Formula 1 simulators are a joke

[Tim.Wright]

Even if testing was opened up again, I don't think many teams would choose to test a lot. With the money spent on it, simulators have improved quite a lot and are getting ever closer to the real thing. You dont need to send a large group of people away for a week to test a few new parts anymore. No, you can do that at home and test those parts in a few hours.

Teams would only test in real life to confirm the simulator, but that's it. Timewise it is much more efficient to use the simulator, and with all the data gathered the simulator will become closer and closer to real life, making real life testing less and less important.

I wouldn't agree with that at all. Simulation technology is very good but like I mentioned before, outside of F1's testing ban, no-one are putting parts on cars without physical durability and performance testing. Its simply not yet good enough. Simulation helps give you a direction but it is in no way a replacement for testing just yet.

How many times have you seen teams bring parts to friday practice that were in the bin before qualifying saturday. Quite often... Its a sign of the accuracy of the development process when phyical testing is removed from it.

If testing was opened up, every team will test because its simply a necessary part of the development process.

[beelsebob]

Source: http://www.autosport.com/news/report.php/id/111960

A slight exaggeration, perhaps, but its time that somebody in F1 admitted the limitations of simulators.

Heh, this isn't him saying "simulators suck", this is him saying "we have more money than most, and would benefit if the rules were changed to use on-track testing, so we're making up reasons why we should do that"

[wesley123]

Even if testing was opened up again, I don't think many teams would choose to test a lot. With the money spent on it, simulators have improved quite a lot and are getting ever closer to the real thing. You dont need to send a large group of people away for a week to test a few new parts anymore. No, you can do that at home and test those parts in a few hours.

Teams would only test in real life to confirm the simulator, but that's it. Timewise it is much more efficient to use the simulator, and with all the data gathered the simulator will become closer and closer to real life, making real life testing less and less important.

I wouldn't agree with that at all. Simulation technology is very good but like I mentioned before, outside of F1's testing ban, no-one are putting parts on cars without physical durability and performance testing. Its simply not yet good enough. Simulation helps give you a direction but it is in no way a replacement for testing just yet.

How many times have you seen teams bring parts to friday practice that were in the bin before qualifying saturday. Quite often... Its a sign of the accuracy of the development process when phyical testing is removed from it.

If testing was opened up, every team will test because its simply a necessary part of the development process.

True. That's where I said as confirmation.

Testing will still be a part of the development, but it wont take a role like it did many years ago.

Teams will use it to confirm what the simulator says, which is what they mostly use the fridays for these days. Does the part work like the simulator said it would? No -> okay, why not and what is different? Then they'll solve this and slowly the simulator will improve more and more, making testing less significant.

It will stay a part of development, but it's significance in 'making the difference' will be reduced over time when the simulators improve.

[Jonnycraig]

Arriving imminently:- "wind tunnels are useless and can never truly recreate on track testing"....

[Jersey Tom]

Teams would only test in real life to confirm the simulator, but that's it.

Except there are things which you just cannot do in a simulator, and "real" testing is more than just a confirmation event.

[Lycoming]

My first overwhelming impression is that he's just pissed that he has to use this giant video game when he has a race track sitting all but unused right next to it.

He argues that they're spending so much money to develop the simulator that they may as well just do track testing. Somehow I don't think the cost of the simulator is as high as that of bringing back dedicated test teams. I also think that if testing were allowed, they would still be using the DIL sims alongside track testing, which kind of nullifies his argument as they'd still be spending more cash. With new tracks being built seemingly every year, one of the useful advantages of the simulator is it lets the drivers learn a track layout before its even built.

[shelly]

As Tim has noted above, a lot of people still make confusion between simulation, driver in the loop simulators, and what can be substituted by testing.

Even if unlimited track testing was allowed, teams would still use several physical testing tools like:
-wind tunnel for complete car 60%
-radiator wind tunnel
-engine dyno
-dynamic engine dyno (with simulation of airbox and radiators)
-7 post rig
- Instron machines for component load testing
- crash testing facilities
These are "lab" tests with real parts, except for wind tunnel - only suitable 100%wt being windshear in US.

Then there are the software testing tools:
-finite elements for loads
-finite element for crash
-cfd
-vehicle dynamics software
-dynamic systems simulation software
- driver in the loop simulators
-race strategy software

Banning track testing altogether it's probably too much; but it is not correct saying that you need track test to "start cfd and wind tunnel", or that banning track testing means stopping physical testing completely. You need some track data to correlate to develop your software, to refine you maths model and so on.

Then, if free to allocate its resources, every team chooses the development tool that is better for them. Teams used track testing because they could afford the cost and because it was the quickest way to get on top of tyres behaviour in times of tyre war.
That meant also that they did not build an accurate mathematical model of a flexy wing or flexy floor - for example: they built one or two prototypes and test it on the rig or on track - and put them in the bin if the performance was poor or the fia tests were not met.
It is not correct to say that banning tests means more money spent because of wrong aero packages thrown away and not raced: it happened routinely with track testing, only not during FP.

With the ban on testing, teams are forced to build better models and to refine simulation tools - and this is a technological development that has a good impact on automotive technology, more than track testing or wt testing.
I like this type of challenge more, compared to the scheduling, production and analysis of testing - I think they can go hand in hand but that it is right to limit tests in order to make the development of better simulation tools profitable.

