Reaper's "ReDefault" Custom Setups

Reaper's ReDefault Setup Series
or
Something Needs to be Done About the Default Setups

​

WTF is this?

The ReDefault Setup Series is my grand attempt at creating new "default" setups for AMS2's many cars (hence the name ReDefault), with a focus on being consistent and race-ready above all else. This project was born from my own frustrations with the inconsistency of AMS2's default setups, which vary from ok to making certain cars completely undriveable. I originally only shared this on the racecraft.online Discord - but with the number of setups recently passing the 20 mark, I felt it was the right time to share this here as well, in order to hopefully spark a discussion on the state of AMS2's default setups (as well as showcasing the setups themselves).​

Why should I care?

The default setups represent a user's first experience with a car in AMS2 - it will shape how they feel about a certain car and will form an opinion that'll be hard to change later. How many times have you read a post asking "Why is X car so bad?" or "Why is Y so slidey"? Well, I'm here to tell you it's (usually) not the car that's at fault, but the default setup.

As a player's first interaction with a car, the default setups should represent a car's particular characteristics in the best light while trying to minimise its shortcomings. They should serve as a consistent baseline that any player, regardless of skill level, should be able to handle. Many of the current default setups do not come even close to meeting these requirements. Some setups are fine, but the vast majority aren't.

These negative impressions that players form about a certain car actively hurt the game's image both inside and outside the AMS2 community. Think of default setups as a "demo" for a car - if the demo sucks, the player is going to assume the car is broken (see the myriad of "GT3 is slidey" posts).​

So what's wrong with the default setups?

Over my many hours of tinkering with setups, I've managed to identify several reoccurring issues with the defaults. They are listed below from what I consider to be the most impacting to the least:

• The toe... What's up with this? Basically every car has some degree of toe (usually -0.3 and front and 0.3 on rear). I get in real life it has its uses, but in AMS2 toe causes some very weird behaviour. The toe values on FWD cars are especially egregious and cause them to be an oversteery mess.
• Open differentials. Like the title says, the differentials are just too open. I'm mainly referring to preload values as number of clutches depends heavily on each car. This is part of the reason why people complain about AMS2 being slidey, especially in regards to the GT3's. Open differential results in a lot of sliding when accelerating at low speeds, and increasing the locking of the differential has been a massive contributor to making the ReDefault setups feel much better than the defaults.
• Brake ducts / radiator. I don't understand why these are the same on every car. Reiza has access to the exact car data and would (theoretically) be able to determine rough "optimal" values for these. For example, rear ducts can be more closed than front in almost every car, yet this is not represented at all in the default setups.
• No variants. This mainly concerns high downforce cars, especially Formulas. These cars never have a single setup for all tracks, and suffer massively if the downforce values are not adjusted on a per-track basis. 3 different aero levels (low, mid, high) would serve as a good solution to offer consistent and reliable setups for any track you might run.

​

So what's up with this ReDefault thing?

Like I mentioned before, they are my own custom setups - completely built from scratch, without using any telemetry apps (except what the AMS2 HUD offers), and with a focus on consistency and being race-ready. You can read the full details behind the process in the linked document, but point is these aren't some complicated, pouring-entire-days-into-building and constantly-looking-at-telemetry setups. Some of these I've even managed to knock out in less than an hour, and still get a much better result than what was offered by the default setup. But to not be dishonest, I'll admit some of these (especially GT cars) do take quite a bit more time.

I strongly urge you to try out the ReDefault setups when you have time, and compare them to the default setups. You'll see what I mean when I spout my many grievances about the state of the default setups. Of course, not all default setups are bad (e.g. LMDh is decent), but most are, and the ReDefault setups try to fix these issues and provide a new and consistent baseline experience.​

Where can I find the ReDefault setups?

Link is HERE.

