GTRevival Is Now Project Motor Racing, Straight4 Secures Publishing Deal With GIANTS Software


GTRevival is no more - the Straight4 Studios title will now officially be called Project Motor Racing. And the studio partners with an exciting new publisher for the title.

The first project of Straight4 Studios has a new name. After being initially announced as GTR Revival, which was later shortened to GTRevival, the title currently in development by many former SimBin team members from the days of GTR and GTR2 now has a new name - it is going to be called Project Motor Racing.

Not only does this likely reflect a change in direction for the game content-wise, it also connects to the Project CARS franchise, which several team members around Studio Head Ian Bell also created. However, this is not the only bit of news that @Michel Wolk and I learned when following an invitation to Silverstone by Straight4.

Project-Motor-Racing-Straight4-Giants-Michel-Porsche-956.jpg

Can you tell that Michel enjoyed our Silverstone trip?

When we arrived at the track, we did not know what to expect. There was a track day for some of the most exclusive and wildest cars on the planet, the "Secret Meet", where even personalities like Adrian Newey or Zak Brown were present. The former even took to the track himself, driving a Ford GT40, an Aston Martin Valkyrie and a Leyton-House CG901, the F1 car he had designed himself for the 1990 season.

In one of the pit garages, there was an old friend from the GTR and Gran Turismo days waiting for us, the Lister Storm. Next to it were banners with the Straight4 Studios logo and that of the new publisher: GIANTS Software. And they really are giants in the simulation genre, just not in sim racing so far.

Project-Motor-Racing-Straight4-Giants-Announcement.jpg

Image: Straight4 Studios / GIANTS Software

GIANTS Software Partners With Straight4​

The Swiss publisher became famous and successful with their Farming Simulator and will now go from a comparatively leisurely pace to top speeds on the virtual racing tracks. We had the chance to chat with GIANTS CEO Christian Ammann about the project, and he is excited about the new adventure.

"With all the capabilities in-house, a successful history of strategic brand alliances, and an infrastructure proven through multiple projects, this partnership of combined strengths marks another milestone by expanding our genre expertise", Ammann says about the new partnership. "We started to self-publish our titles in 2001. That worked really, really well. So we decided to also publish other titles. Of course, we were looking into simulation titles, and sim racing is a very interesting market. It's also games we like personally."

Similarly, Bell is looking forward to realizing the new alliance's potential: "Our partnership with GIANTS is the last piece of the puzzle for the development of Project Motor Racing. It’s fantastic news not only for our studio, but the sim racing genre as a whole. Those who are familiar with GIANTS’ best-selling franchise will recognise why this partnership is going to refresh the sim racing genre in ways that the community is going to love."

Project-Motor-Racing-Straight4-Giants-Ammann-Bell.jpg

GIANTS Software CEO Christian Ammann (left) and Straight4 Studio Head Ian Bell. Image: Straight4 / GIANTS

What To Expect From Project Motor Racing​

Of course, we also wanted to know more about the game's direction. The Lister Storm is a first indication of the content of Project Motor Racing, and while this rare and legendary V12 racing car was scanned live on site and confirmed as the first car in the game, we tried to get a little more out of Ian Bell about the content and features of the new simulation.

"It was GT Revival up until the point where in building the assets, we decided that we were getting a bit bored with only GT. And don't get me wrong, we had about 80-90 GT cars in there. Pretty much every GT car you could ever think of", Bell told us. "We're not listing the content as of yet, but we're way into the hundreds now, in terms of car count, we've just kept going and going. So we kept adding more and more and more, from interesting areas. And alternative series that we find interesting, that aren't called GT. But we will we will announce soon."

Project-Motor-Racing-Straight4-Giants-Lister-Storm-Scan.png

The Lister Storm that was scanned at Silverstone (chassis SA9STRM1B1B053122) is mostly known for its 2003 FIA GT campaign in the hands of Jamie Campbell-Walter and Nathan Kinch, who raced the car in the final four races of the season and took the win in Anderstorp, Sweden.

Bell also confirmed that PMR is indeed going to be a realistic simulator that will focus on both singleplayer and multiplayer. "It’s like picking between your two favorite children. I can't do it because I love a single player for the fact that it doesn't tie you into a system where if you're not social, if you are uncomfortable driving, you can still get on and have great fun in the game. So you need, in my opinion, a great single player career mode, which we're really pushing to hell and back.

