Automobilista 2 V1.5.5.0, Le Mans & Endurance Pack Pt1 RELEASED - Now Updated to V1.5.5.6

He did try AMS2 quite a bit, including wet racing, and gave his impressions on what feels real and not. Like with any sim, he said some stuff in AMS2 are good, some are bad. Simple as that.

The fact he plays more iRacing probably has to do with the MP population, since playing with bots is just not as good as playing vs humans (the fact people don't have spare time to play online doesn't make bot racing as "realistic" as a proper online session, given AI has a very clear ceiling and too many quirks).

In AMS2 there is no organized racing going on officially, so its only natural people look for sims where there is a rich MP scene.

Given Reiza has deliberatedly said MP isn't their focus, and clearly has not given the same attention as they had with content and AI, it is what it is. A pretty neat sim, but fated to single player and random bizarre online lobbies.

 

That’s true; moreover, Renato has described all the many improvements made to date since AMS2's initial release as going after “low hanging fruit”, which seems to indicate that additional, revelatory improvements to the sim (e.g., a total multiplayer renaissance) will probably be very, very hard to realize. Arguably, no developer in simracing outpaces what Reiza delivers in terms of sheer product development relative to the overall monetary investment expected from its users, but it is fair to temper expectations about what AMS2 can ultimately deliver. In the big picture, switching back and forth between an array of sims to leverage the best features of each one is hardly as hardship.

 

Have you also been in the iRacing forums for the past 10+ years posting every day or two that they did not have AI/a single player option of any kind (until fairly recently) or dynamic weather (until a few days ago)? How was the reception over there to those repeated posts?

 

I think you are sharpening what he said a bit too much Fabiano. Eventhough the outcome for this sim are the same. At least in the near(?) future.
What he said (or rather was implying) was that because of the way MP is integrated /created in the Madness engine then MP is not getting Reizas highest priority.
And as I interpret this then if the MP part in this Mad engine had been created with more promissing possibilities then Reizas priorities would have been different.
But I agree with your overall description.

 

Definitely. I like them all, AMS1, AMS2, rF2, LMU, AC and iRacing. None is perfect, but all of them are fun.

All I was saying is that people like Vitor don't spend too much time on AMS2 because we want to go racing vs humans, since AI still doesnt race human-like in any sim.

 

You seem to think if you repeat yourself over and over and over, day after day after day, that you will have some effect on the outcome. I'm pretty sure the outcome will be yet another warning from Renato that you are pushing it. Maybe take a break for awhile before that happens?

 

The thing is: people want fun. The most played car game in Steam is BeamNG, and is one of the best selling titles. And comes Forza series, etc.

AMS2 sells good in Steam too, because has the balance of Fun/Realism, and Reiza understand that sales = performance, graphics, good SP, easy MP and constant updates and new content.

 

That's because it's easier to get those sweet YouTube views by making 10 minute "OMG Extrme crash compilation you WON'T believe !!11!1!" videos with annoying electronic music.

 

People that err to sims find fun in replicating reality to the fullest. That is where they find their fun.

 

At least for me, it's all about realism and immersion.
On all aspects, visuals, sound, physics, feel, how physics translates to motion rig.
And I am on AMS2 because it is a winner, for me, over all these aspects.

 

Apologies if it has mentioned previously, but I would like to make a point about how players interact with the weather system.
Let's say you want to set a specific weather combination throughout the race. Excluding the fact that you can't set a specific time length for each weather slot, it works fine.
If you set random weather, you still can get torrential, red flag level of rain, and that's suboptimal, to say the least. It would be great to:
1. Give a certain weather a probability, to simulate a climate or a weather forecast.
2. Exclude certain weather types. (Set its probability to 0%)
With random weather you often get the feel that the game picks a random weather type from the list, applies it for 1 hour, an then repeats the process. It feels very artificial. Moreover having no predictions is just not how it works in real life - there's always the possibility to check a weather forecast.
The clever folks at JustRace have their own solution in the works, which makes things closer to what I expect, but that's just for multiplayer. I'm sure Reiza know about them and how to contact them, if needed.
For me this is more important to any physics feature the game might have, because it affects radically how one approaches the race. Still I would like to see a more marked rubbered wet line effect. At the moment the racing line is still faster in the wet.
To conclude, either knowing the weather ahead of time with player selected slots or choosing the "broken" random slot is not optimal. There needs to be a balance between predictability and randomness.

