Home | Wiki | Discord | Dev Stream | YouTube | Archived Forums | Contact

Automation Autoshite Challenge (Cancelled)


#1

Automation Autoshite Challenge

Open Beta Challenge!



This thing’s grim as fuck.

What is this?

It’s a car shopping round. Except isn’t.


The year is 2019.

Meet Andrew, he’s an average guy working average job in the north of England.

But Andrew has a nasty habit.

He’s a full blown obsessed shiter.

Andrew has a fleet of shite cars lying around his front yard like an ornament. From basic old hatchback to a really obscure French saloon, he has it all.

The problem arise when his daily driver, a shit old ex-minicab bought for £200, conked out on the motorway this evening.

And pretty much immediately, Andrew grabbed his phone, and start browsing Gumtree, Autotrader and eBay.

Andrew vowed to himself ages ago that he’ll never spent more than £1,000 on a car, ever. Why would you spend that leasing a new car when a perfectly serviceable old chod is available everywhere?


Wait, £1,000? How does that work?

It works with the wonder of Excel Spreadsheets.

A long time ago on a CSR, @Dragawn created a spreadsheet that simulated the used car price. It’s a neat little kit that’s very good for representing used car market value. But that was outdated and wouldn’t work.

So he created a new spreadsheet. Which simulates used car market even more accurately.

It uses in game value to calculate 3 scores for each car. Daily driver value, classic car value, and sports car value. This allows old piece of junk to be worth a few hundred currency while a cherished classics can have appreciated value. Also, sports model can have more value than basic old piece of shit car too.

This allows you to create something that’s utterly worthless. Perfect for Andrew.

More than that, I will also be utilising my own Insurance Rate Calculation, which is much simpler. But also created additional challenge for you. This utilised stats such as Sportiness, Prestige, Horsepower and more to simulate UK Auto Insurance purchase. Do not worry that you’ll be paying £3,000 for a £200 car, however. As Andrew is not 17 year old.

This sounds complicated.

You’re right, it is.


Rules.

Only a few simple rules.

Trim Year: 1970-2006.
The car must be road legal to the trim year. This means a few things.

  • Your car must have all the required fixture, headlamps, taillamps, indicators, door handle, visible fuel filler cap, mirror.
  • Leaded Petrol is only permitted for any car made before 1989. But you may go Unleaded before then.
  • Catalytic Converter is mandatory post-1989.

Cost.

As stated, the budget is £1,000. Using this spreadsheet to determine the price of the car.
Price Calc.zip (10.5 KB)

Also, yearly Insurance rate is capped at £500. Using this spreadsheet.
Insurance Calc.zip (8.1 KB)


Judging

The scoring will be done using anorther formula table that has been through a very thorough testing. The concept used is very similar to the one for demographics formula in the game. Courtesy of Leonardo9613

It utilizes the bell/normal distribution, a concept that is widely used in statistics, to represent the behaviour of several natural systems. By calculating the integral of the curve up to a certain point (the area underneath the curve), it is possible to determinate the percentage of items that are before that car. For example, in the old-fashioned IQ tests, if a certain subject had a score of 130, it would theoretically mean that the subject had a higher IQ than 97.8% of the population. This method can be used for comparing cars, because the vast majority are very close to average, whereas the few outliers will get their score accordingly.

In this case, for this challenge, the automatic scoring sheet calculates the standard deviation, the average and all the coefficients depending on the cars tested, therefore the formula only works with more than one car. If, by chance, all cars are very closely matched in a certain area, the sheet adapts itself to have a higher standard deviation value, thus dampening the otherwise very big variations that could follow.

The scoring works as follows: a value of 50 means that the car is perfectly average, 84.1 means that the car is one standard deviation better than average, in other words, a lot better than the class.
The final score is an weighted average of all relevant stats. The weights were set after extensive testing and are set to represent the relevance of each stat for the buyers.

So what exactly is Andrew looking for?

Everything, basically.

There are priorities to this purchase, but basically every stats is going to be looked at.

High Priority: Comfort, Safety, Prestige, Performance (Acceleration, Speed, Lateral Grip, everything), Cost (To buy, to service), Reliability. Maybe more.
Medium Priority: Drivability, Seat Count.
Low Priority: Design, Practicality, Utility.

There are 2 ways you can go with this. Either specialised the car in specific area, or make a jack of all trades. The cars are compared relative to each other.

