Ok,

I'm calling on the encyclopaedic knowledge of this forum

I remember watching a youtube video that someone had posted on this forum, I cant remember who as it was years ago. Many many years ago.
I have googled and searched on youtube, to no avail.

The video was relating to the complexities of simulators, and how a real time racing simulator is more awkward than any non realtime simulation.
There was also talk (IIRC) that because cars have 4 wheels (unless ive been driving it!) it becomes much more difficult than if there were only 3 wheels.
The video was an interview with a chap who (IIRC) was a maths or physics professor working for microsoft and it was quite along video lasting about 45 minutes.

I have had many heated debates with people and have mentioned things I saw in this video and been told i was talking "hogwash" as i've never been able to produce a link to it.

If anyone knows the video I'm referring to, please can you provide a link. It was fascinating and like to be able to watch it again and show some of the naysayers who think that we can get perfect simulations "as PC's are so powerful these days". I'd like to be able to point them at the video and let them listen to the words of someone who knows whats really involved in simulating a car and tyres.

Thanks guys I'm counting on you!

Ah...thats the one...

You sir are a gent, thank you.

Time to get my mind blown again!

Video might be hard to find without more details.

Quote from gandlers :

The video was relating to the complexities of simulators, and how a real time racing simulator is more awkward than any non realtime simulation.
There was also talk (IIRC) that because cars have 4 wheels (unless ive been driving it!) it becomes much more difficult than if there were only 3 wheels.

This hints as if the video was about some fundamental problems in physics simulation. Collisions between objects can be calculated with vectors, newton physic laws and all that:


Problems appear when multiple objects are in contact with each other.
For example in which order you calculate the new position/velocity objects can influence the outcome.
Or the problem has multiple solutions.

Another thing is that computers calculate physics in "steps" (for example 10 times per seconde) and thus objects move in steps too. This can cause them to clip into each other in "impossible" ways. Even if you calculate a million times per seconde this problem never goes away, it just gets smaller.

This is is not limited to car simulations, it has to be considered even for seemingly 'simple' problems:
http://www.euclideanspace.com/physics/dynamics/collision/practical/index.htm
http://www.euclideanspace.com/physics/dynamics/collision/multiple/index.htm
https://en.wikipedia.org/wiki/Three-body_problem

/edit:
Ok, I did not expect that someone would find a video based on that description
Maybe the links are still interessting if you want to read more into it.

Thanks gutholz

but bbmman has saved the day, that link he posted is the video I was referring to.
I'll follow those link you posted, as everydays a school day.

Thanks