Project CARS 2 General Discussion Thread - Out Now on PS4/XB1/PC
- Thread starter jake2013guy
- 12,209 comments
- 1,210,790 views
- 297
- Bolton
- 99
- Middle of England
- tamora350
Can't be true, chewy would have chosen May 4th
Haha! I'm sure he would! The date doesn't even make any sense, but it's in a respected journal that I will lose all respect for if they've teased me like this.
ETA: Dammit! c172 just burst my bubble and all faith in journalism. #FakeNews
Ian Bell
Premium
- 729
- Ireland
We've rewritten the rendering engine from scratch. We didn't 'absolutely' need to and could have tweaked what we had but the render coders felt that we could extract a little more power and flexibility from it as well as improved looks and we need all of this to make LiveTrack 3.0 shine in all weathers, seasons and times of day. It's an insanely efficient rendering system now that gives us all of the flexibility we need, both now and going forward in the series with new more powerful machines.
- 99
- Middle of England
- tamora350
They pulled it out of nowhere, or to be very fair to them, from nothing we told them. We as a development team gave out a collective 'WHAT??' when we read it
May 6th (2015). Turns out they just re-used the PC1 release date! Forgive me for getting too excited. Any chance you could narrow down the launch window at this point?
ETA: Not to give EDGE too hard a time - They gave you a great write up and seemed like they really got what you were going for - a challenging sim for the hardcore that would not alienate, and could also be really enjoyed by the more casual player, with a little help/practice. I like the noises being made about the career mode, too. Can't wait!
Last edited:
Ian Bell
Premium
- 729
- Ireland
We are going to release a marketing video explaining the details of the new tyre system soon.
But here's a head's up on stuff that isn't secret.
In short it's generations beyond what anyone has shipped thus far. It's also the first fully dynamic node based tyre model in a commercial game. Nothing is baked. It simulates things like surface rubber being removed from the tyre based on the temperature, both external (surface) and internal (fed via the thickness of the rubber, the abrasive properties of the surface, the heat generated from that, the heat dissipated from the brake discs based on their proximity to the tyre, the amount of heat generated from a softer tyre pressure and compound compared to a harder tyre pressure) and that's just tyre wear...
Much of the below we've already published but it got little coverage so:
For the tyre carcass we measure and simulate how the elastic behaviour of the compound changes with speed, temperature, and pressure. How rolling resistance changes with speed, temperature and pressure. We measure sidewall buckling at low pressure and how that varies among Bias Ply, Radial, or Hybrid constructions. We include Gyroscopic Effects and Dynamic responses such as vibration, telescoping, and twisting live and all at 600 times per second as per the tread simulation.
For the tread simulation itself we utilise our finite difference simulation of the contact patch, with the tire tread “flowing” through the contact patch. The whole tread itself is discretized into elements much like the carcass, but the contact patch itself is a finite difference grid bringing:
Flash Heating, which is the change of temperature in the outermost rubber layer through the contact patch.
A Componentized grip model. Each component is affected differently by road surface conditions, wetness, and temperature;
Deformation – the rubber deforms in and around asperities, resisting sliding motion;
Adhesion – the rubber bonds to surface rubber and materials in a realistic manner;
Tack – the sticky tacky grip you can feel on your shoes when walking a rubbered in track, related to adhesion;
Tearing – the ripping of rubber from the tire;
Cut – grip from the geometry, edges, grooves, and siping of the tread, with particular effect in dirt and gravel;
Tread channel depth and water handling;
Discretized and temperature sensitive wear; Curing and; Temperature sensitive elastic properties.
The carcass and tread simulations are coupled such that there is no roughness or “stepping”, while still preserving the detail of both simulations. The contact patch size, shape, and pressure distribution is determined by the carcass simulation and is used by the tread simulation. The forces on the tire from the road surface are simulated in the tread simulation and transferred as external forces to the carcass simulation.
And then we have the heat transfer simulation which handles heat flow between brakes, wheel well, rim, carcass, and tread layers. How the heat transfers amongst tread elements, from tread elements to the road surface, and from the tread elements to the air are handled directly by the tread simulation (including advection and evaporation). The pressure of the tire is maintained by the carcass simulation via the ideal gas law.
