Race and Vehicle Simulations

After a quick debriefing about his visit to Kunos Simulazioni and hands-on experiences with their Assetto Corsa title, there was one detail that Jon revealed that struck me more than any other. It wasn’t the effective confirmation of a delayed release, nor details about how the team are toying with the introduction of “live track” functionality. It had nothing to do with graphics, sound or UI. It wasn’t even anything about how the title drove, though my appetite is very much whetted. No, what jumped out at me and immediately caused me to send an email demanding confirmation was the inclusion of BMW’s incredible E30 M3 Evo DTM car. The E30 M3 is a car I love, a car I adore, a car that ignites a flutter of passion from merely reading those three numbers punctuated by two letters. I knew of the road car’s inclusion and was excited by that, but a competition version was what I really craved. A childhood spent watching E30s battling away in rallying, rallycross and touring car competitions on the TV, and in the flesh (my father rallied in an ex-Vatanen, ex-Prodrive works M3), has ingrained BMW’s original M3 deeply in my emotional subconscious.

Assetto Corsa (AC) is a title I am very much looking forward to, and have been for a while. Prior to my knowledge of the DTM E30’s inclusion, I saw a solid line up of machinery and accompanying playgrounds to enjoy them in; the road going E30 firmly on top of my “to try” list. But confirmation of the presence of the competition car just a few short days ago has transformed my view and approach to the title. What it has done is stoked the embers, and raging fires of “want” are winning in the battle against my best efforts to temper expectations.

It stands to reason that a car which holds such emotional appeal would be a warmly welcomed inclusion. Many of AC’s so-far announced cars hold more appeal for me than many of the vehicles that frequent other titles (Pontiac Solstice, I am looking you!), but in the E30 racing machine I see a car that there is every chance I will never want to jump out of. I can very realistically see myself driving nothing else in AC not just for hours or days, but weeks, and months. So, it’s all good, right? Roll on AC and the good times? In many ways, yes. But I foresee that as the weeks and months pass, that a problem will emerge. The problem? I’ll want more.

More? Surely there’s an obvious remedy for this problem: drive the other cars. This has for a long time been the solution to sim racing longevity. There was a time when content needn’t be anywhere near as plentiful as is now deemed necessary. Grand Prix Legends (GPL) gave a smattering of content by today’s standards, yet few can doubt its staying power. Even as included content levels increase with each generation of titles, the baton is soon passed on from developers to modders to appease the community’s desire for more and extend a title’s lifespan. Whilst I don’t imagine many would complain about increasing choice, it’s not just the quantity of content that is increasing with passing titles. Whilst I’d love to say it’s the quality too, I feel this is perhaps not always the case (in relative terms, at least). What has undoubtedly grown is the complexity and detail of the content, and the size of the task involved in getting a vehicle properly implemented within a title.

This increase in level of detail can easily be seen by casting an eye over the screenshots of upcoming titles like AC and PCARS. In terms of visuals, we are seeing orders of magnitude in increased model complexity. What once would have been ignored, and a few years later maybe drawn on in a texture, is now created in full polygonal glory. Nuts and bolts and rivets brought to life by the artists’ loving touch and skilled hands. Ten years ago, West Racing released renders of their in-game models for the never-to-be-released (as yet, at least) Racing Legends. They were greeted with a mixture of sore jaws from impacts with the floor, and laughter and derision for thinking such models could run in real-time. Today we have models running in-game, on our humble home PCs, that show levels of detail that eclipse those on show in those renders (well, sort of).

The noises coming from the mod teams working on content for rFactor2 reflect that it is a big jump from what was involved with content creation for its predecessor. It’s not just the graphics side of things that have moved on; ever more complex tyre and aerodynamic models have had a significant impact on the difficulty, knowledge and understanding encountered and required to get a car working properly. Whilst this undoubtedly has something to do with a new title and a new learning curve (modders have had a long time to get used to the original rF infrastructure), you only have to look at the ongoing troubles experienced by the iRacing boffins in getting the Lotus 49 to work properly to see that there is more going on than simply learning new software.