Also I think that simulators an their technology could be marketed to get in contact with fans (for example in big cities) replacing in part the pilgrimage to circuits where people get only a distant glimpse of a car and the echo of its noise (that said, I have been one of those pilgrims and I have liked it)

[MOWOG]

You can watch online videos about people having sex for hours, days, weeks, months or even years, but i suggest doing so is not quite the same as having actual intercourse.

Your mileage may vary.

[timbo]

It's all about balance. It seems reasonable that development without track testing might result in more dead-ends which would raise the cost ultimately. Arranging tests in a more policed manner, i.e. staying on track for a day after GP weekend may be more cost effective in terms of development, than having no testing at all.

[Moxie]

Sour grapes, and utter nonsense. This is an engineering competition after all. Monty would be better off keeping his mouth shut and engineering.

[marcush.]

Why would you test?
say you got a perfect idea of what you are doing and you have exact knowledge of the environment and boundaries yo are working in -physical and actual testing is reduced to confirmation of calculationsand of course actual driver training.
I don´t think you can learn that much from actually pounding around a race track- you just cannot test reliability into a system and think it would make for a fail safe approach to development.It is not.
Reliability is a design task not a testing task .
Testing for setup purposes has more of a place as models do deviate from real world significantly because the sheer magnitude of input variables is not manageable in all those simulations....you need now and then some reality check under conditions you hope are as close to controlled as possible.
We see Formula 1 teams being reluctant to clock to many laps for fear of tyre usage or simply by ignorance...and establised habitus(Being the last to go for a timed lap..)Very often this is more a thrill factor than sharp engineering or would anyone really believe another five cars doing one more lap on a track would have a significant positive influence on grip levels and this would justify to take the risk of not getting a second chance ,being late at the line or a simple yellow flag somewhere on the track?

simulations do not need less but more physical testing -you need to correlate your calculation models and prove that your sensivity is correctly modeled for all incredients of your sim.
It is very surprising how close simulations get you to ballpark figures ,but it is also astonishing that variables with huge impact in simulation do not stand the proof of reality and other factors you would not even have included into your sim may well destroy your whole weekend or a whole season.

[Pierce89]

To be fair, can't really blame him. They're there to win, would be silly not to try to play to your strengths. Everyone's doing that to some extent.

Why should they spend huge money on a DIL simulator when they can test the car and real data instead of a close approximation?

"Real" data from track testing has its own shortcomings.

No doubt. But personally, if I ran an F1 team it would drive me crazy to own 2 tracks and not be able to use them. It is f1 after all.

[DaveW]

F1 teams have been using DIL simulators for many years for a variety of tasks that could be accomplished with a simple simulator costing, say, a 4 figure sum. Now, the ban on tracking testing has caused (many of) them to invest in simulators costing, say, a 7 figure sum. It is sensible to ask what has been achieved by that huge increase in cost.

A significant part of the cost increase is in providing improved cuing (visual and motion). Modern simulators are very good, but they are not perfect. A test pilot who was involved in one of my projects once stated "flying a simulator is easy, you (mentally) turn off aircraft and turn on simulator." Which was a simple way of saying that different cues are required to control a simulator at the limit. It adds credence the rumours that some F1 drivers are physically sick when driving simulators - probably caused by a mismatch of cues, caused both by transport delay and bandwidth. Until that can be overcome, how can a DIL simulator be used for serious engineering development?

Sour grapes, and utter nonsense. This is an engineering competition after all. Monty would be better off keeping his mouth shut and engineering.

That is an unkind statement, at best.

I am not an aerodynamicist, but I read that full CFD computations take weeks to complete using serious super-computers. Even tunnel testing is rarely (if ever) conducted at realistic ride heights (for fear of lifting the belt & destroying the tunnel). It follows, kind of, that the aero information used in a DIL simulator, in the absence of track testing, requires a substantial level of engineering guesswork.

Mechanically, an Adams-based simulation (which is in itself a very idealised model of the real thing) has to be further simplified to execute in real time.

Tyres & track maps are other issues that leave much to be desired.

If serious engineering is to be accomplished using DIL simulators, then accurate cueing must be achieved, and the fidelity of the models must be taken to new level. Failing that, track testing is a solution. I think that "Monty" was simply stating the obvious.

[shelly]

I agree with you dave, but to me the interesting part is to "take the fidelity of the models to a new level".
Cfd for example is still rapidly evolving - back in 2008 mclaren were able to run 50 (fifty) full car simulations per week; wind tunnel technology has also improved (running realistic ride heights no longer an issue, with the help of vacuum suction under the belt for example).


The problem with that is that everything in f1 tends to become an arms race; then the wt hours and the teraflops need to be limited, because otherwise the richest teams wouls spend big amount of money there.

Anyway, I would prefer teams spending big money in using latest technology and maybe trying to put some advancement in it (for example pushing for a LES cfd on a full car) instead of spending that money (or more) in endless track testing; still I think that some track testing is needed (like next year).

[Just_a_fan]

The answer to his moan is to say "fine, we'll bring back some track testing but it must be at designated circuits and those don't include your two circuits." The cost of testing will be the same for all teams.