Latest Roadmap Update - January 2020

[Paul McC]

Then why not just race the same fecking car for fecks sake!!! My fecking god I am literally going to go out and kick a baby in the face now!

 

[davehenrie]

yes, unfortunately, there is rarely any notation that any particular skin works with version xxxx.xxx.

 

[whitmore]

Now a select few (invited participants) are the focus of the developers. The boat has really sailed on this sim.

 

[LokiD]

I’ve gotta say that I will pipe up when there is a bug and it frustrates me and I do it with this title a lot because I actually love the pants off this sim even though it frustrates the hell out of me sometimes but, I’m also more than happy to praise it aswell because it’s awsome! Yes there are areas where the ai could be improved like they don’t always go for the overtake on slipstreams (at lemans if you move over they don’t change their line to take advantage of your slipstream as you would do on straights)and some of the Reiza pack ai is very unvaried (the MCRs for example with 16 on track are sometimes less than a second apart in lap times, like a train). However the ai in general in this sim are far from rubbish! I would say the ai in this sim at their best are the best ai I have ever raced! In fact I have had some of the best races I’ve ever had (on or offline) with them and that’s really saying something! They’ll never be human but feck their good!

That's what frustrates me the most the ai only need a few little tweaks as you mentioned not to mention better rain pit strategy. It is awesome when it works on I find small tight tracks. Atm stay away from tracks with long straights. And yes I bought lemans lol

 

[Will Mazeo]

LOL... happens more often than you'd think. "League Edition" car sets are exactly that!

League editions are mostly used in cases like F1 or historics where difference between cars is too big, the fact that there is something called "Balance of Performance" alone makes using same physics unecessary (if you are doing a series that has it IRL), heck with a bit of knowledge you could BoP historic cars too for a league

To adapt a BOP to the lap times of aliens makes no sense, because they often take a risk while driving, which no driver would take. In addition to that you use exploits like the missing downshift prevention or setups which nobody would use to reduce the lap times. I don't need to talk about the extremely many practice laps anymor

Aliens are the only way to BoP, the average Joe cant do it, they'll always think a car they dislike is the fastest and the car they like is not fast enough. But you need an alien that can adapt to any car and have setup knowledge to find the flaws, not all aliens are like this.

Even in real life BoP tends to be a constant process. Sometimes series end up taking the easy option and developing a base, then throw different bodies on top (which are more easily balanced) and call them different cars - Supercars for example. NASCAR I don't think does this exactly, but I gather teams can inspect each others cars at certain times and basically copy them, and again they have different 'brand' bodies thrown on top that are nearly identical anyway.

The issue I think is that this shouldn't be a major issue for the end user - if the initial physics are close, you can close minor gaps (let's say up to 1s/min) with ballast and power tweaks that are barely noticeable. There shouldn't be any real handling changes.

Part of why BoP has become a swear word is that the BoP prior to the Le Mans races also slowed the cars (because, surprise surprise, aliens were doing faster laptimes than the guys testing during development, and it doesn't help a simulation's credibility if your high profile race has unlikely laptimes being set) and that pushed those cars back towards others not involved in the BoP process.

As long as there's a formulated, logical and incremental approach to BoP it shouldn't involve any major enough changes to cause issues. But it takes a decent sample size and some well defined goals, which is one reason individuals shouldn't really try to judge them either. No matter how good someone is, you can't be sure they're getting everything out of every car. You need a number of drivers of similar level across the different cars, all pushing their driving and setups to the limit, in order to judge performance - and again, you need to decide if that's single lap performance or race performance.

Yes, this. At least make 2 BoPs one for top speed tracks like Monza, another for everything else (which happens in GTE cars, they dont have 4 BoP tables IRL, they are boped from track to track and then have a LM specific BoP, S397 also made this). Then leagues can use success ballast easier

 

[stonec]

IRL the GT3 and GTE are not all uniform and do not all run in the same decimal!
in the same second it's already extraordinary.
for uniformity there are cups and monoseries!

