Is the isiMotor really what makes it's children titles not popular?

After a recent comment on the Steam forums saying how all ISI powered games are a flop, I decided to think about it and see if that's actually the case. Here's where I got:

(the vast majority of my reply to such topic was pasted - some parts were omitted for the sake of keeping the discussion here in RD clean).

(all of what was written bellow is what I perceive to be how the described events turned out. I can be wrong in one or multiple subjects, so don't be afraid of doing research to correct me on things)

(I ask that, if you find who made the original comment -which I censored-, do not bring this person up here. The point of this thread is to attack the argument, not the person)

----------------------------------------------------


ISI-powered games (which includes rF2) aren't a flop because of the engine itself, but because of developer/publisher bad decision making.

There's several reasons a lot of sims use the engine from rF1 as a base, the most common one being a great physics starter, but there's also the fact that it's a known piece of software that was used for several years in many great titles, it's historic relevance, and so on.

ISI did great titles in the past, like F1 Challenge '99-'0 (and some older F1 titles) where you got consistency in what cars and tracks to expect.
But then, for some reason, they decided not to do the same with rFactor 1.
Even some of these old F1 games were moddable, and for whatever reason ISI decided to leave most of the content creation for rFactor 1 in the hands of the community. This led to a big inconsistency in mod quality, of 1000 mods you could count on very few of these to be any good.

After rFactor 1, the glorious GTR2 came about. It was revolutionary in quite a few features, and content. It had all cars and tracks for the FIA GT series that spanned 2 full seasons, full 24h day/night transitions, rain, some basic real road technology, a driving school, animated pit marshals, and a bunch of features and consistency/focus not yet seen in the sim world for at least 12 years. It used the same isiMotor as a base and it was a success, still regarded today as one of the best simulators ever made (made by what is now SMS, the makers of Project Cars 1&2 ------ SimBin, the developers of Race07, Race On, and then later RaceRoom, had almost nothing to do with the development of GTR2).

Then a year later, SimBin used the same base engine for Race07, and while it still had great physics, graphically the engine started to show it's age, specially on the same year Crysis was released.
Still, the game is regarded as a great product.

Now in 2008, iRacing was released. No more bad cars, no more bad tracks. Not only it was consistent in content, graphically it was already surpassing anything isiMotor related that was released to that day.
I think many people migrated to iRacing because the overall experience was better not only graphically but in terms of cars and tracks as well, not to mention in multiplayer.

Then in 2011 Kunos started development of Assetto Corsa with an in-house engine. Physics and feature wise the title was a downgrade from their previous, netKar Pro, but graphically AC was step in the right direction. Still, the title lacked many core features even several years after it's release.

And yet again, people were migrating more to promises and shiny graphics rather than staying at stable core and good physics present in the rF1 engine.
It's understandable, though. If companies want to sell a bit more titles, they have to attract the eye of the audience as well.

At this point it is said that ISI had already started development on rFactor 2. In fact, there are records of development starting in 2009.

In 2012 the open beta of rFactor 2 was released. It was a huge step forwards in physics, being at least a decade ahead of any competition - sadly, ISI focused so much effort into 'just physics' that the title was broken in a lot of ways. The pricing model was horrible, it lacked documentation for modders which were now looking for an easier platform to mod since modding for rFactor 2 was too complicated because of the massive advance in physics (and so parameters as well).
That mentality of "let the community produce content" did work in 2005 because there was not many sim platforms to jump around, but it certainly didn't work for rFactor 2, specially after the release of iRacing with many official cars and tracks and later with the release of Assetto Corsa which also revolutionized sim racing in many, many ways.

In 2013 RaceRoom was released. Being regarded as merely a 'hotlap simulator' it lacked content, features, even multiplayer.
And even now, after 5 years of development, RaceRoom (the company, not to be confused with Sector 3 or even the game's name) somehow still has a poor pricing model, stupid online DRM, and not many features for the game. Users can't even change the tire pressure, the setup part of the simulation is still very basic, more basic than GTR2 in 2006.