What would allay my fears for my long-term relationship with the E30 is not more content; I don’t need or even really want more content. Whilst I’m not sure I would go quite as far as to say current titles ship with too many cars, I do think a fundamental shift in the approach taken by racing sims could see (and indeed would likely necessitate) a reduction in quantity, whilst at the same time leaving the player with no less to see and do. What I want is an increase in quality. A big increase in quality.

I don’t look at the beautiful models being produced by any of the major sim teams out there and wish they had higher polygon counts, were better textured, or just somehow better captured the likeness of their real-world counterparts. Kunos Simulazioni, SIMBIN, Reiza, iRacing, SMS, ISI… some may do this or that better than the others, but not a single developer I can think of struggles for talented and competent artists doing a superb job of serving up gorgeous slices of virtual car cake. Whilst on the physics side variability might be a bit more critical to the overall experience, I don’t doubt that each team has their knowledgeable, passionate individuals or teams trying their best to shoehorn the experience of handling a car into numerous lines of code. Whilst we can always wish for better graphics, sounds and physics engines, or more accurate tracks, the fact is the teams are doing their best. Some do better than others, but wishing for more isn’t likely to achieve much. These are things that have continued to evolve and develop over the history of sim racing, and will continue to do so for the foreseeable future. But there are areas where dev teams could improve. There are areas where I think, without fail, every single race sim developer could do more, and do it better. Because, frankly, they aren’t really doing it at all.

My fears for my adventures with AC’s E30 depend not on running out of tracks to drive around. It has nothing to with wanting more power, or a new challenge, or anything that another car could offer. My fear is that as my connection to the car, and my understanding of how it behaves and how it responds, increases, so too will my desire to jump under the bonnet and explore and tinker. I don’t mean adjusting camber levels or tyre pressures. I mean I’ll literally want to explore the car. I’ll want to look after it, to maintain it. I want to care for the car and feel like that care is rewarded with its smooth running. I want to check oil levels. I want to warm her up before thrashing her on track. I want to know her history, what she’s been through, where she’s been. I want her scars and imperfections not to be wiped as I get out of the seat, but to tell the story of our time together, and instead be signs or pride, or hard-fought battles. I don’t just want a simulation of driving a car on track; I want a simulation of a car.

My first car was a Peugeot 205 Rallye; K220 HWX. It was an absolute shitter. It cost far more to run it for the (all too brief) time it remained on the road under my ownership than it did to buy. It leaked, it squeaked, it wouldn’t start. But I loved it. The short ratio gearbox was sweet-as-a-nut, the 1.4l engine might only have packed 75 horses under the bonnet, but it was relatively torquey and the car weighed about as much as a crisp packet. It had shit, cheap tyres on it that offered limited grip, and anyone who has driven a 205 in anger will know why their lift-off oversteer is revered by car journalists the world over.

Getting her going became something of an art form. I eventually learned that the DIY choke pull it came with wasn’t quite set up right. You needed to pull it all the way out, then back in a few millimetres until it gave a slight click. Then with a turn of the key, and an ever so slight squeeze of the throttle, eventually you would be in business. When the old girl would eventually start on a cold morning, the next ten minutes would be spent waiting for the heater to slowly dissipate the condensation from the inside of the windscreen, and often a small pool of water in the passenger foot-well  Once visibility was sorted (often with the assistance of a jumper pulled over the back of the hand), off you’d go. You needed to be careful though; those first five minutes of use were crucial. Watching the engine temperature needle slowly but surely rise, and hearing and feeling the engine’s characteristics slowly evolve, you’d need to carefully feed the choke in. Leave it out and she’d start to run rough and have no go. Push it in too quick and you’d be stalled at the first stop. Oh, and you’d better hope that wasn’t on a hill, as the handbrake didn’t have enough purchase to stop her rolling backwards. Walter Röhrl might be the king of heel and toe, but the two-footed-three-pedalled dancing I used to have to do to keep the old girl running would have made for YouTube gold. Once she was warmed up and the choke fully pushed back into the dash (or at least as far as the ill-fitting, taped-on handle would go), it was shits and giggles aplenty. Except when you touched the brakes in the rain, as water would pour in through the roof lining and down the inside of the windscreen.

