Weeding Engineering/Rossini Automobilli S.p.A

Like most people, I’ve wanted to design my own sports car - I’ve got a very large number of drawings of the same supercar concept that I want to build one day.

Most of the time, I design my vehicles to a market of one person; me. However, as that only requires three cars to cover every car I want, most of the thread is going to be cars that I’ve designed for even more imaginary brands, or cars I’ve designed to amuse myself, or cars designed for various challenges.

The current sub brands are:
Falcon Cars Inc. - a High end luxury/sports brand, starting from 1987
Riviera Automotive - Luxury front-engined cars, from 1946 to present
Boare Commercial Vehicles - a brand for vans, pickups and other vehicles of that nature, '46 to present

But, I mostly specialise in the hypercar and luxury premium markets.

There will also be any test results I fancy posting somewhere having assembled the data I have for the Steam forums, which I’ve come here from…

1 Like

A copy of a post I made concerning laying out a design brief for a car, with the example being the Luxury Premium market.

"Obviously, Automation with the Unreal engine has the big advantage that you have this wide open sandbox - but, at some point you have to actually sell your constructions. So, with the latest iteration of the campaign due soon, some ideas and principles to practice with:

You might think you have a fantastic car; heck, you might actually have a fantastic car.

But, if it costs beyond what people can afford, you won’t sell any of them, and equally, if the market has moved on because you took too long designing it, you won’t sell them.

When you are designing a car, if you follow the order the game presents things, you don’t actually get to these key elements until the very end of the process - the markets screen.

Now, don’t be put off because it’s this huge load of boxes with names and numbers and stuff, and you haven’t got a clue what any of the icons at the bottom mean.

The first thing to check, is which country are you looking at; it defaults to Gasmea.

Next, is the two numbers below it - margin, and price.

Margin - this number is the amount, in increments of 10%, of the production value of the car, added to the production value of the car, to give the “for sale at X” price. (because it is per car, it can be deceptive, and you need to have an idea for how many cars you expect to sell; a 10% margin on 2 million cars will likely be viable and make a profit, but a 10,000% margin on 1 car will almost certainly be a loss)

Price - this is the amount people will have to pay in order to buy the car, and is composed of the materials/labour required to build the car, and the overheads involved, and your margin.

Obviously, these numbers are kind of important. Even more important though, is to look at what the situation will be like when the car goes on sale; you have to remember it takes time to design a car (real world, typically 3-5 years) and so, you need to consider the lifespan of the vehicle as it is on sale, as well as what your competition will be like at launch and the time after; remember, some of these markets have 40+ competitor cars in them.

It is no good merely being competitive with the cars already on sale when you design your new car - take that approach, and your car will be out of date when it goes on sale. That is not good. You want at least 110% AFTER adjusting for markup and the lifespan of the car’s on sale time, as that will mean it will still be competitive when the replacement becomes possible.

This can be checked by adjusting the trim year at the top - just remember to put it back before you start revising the car, as options available change over the years.

The most effective way, is to do market research. Automation has 60 demographics of car/van/pickup buyer, tracking differences in best part of a hundred variables. (see the detail stats screen for the whole lot, not just the 14 displayed at the top left side)

One of my preferred markets is the luxury premium market - this, in my view, is one of the best markets to target; low competitor count, with the richest customer base, with really simple design briefs.

First off, a deeper look at the market:

Hover your mouse over the luxury premium box in the markets page. The left panel is about how suitable your car is, the second panel is about the market itself.

With the luxury car market, the 1st priority is comfort with 35.8%, the 2nd is prestige with 31.8%, and 3rd is safety, with 14.8%. After that, there’s only that they like big multicylinder engines (which you were doing anyway to get the high prestige) and they like a bit of drivability, and they like a large size. (which with the first 3, you were going to end up with anyway)

So, you only actually have 3 elements to worry about - and they are all easy elements.

Comfort - this is how nice the car is to the occupants; they don’t want to be suffering with a loud, harsh engine note, they do want to have a lovely spacious and luxurious interior, they do want the car to do as much for them as possible, they want suspension that floats over the road; you get the picture.

What this means for us? Big, quiet and yet melodious V12 or V16, Automatic gearbox, luxury/hand made interior, soft suspension. Oh, and 4 seats - don’t do 5, that’s less comfortable.

