Automobilista 2 V1.2.0.0 & Racin´ USA Pt1 RELEASED - Now Updated to V1.2.1.4

I don't think they aren't affected as much as I am, I do my fastest laps later in the race, And most of the time, my fastest laps in the race are faster then my qualifying lap.
(for some reason I can hurt my tires a lot without loosing time.)

 

It depends a lot on the format of the race week end. If you run no practice and short quali the track will become fast only during the race, so your best times will likely come there.
If you run a proper week end with some significant practice and quali you won't be able to go faster in the race. For the loss of performance during race is the same thing: tires usually have a couple of laps of best performance then they drop a few tenths. If you race for 10 laps things will be mixed because the car will be very light all along and together with track rubbering may become faster towards the end. If you have a race 70 minutes long where you load the car to the max things will be different and there will be drops in performance that are significant.
Also formula cars have much more sensitivity to additional weight (50-100kg of fuel vs 600 kg for the car is a lot) vs GT3s where 50 kg over 1300 kg are peanuts.
That being said, I still think the soft tires, especially when racing hot tracks, should deteriorate much faster and should be very hard if not impossible to drive them for one full stint (thinking GT3s, GTEs and DPis especially, but not only). Right now it's too easy to keep the tires in good shape (not mso much in terms of temperatures but rather in terms of wear and degradation of performance) even at full load with over 40C track temp while it should be impossible.

 

Well that could be the result of a variety of factors. Were your tyres and brakes up to temperature and pressure in qualifying? In my experience, as evidenced by data collated, in circumstances where variables are kept as constant as possible the ai will post faster lap times than the fastest lap in qualifying (either theirs or the players). Im not suggesting that you are mistaken, simply that the list of parameters and variables (if any) need to be posted if your input is then going to be possible to recreate and reproduce (see my previous post listing lap times, circuit, series and variables).
Otherwise this will rapidly degenerate into a he said/she said which serves no-one.

 

Is the white ferris wheel missing at Silverstone circuit in AMS2?

 

You sure that you don't have tyre wear set to accelerated or higher? Because that is a bug that has the effect you're talking about.

 

I've done a race yesterday at Brands Hatch in GTEs. 70 minutes starting from no quali and no practice so green track at the start. AI at 100 skill and 30 aggression. AI starts very competitive but as the time goes they get slower than me. In facts I pit around mid race changing tires as well and they do the same but likely take no tires because they get real slow. Towards the end of the race I am much much faster. All settings are real and 1x.
Regarding AI behavior they need adjustment in that track: at the Druids hairpin they divebomb you from more than half a second back every time. Even a slower car that I overtook the lap before found suddenly some magic grip and divebombed me from another county into the hairpin on the next lap. When you overtake them at Druids they almost always turn into you, no matter what they go for their line.
At Hawtorns corner they seem to drive the wrong way around: the enter savagely fast and park it on the exit. The result is also that if you are in front of them the will ram you or divebomb you if there is an inch of space at the apex.
Often, after they parked it on the exit of Hawtorns I get alongside going into Westfield and there, if you are not basically with your nose ahead of them, they invariably go for the apex completely ignoring your presence. If instead you are even with the AI car, they just resist on the outside drive in the grass and then jump right in front of you going to Sheene Curve as if you are not there.
Stirling corner is pretty much the same: either you are ahead (or neck and neck at least) or they will just ignore you and go for the apex.
Things really get more difficult when at the end of the race you are much faster than them.
I think the fast corners of this track are really a challenge for AI to respect the human.
Also some of the AI characters are in need of a ban as they are the worst offenders: Lovric and R. Lopez are absolutely ruthless and will punt you, ram you and drive you off the track or loop your car by turning into you from driving on the grass at any chance they are at 0.5s or less.

 

Not discounting or dismissing anything you are saying, but try the same scenario except with "Aggression" set to 85 or 90. This is counter-intuitive, but I think the issues you describe will actually improve (lessen) with a higher setting. Anyway, if you're willing to try it, please report results

 

I may try it, but I also have to report that tonight we did the real race and some players including me experienced a ridiculous amount of tire wear.
I did four test races with same date, same setup and same track conditions, 37 C, with same amount of racers in the past days. All progressions and consumptions set to realistic and 1x.
In the past days I could do the race with soft tires and at half race mark (35 minutes out of 70) I had approximately a 75% percent residual in the worst tire.
Tonight we did the actual race and some players (but not all) experienced 100% consumption of tires in less than 30 minutes. So whereas with the same settings yesterday I had 75% residual tonight I couldn't even reach the halfway mark: in a 70 minutes race I got shown the "pit for tires" board twice, once in the first 35 minutes stint and once in the second 35 minutes stint, in both cases tires lasted less than 30 minutes on a car loaded with 75L.
It is not the first time that something odd happens with tires and their consumption TBH.
To be noted that test races happened on green track that should wear tires much more, while tonight the race event was on a fully rubbered in track.
Another odd thing that happened is that I noticed that compared to my early races mysteriously tires were running at 72-75C whereas in the test races with the same track temp and the same setup they were running much hotter, like 90+. In 37C track temp it does not seem realistic in Brands Hatch to have left tires at 75C and especially it is not consistent with all tests I have done.
Frankly speaking it is quite frustrating to spend time preparing a race and then have bugs like these hamper your efforts and multiply 4x your tire wear out of no reason.
Also, at the start I got another car coming side by side and the two cars sort of clinched laterally and we couldn't take them apart until one of them was literally shot off the track laterally (mine). It was clearly showing in the replay as if events regarding the cars swiping each other were relayed with a huge delay to both of us and events for each car were out of sync. Not sure why the two cars should "clinch" either: I would expect physics to "bounce" them apart not make them stick to each other...