But had one day I woke to find a sparkly, as-new 205 waiting outside the house for me to start at the first time of asking, whilst on the one hand I would have been rather pleased (after, you know, I’d gotten over the improbable turn of events), I would also miss and bemoan the passing of my old partner in crime (literal, and metaphorical). Yet this is what happens every time a simulation is started. Your brand new, pristine, faultless car is there waiting for you to bin at the first turn, press Shift+R, and repeat as many times as your desire, or inability to negotiate a turn, demands.

My 205 was a shitter of massive proportions in so many ways, and fantastic in so many more, but she was my shitter. What I had with my little yellow 205 was an emotional connection. It wasn’t just because it was my first car either. Every car I have owned has given me that feeling to some degree or another, even the dullest: a plain old 5 door 1.6 Focus whose lack of overtaking ability was surpassed only by its impressive inability to stop.

What was the last sim title to give me anything even approaching this level of connection? Has there ever even been one? Two examples spring to mind. Well, sort of. The first was the Ferrari in GPL. More than anything, I just loved the sound. I can admit that in all the time I drove GPL (which is nothing compared to many of the old hands in the community, but a reasonable chunk of ‘should-have-been-homework-time’ nonetheless), with the exception of one run out in the Lotus to see how it compared, every single spin, crash, lap, spin, crash, race, spin and crash I ever did was in the Ferrari. Here was a sim with just eight cars, and yet I only ever felt the need or desire to drive one of them (we don’t talk about my time with the Lotus; it was a mistake. A moment of madness that the Ferrari 312 and I put down to stress, put behind us and moved on).

The other title to offer me a glimmer of car love is slightly less highly esteemed in sim circles, Need For Speed: Shift. Shift was a title I reviewed upon release for AutoSimSport, and as a result it was a title I needed to spend some time with and persevere. There were things I didn’t like. There were things I hated. But it also wasn’t, I don’t think, half as bad as some people made out. Indeed, I found things to like. Take the right car and try to remove as much of the “game” as possible, and it offered a fairly fun experience to my mind. But how good or bad Shift was is neither here nor there. The reason Shift provided an increased sense of connection to vehicles could only be attributed to a few simple factors: you chose and bought a car; you could upgrade it; it had a working odometer. I can safely say it must be down to those small, seemingly insignificant things as they are only elements that differentiate the cars in Shift from other PC titles past or present. Of those three elements, you can scrap upgrades from the list too, since I didn’t upgrade my favourite car at all. I chose the colour of Walter (white with black decals), my Porsche 911 GT3 RS, handed over some imaginary cash, and it kept a record of my mileage. Given in-game money was easier to come by than rain in England, I can only conclude that the tiny, simple inclusion of a working odometer was all it took to transform this sim car (in far from the world’s greatest sim), from simply another car in another title, into something personal to me.

In a previous article, I argued that there is a lot of room still left to be explored when simulating the “racing” part of racing sims. I think that there also remains a huge amount to simulate in the “vehicle” part of vehicle sims. But here’s the rub: vehicle sims of the non-racing kind are doing these things already.

OMSI Bus Simulator is a title I expect few of you are likely to be familiar with. It is a title I am only aware of due to Jon picking it up and sending many, many emails about it. In brief, as the title might suggest, it is a bus simulator. More specifically, OMSI sees the player transported back to 1980s West-Berlin, and running the Omnibus line 92 through the Spandau district. Jon bought this during a period of sim experimentation which also saw him dabbling with railway and underground simulators, and when he told me of his latest acquisition, a bus simulator, I was slightly puzzled.