Then Assetto Corsa was released. Shinny graphics, good documentation for modders and easy to mod (way simpler physics than rF2 or even rF1 from 2005 according to some known personas here in RD), a large community started growing around it. But to me, if it wasn't for the good price and moddability AC would end up close to what R3E is now, perhaps even worse considering how long it took for AI to come and be perfected, for instance (and that is, with a higher budget than what I think Sector3 has had).

A year later, SMS, which had developed GTR2 and was picked by EA to develop a few Need for Speed titles, developed Project Cars. Shinny graphics (probaby the best graphics of it's time), lots of content. The hype was good; the title, however, not so much. Shady physics, bug infested, players were not happy about it.

Still a year later, Automobilista came to be. Being regarded as one of the best racing titles out there, it lost a good deal of potential because Reiza didn't have money to license known tracks/cars, and the fact that they decided to use an ever graphically-aging engine didn't help the title. Still, it's regarded as one of the best when it comes to physics, force feedback, car/track polish, and sounds.
Reiza also did a good job with the isiMotor2 engine.

Again a year later, SMS released Project Cars 2. While improving a lot upon the first title, it's still quite buggy, physics change with every update and can be wonky at times, and their sound engineer should be fired.
Still, player count continues to increase. It has a name for itself, multiplayer is well populated, graphically it is one of the most advanced titles, and it's loads of fun - specially when you go up on it's Competitive License.
It took SMS many years to make such progress with their own engine. They're very capable (as proven by GTR2), but to get to the level of consistency of the isiMotor2 was not an easy task.

2018 comes by, Kunos decided to use the Unreal Engine 4 for their next title, a smart move I might add. A very good engine, it has all the features a game developer might want.
Most of the engine and it's features are already developed.

In conclusion, I don't think it's the isiMotor that makes titles flop, I think the developers play the biggest role if this happens (same for the success of their titles). SMS developed what is considered one the best sim racing titles, using the isiMotor - and the same company developed what is considered one of the biggest disappointments of the genre with the release of Project Cars, a sim that uses it's own engine. Funny how when they stayed with a good engine base they did good progress, but when decided to create their own engine the game was a flop in reception - it had pretty graphics, sure, but the physics side of things can generate heated debates to this day.

Same goes for Kunos. Stefano and his team did a good job at developing a graphics engine that, while looking good, can only be lit by one source of light, so you have no night racing. It also lacks rain, among many other features.

The isiMotor was a powerful tool for those who have chosen to use it correctly, and it was also the doom of titles that used it poorly. The main problem for the isiMotor2, nowadays, is graphics, it still uses a DX9 engine. (somehow Sector3 did a great job on it graphics wise, it looks good and even more natural-looking than AC at times, and the physics are also good).
Physics wise the isiMotor2 is aging as well, but that is to be expected from a 13 years old engine.

And for rFactor 2, if ISI had focused a bit more on content, engine optimization, and graphics, rF2 could be the best sim ever made. I just hope S397 makes the right calls now and continue optimizing the engine while bringing content. The most important part, which is the physics, we already have and to me there's nothing that comes even close to it.

So what's your take on this?
 
Last edited:
Sorry but people didnt make mods for rf2 cause its physics was to advanced. Rf2 was a turd for years, that's why. No modder would want to spend 100s of hours for to work to look like crap in game.
 
A lot of this reads like pretty typical rf2 defense statement. The game is not a problem, it is the people who "like shiny graphics instead of physics". Or: The poor quality of tools and documentation is not a problem, it is the lazy people who want everything on silver platter. The lack of documentation is not a problem, it is the physics which is too advanced for the lazy simple minded modders. In other words: the game is too good for people to use it. The fault must always be somewhere else except in the game itself. Empty online is not a problem: join a league!