Also, the types of car will not have much importance as long as the stated stats are okay. This means anything goes, Sedan, SUV, Van, Shitpost, Convertible Sports Car. As long as it fits in the budget.

You can either go with older, more luxurious car, or more modern but cheaper model. There are advantages to both.


Submission.

Basic naming system.

Engine name: SHITE - [forum username]
Engine variant: [engine name]
Car name: SHITE - [forum username]
Car trim: [car name]

As the ruling is extremely lenient. No resubmission will be allowed.


Inspiration?

Look at the prime chod in the top picture. Alternatively, anything that is going to appeal to Autoshite forum.


This challenge is for now cancelled. Sorry. It might come back one day.


#2

I suspect the absence of PU/ET and material cost limits is due to the fact that the price limit of a thousand pounds (used) is enough to deter someone from over-engineering their entry too much.


#3

That is correct, partly.

I expect over-engineering. The used market means that you can get really advanced older car for cheap from depreciation. That is indeed one of the way to make a car for this.


#4

may i suggest putting the tables into some google drive documents?

not everyone has Microsoft Excel and thus is incapable of doing the calculations because the OpenOffice knockoff won’t cut it


#5

Will do!

Edit: Added to the original post.


#6

you also should add a “copy me to edit” thing to avoid others revealing their stats


#7

I think for those who don’t wish to reveal their stats will already copy the spreadsheets to their account.

But if some want to do it in there and don’t care much about revealing the stats, I am not going to stop anyone from doing so.


#8

Oof, even my total shitbox is too expensive image

I have to create a total small shitty car


#9

is the displacement rating for insurance n litres or cc?

EDIT:
also, how does price alter the clients preferences?
is he willing to sacrifice some stats for it being cheaper or will he deffo spend the 1000 Pounds?


#10

Displacement in litre.

Every stats is being looked at with the priority stated in the brief. Every entries are compared relative to each other. With 1,000 pounds you definitely have to sacrifice something or making a jack of all trades.

Another thing to note is that price is a judging criteria as well, so you definitely also have a reason not to max 1,000 out.


#11

rowaneyes3

Don’t suppose you want any help from a guest judge there, conan?


#12

Do you want to judge the styling of these cars? :joy:


#14

Even though with the calculations as they are, this is literally not possible. And then you even go and contradict yourself like this:

It seems to me as though you have an idea of what kind of car you want already, and you’ve shaped the calculations to make people to go in that direction.


#15

Well, yeah. I gotta have vaguely an idea of what sort of entry I’d like to see. Else this challenge scope would be very wide. From what I see on Discord people are already making all sorts of car, Yugo, Smart Roadster, LDV Van, Chrysler LeBaron. I quite like the variety that spawn from people’s interpretation.

If you’ve found the calculation to have issues, that will have to get fix though. So, I need to know what this issue is and what can be done.

The challenge is not started yet so anything can still be fix.


#16

you have to sacrifice everything a car that sold to the budget family market in arhana in the 80s is like 4k£ on the calculator XD


#17

Perhaps, not sure if that may give an unfair advantage to those that know me and my tastes though…


#18

26p to spare, i think i’ll get a freddo


#19

on hold

damn now what I’m going to do with this perfect proletariat car from 1996?


#20

I’m sorry to anyone who’ve put effort into making a car in advance. But I have made a decision to cancelled this challenge, for now.

There are a few reasons for this. It has become obvious that the challenge has too many errors and design faults, while most of them can be fixed pretty easily, some is much harder. I don’t think it’s impossible to fix all the faults but it’ll take time.

The issue is, in a few days it’s a possibility that I won’t have internet AT ALL (going into rural area). And if the challenge still has teething issue at that point it’d be difficult to remedied stuff. And if I wait for when I return all interest will be lost anyway so I might as well start from scratch.

This challenge might come back when (and if) all the bugs and concerns are sorted out. I hope the used car calculation get implemented elsewhere soon as it’s a nifty bit of kit (contact @Dragawn if you’re interested).

Hell, if anyone is interested in the idea, they can use it for their own challenges. I would love to see the idea of making utter shitbox being realised (properly, that is).

So long.


#21

Just delay the opening of submissions. I’ve got two crap cars, both lore, and cheap as chips. Got one I was hoping to offer Andy that’s blown his budget clean out of the water…