On the surface, OMSI holds no real interest for me. I’m not a particular fan of buses, I have never felt much of an inclination to be a bus driver, and generally don’t particularly enjoy driving in city traffic. I expect, were it well done, that it might hold a level of novelty value for me, but there was very little in the way of incentive to go out and buy it. But as conversations between Jon and I continued over the ensuing months, OMSI kept cropping up over and over. Sometimes it was a simple “Just taken my bus out on the night route; traffic was nice and quiet”, and not a lot more to it. But often, OMSI came into the conversation when talking about racing sims, and specifically where we felt they were lacking.

Bus simulations may hold little obvious interest for me, but what I have learnt from Jon’s experiences with OMSI are that simulators in other areas are taking certain things a lot more seriously than racing sims. OMSI might not have to tackle the same on-the-edge tyre behaviour that every racing sim practically lives (and, sometimes, dies) by, but it is doing things that go way beyond what racing sims do in other areas. Many of these things, if they were announced in a racing sim, would be dismissed by a significant number as gimmicks, not important, inconsequential. But individual fleet numbers on the different buses; fully persistent features like accruing dirt and grime; fully modelled systems like the gearbox, heating system, pneumatics and IBIS (Integrated Board Information System- the thing that tells passenger where you’re going) all contribute massively to the experience of hustling your double-decker around historic Deutschland.

If, after stepping out of a bus in OMSI, you turned your back and had some fictional stationhand park your bus amongst a dozen “identical” vehicles (to clarify, this doesn’t happen; you must park your own bus), you’d be able to identify “your” bus. That shiny E30 in AC? Well, it’s just another shiny E30. You’d have no reason to care if you picked the wrong one, because it would be exactly the same as any other.

But it’s not just vehicle identification where these features matter. The IBIS is obviously pretty genre dependent, but whilst the fully modelled heating system might take on bus-simulation-specific importance in some respects (I gather the passengers can get rather cranky if the bus is too hot or too cold; no, I am not joking), it is also essential in that it guarantees your windows are clear to see through as you negotiate the city traffic. Anyone who has watched a reasonable amount of motorsport in their time (or anyone who drove my old 205) will know just how debilitating a critical lack of effective condensation dissipation can be to the pace of a driver. Yet I don’t know of a single racing sim that has modelled misting windows and functioning heaters. Many titles don’t even model a windscreen. Hell, when iRacing finally introduced windscreen dirt build-up (something which was in NR2003) after a number of years it divided the community.

When looking into other simulation genres, flight sims seem to be an area that take a drastically different approach to racing sims. The level of detail and modelling that go into a top quality sim-plane simply dwarfs that of the comparative car model. The fully interactive cockpits, the full system modelling… There is very good reason that these individual planes have been sold, in some cases, at a price far higher than any racing sim. There’s also a reason that these individual planes come with huge manuals and guides: planes and cars are very different things, and so are the arts of pilot and driver. Is there any point drawing parallel between the two? I certainly think there is, and the differences are as important as the similarities.

First of all, the vehicles themselves. Planes, as vehicles, are on a much grander scale than cars, both in a literal physical sense, but also in terms of complexity. Even the most complicated car dash effectively has a relatively meagre smattering of buttons and switches, that typically do relatively easily understood functions. Of course, there are exceptions; I doubt many can look a modern F1 wheel square on without feeling a slight sense of intimidation at the prospect of mastering its numerous functions in the heat of a race. But that is perhaps the extreme, and compared to the myriad of dials, screens, buttons and switches adorning the cockpit of a modern plane, there is no real comparison. There is quite simply a lot more to flying a plane than driving a car, and this brings us onto what you do with these machines.