But we're 'simcade'
Edit, nearly forgot. We've also written a new drivetrain, throttle and differential simulation model for pCARS2.
Last edited by a moderator:
Ian Bell
Premium
- 729
- Ireland
But, but.. That's all nice, but the people really want Adaptive Tessellation.
Nah, seriously. I guess this answers the second question
There's always some pressure to 'gamify' things such that the harsh realities don't overly affect enjoyment. So maybe we give those old tyres a little more headroom... I want it to be clear though that what we are doing is generations ahead of the competition and I know what they are doing. 
- 4,382
We are going to release a marketing video explaining the details of the new tyre system soon.
But here's a head's up on stuff that isn't secret.
In short it's generations beyond what anyone has shipped thus far. It's also the first fully dynamic node based tyre model in a commercial game. Nothing is baked. It simulates things like surface rubber being removed from the tyre based on the temperature, both external (surface) and internal (fed via the thickness of the rubber, the abrasive properties of the surface, the heat generated from that, the heat dissipated from the brake discs based on their proximity to the tyre, the amount of heat generated from a softer tyre pressure and compound compared to a harder tyre pressure) and that's just tyre wear...
Much of the below we've already published but it got little coverage so:
For the tyre carcass we measure and simulate how the elastic behaviour of the compound changes with speed, temperature, and pressure. How rolling resistance changes with speed, temperature and pressure. We measure sidewall buckling at low pressure and how that varies among Bias Ply, Radial, or Hybrid constructions. We include Gyroscopic Effects and Dynamic responses such as vibration, telescoping, and twisting live and all at 600 times per second as per the tread simulation.
For the tread simulation itself we utilise our finite difference simulation of the contact patch, with the tire tread “flowing” through the contact patch. The whole tread itself is discretized into elements much like the carcass, but the contact patch itself is a finite difference grid bringing:
Flash Heating, which is the change of temperature in the outermost rubber layer through the contact patch.
A Componentized grip model. Each component is affected differently by road surface conditions, wetness, and temperature;
We have Deformation – the rubber deforms in and around asperities, resisting sliding motion;
Adhesion – the rubber bonds to surface rubber and materials in a realistic manner;
Tack – the sticky tacky grip you can feel on your shoes when walking a rubbered in track, related to adhesion;
Tearing – the ripping of rubber from the tire;
Cut – grip from the geometry, edges, grooves, and siping of the tread, with particular effect in dirt and gravel;
Tread channel depth and water handling;
Discretized and temperature sensitive wear; Curing and; Temperature sensitive elastic properties.
The carcass and tread simulations are coupled such that there is no roughness or “stepping”, while still preserving the detail of both simulations. The contact patch size, shape, and pressure distribution is determined by the carcass simulation and is used by the tread simulation. The forces on the tire from the road surface are simulated in the tread simulation and transferred as external forces to the carcass simulation.
And then we have the heat transfer simulation which handles heat flow between brakes, wheel well, rim, carcass, and tread layers. How the heat transfers amongst tread elements, from tread elements to the road surface, and from the tread elements to the air are handled directly by the tread simulation (including advection and evaporation). The pressure of the tire is maintained by the carcass simulation via the ideal gas law.
But we're 'simcade'
Edit, nearly forgot. We've also written a new drivetrain, throttle and differential simulation model for pCARS2.
Wow... you guys are going all out innit?
With all that's going on under the physics engine and the improvements you guys are making to the visuals and car audio, I hope CARS2 can run with most of it's bells and whistles on current consoles, if not all.
Ian Bell
Premium
- 729
- Ireland
Wow... you guys are going all out innit?
With all that's going on under the physics engine and the improvements you guys are making to the visuals and car audio, I hope CARS2 can run with most of it's bells and whistles on current consoles, if not all.
Yes in a Spinal Tap sense, this one goes to eleven.
It's so insanely complex though that I'm glad we have so much time remaining to get it all polished and solid.
- 99
- Middle of England
- tamora350
We are going to release a marketing video explaining the details of the new tyre system soon.
But here's a head's up on stuff that isn't secret.