One thing you forgot to mention at all is that due to its age how hard the isimotor is to develop. There is a reason why it has taken years for s397 just to rewrite the ui code. Lots of code written decades ago for totally different hardware using nowadays outdated development practices (that might have been the way to go back in the day).

And like magzire said rf2 was not exactly the best sim ever when it launched. And many dubious business decisions later it is no wonder Isi was happier to kill the game instead of keep developing it. S397 jumped in to save the game but I'm still not convinced the game can be saved. S397 has been done great things but also repeated many of isi's mistakes. Documentation for modding is still dirt poor (good luck building a car or track using mostly rf1 docs and waybackmachine). Killing gjed overnight with no warning at all is also very bad and strange thing to do. Now all official files are also decrypted meaning modders have no access to any new information. Devs are not sharing and when new features are added modders don't even know they exist.

If rf2 can survive I consider it a small miracle considering where it came from and what it is. Is it fault of the engine? If you put that question slightly differently and ask "does the gmotor engine severely limit what a developer can do with it" then the answer is yes.

People did not choose ac over rf2 because of shiny graphics. People chose ac over rf2 because ac was a better product. It was a better modding platform, it had trillion times better mod support, it had interesting content, it was better to use and unlike rf2 it had a future. People like to pretend when comparing ac and rf2 that only ac has limitations. Every game has limitations. Ac has lots of physics limitations, no day-night cycle, no rain, no multiclass. But rf2 has lots of limitations as well. Fiddly, outdated, hard to work with, poor ui, hazy future, poor online numbers, expensive dlc (whats going on with the formula e dlc, is that a dead end for that?), poor fps, very limited modding capability (just look at what content manager has done for ac), super bad tools and documentation.. Maybe it is not actually a choice between shiny graphics and best-in-galaxy physics? Maybe it is a choice where many things are considered and when you put them all together just having little better physics doesn't cut it for those people who don't see the tradeoffs worthwhile.

And I'm sure developers know this as well. Is the rf2 engine a good buy for a developer? I doubt it. It might be one of the very few options though which could guarantee a future for itself. What else is there? Ac engine, rf1 engine and... that's it? Is it better to pick one of those or pick nothing at all? Looking back how well rf engined games have sold recently it doesn't look like the best thing ever.
 
Sorry but people didnt make mods for rf2 cause its physics was to advanced. Rf2 was a turd for years, that's why. No modder would want to spend 100s of hours for to work to look like crap in game.
That too I can agree.

A lot of this reads like pretty typical rf2 defense statement. The game is not a problem, it is the people who "like shiny graphics instead of physics". Or: The poor quality of tools and documentation is not a problem, it is the lazy people who want everything on silver platter. The lack of documentation is not a problem, it is the physics which is too advanced for the lazy simple minded modders. In other words: the game is too good for people to use it. The fault must always be somewhere else except in the game itself. Empty online is not a problem: join a league!
I agree, but let's make some clear from the start before I make my points: I'm not, in any way shape or form, a defender of rFactor 2. I'm one who made numerous reviews stating how physics alone won't make a sim when this is broken in so many levels :)

rFactor 2 was surrounded by bad decisions from ISI since the beginning, the platform itself was definitely a problem in many ways and that was a major contributor to why people chose not to use it. As I've stated in my reviews, physics won't matter when there's no good documentation for modders, when the sim is poorly optimized, when pricing schemes are shameful, and when the competition has achieved quite a stable platform from the get-go with way better documentation for modders. Many things were decisive for rF2's flop, and in my opinion this was, in many ways, well deserved. The game was not too good for people to use it.

And I'd never think modders were lazy, to be honest. I can imagine going from one platform to another with no good documentation, this can make things quite painful specially when modding for such platform was a way deeper task than to mod for rFactor 1. So when a more welcoming platform comes, it's inevitable that modders will chose that instead.
If I were a modder back in the day I'd probably chose AC as well.