Prestige - this is a complicated set of factors that could be summarised as what you would put on the Top Trumps cards; it’s how impressive the car is, it’s how advanced it’s materials are, it’s how big and powerful the engine is. Basically, a load of stuff you were going to build up anyway. If you really struggle, just whack up the quality on something important.

Safety - people like the idea of car crashes not being dangerous. This is aided by having an awful lot of metal to use as energy absorbing material, and so, just put in the advanced version of the latest decade and you should be fine. (you can save costs by using a lower grade, but remember this is important for how good your car is percieved, so cut costs elsewhere really)

Drivability - as long as it isn’t boat like, it’ll do. And given some of the luxury cars over the years… No one is expecting a sports car here.

Engine - as said, earlier, they like big, and they like lots of cylinders, which you already did for the comfort boosts.

You could probably design an ideal luxury premium car by accident - there are three general methods on how to design a car; one is to do what you feel is right (with practice you’ll get better at knowing it) 2 is to identify what each option you have at each stage does and how it impacts the overall vehicle, and 3 - fill all the component slots until the market screen opens up, select a market, then build whatever things increase competitiveness, and try not to sacrifice affordability. (it’s no use having a brilliant car that no one can afford)

The same process works for any other market - you identify what market you want to target, then you figure out how many cars you can sell, and for how long you can sell them, and in which markets, then design the car.

There will be times when one car and one trim will adequately cover multiple markets - there will also be times when you will need to add extra trims to better cover other markets; just remember to check through profitability - it is not much use adding another trim if not enough demand is present to be worth the extra designing time and factory costs.

I know the mathematics is likely boring, but, if you bear with it, it means you get reliable income, which means you can fund all the crazy stuff that you actually became a car designer to do. :)"

Another copy and paste - I would particularly be interested in someone else verifying the data if possible.

"Using the normal 2010 supercar body shape as the base, without changing the form, and using a 2400 bhp engine I made earlier, and setting the car to be relatively light at 1363 kg:

Firstly, no idea how to actually tabulate it properly with Steam, so the presentation looked ok in the editor…

To keep as many things constant for verification of data, all fixtures are being placed without being rescaled, and are listed from left to right and top to bottom, and all downforce/lift numbers are quoted at top speed of the scale, while gearing is kept constant to the original car pre-testing, and relevant qualities are at 0.

First batch are varying the front lips - all placed at the bottom and centre of the front bumper

front wing | rear wing | TS (mph) | Cd | Fr Df | Rr Df | Fa (m^2) | m (kg) | Under tray |
N/A | N/A | 282.6 | 0.478 | -125.7 | -133.0 | 1.14 | 1363.0 | N/A |
lip 1 | N/A | 279.1 | 0.494 | 4.1 | -154.3 | 1.18 | 1363.0 | N/A |
lip 2 | N/A | 279.1 | 0.494 | 4.1 | -154.3 | 1.18 | 1363.0 | N/A |
lip 3 | N/A | 279.1 | 0.494 | 4.1 | -154.3 | 1.18 | 1363.0 | N/A |
lip 4 | N/A | 279.1 | 0.494 | 4.3 | -154.8 | 1.18 | 1363.0 | N/A |

Second batch are varying the rear wing - all placed as close to the back of the car as possible, at centre (note: while all 15 base game wings were tested, one of the wings got missed in compiling the data set, although the data suggests that shape is not significant anyway)