"At the same time. We also believe we need an iRacing style standard or better multiplayer mode. So there's a reason why we're not shipping at the end of 2024, like we planned a couple of years ago, we've added so much. To try to do the best in every area is what we're aiming for."

Furthermore, VR is a core element that Straight4 has in mind in development of Project Motor Racing. Bell continues: "We couldn't possibly not have VR. It's crucial for us", the Studio Head said referencing the VR capabilities of the Project CARS titles.

All of this combined sounds rather promising. We cannot share any moving images, screenshots or more information about the technical basis yet, but we assume that this could happen in August, possibly at gamescom.

Stig-approved Handling​

As for Project Motorsports Racing's physics, we cannot say anything yet either, but we did have a pleasant and very interesting chat with Straight4's handling consultant - none other than the former Stig on Top Gear, Ben Collins, who drove the Lister at Silverstone to collect both footage and data.

The cars "look great. They sound great. But then how do they drive? How do they feel? What's the feedback through the steering wheel? All of that stuff we finesse", explained Collins. "And I've got the real world experience to, to bring it in so I can figure out, you know, what it should be handling like. And in the case of [the Lister], it's really quite unique, although it's front engine, rear wheel drive."

Project-Motor-Racing-Straight4-Giants-Ben-Collins.jpg


Its engine may technically be front-mounted, but "a long way back towards the middle of the car where the driver sits. So you get really, you know, really good handling, almost like a mid-engine car. So unless you've driven it, it's quite hard to be really sure. What would it handle like? And you might make something that handles evil because you think it looks badass, but actually it's quite tame. So I'll try and bring as much of that into the game as I can."

Interestingly, Collins - who recently started a sim racing YouTube channel himself - also pointed out a seemingly common problem that sims apparently get wrong frequently. "The biggest problem with sims is that nearly always the cars a too difficult to drive, and that there's a massive drop off in grip, either the front or the rear or both." How this translates to Project Motor Racing will be interesting to see.

What are your thoughts on Project Motor Racing as the new name, the publishing deal with GIANTS Software and the comments about the development of the sim? Let us know on Twitter @OverTake_gg or in the comments below!
About author
Yannik Haustein
Lifelong motorsport enthusiast and sim racing aficionado, walking racing history encyclopedia.

Sim racing editor, streamer and one half of the SimRacing Buddies podcast (warning, German!).

Heel & Toe Gang 4 life :D

Comments

Good news to me. Love Giants Software. Although i doubt they will take part in the development other than the equipment needed for 3D scanning etc. But i do hope they will make it as moddable as Farming Simulator. It will be just EPIC.
 
So, they're partnering with the people that make Farming Simulator? Get ready for a million pointless "features" that take half an hour before you can even start the engine. And if you crash in multiplayer you have to sit there and wait for another player to come pick you up with a crane and then for yet anohter player with a truck with an open bed to take you back to the paddock.
One can only hope. You still have NFS, Forza, Solar Crown or whatever arcade racing game you like. Why can't we have something we like? But don't worry, it won't be that challenging. Even you can play Farming Simulator. Giants Software has made it very...."accessible" for people who can't deal with realism.
 
Uhmm ok… So who is that ‘100m driver’? Or did i miss it and am blind? :O_o:
Not going to disclose the name for a number of reasons, but original post regarding the "100m driver" is here if that's what you were looking for.

We've also had guys think they should be able to get to power 30 meters earlier than in real life; the 100m story is just one of many like it. And once you get to the amateur end of the spectrum it gets a bit crazy (and mind you, the "amateurs" in pro racing are still quite capable). Some good stories there.
 
Working with *actual* amateurs, as in Jim and Bob racing from their garage, isn't generally the best way to get usable feedback either.

They might be onto something sometimes, but it requires a high level of car behavior sensitivity and vocabulary to accurately convey it. That and "bumpy" or "smooth" or "slow" or "fast" means different things to different people in my experience. If you just do what one person suggests, no matter how capable they are, it'll inevitably be wrong for someone else. Can't comment on if this applies to elite level motorsports or not, but just my 0,02.
 
