Physic discussion thread

I wouldn't change the championship or default at all. I do understand that it's like this in real life. But random online races could really use an "alternative regulation" if you will. Teams may weigh the turbo question when considering a whole season with altitude differences. Random races are one off.

But this discussion we're having seems more part of a bigger online gameplay discussion and not physics . Fair comment from you, but maybe we should leave this subject altogether.

 

and Im not sure how my post on this video subject ended up here? I responded on Alonso thread not here. Sry if my error, but not down to me as I recall. I know renaldo closed a vid thread completely this evening, and I dont blame him
EDIT looks like stuff got moved here rather than deleted

 

That was about my turbo balance of power discussion which turned into a bigger thing and not your posts .

Ralonso's thread now redirects here. I think that's why you're confused . And this thread was moved to AMS2 / General from General / General.

One more thing about this. The Vee and both Trainers are simply fantastic fun to drive. Vee feels like a clumsy budget race car while still being exciting to drive. Trainer is pure open wheel bliss. Both cars transfer weight nicely affecting over-/understeer in a smooth, predictable and nimble way. I read a real life Vee guide once that mentioned them stabilising on throttle. Certainly is the case here with both of these cars

I think we can live with some positive feedback once in a while .

I really hope we can get the F309 LSD working as well as this while also having throttle oversteer. Like in AMS1. That will be amazing in VR.

 

Best get up to my room and have some fun (online beta)
Agreed about Trainers they were a blast 2 days ago. old school skills communicative
have fun.

 

I had a go in this horrible little thing (Can you tell I'm not a fan? Lol) at Oulton as you mentioned earlier.
Although I don't like the 'Overall' handling of it, I think it actually feels quite good on the front end when running over crests, goes light just the right amount IMO at all parts where I'd expect it to.
I think the issue is that feelings like this can cross from 'Physics' > 'FFB' etc very easily...
Hence glad of the graphs/data posted by others here (Would love the time to myself) even if some on the forum don't seem to understand/appreciate it.

Regards it's other physics, I feel like there isn't quite enough understeer at the extremes (I massively dislike understeer but I do feel like it's missing a little here).

 

@CrimsonEminence Most definitely ha.

 

You mean "tippy, tail-heavy, open differential, under-powered cars with 'bad' tyres"

 

It’s a lovely piece of scrap metal and bug parts . But yeah fair enough, I can also see how it’s a hateful car.

No graphs this time as there wasn’t anything clear to show. Maybe correlation between front suspension travel, steering angle and lateral G. But even then I think it’d miss a variable to possibly pin point any kind of issue if one existed.

I have no comment about the right amount of understeer. I’ll leave that to people who know tires and such. To me 1.0.4 was a milestone for this car, because Reiza fixed the open differential implementation in that update. Vee used to be a spool car by accident. That was an interesting car to drive.

I hate understeer too when the car is on rails and you have to start exit acceleration super late. But I do like a car that has as much (off) throttle steering as these. It’s like another steering wheel. Having the ability to also throttle steer during exit is the dream.

Don’t forget the rich man’s 911 engine!

 

Ok, since my post was moved here for some reason and I didn't get an answer, let me ask different question: can you guys please check and report back if BMW M1 Procar (or any high powered car) behaves in a way I described? I'm not sure if it's something on my end, or is it the way AMS2 runs?

It doesn't seem right that a car that has close to 500hp has trouble breaking traction from a dig or low speed.

 

I've actually wondered to myself how one would go objectively stating that certain cars have "better" physics models than others. Subjective evaluations like "I like how this car feels" don't feel accurate since force feedback, driving style, and preference towards certain drivetrain and engine layouts can bias them.

Are there basic criteria for different components (tyres, suspension, drivetrain, aerodynamics, brakes) that one could use for evaluation? Or at least state that certain cars don't act according to the basic rules of racing cars?

Personally, some of my favourite cars in AMS2 tend to be those that allow some amount of controllable slip while still requiring finesse to drive (FTrainer, Procar, Formula 3, Cat Superlight), since those cars "feel most at home" with the Seta tyre model. But that is no guarantee that those cars are well-modelled, physics-wise.

 

The latest update rendered some cars’ differentials quite open on acceleration which stabilises the car rather than spins it. All part of the ongoing clutch LSD mystery we hope will be solved soon. But it should still burn the other rear tire.

You should be able to spin the CLK GT1. If not, there’s something wrong with your installation. A few people complained after 1.1 release that their cars were impossible to spin. If so I would recommend verifying files via Steam and deleting the Ams2 folder in your documents (back it up first in case you need to check some settings etc).

 

Ahh that is not fully correct.
Eventhough you can interpreet this "completely no braking" from my slightly humorous description then I have to brake (hard) as example on Spa after the long straight.
But even there I need to end braking before I brutally shifts down 4x to get the car to rotate first right and then left - still in too low a gear to be able to use the coast oversteer technique.

