Hello, and welcome to the big Crowd Sourced Challenge Guides post! Inspired by this post, I figured that it’d be good to collect a bunch of useful resources in one spot. What goes here:
Typical host expectations on realism.
Features that are important to include in a challenge.
Useful resources for challenge hosts.
Guidance on how to host challenges.
I will be including an index of posts to make them easier to find, and some links to other useful forum threads. Feel free to comment or DM with extra suggestions!
To start off, let’s list some of the essential challenge elements. These are the sort of thing you should specify in every challenge. Consider this a checklist of what to include when posting your challenge:
What are you looking for people to make?
– Can you specify what sort of cars you want? Econoboxes, luxury sedans, sport cars, utes, whatever. Make it clear and obvious.
– I would caution against making it too set in stone. Your challenge might be “make a perfect replica of a VF Commodore”, but that doesn’t give people too much freedom to exercise their creativity.
– Some challenges exist that allow anything to be submitted, taking anything from utes to sedans to roadsters. Shitbox Rally, Cool Wall and others. Generally, you should still make it clear what a successful entry looks like, which leads us into…
What will be used in judging the car?
– Which stats will be most important? Which stats will you look at, which will you ignore? Selecting these well is an art in and of itself, which will be included elsewhere.
– Will you test the car in BeamNG, or just Automation? Tuning for the two is rather different, and not everyone has BeamNG.
– Will you be judging interiors? They are fiddly and many people prefer not to, but there’s a lot of potential in there.
– How important are aesthetics?
What sort of engineering is allowed?
– What year do cars have to be? There are four years - two engine years, two chassis years.
– What techpool is allowed? That’ll be discussed elsewhere, it’s a massive can of worms.
– Are you going to restrict quality?
– How much can cars cost? You can set a hard limit (binning cars that cost more than a certain amount) or a soft limit (needing a reason to cost that much) or both.
– What fuel is allowed? Consider what was on the market during that era and locale, what would have gone into the car.
– Is a catalytic converter needed? Does it have to be a three-way? (Three-way is almost always better than two-way anyway but…)
– Are particular parts banned for realism or balance reasons? It’s common to disallow race parts, for instance, and also to ban V16s.
– Are you going to force any minimums or maximums? A minimum number of seats, a minimum safety type, a maximum loudness?
What sort of fixtures are required?
– Most cars should have headlights, indicators, taillights, reversing lights, mirrors, a gas cap, wipers, etc. This is practically an unwritten rule.
– Are you going to specify a particular size and shape? Until 1983, the USA mandated a particular size and shape for headlights.
– Do you need to have a particular bumper? This was common in the US too.
– Are aerodynamic fixtures allowed? How many can you have? For 50s challenges, restricting them is common for realism.
What should people name their models and families? It’ll make it a lot easier for you if everything has a consistent naming scheme, so you can spot them in the car list.
Do people have to post an ad in the thread, or can they just submit quietly?
When will submissions open? It’s common to have a short gap between posting and opening, to allow for rules changes.
When will submissions close? Ideally, you should specify a date, time and time zone to make it clear.
There’s more to hosting than this, of course, but answering the big questions makes everything smoother.
Position of the driver’s side depends on the market, but I’ll always assume that mid-hand drive (in which the driver sits on the centerline) is allowed unless the host states otherwise.
Okay, so we have the barebones of a challenge, but how do we make it actually fun? Let’s go specifically with how to make engineering fun. Not all stats are created equal. Some stats are just better to include in challenges than others. But which ones? I’m going to evaluate just about every stat with a view to creating a challenge that provides an enjoyable engineering experience and gives players choice and agency.
Doing it yourself
Let’s start by reviewing my methodology, to compare a “good” stat to a “bad” one. This is all opinion, of course, and it’ll show how I formed that opinion. So, let’s start by going into the detailed stats pane, and looking at the driveability of the Sarek II Twin Offroader I made for LHC4.