Lets say we have two hypothetical sims to choose from. All we know is the results of two tests.

The first test is running a 10 race series with 20 of the top sim racers in the world, where we know their "rankings" relative to each other via some reliable measure.

Simulation A ended up with the 20 drivers finishing almost exactly as the rankings would predict.

Simulation B ended up with the 20 drivers finishing with almost no correlation to what the rankings would predict.


The second test is running a 10 race spec series with the exact real life drivers of the real life version of that spec series.


Simulation A ended up with the 20 drivers finishing with seemingly no relation to how well they did in real life.

Simulation B ended up with the 20 drivers finishing almost exactly in the same order as they did in real life.

Which simulation would you prefer?

Which simulation is more realistic?

(this thought experiment is not claiming it is possible to make sims with these results. The idea is to poke around at the idea of what "realistic" even means)
 
Last edited:
Lets say we have two hypothetical sims to choose from. All we know is the results of two tests.

The first test is running a 10 race series with 20 of the top sim racers in the world, where we know their "rankings" relative to each other via some reliable measure.

Simulation A ended up with the 20 drivers finishing almost exactly as the rankings would predict.

Simulation B ended up with the 20 drivers finishing with almost no correlation to what the rankings would predict.


The second test is running a 10 race spec series with the exact real life drivers of the real life version of that spec series.


Simulation A ended up with the 20 drivers finishing with seemingly no relation to how well they did in real life.

Simulation B ended up with the 20 drivers finishing almost exactly in the same order as they did in real life.

Which simulation would you prefer?

Which simulation is more realistic?

(this thought experiment is not claiming it is possible to make sims with these results. The idea is to poke around at the idea of what "realistic" even means)
We'll never get a truly realistic sim though, until every last aspect of physics that affects a car or bike can be acurately modelled. How do devs do the grip from tarmac? Is it the same for every track or do they tweak them to what the drivers etc say (track a is slippery, track b is grippy). Then you have tracks that have sections of tarmac that offer different grip, is that simulated? How accurate are the grip settings for tarmac in sim? Then you have the way temperature affects grip, clouds can shade and cool the track. Bottom line is NO current sim "game" is real, they all have areas that are better than other sims and have areas worse than other sims. Then we can get to wind, which I'm 100% certain NO sim accurately models yet and in the aero world that is vital for realism.

TLDR, arguing over which sim is more "realistic" is stupid!!!
 
Last edited:
We'll never get a truly realistic sim though, until every last aspect of physics that affects a car or bike can be acurately modelled. How do devs do the grip from tarmac? Is it the same for every track or do they tweak them to what the drivers etc say (track a is slippery, track b is grippy). Then you have tracks that have sections of tarmac that offer different grip, is that simulated? How accurate are the grip settings for tarmac in sim? Then you have the way temperature affects grip, clouds can shade and cool the track. Bottom line is NO current sim "game" is real, they all have areas that are better than other sims and have areas worse than other sims. Then we can get to wind, which I'm 100% certain NO sim accurately models yet and in the aero world that is vital for realism.

TLDR, arguing over which sim is more "realistic" is stupid!!!
And even if all that would be simulated, there are simply more things that cannot be simulated that make simracing and sitting in a real racecar different. The heat in the cockpit, the feeling of the forces, the danger & responisibility, etc. In that scope having spots on track where I have 5% more grip just like IRL because of the sun or different tarmac is just useless. Especially when tracks and conditions are constantly changing IRL.

I would prefer dev time going into actually making quality video games (yes, our racing simulators are, in fact, video games).
 
And even if all that would be simulated, there are simply more things that cannot be simulated that make simracing and sitting in a real racecar different. The heat in the cockpit, the feeling of the forces, the danger & responisibility, etc. In that scope having spots on track where I have 5% more grip just like IRL because of the sun or different tarmac is just useless. Especially when tracks and conditions are constantly changing IRL.

I would prefer dev time going into actually making quality video games (yes, our racing simulators are, in fact, video games).
Yeah, modeling the things Damage Inc mentions is possible and is done fairly meticulously at the pro level, but those are all fairly small variations relative to the general perception of driving. The +-5% local track grip won't really affect the perception of a driver who wants the car to have +20% to feel correct, it'll still feel off.
 