EDIT: A massive overhaul of the ReDefault is in process. Currently there 12 setups available. Marked in bold are the ones that I would say offer the greatest improvement from default:

• Copa Classic B | Puma GTE
• Copa Classic FL | Passat Classic FL
• F-Inter | Formula Inter MG-15
  
• FTRS | Formula Trainer
• FTRS-A | Formula Trainer Advanced
• F-Vee | Formula Vee Fin
• Ginetta Supercup | Ginetta G55 GT4
• GT4 | Mercedes-AMG GT4, Chevrolet Camaro GT4.R
• Kart Rental | Kart GX390 Rental
• Porsche Carrera Cup | Porsche 911 Cup 4.0
  
• Stock Cars 1999 | Chevrolet Omega Stock Car

​

 

I'm not so sure about the attitude in this post, I don't think this level of "look guys how I easily fixed your broken game" is warranted even if the setups were objectively better. Also, if you require extreme values to "fix" a car's behaviour, maybe the car actually has physics flaws and you're pasting band-aids over them with this.

Anyhow, this post is not too timely as the 20+ setups will most likely become outdated in the foreseeable future.

 

Sorry to say, but looks like someone is about to remake all his setups...

 

Not sure how you interpreted it as that being the message. What you've read about is the cumulative efforts of 2-3 months worth of work, with many setups requiring multiple passes and revisions and even some complete redo's as I gained experience with the setup settings and how each one affects a car's behaviour. The point I'm trying to make is that many default setups have big issues and are wildly inconsistent in quality and care between cars - and that's especially bad considering they are usually a driver's first experience with a particular car (see "GT3 slidey"). My ReDefault setups are a glimpse and an example of what we could have, while pointing out some common flaws to look at with the current defaults.

Maybe you hyperfocused on me saying some of these setups took less than an hour, but I've included that line to punctuate that fixing the default setups isn't an insurmountable task. This can be done, especially with access to the game's inner workings, and it's an issue that I've happily contributed to in trying to fix. I kinda get the feeling you came in here looking to argue and that is very unfortunate.

Readjusting a setup from 90% back to 100% takes considerably less effort than building it from scratch. Sure it'll take a week or so to go over everything but it shouldn't be a big hurdle. Considering most of what I've described above has to do with suspension geometry and differential settings, not with tyre settings, I'm not too worried.

 

I'm actually intriqued to try some of these out (also in the beta with physic changes). It could be also a preference thing, but I'm interested to try out alternatives to current defaults just out of curiosity.

Could be that Dicra is right that even if these work now, they might not work with 1.6 or at least would require revisions.

I personally think the current defaults are often very approachable, safe and easy to drive (depending on track and conditions a bit of course). I haven't really understood the critique about them (to the degree it's given) or would agree on some of the critique given in the op of this thread.

If I compare to something like ACC or rF2, the defaults those on those are often way stiffer and to make the car harder to keep on track (even if for some it might lead to more "predictable" behavior to what they are expecting) IMO.

 

To give credit where it's due, some default setups are fine for what they are. Not exactly to my taste and they aren't amazing, but I would feel comfortable giving them to a beginner: the Merc GT3 (Gen 1) and BMW M Hybrid are probably the best examples of "good" default setups. But others are egregiously bad and are essentially undriveable (especially in longer race conditions), like the Merc GT3 Evo, Formula Trainer Advanced and F3 which I've seen time and time again cause headaches for beginners.

 

Just did a 30m Qually / 30m Race at Nürburgring Sprint 2020 against the AI in the Formula Trainer Advanced using your tweaks and the car is indeed night and day different. Lots of confidence to slide on edge with a much bigger margin. Awesome stuff, man. Cheers!

 

I will definitly try the Mercedes GT3 Evo Setup! Like all GT3 Base Setups the car tends to float when throttle is applied.
Someone on this Forum suggested to me that reducing the clutch will help with this and it did, so I‘m looking to forward to try your setups!