In a car out on track, the experience is primarily led by the directional inputs of the driver; the throttle, brake and steering. Gears and clutch are just ways of managing acceleration, braking and changes of direction. In a plane, it’s quite the opposite. On a flight of any real distance, you aren’t there making drastic changes of direction. You’re not making split second decisions about throttle and steering adjustments. As much as directing it, you are managing the plane. It makes perfect sense then that racing sims should be so focussed on the immediate, and similarly that flight sims see such a focus on complex underlying details of the aircraft. But to me that doesn’t form an argument for why racing sims can’t and shouldn’t embrace more of this side of things.

The pilot in a Boeing 747 might have a lot more time on their hands under normal conditions to fiddle around flicking switches on his or her dashboard than the driver grappling with their E30 does, but similarly the 747 pilot has a lot more switches to flick, and a lot more need to flick them. Why can’t a simulated E30 have a working everything that a real E30 has? If flight sims can tackle the complexity, scope and scale of a Concorde or the like, why can’t race sims tackle the simpler (I’m not saying simple!) task of realising the full complexity of a given car? If the flight sim modders and community can achieve such things, offer them at the respective price point, and find a market for them, why can’t cars be recreated in similar levels of detail and sold for prices reflecting the level of work? Perhaps more to the point, why aren’t the community willing to support such an approach?

Another “sim” genre seems to also offer an interesting point of comparison. Space sims (yes, something of an oxymoron, so let’s just consider them simulations of what life in space might be like) look set for an interesting future. Not long ago, Star Citizen (from the creator of the Wing Commander franchise) smashed through its Kickstarter campaign target, and Elite: Dangerous, the next instalment of the highly respected Elite series, has at the time of writing secured half of its £1.25 million target with some time yet to go. Further to these, the Minecraft team have 0x10c in the pipeline. Whatever becomes of these titles, the hype and activity around them right now, and the backing they have received, at least demonstrates one thing: there is demand.

Many have for years bemoaned the lack of a modern version of Elite. It isn’t just about having the Elite moniker either, but the type of gameplay and approach. Early feedback and comments I have seen accumulating around each of these titles seems to suggest quite strongly that people want a deep, detailed, full experience. There are requests for full ship modelling; not just graphical, but systems and technical aspects. Here you have a genre “simulating” something that doesn’t exist; there is effectively a blank canvas that, provided they obey the known laws of physics (and of course, who knows what our understanding and technology will enable in thousands of years from now), no one could call unrealistic. And yet people are demanding this level of detail. They positively crave it.

Of course, the world of racing sims has also had its own foray into the world of crowd-sourcing with SMS’s PCARS title. A big difference between the approach taken there and that seen with the aforementioned Kickstarter initiatives is one of the crowd’s involvement beyond putting up the cash. This could have gone either way, and we won’t know until the final product is delivered just how it has turned out. What the PCARS project has successfully shown is that people are willing to put money up front for a non-existent racing sim, in the hope that in return they get the product they want. The “inclusivity” offered to investors, and the way the project seems to be heading, does not, however, provide an answer as to whether or not people are willing to pay for a full depth, no compromise racing simulation (no, this isn’t a slight against PCARS). In the Kickstarter video for Star Citizen, its creator Chris Roberts makes it quite clear the title is aiming itself as a title that simply could not be done on console; a celebration of what the PC is all about as a gaming platform. But conversely, in some ways PC racing sims seem to be moving away from this and towards the console titles.

Today’s simulators seem to be taking increasingly large steps to broaden their appeal. GPL was truly a formidable title. Despite the presence of lower powered cars and variable difficulty levels, GPL didn’t really have a learning curve as such; it had a learning line. That was vertical. Although quite primitive in many respects when compared to the best on offer today, it was an uncompromising title. It was difficult, it was challenging, and it felt pure. Its physics simulations might not have been wholly accurate by today’s standards, but you got the feeling it was trying its best. GPL was not just a benchmark and genre defining simulation, it was also a fantastic game (very much in keeping with the Elite series, which had a very steep learning curve and immersive experience). The (now sadly quite rare) full line-up of cars, tracks and drivers that comprise a full season provided the game with soul and with a narrative. It didn’t need CGI cut-scenes or RPG elements, it just put you in race after race with the same familiar names: your team-mates, your respected peers, and your fiercest adversaries (often not mutually exclusive).