Yep, certainly, you can get the BoP to work if you are allowing a large enough tolerance as in real world. The problem as @davehenrie pointed out is that sim racers quickly figure out which car is a couple of tenths faster under given conditions and given tracks and then the complaints start and the field becomes homogenized. In real world drivers are under contract and don't jump back and forth between cars.

Single physics with different looking cars! Really, REALLY! Get the feck outa here! Never heard anything so fecking ridiculous, this is a sim! A sim with different cars but same physics, yeah there’s something no ones buying! I had a fecking great race with the a.i. The other day, bloody fantastic! Was with the Aston GTE, bop was perfect guess why.... I was racing 15 other Aston’s! Simple as that! Want exactly same performance and a total fair race, use the same car!!

Geoff Crammond's legendary Grand Prix series had single player physics in use for all different F1 cars. Anyway, I think you misunderstood my point, I didn't suggest they just remove all physics differences in the base game, it was just an idea to have a single locked set of physics for official events if BoP cannot be worked out. If everyone chooses to drive the one fastest car in the series, then that is the same thing as single physics anyway.

 

[mantasisg]

Now a select few (invited participants) are the focus of the developers. The boat has really sailed on this sim.

This looks like it more and more as time goes by. I have started worrying a while ago that this what the "mysterious" competition system is going to end up being, and probability of that is increasing. What I am dreaming of is very flexible and well working online platform for various sorts of usage, and with some kind of statistics leaderboards to pursue (positions, race craft - incidents avoidance and so....). But I will not be surprised if it is going to be WIP sort of thing that will hold several online races per month which only aliens will participate, and others will be just able to pretend like they are participating by trying to qualify by doing hotlaps, and never actually racing. And it is of course going to be 95% gt3 cars at six same tracks. To me it would be extreme reduction from what simulation could offer, but it would be nothing new, because it already was like that in simracing world and in rF2 especially.

 

[Sebastien Sestacq]

Yep, certainly, you can get the BoP to work if you are allowing a large enough tolerance as in real world. The problem as @davehenrie pointed out is that sim racers quickly figure out which car is a couple of tenths faster under given conditions and given tracks and then the complaints start and the field becomes homogenized. In real world drivers are under contract and don't jump back and forth between cars.

that's why I don't like the championship on iracing (everybody take the same car at 80% with the same setup that we buy from the most famous teams...), where people change car at each race to get the best on this or that track, like in reality we should choose a car for the whole championship and deal with its qualities and flaws

 

[Adam Helyes]

About BoP
There is a "weight table"? How many kilos the BMW, Aston? etc

 

[Will Mazeo]

that's why I don't like the championship on iracing (everybody take the same car at 80% with the same setup that we buy from the most famous teams...), where people change car at each race to get the best on this or that track, like in reality we should choose a car for the whole championship and deal with its qualities and flaws

These championships should be like a career mode, you have some virtual credits at the start of it you buy the car, spare parts, pay transport fees, make some extra virtual credits buy finishing races so you have more for spare parts or chassis. Everyone would probably go bankrupt before the end of the season but hey it'd be fun (may help to promote clean races if you arent getting new stuff every time you hit esc)
Maybe one of S397 leagues should be like this @Marcel Offermans, FCY, budgets, no new parts every time we go back to garages, hardcore mode. I'd join this one for sure

 

[Remco Majoor]

I always find the comments interesting on these roadmaps. It's like people forget the state rF2 was in when S397 took over from ISI, the amount of content that has been released etc.. They fixed the join lag in this road map, and people complain about them not fixing bugs. That was literally the most game breaking bug for online racing. People saying that they only do stuff for the aliens, while in their road map they said that the challenge series will have multiple splits so everyone can join. What more do people want? They are Implementing bit by bit, testing stuff out.

AI indeed sucks, but that is a compromise S397 has to make with their relatively small team. They can't fix everything at the same time.