 

Look man, the quite aggressive way you descripe defaults is already a big indicator of you having some fundamental misunderstanding of car setup and why defaults are what they are. Given there´s currently only 1 setup per car it cannot be optimized for basically anything instead it has to be a decent baseline for every single track, condition, and hardware.
People play with high end ffb wheels, low end ffb wheel, wheels with no ffb, gamepads, some even with keyboard- in dry and rain, hot and cold, low grip track, high grip track, smooth and bumby tracks, short races, long races, you see where this is going.
Now you say you´re creating alternative default setups "default setups should represent a car's particular characteristics in the best light while trying to minimise its shortcomings..." and then you say "I don't think there's a single ReDefault setup that doesn't have the ARB and slow bump/rebound cranked up to the maximum allowed values" I find those statement´s kind of contradicting, but wait there´s more "An important thing to note is that the ReDefault setups are, at the end of the day, made and tailored to my personal style of driving"

Then the more technical stuff, you make quite interesting statements on this department also "Softer/stiffer settings on one half of the car seems like a good solution but they cause incredibly inconsistent behaviour during racing. A car should aim to be as neutrally balanced as possible, and I have seen much more success in creating consistent handling with symmetrical suspension values in front and rear" If we take the BMW Lmdh low downforce 3/3 fully tested setup spring rates as a small sample you´re screenshots show 28 clicks both front and rear, but that´s not symmetrical it equals to 220 N/mm on the front and 185 N/mm at the rear (highly suggest everyone to use rates instead of units in suspension setup screen) to make the difference more pronounced weight balance of the car is set 45,5 / 54,5 so on a 1,12 metric ton car the rear axle holds over 100kg more weight - with softer suspension.
In general when you set the cars to basically max stiffnes they have no mid corner performance especially in medium to slow speed and slow speed is usually where the lap time is gained. Few cars I tried all felt like twitchy understeer into snap oversteer cause the front axle was never set up for perform but instead to react fast and balance the lack of rear, to be fair that´s quite common in ams2 time trial setups but it certainly holds no ground in full distance racing nor is it an approachable option for newcomers.

Overall, I think you are not really doing proper setup work but instead trying to apply something that worked in one situation into every situation.
Not trying to make it sound like some a-hole bashing the whole thing as you´ve taken some effort and don´t ask any money for the setups but there´s bit of learning and self reflect on the communication side needed.

 

I don’t see a problem with maxing out available setup values. Bumping up the numbers to an extreme value is quite normal in modern sim games.
In ACC you max out the chamber, the rear right height & the rear wing. And don‘t forget the crazy values for the rear toe.
In Rfactor2, often claimed to have the best physic enginge available, you have to run chamber as low as possible (at best 0 degree if possible) and you have to detach the rear anti roll bar.
In LMU even the well praised coach dave academy setups, normally behind a paywall, use these settings on all of the Hypercars.

Non of these settings would work in real life.
All I want to say: It‘s a game at last, so choose whatever setup settings fit you best!

I believe the foundational physics of the Madness Engine are also (based on) ISI Motor, so maxing out the values maybe the right thing to do!

 

Maxing out values is not a problem the range is there to be used.
When the fastest way in same class cars is to min max same settings in all situations then there´s a problem with the simulation.
When setting over 200 different cars with the same stiffer the better philosophy then it´s a problem between the seat and the wheel.

 

Hey there, thanks for the long message, I love receiving detailed feedback like this be it encouraging or constructive. Let me break it down point by point.

1. Never did I say these setups should serve as the definite new AMS2 defaults. God no. Sure, I call them "ReDefaults" for the panache factor, but the original intention of this thread (besides sharing some custom setups) is to bring light to some issues with the default setups. I never once claimed these were the be-all-end-all answer to this issue, but should serve as a tool to put in perspective some of the shortcomings of many of AMS2's defaults.
   
