff

[GrandAxe]

The attempt to limit CFD is sound, though I suspect, is impossible to police.

The bigger teams are so well financed and resourced that they can buy engineering and coder services to tailor any CFD software of their choice to the parameters of their wind tunnels and their supercomputers (Merc used to have a supercomputer, but I think they spun it off as a subsidiary biz some years ago - not so sure though).

The sort of coupling above would result in very tight correlation between CFD and wind tunnel - with that, the wind tunnel can be accurately simulated. Result? The restriction on wind tunnel time and the cost caps that are based on it would become meaningless.

Some years ago, Richard Branson tried to compete in F1 based on a car whose aero was produced by CFD only. No wind tunnel. I can't remember the name of the team now, but the effort failed.
These days, the outcome would most likely be different. Technology has come very far since then, particularly in areas like Machine Learning and computation speed.

EDIT: Richard Bransons team was called Virgin Racing. Here's their CFD only car (Wikipedia):

[Sierra117]

allow their fans to install nodes to donate the cruncher resources of their computers and phones. Then it would mean that popular teams that engage fans would get more iterations than the team that doesn't engage.

As long as we get paid for our component and power usage, that is ...

[Zynerji]

allow their fans to install nodes to donate the cruncher resources of their computers and phones. Then it would mean that popular teams that engage fans would get more iterations than the team that doesn't engage.

As long as we get paid for our component and power usage, that is ...

Why? the entire point is to support your favorite team. You buy hats, shirts, merchandise already. This would be very similar in that you donate a small percentage of computing power. You wouldn't be REQUIRED to participate, so why would you get paid?

[Jolle]

The attempt to limit CFD is sound, though I suspect, is impossible to police.

The bigger teams are so well financed and resourced that they can buy engineering and coder services to tailor any CFD software of their choice to the parameters of their wind tunnels and their supercomputers (Merc used to have a supercomputer, but I think they spun it off as a subsidiary biz some years ago - not so sure though).

The sort of coupling above would result in very tight correlation between CFD and wind tunnel - with that, the wind tunnel can be accurately simulated. Result? The restriction on wind tunnel time and the cost caps that are based on it would become meaningless.

Some years ago, Richard Branson tried to compete in F1 based on a car whose aero was produced by CFD only. No wind tunnel. I can't remember the name of the team now, but the effort failed.
These days, the outcome would most likely be different. Technology has come very far since then, particularly in areas like Machine Learning and computation speed.

EDIT: Richard Bransons team was called Virgin Racing. Here's their CFD only car (Wikipedia):

https://upload.wikimedia.org/wikipedia/ ... z_2010.jpg

It's not an attempt, for years now (I believe 2008-ish) the amount of CFD is limited by the amount of gigaflops of data you may process.

[Zynerji]

The attempt to limit CFD is sound, though I suspect, is impossible to police.

The bigger teams are so well financed and resourced that they can buy engineering and coder services to tailor any CFD software of their choice to the parameters of their wind tunnels and their supercomputers (Merc used to have a supercomputer, but I think they spun it off as a subsidiary biz some years ago - not so sure though).

The sort of coupling above would result in very tight correlation between CFD and wind tunnel - with that, the wind tunnel can be accurately simulated. Result? The restriction on wind tunnel time and the cost caps that are based on it would become meaningless.

Some years ago, Richard Branson tried to compete in F1 based on a car whose aero was produced by CFD only. No wind tunnel. I can't remember the name of the team now, but the effort failed.
These days, the outcome would most likely be different. Technology has come very far since then, particularly in areas like Machine Learning and computation speed.

EDIT: Richard Bransons team was called Virgin Racing. Here's their CFD only car (Wikipedia):

https://upload.wikimedia.org/wikipedia/ ... z_2010.jpg

It's not an attempt, for years now (I believe 2008-ish) the amount of CFD is limited by the amount of gigaflops of data you may process.

Last time I saw numbers, it was like a 20 TFlop limit, and now you can get that in a full tower Desktop with 4 GPU's.


That's why uncapping CFD, but only in a distributed manner, is the best way. It directly prevents AMD/Intel from coming in and building Super Computers (raising costs), and relies on the fanbase to donate everything over the 20 TFlop limit.


I don't like the even distribution model, as I'm sure that people would want to donate extra time to Williams (for example) to help them catch up. It would be interesting to see if the fanbase would react in such a way to prevent a single team from completely dominating (year after year), and switching their support to the underdogs...


Lots of fan engagement is out there if Liberty were to really dig into it.