But if it really isn't your cup of tea yet, and you can't wait, go to a different simulator, until you find this one better again. Tonight I will be driving the Bathurst 333 in Raceroom (which funny enough has the same join lags). Next week I'll be back in rF2 driving the VEC, enjoying it to the max, because the racing is awesome.

 

[hitm4k3r]

League editions are mostly used in cases like F1 or historics where difference between cars is too big, the fact that there is something called "Balance of Performance" alone makes using same physics unecessary (if you are doing a series that has it IRL), heck with a bit of knowledge you could BoP historic cars too for a league



Aliens are the only way to BoP, the average Joe cant do it, they'll always think a car they dislike is the fastest and the car they like is not fast enough. But you need an alien that can adapt to any car and have setup knowledge to find the flaws, not all aliens are like this.



Yes, this. At least make 2 BoPs one for top speed tracks like Monza, another for everything else (which happens in GTE cars, they dont have 4 BoP tables IRL, they are boped from track to track and then have a LM specific BoP, S397 also made this). Then leagues can use success ballast easier

I think there is a way to create a solid BoP that doesn't require hours and hours of testing from aliens and is quite often overlooked. And that's simply by using AI, but with player physics as this would give you the most controlled way of testing. You could even let this happen without any human interaction. One could make little adjusments and arrange test runs throughout the night, while you evaluate the results the next day and make adjustments accordingly. There are even parameters like the AI limiter that could be used to aid the whole process. Another thing that could be used as a result of this are special BoP settings that would be shipped with the upgrades for a certain type of track or simply the option for league admins to adjust weight penalties and similar variables on the fly with the new UI. I think AC allready offers this, so I am fairly sure that this would be a possibility. Anyway, I think it would help S397 to get a good baseline and leave the rest of the BoP up to the leagues. All this back and forth is taking lots of time and I suspect evaluating this approach in a scientific way isn't as easy as it sounds, especialy the more cars, human error and track types are involved.

 

[davehenrie]

I think there is a way to create a solid BoP that doesn't require hours and hours of testing from aliens and is quite often overlooked. And that's simply by using AI, but with player physics as this would give you the most controlled way of testing. You could even let this happen without any human interaction. One could make little adjusments and arrange test runs throughout the night, while you evaluate the results the next day and make adjustments accordingly. There are even parameters like the AI limiter that could be used to aid the whole process. Another thing that could be used as a result of this are special BoP settings that would be shipped with the upgrades for a certain type of track or simply the option for league admins to adjust weight penalties and similar variables on the fly with the new UI. I think AC allready offers this, so I am fairly sure that this would be a possibility. Anyway, I think it would help S397 to get a good baseline and leave the rest of the BoP up to the leagues. All this back and forth is taking lots of time and I suspect evaluating this approach in a scientific way isn't as easy as it sounds, especialy the more cars, human error and track types are involved.

AI respond more to weight changes than humans. I started modding after testing one of the SCC mods and always found the same car worked it's way to the front.(Saleen) Turns out the team had given that car about a 2kilo weight reduction. That is all it took. Weight would be more of a deterent to humans if it had a greater effect upon things like the brakes. Nothing like running out of brake pads with two laps to go....

 

[Russell lowe]

I always find the comments interesting on these roadmaps. It's like people forget the state rF2 was in when S397 took over from ISI, the amount of content that has been released etc.. They fixed the join lag in this road map, and people complain about them not fixing bugs. That was literally the most game breaking bug for online racing. People saying that they only do stuff for the aliens, while in their road map they said that the challenge series will have multiple splits so everyone can join. What more do people want? They are Implementing bit by bit, testing stuff out.

AI indeed sucks, but that is a compromise S397 has to make with their relatively small team. They can't fix everything at the same time.