N/A | Wing 1 | 268.5 | 0.525 | -163.7 | 213.1 | 1.26 | 1363.0 | N/A |
N/A | Wing 2 | 268.5 | 0.525 | -167.1 | 216.1 | 1.26 | 1363.0 | N/A |
N/A | Wing 3 | 268.3 | 0.525 | -169.0 | 218.0 | 1.26 | 1363.0 | N/A |
N/A | Wing 4 | 268.1 | 0.525 | -170.1 | 218.8 | 1.26 | 1363.0 | N/A |
N/A | Wing 5 | 268.1 | 0.525 | -170.6 | 219.6 | 1.26 | 1363.0 | N/A |
N/A | Wing 6 | 268.0 | 0.525 | -170.7 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 7 | 268.0 | 0.525 | -170.5 | 220.1 | 1.26 | 1363.0 | N/A |
N/A | Wing 8 | 268.1 | 0.525 | -170.6 | 219.6 | 1.26 | 1363.0 | N/A |
N/A | Wing 9 | 268.0 | 0.525 | -170.7 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 10 | 268.5 | 0.525 | -167.7 | 217.0 | 1.26 | 1363.0 | N/A |
N/A | Wing 11 | 268.0 | 0.525 | -171.6 | 220.7 | 1.26 | 1363.0 | N/A |
N/A | Wing 12.1 | 268.0 | 0.525 | -171.8 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 12.2 | 268.0 | 0.525 | -170.6 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 12.3 | 268.0 | 0.525 | -170.9 | 220.7 | 1.26 | 1363.0 | N/A |
N/A | Wing 12.4 | 268.0 | 0.525 | -170.6 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 13 | 267.4 | 0.525 | -176.5 | 225.5 | 1.26 | 1363.0 | N/A |
N/A | Wing 14.1 | 268.0 | 0.525 | -170.6 | 220.2 | 1.26 | 1363.0 | N/A |
N/A | Wing 14.2 | 268.0 | 0.525 | -170.7 | 220.3 | 1.26 | 1363.0 | N/A |
N/A | Wing 14.3 | 268.0 | 0.525 | -170.5 | 220.1 | 1.26 | 1363.0 | N/A |
N/A | Wing 14.4 | 268.5 | 0.525 | -171.8 | 221.6 | 1.26 | 1363.0 | N/A |

Commentary - it is highly interesting that the game records no variance with Cd, frontal area and mass, despite being able to distinguish different wings in effects on top speed and front/rear downforce/lift.

Third batch - this time varying the undertray.

N/A | N/A | 282.6 | 0.478 | -125.7 | -133.0 | 1.14 | 1363.0 | N/A |
N/A | N/A | 282.6 | 0.478 | -127.2 | -132.6 | 1.14 | 1368.0 | Offroad skid |
N/A | N/A | 286.3 | 0.459 | -123.4 | -129.1 | 1.10 | 1379.6 | Semi-clad |
N/A | N/A | 289.4 | 0.441 | -116.2 | -120.7 | 1.06 | 1363.0 | fully-clad |
N/A | N/A | 288.8 | 0.446 | -83.7 | -85.8 | 1.07 | 1396.2 | Downforce 0
N/A | N/A | 283.7 | 0.470 | 26.0 | 29.0 | 1.13 | 1396.2 | Downforce 50
N/A | N/A | 272.1 | 0.524 | 148.4 | 155.3 | 1.26 | 1396.2 | Downforce 100

After further testing using Lip 4, wing 13 and the downforce undertray, can confirm that scaling of parts does not effect downforce/lift generated.

Fourth batch - this time with all the trimmings one expects a car to have - exhausts, mirrors, lights etc. 1st car nothing, 2nd car with extras.

N/A | N/A | 282.6 | 0.478 | -125.7 | -133.0 | 1.14 | 1363.0 | N/A |
N/A | N/A | 282.6 | 0.478 | -125.7 | -133.0 | 1.14 | 1363.0 | N/A |

Fifth batch - placing a Lip 4 on the sides of the vehicle, with the intent of it acting as a side skirt. ( to clarify, again not altering the scale)

N/A | N/A | 279.1 | 0.494 | -78.2 | -74.6 | 1.18 | 1363.0 | N/A |
N/A | N/A | 269.1 | 0.539 | 188.6 | 203.3 | 1.30 | 1396.2 | Downforce 100

Sixth batch - adding a modded rear diffuser to the 2nd fifth batch car. (the mod in theory counts as a lip, so this could be interesting)

N/A | N/A | 282.6 | 0.478 | -125.7 | -133.0 | 1.14 | 1363.0 | N/A |
N/A | Diffuser | 266.1 | 0.555 | 163.1 | 312.1 | 1.34 | 1363.0 | Downforce 100

Commentary - owing to only being able to ensure consistent placement by eye, I cannot be sure differences are substansive to wing itself, rather than very slight differences in placement, with the small scale of difference, non-differences in 3 tracked variables that should be affected, despite massive dimensional deviation, and the knowledge that scaling does not affect downforce/lift, implies to me that the game treats wings quite simplified, barring outliers like the ducktail (14 in game, 13 in the list above) however, even that outlier is within less than 1% of the others.

