Fifth Gear Tries F1 Simulator of Cruden

[Bram Hengeveld]

Bram Hengeveld submitted a new blog post:

Fifth Gear Tries F1 Simulator of Cruden

The team of Fifth Gear have tested a GBP 165,000 simulator that is used by multiple Formula One teams to help develop their real life racing cars. Driving force behind these simulators is the Dutch company Cruden that we all know here at RaceDepartment as the creators of the popular free to play simulation software RACER.

Powered by ten pc's this machine simulates a F1 car with 98% accuracy. I was lucky enough to test one of...

Continue reading the Original Blog Post

 

[Mr Whippy]

The thing is Davesta, the Cruden sim is only as good as the content and motion cueing you use.

Those both vary depending on what the driver/user wants.

Any motion sim is flawed because it doesn't really generate the real forces, it just tries to articulate the appropriate ones that are important and physically possible, for the needs of the project at hand.
Tipping forward to generate decelerative g forces requires a pitching force that isn't there in real life, and the linear jerk to generate the same force can't last long vs a real 3-4g deceleration force in a braking zone over a few seconds! The sim would have to move 100 metres or so haha!


I think half of your problem with the Cruden sim is lack of understanding of what it's designed to provide you with, and the content to deliver that to a quality that real high end users will have access to.

Most of the really high end teams that use the Cruden sim have the physics done externally and Racer just provides visuals and the motion cueing elements for the sim movements. Everything from the engine to the tyres is done by by the client and no one, especially students who will run off with such IP to other teams, will have access to that.


So to judge the Cruden sim on how well it'd work for the high end teams, when you can't have experienced it in those situations, is a little unfair.

Since the Cruden sims are probably at their worst with random content at a university I think it's unfair to judge it on that vs iZone where they are charging people to use their kit and so have well balanced and well made commercial content.


All this info is gleaned from the videos on their website videos and my knowledge of racer.nl top to bottom, almost haha.


I bet if you had access to a really well made track (visually and topographically), and a super detailed Racer physics car model you'd be much more impressed.


Does iZone use rFactor engine on their sims?


Dave

 

[Davesta]

.

 

[Mr Whippy]

Yep I guessed it was rFactor hehe, you can spot it a mile away somehow

It is hard to judge stuff, but all considered the simulator hardware that Cruden offer is amazing. The Racer software is ultimately limited, just like the rFactor software... I pull my hair out every day because we don't have true wheel dynamic kinematics, tyre lat/long spring/damping variables, proper awd, quality mapping (ie, 3d table generation with input/output capability like OEM's use for say gearbox shifting logic tuning)
BUT, I know rFactor is full of holes too, gaps, good old bugs, areas you have to fudge around, etc etc... I know the kinematics is good but most people still fudge in unreal suspension rather than using real measured linkage locations... just like we in Racer fudge in some real life effects into hard-coded variables because we can't properly tune those real effects.

But once those are brushed aside, the raw hardware that Cruden offer tied into the external physics has no reason to not be absolutely amazing in every visual and physical aspect if the client wants it to be.
Problem is, mere mortals will never get to see or experience it, or probably even be told that it actually does exist hehe

Dave

 

[Davesta]

.

 

[Mr Whippy]

Racer's biggest weakness is it has never been 'finished' wheras rFactor has been.

Racer is perpetually changing and growing, but in amongst that mix some features have never really been finished.
Ie, the current castor/kpi system in Racer frustrates me because it really should just be angles and offsets and that is all it needs to be.
But right now we have a half-implemented angles system that isn't functional, while still having to define the scrub radius and caster trail, and the inherent camber with steer with linear values that we calculate out.
Simply defining the angles should cover these issues right away. If you are unaware of how to do that elegantly, or using legacy cars without good values... or any values, then most cars will feel really quite pants.

But then rFactor has a fantastic system for joints based suspension, but even pro-made content I've seen has terrible kinematics that are completely stupid.
Since rFactor has no proper dev/debug environment that I know of then developing good content becomes a task of making your own tools in Excel or similar (like Niels has done), and making sure that everything stacks up there as there is no way to do it in rFactor.


Pros and cons to everything I guess.


Racer's biggest weakness is that it never needed to be finished, so no one ever made finished content, so you've never experienced a well-rounded package


Hint hint Ruud... please start finishing up some of these essentials for us... they have been half done for far too long

Dave

 

[Edward Raprager]

I know this thread may be a little old but I just couldn't resist!

As a Student at UoH who uses this simulator daily, I can say that for the purpose it is intended it does a brilliant job! We develop our Formula Student car on it and it provides us with invaluable information we couldn't achieve even going and taking the car testing.

I agree it doesn't have everything, NOTHING does! But it does everything we need it to. We are not an F1 team, we don't need 99.999999% accuracy. What we need is a tool that allows us to learn and predict how our car will behave to get us close by the time we reach the first test day. It is mainly used for the areas of the car that are not tunable. We will always carry extra springs and spanners to fine tune the balance at the track.

As for the motion. It is merely a cue to help the driver feel the limits and reactions of the car. It is by no means meant to be identical to the real thing. Sadly we don't run the Cruden to its full potential either. Mainly for safety but also because there is no point. As it is it does its job very well there is no need to chase performance on the simulator when it provides no extra performance on the car!

And as a member of UH Racing I can say that we could not do our job if it weren't for the excellent support we receive from the rest of the University, even beyond the Engineering Department!

 

[Mr Whippy]

A great post there Edward!

I agree the Cruden sim doesn't do everything, but it can do enough in many different areas that it becomes hugely valuable vs static devices.

Just the fatigue element alone was what amazed me, so for example if you were a race car driver getting familiar with a track, just knowing how hard you may have to work vs another circuit would be very valuable, while also avoiding gym membership fees

You can quickly see why they are good value for money when you consider ALL the things they can do.


I'm really excited to see what happens with the Oculus VR. I guess Ruud is too which is why he added support for it early on.
Combining the enclosed visual world possible with VR, with the g-forces and motion could be very impressive stuff!


All we need now is a way to feed the correct signals into our inner ear sensory organs to 'fake' the motion and we can get rid of the simulator