Yeah, modeling the things Damage Inc mentions is possible and is done fairly meticulously at the pro level, but those are all fairly small variations relative to the general perception of driving. The +-5% local track grip won't really affect the perception of a driver who wants the car to have +20% to feel correct, it'll still feel off.
I'd be very interested to know how tarmac grip levels are calculated! Although we are talking about Sim games that have to run Ai and only on one PC. I'd imagine (prepared to be wrong) that pro sims have multiple PC setups? How is wind calculated in the pro sims, is it dynamic within the software or is there an engineer, physically tweaking the numbers as the driver goes round?

Bottom line is, as we know a sim game is a balancing act with the computational power available to the average end user. I still think it's impossible for modern tech to simulate everything and have a coherant "game" there as well.
 
I don't think computing power is really a bottleneck for the kind of things that people would want/think they want in their consumer sims. It's more about having someone who has the dataset and know-how to translate it correctly. Being able to run your sim at 100kHz does nothing if you know less than what you don't know, you'll just get garbage that runs smoother.
 
I'd be very interested to know how tarmac grip levels are calculated! Although we are talking about Sim games that have to run Ai and only on one PC. I'd imagine (prepared to be wrong) that pro sims have multiple PC setups? How is wind calculated in the pro sims, is it dynamic within the software or is there an engineer, physically tweaking the numbers as the driver goes round?

Bottom line is, as we know a sim game is a balancing act with the computational power available to the average end user. I still think it's impossible for modern tech to simulate everything and have a coherant "game" there as well.
This is pretty true, especially if only running a single car:
I don't think computing power is really a bottleneck for the kind of things that people would want/think they want in their consumer sims. It's more about having someone who has the dataset and know-how to translate it correctly. Being able to run your sim at 100kHz does nothing if you know less than what you don't know, you'll just get garbage that runs smoother.
I don't think it's entirely uncommon to have separate computers running different parts of the software on the manufacturer sims, but it's fairly unnecessary if AI is out of the picture. Physics rate is the biggest cost there (e.g. 1kHz is approx 2x more computationally intensive than 500Hz, etc), things like track grip etc are generally a non factor in that regard (tiny percentage of overall real-time computation). Consumer sims have more than enough headroom for features like that.

When possible, you'll take PSD measurements of the surfaces around the track, which can be used as direct input into the tire and track models. That'll get performance in the ballpark; things like rubber buildup, oils, etc are usually either hardcoded or ignored (as they're highly variable in real life and thus aren't super useful to model in a sim where you generally want repeatability). If the tire model isn't advanced enough (which is pretty common) to utilize PSD data, it'll typically be done via correlating the car model to real data on track. Given some base assumptions about vehicle accuracy, you can get fairly close with localized track grip adjustments based on sim vs real cornering speeds etc.

Definitely no one turning knobs for wind. It's not usually heavily considered in my experience (probably much more so in F1) as again it's heavily variable in real life, so simple models (to show broad trends) are usually fine. We have maps for aero sensitivities vs. wind velocity (as velocity gradients mean it's not the same as adding wind velocity to car velocity) but that's overkill for most applications. Most high level racing cars will have pitot tubes that allow for fairly accurate wind speed/direction analysis that you can fairly easily directly import into the sim if doing correlation work (but simple directional models also tend to get pretty close if the track doesn't have some extreme features that direct the wind unusually).
 
Last edited:
I don't think computing power is really a bottleneck for the kind of things that people would want/think they want in their consumer sims. It's more about having someone who has the dataset and know-how to translate it correctly. Being able to run your sim at 100kHz does nothing if you know less than what you don't know, you'll just get garbage that runs smoother.
I disagree, because we want Ai and that's a big factor in what processes need computational power. Also a large % of consumers want better graphics than what we see in pro sims.

This is pretty true, especially if only running a single car:

I don't think it's entirely uncommon to have separate computers running different parts of the software on the manufacturer sims, but it's fairly unnecessary if AI is out of the picture. Physics rate is the biggest cost there (e.g. 1kHz is approx 2x more computationally intensive than 500Hz, etc), things like track grip etc are generally a non factor in that regard (tiny percentage of overall real-time computation). Consumer sims have more than enough headroom for features like that.