Conclusions - more investigation is definitely necessary into how the placement of fixtures affects aerodynamic properties, in particular, I feel the interaction of using lips as side-skirts definitely needs more research, but I feel the testing I’ve done shows that what model of fixture it is and the size of that fixture, does not appear to be a significant factor compared to what type of fixture the model of fixture is."

And lastly, an overview of the only car I’m semi-happy with as far as transfer to BeamNG goes - the current iteration of the car mentioned in the OP, currently known as the G57 JGW, for the plate I’d quite like to put on the actual car if/when it gets built.

I learnt quite a lot about the differences in the aero-modelling between BeamNG and Automation from this car, and even un-optimised with a keyboard driven by someone with dyspraxia (that is not a good combination) it can lap the Automation circuit in 2:03.689 and Hirochi Raceway Long in 1:30.895, both from standing start, and more importantly, clean laps.

The handling is… unpredictable at low speeds, but it settles into understeer at high speed - despite a 724 bhp NA 4.6L V8, top speed is aero limited (1.5 tonnes of downforce in Automation, with about half coming from the monster 4 element rear wing; not sure in BeamNG though) to 157 mph.

Fuel consumption is poor, dropping to 4 mpg on track, but holds at a steady 17 mpg if driven sensibly.

The new smallest car in the range - 2.3m wheelbase, with relatively minimal overhang. Now featuring Chrome wheels and window trim. More importantly, with a revvy 2.0l NA V12 delivering 270 bhp in fuel economy trim -25 mpg according to Automation- and a lower downforce package compared to it’s larger relatives; still nearly 400 kg at top speed of 124 mph though.

Handling is still awaiting finishing touches though, it isn’t as stable as the bigger relatives.

Downforce verification testing.

As is well established, the level of cross-over between Automation’s simplified treatment of fixture induced aero (which only factors location of the aerodynamics component relative to the centre of the car) to BeamNG is unknown.

I will endeavour to establish how much crossover there is, and in particular, which fixture is best, with best defined with three variables; most downforce, lowest drag, and downforce/drag ratio.

Methods:

Testing drag is quite simple - using a car with a very, very big engine to enable very high speeds, the effect of varying aerodynamic devices on drag can be understood from the change in top speed of the vehicle. For this test, the best wing is the one with the highest top speed.

Testing downforce is more complex - from first principles, we know that downforce pushes down upon the car, and therefore generates additional tyre loading, and therefore increasing cornering grip. Therefore, if one was to take a car of otherwise fixed status over a predefined circle, the downforce can be quantified by the variation in speed at which the circle can be completed. An additional method that has less accuracy but better verification of data, is to use a vehicle with soft easily compressed springs, and very high ride height, and compare the change in height between stationary and top speed; an increase in downforce leads to a reduction in ride height. For this test, the best wing is the one that produces the largest change in ride height.

For the best downforce/drag ratio wing, the process compares the ratio of the speed lost with the reduction in ride height - the ideal state is a wing that produces a very high amount of reduction in ride height, with a very low reduction in top speed. which means, bigger numbers are better.

Known difficulties:

Precision of measurements - while speed is easy to quantify, BeamNG’s speedometer only runs to a precision of +/- 0.5 mph; this variance is far greater than the variance on the majority of wings from the previously obtained Automation derived data tabulated above, which at the outset of this test, risks making the entire test invalid as a means of comparing individual wings with other wings, although general principles
may remain valid. Of far greater significance is measuring the variance in ride height; while it can be quantified to a much greater level by tracking the pixel count, the method is tedious and prone to counting error; an alternative method of measuring via a ruler or similar device can be used, but has the problem that precision is, at best, +/- 0.25mm. (you can tell roughly halfway between two mm markings)

These two flaws are serious, and therefore a final conclusion on best wings is suspect; never the less, the general principles will be sound.

The car:

While the base of the car is the same as the car used to obtain the data above, there are substantial changes.

