Automobilista 2 V1.4.5.2 & New DLC Packs RELEASED!

I think there has been a lot of improvement in the AI in the last couple of updates. No more clustering of cars, the spreading is a lot more believable. I think it just lacks a bit of AI vs AI overtaking on the straights and more defending vs human. Otherwise I think we're nearly there. I had awesome races, best in at least a decade vs the AI in all categories with more or less the same values of difficulty and aggression.

 

Renato Simioni, post: 221132, member: 4"]Next update will be late February guys, we´d hoped we might be able to fast-track Bathurst 1983 and Nurb71 by the end of this month but we are not quite there with it all yet.

For 2023, the major updates should typically be deployed only at the end of even-numbered months (February, April, June and so on).[/QUOTE]


I think you should not make announcements about targets you will not meet. This always looks a bit unprofessional.
The community is quite patiently when I see the comments, but I do not like the fact that nearly every target within the last 12 months or so came later as announced.
I'm happy with the progress each time, but the worst thing about the game are the announcements that let us wait without any info... Telling us yesterday to wait another month for things you wanted to deliver last month is just lame.

 

Do you think there's a chance of a hotfix update before next 'major' release? At least to address the broken tracks...

 

I agree; I’ve had some fantastic races, with GT3/Group C/ V8 supercars/ GT4 etc, & there has been encouraging steps forward with the AI. It’s not all perfect (which sim is for that matter), but clearly some classes/tracks need more work than others.

 

As someone pointed the biggest problem with the AI is how they behave among themselves, around the player they behave nice enough, but also they need some adjustments of speed through corners, sometimes too fast, sometimes too slow... Also the behavior of the AI under blue flag needs improvement.

 

Tin tops and prototypes are okayish. AI is not okay for open wheelers in general.

With that said, I always asked which open wheel track+car has been finetuned, never got a direct answer, only evasive ones.

Well, I happened to try F-V10 Gen 1 at Old Interlagos, AI 120 strength and max aggression. This combo is not flawless and the AI clearly can be improved there, but oh my you can see all the potential AMS2 has with AI racing there. What a blast! I advise people to give it a try.

I'm not a very fast driver, so 120 strength might not be enough for you who are way faster, but it did fit me pretty well.

 

Yeah, Interlagos is one of the best tracks for open wheelers, the only dubious corner is the first one, where the AI can go off track to overtake the player, apart from that its very good. On the other hand, Imola is awful, the worst of the tracks I have tried the modern F1's, they are all over the place and go off track a lot.

Thats why I think Reiza should try to focus more time and manpower to calibrate all the car/tracks combos so the AI is more consistent and new users can get a much better first impresion than getting lucky with a good combo on the first try.

I loved the AI from the get go, but I have a higher tolerance than most and features like the AI mistakes, customization and mechanical failures give me more incentive to look past some shortcomings, because the game has that magic feeling that GP4 gave me way back.

 

I mean the "real" Interlagos. If you go off track you die

 

I think you should not make announcements about targets you will not meet. This always looks a bit unprofessional.
The community is quite patiently when I see the comments, but I do not like the fact that nearly every target within the last 12 months or so came later as announced.
I'm happy with the progress each time, but the worst thing about the game are the announcements that let us wait without any info... Telling us yesterday to wait another month for things you wanted to deliver last month is just lame.[/QUOTE]
Hindsight is a wonderful thing.

Targets are targets - they are ambitions which aren't aways met. The post you quoted is a perfect example of reasonable communication IMHO - we aimed for this target but missed - here is the new target.

There are two alternative means of communication that could be taken - both of which are worse than communicating a target release that may, or may not, be hit -

• Saying nothing - then the community complains that they aren't informed of plans
• Communicating deadlines - not targets - we've seen already that targets are already interpreted by many as deadlines and the complaints that follow when they aren't hit.

I guess what you want is for more targets to be hit, but even the best made plans often change. Perhaps we need to manage our expectations better?

 

i completely agree with this

 

can there be an option to turn off the status popups like brake bias changes in vr, since this data is now on the dashboard?

 

It's not yet on the dashboard for many cars...and some will never have it because they don't in real life. But I never disagree with an option to turn things on or off based on user preference.

 

The notifications were already Off by default when the HUD was off. A return to this would suffice.

 

Well since the new update isn't going to come out for another 20-30 days, I want to bring up 1 criticism that going back to Pcar's 2 brought to the light. I'll explain the problem, and then hopefully propose a suggestion.

Issue,

Group C's Low and high downforce models need to become separate cars like in Pcars or in some way selectable so we can pick which model we would like to use. In real life it was common to see teams show up with a mixture of areo packages, especially at races with a lot of privateer entries who maybe couldn't afford the lastest and greatest. Also some tracks like Daytona for example would be a toss up if the HDF or LDF areo kit would be better. Personally, I like the HDF kits on the group C cars much better than the LDF kits at Daytona.

