Automobilista 2 V1.0.4.0 Update Released, now updated to v1.0.4.1

Did they mention the other 50 things not completed?

AI calibration takes a lot of time...when things like the default tires or sounds or tires showing through the sides of a vehicle are not done, you can be very sure the AI calibration is not done, either. There is also no point in doing the AI speed calibration until you are sure the car is "final" in terms of physics, lap times and performance corner-by-corner and straight-by-straight. Otherwise, you have to do it over again when it is finalized.

 

September update today?

 

Finally got to try out the new update today and jumped straight into an online race in the M1. So much fun to race! Thanks Reiza - this game just keeps getting better and better.

 

Since its so nice to read a positive post
I have to agree the car is "fun to race" and glad to hear you had a good online race.
Just my thoughts. I find the car very responsive and more so when pushed hard, giving a lot of feedback as to what it will do, and does not what do. Even in default setup it can raise a grin. But it will bite, though is easy to learn from mistakes. Braking heavy for too long or too much gas on exit can punish but it can be hustled nicely with slight drift and power correction. Default seems very stiff and low esp on dry Historic Tracks like Interlagos Historic long where a setup change makes a difference. It can however run round there at an impressive pace.
A handful in very wet conditions as it should be for such a car. Currently my favorite I will be running online test races for a while in it, and trying some less familiar tracks that suite the machine.
When the well reported too slow AI in this class get a boost racing with them will be better also. As a work in progress choice it a great machine to pick, learn, and have some old school driving fun.

 

I am not the best drive, you could call me slow and untallented. Dont do setups verry wel......
But boy is the M1 fun. I also think the GT4 isch Ginetta is verry good fun. Just doe the record.

 

Thx a lot Reiza for the partial fix of the world movement, it was so difficult for me to drive F1 cars for example due to the extreme movement of the horizon until now, actually didn't drive any F1 cars for quite a while. Today I tested for the first time the Mclaren car from V10 Gen 1, all I can say great work, really amazing car to drive, I only wish you could find some time to optimise the performance because this car has stuttering issues at Snetterton.

 

Imagine all 11,242 forum members owning and playing AMS2! I wish that would happen.

 

I did some more testing today with the Formula Retros. I still cant touch my old TT times at Bathurst that i set with the geared differential. Note Bathurst is a great testing ground because it really punishes understeer, and has a lot of crests, which really upset the F-Retro cars, so when I try to build setups for these cars Bathurst is my starting point. If I can make it drive okay around Bathurst it can get me a good starting point for everywhere else.

Using clutch diff:
Lotus 72 best time: 2:00.3
Mclaren best time: 2:01.5

Previous times using geared diff:
Lotus best time: 1:57.9
Mclaren best time: 1:58.9

I cant even get close to my old times.

The geared diff physics model approximates a clutched limited slip differential far better than the in-game clutch LSD model.

To the Devs, you can drive these cars with the geared diff, run the power bias around 5, and the coast bias around 2-3, and you'll see they behave SO much better at corner entry. They're a pleasure to drive this way.

When you go to turn with the clutch diffs, you can feel the diff fighting the car until it reaches breakaway torque, and the the car turns in perfectly. Real clutch differentials behave somewhat like this, but where AMS gets things so right with transient tire traction, the diff model gets it so wrong for the same reason. The breakaway needs to happen FAR sooner and be far smoother. The kind of sudden snap break we experience makes sense with higher preloads and steeper ramp angles, but with 0 Nm preload 2 clutches, and ramps at 89 degrees, there should be almost ZERO lockup between left and right tire. We should be able to tweak these to behave nearly like an open differential, or like a spool, as that is how these differentials behaved in real life. That's WHY they were used, the extreme adjustability.

With the F-Retros and their gigantic rear slicks, short wheel bases, tiny front tires, and light weight and low aero grip, they are a perfect test vehicle for differential testing. Or alternatively, they are a worst case scenario for differential modelling problems, as they exaggerate everything.

Here is some testing I did with all diffs set as loosely as possible (ramp angles as close to 90 as possible, preload as low as possible, and clutches as low as possible).

This is The "Grass Test". It simulates one tire with traction, and one tire that is slipping. With an open differential, all power will go to the wheel with less traction (This is why you dont use open differentials in racing unless you are not in danger of breaking traction). With a very, very loose clutched differential, it should behave VERY similarly to an open differential, but you may have a little bit of initial bite as the chaotic shock loads work their way through the system, but after that stabilizes, there should be an OBVIOUS redirection of power ot the slipping wheel, and the car should go straight, even though half of it is on grass and the other half on pavement. If the car consistently tries to drive off the course, then the tire on the pavement is trying to push the car around and torque isnt getting split. This is how other sims compared, FEEL FREE TO DO THIS ON YOUR OWN!

rFactor 2: Very obvious redirection of power to the slipping wheel after initial hit. Car goes straight.
Project Cars 2: Very obvious redirction of power to the slipping wheel. Car goes straight. INTERESTINGLY, preload at zero behaved the same as AMS2. I had to set the preload to an imaginary -100 Nm (As low as PC2 will let you take it for this car) for it to behave the way it should.
Assetto Corsa: Very obvious redirection of power to the slipping wheel after initial hit (Couldn't use the 312t since AC doesnt allow a zero preload on it, but it will on the 312 '67)
AMS2:
A) Formula Retro: Both tires spin, car consistently violently veers off track as the spinning tire on pavement gains traction eventually.
B) Formula Classic Gen 3: After initial bite, pavement tire does NOT spin, and the car goes straight! In this car it seems to work near perfectly... And you can really feel it if you try to drive it around. I believe I spun the car in the corner leading to the straight to do the test due to the diff being so loose, so you can probably even see that below:

I dont know why youtube forces this to be such a low quality video. I have a full quality version on my google drive where it's easier to see some of these things.