When possible, you'll take PSD measurements of the surfaces around the track, which can be used as direct input into the tire and track models. That'll get performance in the ballpark; things like rubber buildup, oils, etc are usually either hardcoded or ignored (as they're highly variable in real life and thus aren't super useful to model in a sim where you generally want repeatability). If the tire model isn't advanced enough (which is pretty common) to utilize PSD data, it'll typically be done via correlating the car model to real data on track. Given some base assumptions about vehicle accuracy, you can get fairly close with localized track grip adjustments based on sim vs real cornering speeds etc.

Definitely no one turning knobs for wind. It's not usually heavily considered in my experience (probably much more so in F1) as again it's heavily variable in real life, so simple models (to show broad trends) are usually fine. We have maps for aero sensitivities vs. wind velocity (as velocity gradients mean it's not the same as adding wind velocity to car velocity) but that's overkill for most applications. Most high level racing cars will have pitot tubes that allow for fairly accurate wind speed/direction analysis that you can fairly easily directly import into the sim if doing correlation work (but simple directional models also tend to get pretty close if the track doesn't have some extreme features that direct the wind unusually).
So what you're saying is the pro sims are not 100% accurate/real because 100% accuracy/reality is not needed or desired. That last few % of realism is not worth the effort in attaining?
 
I disagree, because we want Ai and that's a big factor in what processes need computational power. Also a large % of consumers want better graphics than what we see in pro sims.
I think you've lost me. There is a perception that "computers aren't fast enough" to do "realistic physics" and I just don't agree right now. Better modeling for most things wouldn't increase CPU use noticeably, you need to get pretty advanced to start actually needing very high physics rates and so on.
 
I think you've lost me. There is a perception that "computers aren't fast enough" to do "realistic physics" and I just don't agree right now. Better modeling for most things wouldn't increase CPU use noticeably, you need to get pretty advanced to start actually needing very high physics rates and so on.
I mean with everything a game needs, seems clear from Mcf1papa that physics wise a modern PC should be able to handle things fine, but when you have a consumer sim you have to have realistic graphics and Ai etc etc that all take a slice of the pie of performance and more importantly development time.
 
I mean with everything a game needs, seems clear from Mcf1papa that physics wise a modern PC should be able to handle things fine, but when you have a consumer sim you have to have realistic graphics and Ai etc etc that all take a slice of the pie of performance and more importantly development time.
Plenty of sims are already running stuff with modern graphics and lots of AI at high enough physics rates to not really be a limit to reasonable designs. A lot of that isn't really related performance-wise AFAIK.
 
So what you're saying is the pro sims are not 100% accurate/real because 100% accuracy/reality is not needed or desired. That last few % of realism is not worth the effort in attaining?
Not really what I'm saying, no. Reality is highly variable (in ways that aren't predictable, someone could spill oil in a corner, for example, or wind could blow one way one lap and the other the next); that is a useless attribute for a simulator to have in a professional context, as it makes it impossible to tell how much of the lap time/handling/etc difference was from e.g. the setup change you just did vs from track evolution. Modeling random track condition variations would be actively making the simulation less useful.
 
Not really what I'm saying, no. Reality is highly variable (in ways that aren't predictable, someone could spill oil in a corner, for example, or wind could blow one way one lap and the other the next); that is a useless attribute for a simulator to have in a professional context, as it makes it impossible to tell how much of the lap time/handling/etc difference was from e.g. the setup change you just did vs from track evolution. Modeling random track condition variations would be actively making the simulation less useful.
so confirming my point that NO sim is "realistic", even pro sims as they exist within a bubble of non dynamic physics and reality is (as you state) highly variable and dynamic. Every dev will pick and choose (down to personal preference) what areas to devote most time to, sacrificing others.
 
Plenty of sims are already running stuff with modern graphics and lots of AI at high enough physics rates to not really be a limit to reasonable designs. A lot of that isn't really related performance-wise AFAIK.
yes but there are not simulating EVERYTHING are they, that is my point. EVERY sim dev chooses what aspects of the sim to concentrate on at the expense of others.
 

Latest News

Article information

Author
Yannik Haustein
Article read time
5 min read
Views
14,838
Comments
280
Last update
Back
Top