This is less of a issue in GT1 (Where all the cars had Manufacture support and likely would have the optimal areo configuration for the track) and the Formula Series (For much the same reason).

Proposed solution 1, Player selection, AI randomization

Now, I don't know how hard it would be, but the optimal solution would be for the player to pick which body they want and then to just to randomize the AI cars with the selection skewed for 1 version of the car over the other.

So Daytona would be a 50/50 split, but Monza or the Unnamed French Track would be a 10% chance of HDF and a 90% chance of LDF for each AI car. Long Beach would be the opposite with 90% HDF and 10% LDF for each AI car

However, I imagine that sort of solution would be complicated... So...

Proposed solution 2, GTP/Group C split 1

Because of the less restricted nature of IMSA GTP, most of the cars features much more downforce then their european counterparts. With that in mind, we could split the current Group C Class into GTP (Which has indvidual cars with some LDF cars removed) and Group C (Which allows the game to pick the optimal kit by track). We now have enough cars to do this, and assuming we get more Group C and GTP cars, this could make the game more flexable in regards to liveries being removed as new cars are added.

GTP would include the following cars as individual models
C4 GTP (HDF)
962C (HDF)
962C (LDF)
Nissan R90C (HDF)

Reasoning.
C9 Sauber apparently never raced in the US, however if it did, I assume it would use it's HDF kit. (Not that there is a huge difference with the C9)
LDF C4 was used at LeMans only.
Nissan only raced Daytona with the R90c, and they used the HDF kit in 1991.
The Porsche teams were basically doing anything they could to keep up at this point, and becuase the car had been around for a while, I would expect to see both areo kits in a normal race.

Group C would remain unchanged, with the game selecting the optimal areo kit for the track you are running.

 

I'd much rather that we get an entry similar to that for the helmets for the Longtail or LD versions of any car... So that xml's don't have to be changed just because I want to race at a LD track after running at a HD track...

Same goes for the oval versions of cars... They should have their own entries in the xml files so it's swapped to the correct skin for the correct track...

As for Daytona, that's like Monza, you may brake better with a HD package, but you will be smoked on the straights and there's just not enough corners to make use of the downforce and make up that lost time...

However I do like the idea of creating a GTP class with the Group C cars that raced in the states with their differing physics... Thus allowing the Group C versions of the Vette and Nissan to be toned down a bit so that the C9 is still the boss within that class... As I believe the versions of both cars we have atm are more in line with the IMSA GTP spec than a WSC Group C spec...

 

Tbh i quite like the current system as we never have to remember to select the correct kit. This is really painful in rFactor 2 when the ai loads in a different aero kit to yourself, and im happy that reiza avoided this issue with rFactor 2

 

I don't think it's much different for GT1 to be honest, as a lot of the smaller aero differences (we only have the bigger packages right now) were pretty much driver preference:

Here you can see one Merc with smaller front winglets, the other with bigger ones. One of the works Porsches has additional winglets, the other one doesn't. The Zakspeed Porsches in the back both have front winglets, but they generally used different ones to the works Porsches (these are the ones we have in AMS2 right now).
The McLarens in 1997 also frequently appeared with different aero packages to the same track.

It would also help with the Nissan GT1 models which both don't really make sense to be labeled as high- or low-downforce package, as the 1998 longtail car had both more downforce and less drag. At the moment it just feels like a forced-on low-downforce variant as the other GT1s have one. Having the model is cool of course, but it doesn't really make sense right now.
Making these different versions somehow selectable or maybe tied to liveryslots or something, could be a solution for this car at least.

 

I'll be honest I'm very indifferent. I think maybe having it be an option to switch on and off? Though that might just be feature bloat getting in the way of more important fixes that needs to be made.

 

Yeah this is my major concern with the many different variations of cars that we have now...

Plus it's not just the Group C or GT1s that had differing downforce packages out there during the real life races...

There's many F1 teams, especially before the turn of the century, that didn't have a low downforce version for Monza or a high downforce verison for Monaco... And IIRC Minardi didn't after the turn of the century either until Red Bull bought them out...

That's a lot of work/time that could be spent on various fixes, new features and content that we don't yet have...

If it's an easy fix to have an option on the car select screen and opponent select screen for the AI or online lobbies then ok, but there's a lot of features I want to see before that, so if it's labour intensive I'd rather see other features like selecting the skins you want to race against or scalable engine power for the AI come in first...

Or anything related to the MP set up, like the ability to put a password on the lobby so that the lobby doesn't randomly get hosted by someone who shouldn't be hosting it... Not just an entry password but one for the admin/host so that only those with the password can start the server and be the host...