2. The "stiffest ARB and slow bump/rebound" are the answers to 2 very common flaws present in the majority of default setups that I've driven (and AMS2 cars in general - be it a quirk of the engine or simply my style of driving I do not know): over-rotation of the rear especially under acceleration, and oversteer caused by sudden changes in weight balance mid-corner (usually by releasing the brake too quickly). These two things, in combination with other adjustments that are made to correct some of the commonly encountered erratic behaviour (e.g. much more locked differential) result in needing to run the ARB and slow dampers stiff. And on top of that, I do have a gut feeling that AMS2 suspensions (especially the ARB) are much softer than they ever would be in real life. If you watch the cars physically in replays or 3rd person cameras (again, no telemetry), you can see a concerning amount of body roll during cornering (on certain cars - GT3's come to mind) despite using the stiffest values possible. Maybe I'm applying something that worked for one car to all - but I personally have not seen a noticeable enough negative impact to warrant changing this approach. Important to note here that while ARB and slow bump/rebound are set to the maximum, the spring rate isn't, and along with camber and the differential they are the main areas where I tinker with the overall handling and feel of the car (and counteract any undesired effects from the stiff ARB + dampers).
   
3. Damn that BMW suspension thing sucks to hear... Yeah, like any reasonable person I assumed the number values would be the same regardless of their range. And I'm not using absolute values because visual clutter - but I guess that may have to change. That's upsetting to hear.
4. Despite the differing natural weight distribution of the cars (obviously not all cars are a perfect 50/50), I once again have not seen any benefit in having unequal rates - and certainly not to the extreme degrees seen on some of some of the defaults. Unequal spring rates were something I had explored in the beginning, but found it more often than not resulted in unintended behaviour. I've found much more consistent success using other setup elements to balance any behaviour resulting from the car's natural weight distribution. Again, maybe generalising something that worked on one car to all - but I also haven't seen any benefit in not doing so.
5. You're right about low and medium speed performance - and is exactly why the spring rates are almost never the stiffest, and they are always adjusted (along with other available setup settings) to provide a good balance between low speed handling and general responsiveness. Yes that sometimes means a single click off the stiffest setting - but refer to point 2 above for why that may be. As for your opinion on where time is gained I will respectfully have to disagree. You gain much more time by being overall consistent over a whole lap than running a setup focused on performance along a single section of the track.
6. I really am not sure where you're getting the twitchy understeer / snap oversteer from. Could just be wildly different wheel settings and/or driving styles. Understeer is indeed present in some setups, mainly due to the high locking of the differential (but also due to the stiff suspension in some setups). Snap oversteer is one of the main things I try to dial out of any setup I build, so I really am not sure how you're getting that.
7. Also not to be an a-hole, but let's see you put your money where your mouth is. I don't say this in a mean way at all - I would very much like to see how you would go about building a setup for one of the cars that I've also done. Show me my wrong ways as I am always eager to learn something new. We can talk all day about hypothetical best setup practices and how to tune what and god forbid someone runs these settings in real life haha. But at the end of the day, behind all the numbers and theory and graphs, driving the setup is where the difference is made. So, show me one of your setups, I would be more than eager to learn.

At the end of the day, my "stiffer = better" approach is something I gained from hours upon hours of trying all manner of settings. In the early weeks I tried to do things exactly the way you describe them - balance springs and dampers according to weight distribution, adjust stiffness based on characteristics of car and aero levels, etc. etc. And I ultimately found that this "symmetrical stiffness" approach tended to yield the most consistent, "95% applicable" setups. Don't be fooled, this was to my dismay as well - one of those reworks I mentioned was when I found out this approach yielded better results than the meticulous ways I was trying before and essentially wiped my slate clean losing me hours upon hours of previous work.

However, like I've said time and time again. I simply have yet to see enough benefit by deviating from this approach. Prove me wrong.

 

Does your stiffer=better work at very bumpy tracks like Long Beach?

 

Generally yeah. Obviously not as good as flat tracks but again, not enough to warrant a different approach. Once again another thing that I had in the past was softer suspension variants for modern cars (GT3's)... that were later shafted when it turned out the stiff suspension worked almost as good as (or very close to) the softer version.

