That's a really bad video, biased, ironic, smart ass style. I don't like it.
I have enormous amount of admiration and respect for everything the ISI guys are doing. I'm in contact with some team members too, great guys. So to have to comment on such a bad video that puts rF2 vs AC is the last thing I wanted to do and so I will try to not mention rF2 in my analysis. I think rF2 is doing a great job in that situation. I don't think it's perfect, but who cares, it's a racing simulator and it does a ton of other things perfectly. That's all I have to say on rF2.
Please allow me to defend AC though.
First of all. It's not like you put two real cars on the same track/bump and test. No, you put a virtual car on a virtual circuit with some virtual physics, against another virtual car on a virtual circuit made differently with some different virtual physics...
I can hear you scream, "the results should be similar". Well... no.
The most important thing is the circuit. It's evident by naked eye that the bump kerb on our version is quite higher. Add to that, the grass side of the bump is lower in AC, the difference is quite important. Strangely all the test are made from that side of the track (grass to bump to asphalt) and never the other way around... wonder why.
This alone is the most important fault in this whole video. A comparison of two optically similar bumps, that differ in height as much as a normal sidewalk.
Other details.
- Why compare an old late 80s street car with whatever dampers they used back then, against a modern panoz with top of the line dampers, much better chassis, geometry, inertias and so on? Why didn't he choose the X-Bow for example. It's true that even our X-Bow will jump on that kerb but IMHO in a much more predictable way. Remember our kerb is higher and more straight. I'm pretty sure of course that somebody will try hard to make the X-Bow to jump in an equal way... but whatever.
- There is a bug in the current version of AC. I'm sure most of you know. The body of the car goes through kerbs without colliding. It is evident in the video too. This adds to the strange behaviour in many ways, i.e. car does not jump very high at higher speeds (suspension absorbs everything)
This bug was not there in older versions, should be fixed in the next update.
- We have a resolution of 6 vertices on that kerb. This is very low, the tyre goes bumping on practically a straight vertical face from the grass side.
"what? wow you're laser scan the track and then use 6 vertices for a kerb?"
No. We use millions of vertices on the actual track and normal kerbs. On a kerb like that we decide to make economy so that we keep vertices number logical and gaining fps. Why on that kerb or similar? Because that kerb's job is to keep you from cutting. At 6 vertices it does so even better. Decision and compromises. I'm sure rF2 has its own.
- Our low speed physics (under 3-4km/h) sucks. Yes it will manifest in such occasions. What happens is that under that speed the whole suspension/tyres/collision pretty much everything movable, is damped in extreme values. That's why the cars do not bounce nicely when stopped for example. That also means that everything becomes so stiff, that you get those nasty rebounds and jumping at an extreme situation like that.
Why we haven't fixed it yet? We do a racing simulation, such as driving at high speeds against other high speed cars... things like that.
Yes we know that when you park your AC car on a bump it seems silly, but do you prefer we lose a week or two on that, or making better multiplayer? I'm sure people complaining about this will be happy for 2 days of bump riding and then forget it forever, but let us know.
- Do you know most of the bimmers (especially M3's) work on their elastic bump stops most of the time? That design decision makes them comfortable at normal speed cruising and much stiffer when pushing. Being the bump stops much stiffer, means that in extreme situations (such as this) the elastic energy cannot be damped by the dampers quite well. Bouncing is what you might get.
Dunno how the panoz work, can't comment on that.
- Have a re-read of what
@Stereo says in his analysis. It's good and can make you understand some more things.
I could go on and on but I won't, makes no sense. Tomorrow, an existing sim or a new one might do off-road surface displacement modelling. Will this make rF2, AC, iR and all the other sims instantly "not real sims"? Heck I thought we were racing on the asphalt.
AC brought Laser scanned tracks, movable aero parts and whatever other features we added... Did all of these made other sims "not real sims"? Really?
I'm a bit pissed off with that video as you can see
But I shouldn't because fortunately there are guys like you that can discuss nicely about such things and I thought I owe you some explanations. But yeah, buy AC because arcade