  1. Aerodynamics quality - unlike with the car tested above, I have set the AQ to +15. I feel the top speed increase from reduced drag to be considerably advantageous for enabling best possible testing, although the fact it has such a huge impact on frontal area is concerning…
  2. Gearing - has been subtantially altered to accomodate the higher speed enabled by AQ increase.
  3. Front wheel width reduction - in an effort to reduce oversteer, the front wheel width has been cut from 325 down to 155.
  4. The removal of all fixtures apart from the exhaust and door handles - the above testing has confirmed that Automation does not weight fixtures apart from lips and wings for performance; on the assumption BeamNG does count other fixtures, I have opted to delete everything else other than the two essentials. (no one will make a car that doesn’t have an exhaust or a way to get into it, so I feel keeping those is necessary; placement is in the normal places, so any effect should be comparable to “real” vehicles)
  5. Power distribution has been shifted from 36/64 front/rear, to 20/80 front/rear, in keeping with the reduction of the front wheels and increasing top speed.

These 5 changes are the only changes at this stage - it turns out younger-me had tested the first car at the highest ride height possible anyway, and I don’t want to risk messing up the handling by altering the springs now.

The result, according to Automation data is the following:
front wing | rear wing | TS (mph) | Cd | Fr Df | Rr Df | Fa (m^2) | m (kg) | Under tray |
N/A | N/A | 308.0 | 0.364 | -223.9 | -241.6 | 0.863 | 1348.7 | N/A |

I feel it necessary to point out that when comparing the lift quantity to the original car, the original car was quoting at a ~30 mph lower speed; the figures controlling for speed are -187.6 and -200.1 front/rear, which leads to the implication that increasing aerodynamic quality massively increases lift; I’ll have to test that another time, going to -15 aerodynamic quality means the top speed drops below comparable data…

At this point, I now have a vehicle to export to BeamNG.

And, having drained a tank of fuel trying to drive around the issue (which doesn’t take long with the kind of engine needed to do 300 mph) I’ve concluded that there’s nothing I can do to stop it turning left at 200 mph. I’ve also concluded the variance in ride height is so small that it won’t be measurable; no discernable difference in ride height with speed, which frustrates the method I had planned.

So, I now need someone with a steering wheel, because you just can’t do the necessary driving on a keyboard…


I’ll post this in it’s unfinished state, but the end conclusion is I can’t finish it. :frowning:

First off-roader, so the marketability is truly terrible. But, despite that, anything that will climb a mountain at 50-80 mph is pretty good. :slight_smile:

A family of Rossini supercars from 1965 to 2012. Quite pleased with the set; all four can have at least 150% markup in Gasmea (with two able to take 200%+) and keep the hyper market above 100% for ten years. (well, the 2012 one should, but the game ends at 2020; the 1965 one can go on until 1987 with swapping to unleaded fuel, a slight decrease in air-fuel ratio and updated interior features/safety, and the 2002 one stays viable until 2020 too, still high 90s for competitiveness)

All using variants of the same NA 4.0l V12, power levels starting at 260 bhp in 1965, through to 530 bhp in 2012. (not pushed to the limit of performance though, particularly for the '65, the original spec was over 290 bhp, but drivability suffered)

As far as any background lore goes, obviously, copied mainly from Ferrari; another game I play, Motorsport Manager, has their Ferrari copy be called Rossini, and they have a road car factory, so… These four are the top hypercar-category class cars that I’ve done so far, although even with markup they have, well, acceptable, competitiveness in at least one other market. (luxury premium typically) I enjoy making mess around one offs (anything posted with the blue above is; I actually prefer red, but I can’t get a red I like…) but the main goal is to come up with cars I can copy straight in for a prospective campaign.

As these are - for Automation purposes - releasable:

The Rossini Milano SP in 1965 original and 1981 update: (closest in the screenshot)
Rossini 2 - Milano SP.car (24.2 KB)
Rossini 2 - Milano SP '81.car (24.2 KB)

The 1985 Rossini Targa Florina SP:
Rossini 3 - Targa Florina SP.car (23.4 KB)

The 2002 Rossini Diamante S:
Rossini 1 - Diamante S.car (24.5 KB)

And, furthest away in the screenshot, the Rossini LP530 Speciale:
Rossini 4 - LP530 Speciale.car (23.8 KB)