 

Random weather already employs percentage chances based on a location's real weather patterns. Thunderstorms, for example, have a much higher % of occurring at Interlagos than Monza, because of the tropical climate of Brazil.

I agree more user control is needed, but random isn't really random. It's weighted randomness, based on real weather patterns.

 

Please note that I wrote "you often get the feel..." because the weather changes are often very sudden (because of the 1 hour slot limitation) and sharp (from full torrential rain to clear sunny weather).
Perhaps probability of a single weather type alone isn't sufficient. Maybe having a probability for the weather to improve or to get worse is needed. (a weather probability "derivative" if you will)
The way the patterns work (the succession of random weather slots chosen by the game) is often not very realistic. If you are in a region with some chance of rain, what you get is a race with often at least one big weather change with a fully dry slot and a fully wet one. It feels like these weights are not set up properly. I don't know if you ever had races in them but they do have a very random feel to them.
(I was aware of this feature, it just doesn't seem to work properly to me)

 

Nope, people that find fun in replicating reality to the fullest and aren't Max Verstappen will find their fun in anything a lot of people say is more real, even if it looks and feels like crap.

Sad, but true.

 

My point:

Years reviews talking about how no sim can make a wet track that behaves like a real track, that wet track is really complex unpredicatable. And now, IRacing after four years deliveries the most advanced weather system in the galaxy. For this:

Really?

 

Does real weather work for everyone? It just doesn't do any weather transitions in my case no matter the date. I have been trying to simulate the weather of some past F1 races and it doesn't change from the weather at the beginning of the race when it should. Tried to do Spa 95, August 27, 2pm start time, and the whole 44 laps ran on medium cloud, when it should have rained later.

And its basically the same on any track and date, the weather almost never changes on 80+ minutes of racing using real weather with real time progression. Sucks big time having this great feature simply not working at all. Is it just me, or does anybody else have issues with it?

 

Have not used real weather enough on AMS 2 to have an opinion on that, I just set weather manually usually.

I once wrote a custom weather script for Pcars 2 that got its data from some free weather api, applying its own logic about transitions/evaporation rates etc, and fed it to a dedicated server I ran locally and just raced with the AI on it. While it was still limited by the max of 4 slots and weather presets, in general it did a much better job than Pcars 2 itself.

I wish I still had it, but it got lost when I had a sim racing hiatus and in the meantime the HD it was on went "bye bye!"

No clue how hard or how not hard it is with the madness engine, but a lot of the "issues" could be solved by adding more slots I guess, perhaps something like 8, to allow more realistic transitions in longer races, so you can go from rain to light rain, heavy cloud in maybe 45 mins and still have 5 slots left to get the track dry and sunny and back to rain again should you wish to do so.

It could only allow more than 4 slots if the race is long enough to allow at least 10 mins per slot, or whatever, so you do not end up with people making 20 minute races with 8 weather transitions, but as I said, I have no real knowledge of the inner workings of the madness engine, so this might not be feasible at all without re-writing the whole thing.

 

I tried the 24h at Nordschleife 2016. It was a hell of a weekend, rain over rain and thunderstorm on Fri, beautiful start on Sat. 30min later, 2 weather fronts rushed towards the other. Never seen that before....worst conditions on track, hail, thunderstorm, massively spuns at Schwedenkreuz. Race has been stopped for hours.

Ingame it just rains a bit in the Sat evening. No big deal. In reality quite different. I was there from Thu to Sun...it was a hell unforgetable a weekend.

 

I was actually never able to replicate the actual "real" weather experienced in a past weekend (e.g. WEC at Monza, ora F1 at Red Bull RIng...).
Maybe the weather database is not so accurate, or maybe it is sampled 2-3 times per week and then Madness Engine "fills the gaps"....we do not know.

I would be really helpful to have at least a sort of weather prediction/forecast/indication when choosing a certain date and use "real weather", before launching the race/session.
So, to me "real weather" is not really "real weather", at least not for the 2-3 events I tried to simulate.

On the other hand I find the "real weather" maximum strength is the absolutely dynamic and granular condition of weather and track state.
For example, only in real weather I can find the track in status "moist" with some puddles here and there on some areas of the circuit and the rest mostly dry.
And it feels to me that also the sky/clouds states are usually quite different than the 4-5 possibilities that can be chosen in fixed weather.

All in all, I think madness engine and livetrack 3.0 have a lot of options and tweaks under the hood which are not accessible by the fixed weather UI, and this dynamic combinations is what I love about real weather.