We can see here that driveability is impacted by a lot of factors. This gives me, the challenge entrant, a lot of freedom in how to improve this stat. There’s a lot of levers to pull. Even if the host takes some away (perhaps by strongly incentivising use of manual gearboxes or having the challenge early enough that ABS or power steering don’t exist), there’s still a lot I can do. Now let’s take a look at practicality.
Mm-hmm, see that? Far barer, far fewer options. Worse, if you have a heavily restricted set of bodies to use, many of the levers are taped off. This is most pronounced in challenges like Automation Restomod. The body being set means that you can’t really impact width, height or spaciousness that much. There’s a little but not much. So, if you have a base car with five seats and bin anything that drops practicality by more than 10%… You have effectively banned 2-seat cars, and near-banned 4-seat cars. Not great.
My opinions on stats, as of LCV 4.2.42
Broadly, I group stats into four classes. Good stats, black-boxes, limited stats and broken stats.
Good stats are ones which allow for many meaniungful engineering choices, having interplay with many other syetsms. Ideally, some or all of these choices should have impacts elsewhere. Examples are drivability, comfort and sportiness (all of which impact one another), price (which impacts just about everything)… You get the idea. Not all stats here are equal.
Black-boxes are like good stats, but it’s not clear what impacts them. If a stat isn’t in detailed stats, there’s a chance it goes here. Reliability is the biggest example, I still have no idea what impacts it besides quality and certain choices. If you can’t tell what a car does to do well or poorly in a stat, it’s a black box.
Easy stats are ones that don’t have many choices. Practicality to a degree (we’ll get to it next), environmental resistance… Click button get stat. Easy, not a lot of meaningful stuff to be done to compete. There’s one choice, and that’s what you should use. If there’s an obvious right choice, stop and think. Maybe the right choice to maximise one stat hurts another (for example, corrosion resistant steel is expensive). Not saying never use these stats, saying to be careful.
Broken stats are just that - broken. For whaqtever reason, they’re super easy to game, don’t work hor you might think, return nonsensical values. Load capacity and towing capacity are the two biggest offenders - and they impact heavily on utility, and somewhat on practicality.
I’m not going to classify everything here, because that would cause arguments. Some stats clearly go into good, some are a toss-up between good and easy. Most importantly, it changes based on the challenge. If entrants have to use a set number of doors and/or seats, practicality’s already-limited options become even more limited. The fundamental principle here is to give people meaningful options, and to allow them to use those options.
Now, for one of the biggest topics in challenges: realism. I’m gonna be real with you, this one is a crapshoot. As a challenge entrant, it’s hugely at the host’s discretion. Why? I think it really comes down to expectations and communication not being where they need to be. If you’re a challenge host, I implore you to read this post if it’s the one thing you read in this thread. Here, I will discuss three broad approaches one can take to realism. I urge challenge hosts to communicate which they intend to use. WIthout further ado…
Approach One: Go Nuts
Yes, approach one is to simply have no approach to realism. Whatever you get, you judge on its stats and don’t ding it for anything at all. This approach is rife with exploits, like cars having wildly imbalanced downforce to the point that they’d do this on any straight. You can ban the stuff that causes this, you can regulate it, just be aware that realism is a way to rein in exploits.
Approach Two: The Verifiably Real
This is the other end of the spectrum, in which you only allow what companies were actually doing at the time. Here, I will urge caution as well, because there is always some obscure car that did some weird thing, so you run the risk that you would bin a real car for… not being realistic. Some people advocate a “two car rule” where you allow something if you can find two real examples. The question, then, is whether you want to do the research to verify the realism of oddball choices.
Approach Three: Internal Consistency
And now we get to the middle, where you look at the car on its own. Take, say, inline-fives in 2020. Nobody uses them any more, Volvo stopped making them. Are they that unrealistic though? Could someone have actually kept making them? Yeah, sure. Similarly, seven-speed automatic gearboxes were introduced in 2003, but would they have been that unbelievable in 2001 (doable with default techpool)? There’s plenty where it didn’t happen in reality, but where it does make sense. There’s also plenty where it doesn’t make sense, like a transverse-mounted boxer engine (which has severe packaging problems with the gearbox).