And yes, names are in psuedo-Italian…

I’m quite pleased with this one. 293 mph on cruise control, cruising - not even touching the keyboard. ESC off from start to tank empty. (35 miles later!) It is manageably unstable between 270 and getting to 296 mph though. (once at 296 mph and stable, shortshift into 6th and it’ll be fine)

The 0.4-0.5g pulling to the right is mildly concerning, but it’s only a problem if you try to turn…

Handling is much more touring than sporting; braking against such speed potential does leave much to be desired, but there’s very little wheelspin - unless you want it - due to the very long gearing. (1st 120 mph, 2nd 190 mph, 3rd 250 mph…)

But, it’s not marketable…

So, mild thread revamp - having vaguely thought through the background lore ideas, this thread is mainly going to be the Rossini Automobilli S.p.A. lore thread, with the odd blue car that interests me posted as OOC.

And with the fact there’s a possibility I might be the aerodynamicist/bodywork design guy at a British kit car manufacturer coming up, I am not sure about the legalities of hosting what could become copyrighted material (I don’t want it to be - my attitude is the highest achievement that a motorsport engineer can get is to have their innovation banned for being too good, and the second highest is it being copied by everyone else - but it isn’t my place for that decision) if it really takes off. So, the Weeding Engineering (working title subject to change) side is going to be put to one side…

Now, with update 4.2.1., all my old cars no longer work. Which means I have to start fresh.

Which leads back to Rossini; as already mentioned, it’s a Ferrari-like company, only without the intention to stick to just luxury sportscars, although there certainly won’t be anything “budget”.

As OOC background lore - as stated earlier, Rossini is the name of one of the race teams in Motorsport Manager (I suppose most people on here have heard of it) and well, they have a road car factory. Which leads to thinking through the past history of their road-car division. (and occasional track cars)

So, founded in the 1930s by former-racing driver Ezio Rossini (small homage) as a racing team, quitting activities during the war, restarting road car production to fund racing in 1946, with a variety of models all built around the common 3 litre V12, with this V12 powering every car but one the road-car side of the company makes until Ezio’s death in 1987. (the “but one” still had the engine as an optional choice)

The first road car was the 150GT, named for the power output it possessed, and the target market.

14 years later, the 150GT would go out of production, replaced by the 200GT. (same name logic)

Coinciding was the first mid-engined Rossini, the LM240 sports car, intended to dominate GT racing before anyone else caught on, following the success of mid-engined cars in single seater racing. It wasn’t as successful as planned as it lacked straight-line speed and had difficultly adapting the ER1-60’s suspension to a road car body, but it was still the first mid-engined Rossini, and Ezio reflected on it proudly.

In 1984, a now elderly Ezio recognised that in order to bring a new generation of racing drivers into the Rossini fold (and therefore give them the scouting advantage over competitor teams by encountering rising stars first) required a new spec series car, where in equal machinery the very best drivers could prove themselves. The result was the FR150 (Formula Rossini, 150 bhp) - powered by either a 150 bhp (the vehicle is after all, intended to be driven by teenagers; competent teenagers yes, but still teenagers) Inline 4 (the only Inline 4 Rossini would ever make, and the only engine in a Rossini not to be a V12 under Ezio’s leadership) or, for older, wiser young racing drivers, the FR270, a 270 bhp version of the V12.

In 1987, Ezio signed off for production the last car he would design, the 260SPE - the latest in the line of road-going MR Rossinis. His death a few months later however, meant the end of the Rossini V12, as Alberto Rossini disliked the V12, and chose to broaden Rossini’s engines; the new flagship supercars would be given V6s.

In 1990, Alberto decided that Rossini needed to break speed barriers - everything was put on hold for the development of the 900SPR, a 240 mph hypercar powered by an 8 litre V10.

Which, in 1994, promptly near-bankrupted the company thanks to a series of crashes during top speed testing, forcing the car to be redesigned again, and Alberto’s mismanagement meant the company had been losing sales already and now had multiple 8 figure lawsuits. The racing team, thanks to some mildly shady sponsorship deals, was fortunately flush with cash, and so had to bail out the road car division, which went through a period of merely supplying spares for Rossini’s already on the road.

And that covers the cars done so far.

I’ve still not settled on the consistent looks for the three eras I have planned, so these will probably be revised sooner or later.

1 Like