Automobilista 2 V1.1.0.0 RELEASED - Now Updated to V1.1.0.5

Hard to forget Sunday afternoons, and Murrey Walker could be on crewchief!

Maybe i should try a variation on my synth to avoid royalties. ha sry not that good yet.

I want to drive but the lady of the house is in my room . dam pirates of the wats sit caribibian

 

I thought the biggest change was in the Formula Vee. It went from a completely locked diff to the open one we have now. So same as the Cat, but the transformation and meaning is just more dramatic in a tiny open wheel car that is supposed to pivot around its center depending on power/coast.

Oddly before the open diff rehaul, some other open diffs like the Trainer and MCR2000 were way too shopping cart like. Definitely not locked. In fact they seemed to lack any friction and resistance to tires spinning at different speeds which open differentials still have. It's very similar to this situation with clutch LSD where it just seems to work different from car to car.

I feel the same way and certainly many people love the Caterhams. They are very nice to control on throttle... but it's missing the other half which would complete them. As you said the Academy is already fine as it doesn't use the clutch LSD.

I have no idea where I read or heard this so it might be way off, but apparently the differential simulation is complex in Project Cars 2. It could be replaced with a simple formula, but clearly SMS was hungry with the physical SETA tire model so I can see the differential also not just being a simple formula either. If we're still using the same if modified LS differential implementation that is...

And actually is it simple to simulate a salisbury type clutch differential that works with this many cars? Changing the ramps, number of clutches and preload needs to mean something and not just be a complicated way to end up with simple locking values like "50% locking power, 10% locking coast".

First thing to understand about the current LS differential state is that some cars combine the worst of open and locked diffs. They stay locked for too long when turning and open too easily when accelerating. It's madness.

1. Could the first dip be during trail braking? The inside tire (right rear) is unloaded and starts slipping a lot.
2. Then you get back on the power for a bit approaching the second part of the chicane. Diff is locked again so no inside tire slip unlike with open diffs in a situation like this.
3. Then you brake again for the second part of the chicane. The opposite side (left rear) is now inside and unloaded. It starts slipping. This is the second dip.
4. Finally you accelerate out of the chicane. It's a pretty tricky and long exit from slow speed. The left side tire is unloaded as your car leans on its right side. It starts slipping. This is that last peak.

 

A posit: didn't the diff situation get changed a few months back? I am sure things improved after that.

Hi, I don't mind such discussions. You know car manufacturers have been mulling over these things for 100 years.

The thing about complexity of which I am sure you are aware, is its nuance. IF you have 10 spokes you can adjust each spoke on a wheel in relation to how sturdy it makes the tire - so lets look at making for example simplifying his graph....

hold on because the graph lacks context.

The trouble with such a thing is you can't go in an alter the tiny bits if they do not exist and NOR do they go wrong if they're not there 'to go wrong'. In computing and simulation, where they are mostly limited by CPU power, they can do the complexity so long as they are willing to work with it.

So the question would be how useful is that to the engine, and did the madness engine consider this complexity.

If you look at AMS1 its easy to see that generation (of which AMS1 came out at the end) had a narrow scope of the things they would simulate, compared to today, and impoverished models by the same standard compared to what is now.

Meaning - that while what they have now is far more advanced, it would not surprise me IF what exists now is out of convenience. So how much does complexity help here and how would they be able to work with it, AND is it materially any different...in terms of outcome, to warrant such complexity.

^ SO on that point - just like transmissions can skip a bit and blip etc, mechanical things do not always perform exactly as they should...same with differentials.

Sometimes streamlined is good. I think in any case the current system is doing what it needs to do - and what advantages there are to the user-facing side of it, are probably not as many as you may think.

So I think his little oscillation in the signal is not worth worrying about - he cannot isolate it now.

Feature Column from the AMS.

 

Let's discuss it in the right place guys.
Ops... I forgot it's just for beta.

 

Ooh this is what you were wondering about. I'm none the wiser.

However this "grabby diff" phenomenon explains so many comments from different drivers, including myself.

1. When you lift (off throttle), you expect the differential to loosen up and let the car turn more. This is of course multiplied by the increased front tire load due to weight shifting forwards. If the tire can turn the increased load to more grip that is. Well the differential does not open at all when lifting so no liftoff oversteer.