Though for bumpy tracks, decreasing the fast dampers by a click or two is also a solution that works relatively well if you don't like the stiffer feeling.

 

I am not sure I understand the apparent vitriol from some people here. This person has clearly put a lot of effort into making new setups, and he has some good points about the situation with the default setups. Until I started sim racing i had little idea of the complexity of the various ways one can set up the car, the effect that each changeable factor has on each variable is often unclear or hard to parse, and i often feel that I don't really understand what people are talking about when they describe setups and their effect because I don't really feel the same thing they appear to describe. So someone putting a lot of effort in and sharing it with us deserves nothing but praise, and I don't think there is any reason to get defensive on Reiza's behalf.

Thank you Horia M for these and I look forward to trying them. I became convinced this would be useful when you mentioned the BMW M series V8 as I feel this drives well "out of the box" - at least something I can see fits with my own experience.

 

So you are publishing custom setups, which is a nice contribution specially considering in AMS2 there are not many options of the kind.

Calling them defaults is controversial, I could understand the “punch” of the name, but for instance you state the issue with default setups is that there are no variants… The important thing: the tone of your post is criticism to other people’s work because the default setups they developed are not custom. Unnecessary, unfair.

“The fish dies by its mouth”.

 

Yeah I can understand how the name might be misleading. But as the setups came about after my many frustrations with the defaults, and these setups were never track-specific to begin with, the name stuck. Plus... it's a catchy name ¯\_(ツ)_/¯

The variants argument is something I stand by. For certain cars (Formulas) the one-setup-fits-most approach is just not adequate. Reiza included low downforce variants for some cars despite not really needing to do so - why shouldn't the same approach be considered for some cars' default setups? ACC also uses multiple (two) default setups so this surely isnt such a extreme suggestion.

Also yeah, two thirds of the post is criticism aimed at other people's work. I don't want to misinterpret your message but I'm really not sure how this is such a bad thing. Is someone immune from criticism simply because they put a lot of work into something? Should we really abstain from criticising the things we love?

I also feel I've been quite fair with my criticisms. I've broken it down into specific grievances, I've tried to offer on-paper solutions with explanations as to why and how, and I've even provided practical examples of the solutions I'm suggesting in the form of my own setups.

"I don't have a witty proverb to finish this post with." - Abraham Lincoln

 

It's interesting that on some things I have bit opposite experiences.

For example that mention of body-rolls being excessive is something totally opposite to my perception... but I have based on my observations mainly on defaults with or without major modifications. Those are more obvious in rF2 to me as a comparison (on cars like pCup, GT3, GTE, BTCC etc.) , and that in that sim, the cars feel much stiffer and first thing you usually do is to start soften the rear up first (especially the rear ARB to minimun in a lot of cases).

I do think you are adjusting the setups also to counter some of the things that aren't necessarily going anymore to be there anymore with 1.6 because of the physics improvements.

Anyway, it's good to have some alternative options to try, just because of different driving styles. I

ps. there is just too many cars and combination for Reiza to be able to give multiple setup options for each car, especially with occasional physic updates no matter how much they would like to do that. I honestly think it's more up the the players / community to provide alternatives and share them for people. Let's just try to keep the negativety minimum, the default setup doesn't need to "suck" for some other setup to be better.

 

I think it is just a matter of tone, just consider it is probably not a coincidence that many of us think the tone is harsh, which is a shame because your contribution is nice.

I have had little to no time in the rig recently and It will not change for a couple of weeks but perhaps this friday I will check some (prob the gt3 merc).

ps: ROFL on your last sentence, nice one

 

Hello Horia!
Thank you for sharing and the work you have put in here. I'll be testing a few of your setups although I actually find Reiza's standard setups very balanced and accessible and a good starting point.
I still have a few questions for you about these setups.
1. Which base did you test it with?
2. What in-game settings did you use for this?
3. Have your setups been tested with Default or Default+?

Thanks