Oddly, it appears the F-Classic Gen 3 behaves much closer to "normal" compared to the F-Retros. I couldn't begin to guess as to why, as I would assume this should be a global behavior. But the tire variables differ between them so this may be expected.

Now be aware, this video was from BEFORE the update. I havent had time to try again with AMS2, and I can tell from driving the cars that it's GREATLY improved. But it is still not good enough. We are maybe halfway there from where we started (again this is all in reference to the F-Retro).

Note the extreme dartiness of these cars over crests may also have something to do with the diffs being too tight. They behave as if they have spools, whereas in real life if you have one tire in the air there really shouldnt be that much torque flowing through it as the car lands or re-loads on a given tire. It should be eager to direct some of htat power to the slipping wheel with very loose settings.



I dont mean to belabor the point, but Im not giving up on the F-Retros. They feel absolutely AMAZING 95 percent of the time. It's ONLY initial turn-in and crests that make these cars anything less than perfect, but unfortunately it makes them nearly undriveable.

These cars are so good you guys! They're so amazing to drive! Please keep working on the differential and the F-Retros will be my favorite car in all of sim racing!

OR.. alternatively, can we get the geared differential back? The car drives SO much better with it. It's such an arbitrary restriction that, frankly, ruins these cars considering that's the major dealbreaker with them. It's a simple solution, even make it temporary until the clutch diffs are completely sorted.

Also, the M1 Procar might be the best car I've driven in sim racing so far. I absolutely love it!

 

Great post!

 

so, should we expect this month roadmap?

 

this silence is a little bit unusual, I hope it prelude a new wave of good news

 

I will say, this does feel slightly unprofessional, stating the date and here we are two days later with not a word uttered since.

This isn't "they missed a deadline", I understand those are changing goalposts, but doesn't mean they can't communicate an update on one of their channels to say it's been delayed for X, Y or Z reason.

I very much bought into this game because of the comms and transparency of the development team, so the last month in total has been a little disappointing. I understand they're busy with the game and they've delivered a lot so far, so they've certainly earned a free pass for a month (and some studio holiday if people are taking it!) but just a slight niggle.

 

This has got to be the biggest update I've yet seen. The more transparent they are, the more everyone complains about everything. The less transparent they are, the more everyone complains about everything. I feel terrible complaining about the differential issue above for this reason, but I just believe in what they're trying to do as I think this can easily be the best racing sim out there if they get a few things right. But after such a huge release, give them a few days to take a breather.

 

Strange reason to buy a game.

 

As I say, not talking about the update, talking about saying "5th October" for something and not even posting a "it will be a bit later in the week" post. Not about the patches, not about the blog content, not about deadlines - I've not issues with anything there (and those are the important things).

I'm confused by that, I don't see that at all - but I may be relatively new here, so maybe there's a history I'm unaware of. If you're in the beta forums, maybe it's different, but publicly surely we can agree there's been fewer comms for the last month than in usual months?

I'm sure most people playing AMS2 have bought into it for its potential at some point, especially the early backers such as yourself. You've bought into a belief of Reiza's product and their record as a company at producing things that you want to play. I'm a later adopter, but presumably we can agree that the game in its current state is more of an unpolished diamond than the finished article?

So a decision on buying AMS2 has been very much based around my 'buying in' to Reiza as a company; they've got a great reputation and a track record of doing the things they say they'll do, so the vision they've communicated for their game's future is very much a selling point, I'd argue.

---

People seem to think I'm having some big complaint here - to clarify I'm not saying there's anything I'm particularly upset by, just a little disappointed that they've given a date for something and not communicated it changing (I don't mind the target being missed at all).

 

It is highly unusual and a bit worrying that there has not been a clarifying update about missing the blog update on the 5th and the expected bug fixes to the main update, but as someone mentioned above, I am willing to give a free pass based on the history of what this company has produced and released.

I am 99.9% sure that Reiza didn't just decide to take a spontaneous holiday after releasing 1.04 with many obvious issues. Renato already signalled that there would be a patch or two to correct these. They are either working flat-out to make sure the patch is of top (better) quality, or, something critical and untoward has occurred that is preventing both the patch and communications.

Unfortunately, they are in a no-win situation with social media and fans. Miss a release (such as a delay of 1.04 to better polish it) and a whole portion of the audience goes crazy with concern and anxiety. Release WIP stuff, which used to be appreciated as a chance to try things out and help fine-tune them knowing that Reiza would never leave unfinished work unfinished, and get shat on here and on YouTube for releasing things that are not fully polished. If I saw what was happening, I'd just tune it all out and try to focus on what I was doing. Renato has proven to have the patience of a saint over the years, so I am sure it is one of the two scenarios I described above. Either one is a fully acceptable justification in my opinion, albeit unfortunate.

 

That was me too, for anyone thinking I'm having a go

 

Maybe just hoped to get smth signed to be able to announce it and there's been a delay with the signing. Perfectly reasonable possibility.

 

Yep, completely reasonable. But also fair to say "we're having to delay the blog a bit more, we're trying to get some approvals sorted before publishing".

Codemasters did that a couple of times with Dirt Rally blogs in the past.