In short it's generations beyond what anyone has shipped thus far. It's also the first fully dynamic node based tyre model in a commercial game. Nothing is baked. It simulates things like surface rubber being removed from the tyre based on the temperature, both external (surface) and internal (fed via the thickness of the rubber, the abrasive properties of the surface, the heat generated from that, the heat dissipated from the brake discs based on their proximity to the tyre, the amount of heat generated from a softer tyre pressure and compound compared to a harder tyre pressure) and that's just tyre wear...
Much of the below we've already published but it got little coverage so:
For the tyre carcass we measure and simulate how the elastic behaviour of the compound changes with speed, temperature, and pressure. How rolling resistance changes with speed, temperature and pressure. We measure sidewall buckling at low pressure and how that varies among Bias Ply, Radial, or Hybrid constructions. We include Gyroscopic Effects and Dynamic responses such as vibration, telescoping, and twisting live and all at 600 times per second as per the tread simulation.
For the tread simulation itself we utilise our finite difference simulation of the contact patch, with the tire tread “flowing” through the contact patch. The whole tread itself is discretized into elements much like the carcass, but the contact patch itself is a finite difference grid bringing:
Flash Heating, which is the change of temperature in the outermost rubber layer through the contact patch.
A Componentized grip model. Each component is affected differently by road surface conditions, wetness, and temperature;
Deformation – the rubber deforms in and around asperities, resisting sliding motion;
Adhesion – the rubber bonds to surface rubber and materials in a realistic manner;
Tack – the sticky tacky grip you can feel on your shoes when walking a rubbered in track, related to adhesion;
Tearing – the ripping of rubber from the tire;
Cut – grip from the geometry, edges, grooves, and siping of the tread, with particular effect in dirt and gravel;
Tread channel depth and water handling;
Discretized and temperature sensitive wear; Curing and; Temperature sensitive elastic properties.
The carcass and tread simulations are coupled such that there is no roughness or “stepping”, while still preserving the detail of both simulations. The contact patch size, shape, and pressure distribution is determined by the carcass simulation and is used by the tread simulation. The forces on the tire from the road surface are simulated in the tread simulation and transferred as external forces to the carcass simulation.
And then we have the heat transfer simulation which handles heat flow between brakes, wheel well, rim, carcass, and tread layers. How the heat transfers amongst tread elements, from tread elements to the road surface, and from the tread elements to the air are handled directly by the tread simulation (including advection and evaporation). The pressure of the tire is maintained by the carcass simulation via the ideal gas law.
But we're 'simcade'
Edit, nearly forgot. We've also written a new drivetrain, throttle and differential simulation model for pCARS2.
That all sounds fantastic, Ian, but when are we going to get our hands on it? Will there be a demo?
- 1,814
- San Diego CA
- Alex_ONeil
Ian Bell
Premium
- 729
- Ireland
The rule I always follow on PC is to get the most powerful system you can afford that also has some scope for future upgrading. pCARS2 won't have requirements much above what was needed for pCARS1. But to show it at its epic best, always aim for a latest generation GPU (1080) and CPU (that new AMD is looking very good).
- 1,814
- San Diego CA
- Alex_ONeil
- 297
- Bolton
Just take my money now .We are going to release a marketing video explaining the details of the new tyre system soon.
But here's a head's up on stuff that isn't secret.
In short it's generations beyond what anyone has shipped thus far. It's also the first fully dynamic node based tyre model in a commercial game. Nothing is baked. It simulates things like surface rubber being removed from the tyre based on the temperature, both external (surface) and internal (fed via the thickness of the rubber, the abrasive properties of the surface, the heat generated from that, the heat dissipated from the brake discs based on their proximity to the tyre, the amount of heat generated from a softer tyre pressure and compound compared to a harder tyre pressure) and that's just tyre wear...
Much of the below we've already published but it got little coverage so:
For the tyre carcass we measure and simulate how the elastic behaviour of the compound changes with speed, temperature, and pressure. How rolling resistance changes with speed, temperature and pressure. We measure sidewall buckling at low pressure and how that varies among Bias Ply, Radial, or Hybrid constructions. We include Gyroscopic Effects and Dynamic responses such as vibration, telescoping, and twisting live and all at 600 times per second as per the tread simulation.