One thing you forgot to mention at all is that due to its age how hard the isimotor is to develop. There is a reason why it has taken years for s397 just to rewrite the ui code. Lots of code written decades ago for totally different hardware using nowadays outdated development practices (that might have been the way to go back in the day).
Thanks.

Yes, the code is ancient, though that shouldn't bee to much of a problem for ISI back in the day if they hadn't focused so much into the tire model. Code is evolution, and it would take too much effort to re-write most of it for modern standards, it doesn't make much sense. Whatever needs updating should be updated, otherwise it should stay the same. For instance, you ca still see elements of the Quake 3 Engine from 1999 in modern Call of Duty games.

And like magzire said rf2 was not exactly the best sim ever when it launched. And many dubious business decisions later it is no wonder Isi was happier to kill the game instead of keep developing it. S397 jumped in to save the game but I'm still not convinced the game can be saved. S397 has been done great things but also repeated many of isi's mistakes. Documentation for modding is still dirt poor (good luck building a car or track using mostly rf1 docs and waybackmachine). Killing gjed overnight with no warning at all is also very bad and strange thing to do. Now all official files are also decrypted meaning modders have no access to any new information. Devs are not sharing and when new features are added modders don't even know they exist.
Agree 100%.

If rf2 can survive I consider it a small miracle considering where it came from and what it is. Is it fault of the engine? If you put that question slightly differently and ask "does the gmotor engine severely limit what a developer can do with it" then the answer is yes.
Again, agree completely.

I don't know if rFactor 2 will survive, but one thing I wish S397 would do is throw the gMotor engine in the trash and use a new set of graphics engine that is well maintained, which is the Unreal Engine 4. Use it for graphics then plug whatever else they want, like their pMotor physics engine, FMOD for sound, and whatnot.

People did not choose ac over rf2 because of shiny graphics
Sure, but that's not what I said :p Though IMO it was one of the reasons.

People chose ac over rf2 because ac was a better product. It was a better modding platform, it had trillion times better mod support, it had interesting content, it was better to use and unlike rf2 it had a future.
Couldn't agree more.

People like to pretend when comparing ac and rf2 that only ac has limitations. Every game has limitations. Ac has lots of physics limitations, no day-night cycle, no rain, no multiclass. But rf2 has lots of limitations as well. Fiddly, outdated, hard to work with, poor ui, hazy future, poor online numbers, expensive dlc (whats going on with the formula e dlc, is that a dead end for that?), poor fps, very limited modding capability (just look at what content manager has done for ac), super bad tools and documentation.. Maybe it is not actually a choice between shiny graphics and best-in-galaxy physics? Maybe it is a choice where many things are considered and when you put them all together just having little better physics doesn't cut it for those people who don't see the tradeoffs worthwhile.
Again, couldn't agree more.

And I'm sure developers know this as well. Is the rf2 engine a good buy for a developer? I doubt it. It might be one of the very few options though which could guarantee a future for itself. What else is there? Ac engine, rf1 engine and... that's it? Is it better to pick one of those or pick nothing at all? Looking back how well rf engined games have sold recently it doesn't look like the best thing ever.
There are quite a few engines out there that could be licensed, like the CryEngine, Unreal Engine, and a few others. Though, I think there are only a couple of developers that could license these because if they're not big they'd have to create physics from scratch and plug them into the engine. Not that many developers have the knowledge to do this, to create a new tire model from scratch. I think not even Kunos, SMS or S397 could do this, because of the time it would take ; they could license these engines because their own tire models are already done, all left is to "plug" them into the engines. Kunos did this and S397 also needs to follow suit, otherwise the future of rFactor 2 is still unknown.
 

What are you racing on?

  • Racing rig

    Votes: 528 35.2%
  • Motion rig

    Votes: 43 2.9%
  • Pull-out-rig

    Votes: 54 3.6%
  • Wheel stand

    Votes: 191 12.7%
  • My desktop

    Votes: 618 41.2%
  • Something else

    Votes: 66 4.4%
Back
Top