GPL’s AI and league structure (i.e. a proper championship roster of drivers and tracks) was what provided the “game” element. However it might compare to today’s offerings, it was a sim through and through in terms of the driving experience. Flawed and inaccurate? Probably. Tightly focussed on trying to offer the experience of competing in the GP world of yesteryear with little concession to accessibility or crash-bang-wallop fun? Very much so.

AC looks set to be a truly great title, and one I am very much looking forward to. I don’t want there to be any doubt about that. But I look back at some of netKar Pro’s now gone and forgotten features, such as the interactive cockpits and fully modelled start-up procedures, and can’t help but feel sad that there is effectively a backward step going on. AC will no doubt be a big march in the right direction in many areas, and I’m not for a second suggesting it will be an inferior title to its predecessor. But there are features that the devs are not only capable of implementing, they have already implemented them in the past, that may not be making it into an all new sim. This, to me, is a crying shame.

I can sit here and wish the devs would include this or include that, but these things are only going to make it into a title if one of two things happens: you get some devs so focussed and passionate about the subject matter at hand that they won’t rest until they have modelled every little detail they can manage; or the audience makes it clear that this is what they want. The former is something of a difficult stance to adopt. One title springs to mind when I think of such an approach. As I have mentioned many times before, whilst Racing Legends might stand head and shoulders above all others in my mind as the sim of my dreams, the fact it is the sim of my dreams and not a sim on my hard-drive tells its own story. Developers don’t sit in the privileged position of having all of the time and resources their every whim may desire; they are individuals, groups and companies, with mortgages, children and bills to pay like everyone else.

So what do simmers want? I must confess I get confused by the racing sim community. On the one hand, I see no other community so quick and willing to criticise a title for its lack of realism, or for deviating from the well-worn sim path. And yet, with that, I see a community which, on the whole, does not seem to really crave the realism they claim to demand. I don’t really know why this is, but I think people need to consider exactly what it is they do want before being so vocal and vociferous about what it is they do not.

I get the impression that the simulations many in the community wish they could have are those used by the F1 teams. If teams spending hundreds of millions of pounds a year going racing invest time and resources into these simulations, they obviously judge them to be of worth. As such, what better verification of quality and accuracy can you get than that which comes from the people at the cutting edge of automotive design and implementation, running the most tightly monitored and scrutinised cars in history? It would be easy to make an argument that such set-ups are the ultimate vehicle simulators. But I am not convinced they are; the role of the F1 team simulators are not to simulate the experience of driving a car per se, but to provide a tool to help with car and driver development.

F1 drivers in a simulator will not be interested in the simulated form of many of the real aspects of driving. The aim is to work on learning the track, general set up work; they won’t be testing with dirty visors, mechanical frailties and failures, other cars etc. Despite the name, an F1 simulator is, in many ways, not really concerned with simulating. You would expect the physics engine and modelling to be top drawer, but in terms of simulating reality, they only really have to concern themselves with a small subset of the real world’s complexities. In this respect, iRacing’s strive to market itself as a driver training tool does not necessarily mean the product is automatically moving in the right direction in terms of absolute simulation. I agree with the general sentiment within the community that physics are more important that graphics, but I think it’s worth remembering that physics aren’t everything.

Curiously, I seem to detect a sentiment that this stripped back approach is what a racing simulation should be. That anything else is gimmicky; is somehow making the experience a “game”. But I don’t think simulations and games need be mutually exclusive, and indeed think that at times they necessarily overlap. The sims we play are video games, fundamentally.