For the tread simulation itself we utilise our finite difference simulation of the contact patch, with the tire tread “flowing” through the contact patch. The whole tread itself is discretized into elements much like the carcass, but the contact patch itself is a finite difference grid bringing:
Flash Heating, which is the change of temperature in the outermost rubber layer through the contact patch.
A Componentized grip model. Each component is affected differently by road surface conditions, wetness, and temperature;
Deformation – the rubber deforms in and around asperities, resisting sliding motion;
Adhesion – the rubber bonds to surface rubber and materials in a realistic manner;
Tack – the sticky tacky grip you can feel on your shoes when walking a rubbered in track, related to adhesion;
Tearing – the ripping of rubber from the tire;
Cut – grip from the geometry, edges, grooves, and siping of the tread, with particular effect in dirt and gravel;
Tread channel depth and water handling;
Discretized and temperature sensitive wear; Curing and; Temperature sensitive elastic properties.
The carcass and tread simulations are coupled such that there is no roughness or “stepping”, while still preserving the detail of both simulations. The contact patch size, shape, and pressure distribution is determined by the carcass simulation and is used by the tread simulation. The forces on the tire from the road surface are simulated in the tread simulation and transferred as external forces to the carcass simulation.
And then we have the heat transfer simulation which handles heat flow between brakes, wheel well, rim, carcass, and tread layers. How the heat transfers amongst tread elements, from tread elements to the road surface, and from the tread elements to the air are handled directly by the tread simulation (including advection and evaporation). The pressure of the tire is maintained by the carcass simulation via the ideal gas law.
But we're 'simcade'
Edit, nearly forgot. We've also written a new drivetrain, throttle and differential simulation model for pCARS2.
Quick question, not sure if it's already been answered. Will the AI cars have a simplified tyre model and how will they react to changing conditions and has this element been improved/ tweaked for pc2?
- 204
- Johannesburg,South Africa
- Xx_REDRUM_xX
@IanBell
Over in the Q&A thread I read that if it rains in practice and you start qualifying the track will still be wet. That is truly remarkable and so far beyond what anyone else has done. If you guys achieve everything you set out to achieve and manage to do all of it at a high level...man then pCARS2 are going to cause some serious headaches and stress at Turn10 and Polyphony. I really can't wait for pCARS2!
One thing I've been thinking about lately is how do you guys go about managing the speed of the AI. How do you decide the lap times for the AI is fair when the AI level is at 80 compared to when it is at 90. And more tricky, how do you decide that the lap times for track 1 is equivalent of AI level 80 compared to what lap times is fair for AI level 80 at track 2? If what I just said makes sense...
Seems like a very difficult and almost impossible process to perfect.
What are the chances that you can include the AI level slider within the race menu...in other words after you loaded the track and cars etc. Reason being whenever I start a race weekend, or career race going to practice. I spent quite a bit of time to change the AI level, see what lap times they get and how competitive they are...exit, make adjustments to AI level, reload, test, exit etc. etc. until I've reach a nice competitive balance before proceeding through the sessions.
The issue why I do this is, as any racing game ever, the AI difficulty differs from one track to the other, as I've got stronger tracks and weaker tracks pace-wise. The difference is impossible to get right from a development point of view. Just thought if we could dynamically change the AI level within a race session it could go a far way to make it easier for us to find the right AI difficulty for that track.
It all depends though when the AI speed, track behaviour etc. is loaded and whether it is possible to include it within a race weekend session?
Over in the Q&A thread I read that if it rains in practice and you start qualifying the track will still be wet. That is truly remarkable and so far beyond what anyone else has done. If you guys achieve everything you set out to achieve and manage to do all of it at a high level...man then pCARS2 are going to cause some serious headaches and stress at Turn10 and Polyphony. I really can't wait for pCARS2!
One thing I've been thinking about lately is how do you guys go about managing the speed of the AI. How do you decide the lap times for the AI is fair when the AI level is at 80 compared to when it is at 90. And more tricky, how do you decide that the lap times for track 1 is equivalent of AI level 80 compared to what lap times is fair for AI level 80 at track 2? If what I just said makes sense...
Seems like a very difficult and almost impossible process to perfect.