2. When you turn in, you expect a certain response. But you're not quite getting it so you turn in more. I believe this is why some people say AMS2 needs more steering input than expected. And then suddenly the diff opens. You don't gradually get more oversteer. You get all of it at once. And your steering input is too much for this sudden balance change. It's easy to spin at slow speeds when this happens. It feels sudden. I have also seen this happen with the F-Retro at high speeds, mid corner. A very nasty situation indeed.

3. When you trail brake, you except the car to respond with more turn in. But like with point 1, you don't quite get what you expect. The weight transfer does help, but not 100%. So you adapt by trail braking more. The diff snaps open and now on top of oversteer from weight transfer, the inside tire is actually dragging a lot. You get a lot of turn in, because your amount of trail brake is just too much. But you needed it to break the diff open from its syrup filled sticky state.

It just says a lot that none of these things happen with the MCR2000 for example. No sticky clutch LSD, no problem. It's easy to modulate steering input, throttle and braking, because you have that understandable gradual response to everything.

Ok seriously, let's open a thread for this as this shouldn't be a public secret

 

Umh... have a closer look...

Physic discussion thread

 

Yeah what about the latest hotfix. This is a great thread for that. So what they design a game where the cars are meant to be driven fast and thats how the diffs work. Its ten times better than March.

Trying to make it all about mechanics and engineering, is tedium.

There's a lot to be said for placeholders. There will always be rats/bugs/discrepancies in physical things - look at the theory of relativity (or was that gravity lol) guys there's only so much one can drill down to before the picture falls apart.

They will just replace values or systems in 3 months and it will be a moot point.

 

Oh believe me, I love AMS2. Even the cars with most clutch LSD issues. I'm just looking forwards to loving it even more when this dark corner is fixed. And as a bonus it should end the "no liftoff oversteer in AMS2, Madness is a garbage fire" rhetoric. Although it's the internet so maybe not.

 

And I agree with you and am not wanting to invalidate the arguments. I just wonder if the issue has first been tackled at indirectly for a variety of workflow reasons.

I just think while technical aspects are needed for a title like this, that the systems would also be built for tolerance of a wide variety of driving styles and vehicles - so its probably less acute than one may expect. At least for now.

Separate note: The same system that did that was also responsible for a lot of exhilarating driving being done lately. With the right ffb you can get a good response from the throttle through the wheel, in the right cars you can feel the power through the wheel on take off, launches feel and look and 'go' better than ever...

and its not just the diff doing all this, as it comes together - for a game, there's values/constants/coefficients etc,

but there is also sight, sound, feel...those are tools too. The game is ultimately meant to be enjoyed. Its not mean to be a mechanical workbook. Ok on one hand it kind of is and would be nice, yes! But ultimately its meant to make you forget you're at a computer and immerse you - and thats 90% of the puzzle, mate

This patch shows me the challenge is now making things standup to some edge cases and further scrutiny - so its my guess the codebase is ready for such a thing... the game may not be a resounding success with about 800 max concurrent per day, but with what we are seeing now I am sure it can iron further things out and halt this immediate discrediting sometimes is heard from certain segments.

So while it may not as of today have the cred - objectively speaking, but for different reasons than ACC, I don't see why one would not give this title at least 30% of their time if they played ACC for 30%. I guess in the coming months they may owing to features appearing now.

 

Question:
What happened to the showroom, why are the loading times so long now?

 

No lift off oversteer and no full throttle oversteer...

 

Anyone else dislike the raised bumps on the green part of the curbing on spa? No other sim (except maybe PC2?) seems to have this, and I dont see it in real onboard footage of spa either.

 

The madness engine has seen many iterations. They will get to it. BMW challenge on madness engine probably had super naff differentials haha

I spent a good deal of money on this setup - almost 2 thousand dollars - and I do not feel jipped by this game or the others like Rfactor or ACC. So I totally understand you when you say about differential and being close to the actual thing etc. I often adjust the dif settings, depending.

I am pretty sure they will get to it. What I am seeing is a workflow thing, not a 'brain drain' thing. The differential needed to be a repeatable results thing - the guy fronting the project said that basically (iirc) a few months back.

This helps them out the diff thing, its allowed the game to do other things in that time. They will get to it.

 

What is the problem with that?
If it is too bumpy do not drive over it... I think that is the reason why they are there.

 

Mostly the fact that it doesn't exist on the real-life track. They bumps dont seem to affect stability even if you drive over them anyway.