 

The chance of some AMS2 bug/issue coming up and screwing up your race makes it pointless to practice even for public lobbies 5 lap races, certainly not worth spending any practice time for league races.

Same frustration on offline races, since the AI physics is so artificial...

Online and offline experiences are both frustrating, what is left to do in AMS2? :T

 

Are you suggesting there is something random happening? Or did an update change something? Were all tests done online in the same type of session? Or a mix of offline/online and P, Q and R?

We know that there are some issues with the multipliers, but you said everything was 1x. I am wondering, depending on which series you are talking about, what would plausible tire wear be? Either of the original or the later tire wear? Using up a set of tires in a half an hour isn't exactly unheard of in many series.

 

Yes, the AI are affected by track state, but like all their other factors, they use a simplified algorithm and not the same "pure" physics as the detailed player car simulation.

The AI are certainly not calibrated as well for wet as for dry. I am sure the track state will need lots of fine tuning as well (in dry and wet!).

 

Yes tyre wear is accelerated, haven't heard of that bug, is the bug slowing them down or making them super fast initially?

 

I am saying that there are some abnormal tire behavior that randomly happen once in a long while. You go in a session with conditions you have tested many times and suddenly you are missing 2s a lap or the car behavior is completely odd or like this time tires get eaten at 4x the normal rate even though settings are the same.
I've had a similar situation many months ago at Jacarepagua' on the F1 G3M1: there wasn't even all the real weather thing. Server was started more than an hour before the race with a long practice, track was fully rubbered in. I did race in practice and set normal laptimes. When going to quali I was 2s slower and the car was absolutely driving like garbage and had no grip. Temperatures were the same nobody else experienced a problem with track grip and did the same laptimes as in practice. I couldn't get even remotely close to them. Then in the race I started and again the car felt undrivable. No changes were made in the setups other than fuel in all the process and compared to the weeks of tests for that race I had previously.
I went to the pitlane and changed tires and suddenly the car was normal: I was able to set the fastest lap of the race and almost a world record. The first set of tires was a standard soft and appeared to be new but actually behaved like a fully worn set of tires.
Bear in mind that before races I run tests in the exact same conditions over and over again, at times spending even 10 hours on that car/track combo with exactly the same conditions of the race in the working week before.
Often times if you re-race with the same settings one more time the weird things do not show up again.
I've had sudden weird experience in the past with car behavior and FFB and had to delete the docs folder even though there was no update intervening in the meantime to justify any change to FFB or cars physics.
I'm not sure what causes this stuff, although I have a feeling that the caching that is happening in this sim behind the scene may have a big role in this silly gremlins.

 

Honestly 95% of the races I've had were great experience, really awesome. I loved most of them. However it may happen that some crazy stuff appears every now and then. That is why it is important to report. What really concerns me is that some of these things are not happening consistently, specifically situations where like today the physics of something goes completely shambles for unknown reasons and for some players only.

 

So, it is always online when these incidents occur? There is at least some possible explanation (synch issues of LiveTrack, tires, etc.) that could explain it. Offline, there is no explanation that I could postulate.

 

Issues with LiveTrack syncing or track rubbering used to be quite prevalent. I remember many online races that were like driving on ice when the quali had been normal. Haven't really noticed it happen in a long time.

 

Yeah, the accelerated tyre wear causes the AI's tyres to degrade at a much faster rate than the players' (or some other issue entirely) which in turn makes them a couple of seconds slower per lap compared to the player towards the end of the race. Usually from around 20% wear their pace falls off completely. So they get slower towards the end after starting with normal pace.

 

And that is the key question. In fact, all game modes suffer from bugs more than a year after release.

Even TT, where the game gives you weird starting positions (Hockenheim!!) on some tracks, or replays, where the recording loves to show buildings and trees instead of the race, are buggy.

 

Can't really say I've experienced lots of issues in multiplayer recently, except when people insist on picking frustrating cars to drive instead of good cars. I wonder how much of this buggy behaviour we see, cars flickering and bouncing all over the place, is due to bandwidth issues rather than inherent netcode bugs?

Same with offline, personally I am having a lot of fun running a FClassic Gen 2 championship and the AI in those cars is better than ever, while others cannot complete a lap without getting spun. Maybe I've just learned from experience that trying to do fancy things with the AI like using accelerated fuel or tyre wear is a recipe for disaster in every sim. They will always pit out of sequence or pit multiple times or just not pit at all and run out of fuel etc.

A(n) (in)famous sim racer once coined the expression "7/10ths sim racing experience" that means that 7 times out of 10 sim racing is amazing and the best thing you can be doing, and 3 times out of 10 everything goes wrong and it's just terrible. So if AMS2 can be the 95% sim racing experience, it's already much better than the competition.