What are the chances that you can include the AI level slider within the race menu...in other words after you loaded the track and cars etc. Reason being whenever I start a race weekend, or career race going to practice. I spent quite a bit of time to change the AI level, see what lap times they get and how competitive they are...exit, make adjustments to AI level, reload, test, exit etc. etc. until I've reach a nice competitive balance before proceeding through the sessions.
The issue why I do this is, as any racing game ever, the AI difficulty differs from one track to the other, as I've got stronger tracks and weaker tracks pace-wise. The difference is impossible to get right from a development point of view. Just thought if we could dynamically change the AI level within a race session it could go a far way to make it easier for us to find the right AI difficulty for that track.
It all depends though when the AI speed, track behaviour etc. is loaded and whether it is possible to include it within a race weekend session?
That is a very good point actually, glad you asked about this, as I had very similar experience as well in the past with multiple sims. I think the ideal solution to this problem would be some kind of dynamic difficulty setting, that is the game constantly checking our lap times on all tracks, calculating an average and adjusting the AI's difficulty and speed to match that or even making them a bit faster on the highest difficulty setting (within the limits of physics of course) and making them progressively slower and less capable on the lower difficulties.
I imagine this setting as a separate one from the current number based difficulty, with a possible off setting if someone only wants to use the current, fixed difficulty numbers.
What would also be very cool if we could also have another slider that would adjust the difficulty spread, with the possibility of eliminating it completely, so that all cars on the grid would be just as fast as the leader.
mister dog
(Banned)
- 8,396
- Spain
- misterdog
- misterdog7
Well I actually just started now, so that won't be a problem. To elaborate on what the wookie said for me the new trees and upgraded skies stick out the most. As a graphics whore that's probably where most of my feedback will come from in these next months so rest assured I'll be giving SMS a hard time from now on.
Nah it's most of the content already, but all of it WIP of course so some cars/ tracks are further along than others.
- 51,174
- Australia
- SpacedustDaddy
Nice. Thanks for a bit of insight.Well I actually just started now, so that won't be a problem. To elaborate on what the wookie said for me the new trees and upgraded skies stick out the most. As a graphics whore that's probably where most of my feedback will come from in these next months so rest assured I'll be giving SMS a hard time from now on.
Nah it's most of the content already, but all of it WIP of course so some cars/ tracks are further along than others.
Ian Bell
Premium
- 729
- Ireland
There won't be a demo. Demos mean less sales for all games, regardless of how amazingly good they are. Hence very few demos lately.
A very odd set of rules are involved here and I'm quoting from published metrics.
Demos are expensive, and have four cases:
Still buying: I was going to buy the game, played the demo, and am still going to buy the game. No change for the company.
Now buying: I wasn't going to buy the game, but after the demo, I will. +1 for the company but quite rare as those not buying aren't motivated to play or 'enjoy' the demo. They want their preconceptions to be confirmed and look for negatives however tentative.
Not buying: I was going to buy the game, but the demo didn't fulfil all of my desires. I'm not buying now. -1 for the company
Never buying: I wasn't going to buy the game, and the demo didn't change my mind or I didn't test it. No change for the company.
Unfortunately for people who like demos, the "Now buying" group tends to be outnumbered by the "Not buying" group; which means that many companies see demos as spending money to lose customers.
Now, we leave all marketing and commercial decisions to Bandai Namco but the data is in and it's clear. Demos, regardless of how good the game is, cost money and do nothing for sales.
Keep this between us.
Half of this was nicked from Reddit after I recalled reading it there and in two other places.
Last edited by a moderator:
- 385
- Hawkes Bay
- davidb_165
Hi. Can you flat spot tyres with a big lock up?
Ian Bell
Premium
- 729
- Ireland
Yes.
- 385
- Hawkes Bay
- davidb_165
Never see nothing like that 
- 1,452
- 2,319
- Michigan
- FuriousDemon
I think I saw somewhere that GTX 1080 can run it at 4K at 60 fps. Maybe it was some YouTube video. That would be pretty impressive if true.
Similar threads
Latest Posts
-
-
-
Assetto Corsa 2 Will be Called "Assetto Corsa Evo", Coming 2024- Latest: OooAhh Cantona
-