ByTheWay: The only other place where I use brake on Spa is just before the bus stop - where it is even more crucial to use the end brake + brutal shift down (5x ) to get the car to rotate L/R.
Hehe or run directly into the wall

EDIT: Sorry I completely forgot that I also have to brake HARD at the 1st hairpin La Source corner. But here it is even more urgent to use the described special BrunoB Brake Technique

 

I did this, because I wanted to find out how good of a test it is. This is not meant as extra evidence, but as testing of tests. Something to do while is doing their thing .

Car: Boxer in AMS1 and Cup 3.8 in AMS2
Track: Londrina Kart One

First impressions are that this does really test slow speed differential behavior. You can tell that response to off throttle is more immediate in AMS1. I don't think this has to do with actual response time, but more the fact that AMS2 clutch LSD sticks closed for a while in a suspicious way as we've talked about before.

Driving styles are also quite different. The 3.8 feels more slippery to me. But also just the simple fact that the Cup 3.8 lacks turn in, but is maybe somewhat easier to control in a slide.

Boxer in AMS1
I put the playhead where I think it really helps that the rear is open. This might not be the best approach to turn in this early, but it's also not the only corner in the world. The rest of the lap demonstrates how compliant the rear (and LSD) is. Rear tire speeds are different a lot of the time meaning it won't resist rotation off throttle like a locked differential would.



Cup 3.8 in AMS2
Very different story here. It almost looks like the differential is somewhat working, because rear tire speeds difference grows near the dips (around apexes). But this track is so slow and we've seen in earlier graphs that rear tire speed differences only grow at slow speeds. I would say that they stay identical at over 100km/h no matter what you do. Be careful when testing at very slow speed / tight tracks. It might not tell the whole story.

There is a lot of one tire fire though! I've felt that since the Cups were released. This is maybe the result of reduced locking per clutch to improve the situation? But since the differential likes to stay locked mysteriously, it mostly just increases one tire fire / understeer during acceleration.

Have a look at where I placed the playhead. This is a typical "high speed" (I mean it's barely over 100km/h, but still) lift to help rotation in order to make a a turn without braking. There is zero difference in rear tire speeds until much later when it's way too late. It's actually important to look at the graphs here rather than anecdotal evidence, because the 3.8 Cup is quite eager to turn at higher speeds anyway. I don't know how or why, but it masks this issue pretty well.



High speed lift in AMS1
Lets look at a common driving technique where you approach a very high speed corner that's a bit too tight at full throttle. You lift to help initial turn in giving you a better trajectory for the corner. You will barely make it without running wide.

Look at the playhead. It's T1 at Ibarra. 230km/h and I can't make this turn at full throttle. I lift for a tiny moment (look at the throttle graph) while adding steering input. The car rotates and I'm on a trajectory to make the corner back on full throttle. You can see that rear tire speeds are different right after I lift. Quite a lot in fact, but even a little would help.



High speed lift in AMS2
First I want to note that Cup 3.8 has better grip in AMS2. I can make this Ibarra T1 at full throttle. I could find another track and corner where this is an issue. Spa is full of them. But lets compare AMS1 and AMS2 exactly side by side.

Have a look at the playhead. There is absolutely no response from the rear to me lifting. Rear tire speeds stay the same. This car does actually get some liftoff oversteer in this particular situation (no longer in T3). Probably from the weight transfer and tire load? Either way I bet it's not as much as there could be and many other cars aren't as lucky.



Finally, here's a practical example of where you do need to help rotation in many cars at Ibarra. I've driven the open differential MCR2000 here many times and that car responds well to lifting. The 3.8 Cup however doesn't open its differential one bit.



So there you go. Another pile of graphs. I would love to repeat this test in AMS2 if and when there's an update that promises major improvement to the clutch LSD .

 

Thanks for the response, although I have to say I'm disappointed to hear that.
I just checked and I can indeed spin tires driving CLK GT1.

I've also checked Porsche Cup 3.8 (I have over 100 laps in 2016 Cup car IRL, not sure what model the 3.8 is based on in AMS2) and it's the same issue. Flooring Cup car in real life, coming out of a tight corner you can easily spin and loose control, while in AMS2 the car feels like it has 80hp.

Sorry to say, but this renders AMS2 un-drivable for me in the current state. Nobody is really complaining about this? This is not how these cars should handle in real life
I hope Reiza will look into it

 

Stick with us, friends.
Porsche Cup are being worked on while I am writing this. Should be not done 100% final, still, updates are on the way and this is a way to go still... but please, stick with us and lets make it right. We need all thoughts about this. Thanks.

 

Of course we are complaining about this!
Even too much sometimes...

I asked the above test to show you the difference between something 'realistic' like the BoxerCup in AMS1 and something wrong like the 911 Cup in AMS2.

@oez
I suggested a low speed test because at higher speed a lot of aspects may be masked by aero, low engine thrust and intrinsic low speed difference between tires...
At low speed sims can't lie about fundamentals and the issues on the basics becomes much more evident...