But if it really isn't your cup of tea yet, and you can't wait, go to a different simulator, until you find this one better again. Tonight I will be driving the Bathurst 333 in Raceroom (which funny enough has the same join lags). Next week I'll be back in rF2 driving the VEC, enjoying it to the max, because the racing is awesome.

The mic has been dropped, well said sir

 

[Andregee]

League editions are mostly used in cases like F1 or historics where difference between cars is too big, the fact that there is something called "Balance of Performance" alone makes using same physics unecessary (if you are doing a series that has it IRL), heck with a bit of knowledge you could BoP historic cars too for a league



Aliens are the only way to BoP, the average Joe cant do it, they'll always think a car they dislike is the fastest and the car they like is not fast enough. But you need an alien that can adapt to any car and have setup knowledge to find the flaws, not all aliens are like this.

That happens to aliens too. And by the way what i meant was that S397 tried to Bop the lap times compared to the real cars and not the cars under each other and thats stuff in my eyes to use aliens for like i said because they don´t drive like real racing drivers in many aspects.

 

[Will Mazeo]

That happens to aliens too. And by the way what i meant was that S397 tried to Bop the lap times compared to the cars and not the cars under each other and thats stuff in my eyes to use aliens for like i said because they don´t drive like real racing drivers in many aspects.

The part you should have paid attention is this tho, not all aliens are worth using when BoPing

Aliens are the only way to BoP, the average Joe cant do it, they'll always think a car they dislike is the fastest and the car they like is not fast enough. But you need an alien that can adapt to any car and have setup knowledge to find the flaws, not all aliens are like this.

 

[Andregee]

I forgot the word real and added it.

 

[Marcel Offermans]

What I personaly would like to read is, how the updates to older content are comming along. How is the new tech for tracks and it's implementation for older tracks such as Lime Rock, Silverstone, Nola, AMP or the nice bunch of historic tracks coming along? How is the PBR implementation for "older"/free cars coming along? You have such a nice set of cars and tracks, so don't waste the opportunity to offer the first time user a consistent great experience. Even people from the community managed to implement PBR materials to some of the content, so why this wasn't high on the priority list for the content side of things is really far beyond me.

For cars we are now roughly working our way backwards through our content portfolio. The latest we did was the Nissan GT500. Next up are our historic McLaren cars. Internally we've talked about doing the Radical after that. For tracks we are holding off on updates for a bit longer. This has, I think, a good reason, which is that we're still working on a global lighting review that attempts to address some of the issues that have been reported over the years (such as "why does everything have a blue tint"). The underlying issue unfortunately is a bit more complicated. We've decided to go back and recheck all lighting aspects. A big task, but it addresses the fundamental problem. Once that is done, we are confident we can go back and adjust the tracks.

How is the competition infrastructure for the average Joe sim racer with scheduled daily races coming along, because this is the area where you can really offer an alternative to iRacing? It's nice that there are big events, but they are all big endurance events and alot of people propably don't have the time to commit to such events. What is to be expected in terms of AI fixes and FCY issues that were expected to be worked on when the UI appears?

Daily races are at the top of our agenda for the competition system. Beyond the fact that we're working on it (and privately testing certain aspects) there is not that much to say. That is why I decided not to mention it this month and start by announcing some of the other things we're planning (big events, our GT3 competition, which btw is not an endurance event). This month our focus is on addressing feedback we got from the initial release of the new UI (and fixing the join lag). Looking at AI and FCY are on our shortlist, being investigated and discussed.

Sorry for this wall of text, but I think it is important that you guys reach out to the average Joe. This is where alot of the current frustration seems to be comming from. Don't forget that your free content is selling your DLCs aswell if it gives a glimpse of what people can expect from premium content and competitions.

We are reading along. I don't mind people venting their frustration and everybody's feedback is taken in and helps us set priorities for short and long term goals.

 

[Lazza]

@Marcel Offermans that post reads like a roadmap. The roadmap reads like a marketing blurb

 

[Marcel Offermans]

Point taken @Lazza