Whatever a simulation is simulating, the end point is that is needs to be fun for the people “playing” it. What is or isn’t fun is obviously going to vary immensely from person to person, but there seems to be a sometimes strange approach to fun in racing sims. It is quite obvious that for some people, winning is all that matters. iRacing, the title with undoubtedly the best organised, structured approach to assessing performance via numerous metrics, positively thrives on the element of competition. But I’ve heard numerous stories that really make me wonder why people go to the effort. The use of loopholes and exploits provide a case in point, but stories of people deliberately using inferior hardware as a means to obtain a competitive edge seem a particularly striking one. Here people are diminishing the experience, reducing the accuracy of the simulation, for the sake of being “better” at it. It’s not hard to see when simulating a sport that competition comes into it, but I can’t help but ask the question: why bother?

The fun for me comes from the challenge, from the experience. I would rather be slow and off the pace but know I am experiencing something as close to reality as possible, than be winning in the knowledge that I am deliberately diluting the simulation. On a broader scale, I think this behaviour manifests itself in subtler ways, perhaps subconsciously. I think it’s fair to say that some people would rather think they are good at something than face up to the fact that they are not. Many of the people who enjoy bashing around in Gran Turismo, invincible to damage, generally doing a terrible job of race driving would likely not have much fun in the world of netKar Pro. Whilst toning down netKar Pro’s more extreme elements of simulation might make the title have a broader appeal, there are also a number of people for whom such feature reduction diminishes the appeal.

I can very much sympathise with the developer’s position here, and completely understand the AC team’s comments about how “everyone wants to race a 24 hour race, driving at night, under heavy rain, with broken suspension and a puncture. But just one time for a try.“ I get the sentiment, but it seems to me to miss the mark a bit.

I might be in a minority of the whole gaming scale in wishing for more and more intricacy in my racing sims, but I’m also a little disheartened by being told what people do or don’t want; they don’t seem to be speaking for me. Many in the community raised an eyebrow when Codemasters announced they would be dropping cockpit views from some titles as “no one uses them”. But similarly, I raise an eyebrow when I’m told proper that start up procedures, interactive cockpits, deeper vehicle systems simulation, component modelling and management are not being included because it’s not what people want. Of course it won’t appeal to everyone, but that doesn’t mean that it doesn’t appeal to others.

To return to the topic of space simulations, an interesting video recently came to light featuring an interview with David Braben and Chris Roberts about their respective titles in the works. Some of the things they discuss are the issues of trends in genres, and the effect of publishers. Both of their titles are being made publisher free, and in doing so removing the restrictions and demands that accompany going down the traditional development path. Fundamentally, they are taking out the money men who reduce games to the lowest common denominator, and giving themselves the chance to make the games they want to make. In a sense, this is the artist taking back their art. None of the big players in the racing sim market run through the conventional publisher model, but there still seems to be lot of (self-imposed) restrictions on development. Some of this is necessary; it’s called discipline, and it is essential if a title is going be released in a reasonable time frame without the developer going bankrupt in the process. But some of it seems, for want of a better word, a little short-sighted.

I wouldn’t have thought there would be much of a market for a hardcore bus driving simulation. Whilst I don’t know how OMSI has fared in terms of sale figures, the forums are very busy and it certainly seems to be doing ok for itself and gaining plenty of recognition and positive response. How many people want a fully featured racing sim, with as many details modelled and included as can possibly be incorporated? In short, I don’t know. I don’t have an answer to that. But I’m not convinced the devs do either. As people look for what separates one title from another, such a change in approach could at least make a title stand out.

It’s taken decades for Roberts and Braben to put their latest titles forward as propositions for the gaming public, but the millions of pounds being put down as backing show that there is plenty of want and demand. If we can’t have a bit of “Build it and they will come” in the sim racing community, how about a bit of “Show them the architect’s drawings and see if they would like to”?