How To Communicate This
I think that everyone in the community would be served if hosts actually posted what they expect in a challenge. If hosts didn’t just say they wanted realism, but if they specified that they wanted entries mirroring reality, or if they specified that they wanted entries which make sense even if they didn’t actually exist. I’m not going to come up with an exhaustive list of what is or isn’t realistic, because I generally don’t host like that.
Ok, beyond just enforcing Cunningham’s Law, this example does provide a great example of how easily people’s knowledge/opinion of “realism” can differ, and why it’s such a “crapshoot” as you said earlier.
A given approach-two host could easily allow I5’s in a modern challenge. because they can buy one from a major manufacturer. Or they could easily ban I5’s in a modern challenge, 'cause there isn’t enough for the 2-car-rule mentioned earlier, or 'cause the only example is way outside their challenge’s scope. And both decisions would be valid even after a posted expectation for approach two.
And that assumes the same host, with the same background knowledge; another host might not know about the RS3, or might be unaware Volvo stopped making I5’s. And etc, and so forth…
Who the heck in their right mind would actually host like this? It literally just invites salt and stifles creativity. I get wanting to have things be era appropriate, but just yeeting a car for something that “no longer exists irl” (i.e. the i5s you mentioned or anything other than rear engine boxer motored sports cars) is just asinine and will pull so many “um actually” moments.
I personally believe that something closer to the internal consistency model is the best and more common approach to realism in challenges (at least those that aren’t already incredibly prescriptive with their rulesets). Question much more modern engineering choices for the era, car class, and country of origin (like said 7 speed in 2001). Allow obsolete engineering choices to still stand because they may be integral to a brand or be a market specific cost cutting measure and let the stat hit take it’s toll. And, as you said, bin things that are truly out there and make absolutely no sense like a transverse fwd boxer or a 1.8 L V16. Realism =/= only things that actually existed in X year.
One engine I know for sure is controversial most of the time is a 60 degree V8, since there are only 2 examples in history that I can think of, those being the Taurus SHO V8 and a Volvo V8. Is the general consensus that using them in any period that isn’t the 90s-00s means straight down to bin hell?
Not necessarily. They could be used to simulate even narrower v angles of earlier cars that we closer to 45 degree Vs. This also rules out it being chosen for packaging purposes, which the Volvo and Yamaha V8 were used for. There are legitimate use cases for a narrow angle V8 outside of a specific time period. Doesn’t mean they won’t be looked at oddly though.
This is all you need to do - now (almost) all of the cars’ statistics and design choices are in one sheet, where you can pick through them. Personally, though, there are just too many columns to really work with, so the rest of this guide will be devoted to filtering off and visualizing the data you care about.
Step 5: Create a new sheet.
Step 6: Find some data you care about and paste it into the new sheet.
Copy these columns across first and foremost, just for organization’s sake.
Fill the rest of the sheet with formulas to make calculations easier.
Some final words of advice:
This will never totally replace having Automation open. There are stats you’ll need to go through and manually record. Estimated cost appears nowhere on the sheet. But, even with all this setup, I hope that this both saves you time and, more importantly, makes the process of running a big challenge feel less grating and monotonous and more fun.
I’d advise you to not do this until all of the entries are officially in. With this method, there’s no easy way to add in new entries after the fact - you’ll have to essentially do step six again.
Try to make as few changes to the first sheet with the raw data as possible - you’ll never know when you’ll accidentally edit a number on the second sheet and need to grab it from the first.
One of the most powerful tools for analyzing data is filtering. To create a filter, select your data and click the create filter button under Data:
Sorting A-Z or Z-A will sort all of the numbers from smallest to biggest or vice versa while keeping the data for each car together.
Finally, mess around with conditional formatting. It’s a really powerful tool, and a little knowledge goes a long way.
As an alternative to copying across data, selecting a column then right-clicking it will allow you to hide the column. You can’t re-order, but it is good enough to get rid of stuff like game and exporter version.
