My biggest issue with iRacing and a lot of other sims is actually not how a single car drives but how they interact. Netcode aside, the cars spin way too easily after contact. If you look at real life motorsport, the start of a GT3 race is full of contact, little bumps here and there, maybe a bit of bodywork flies off, sometimes you're unlucky and the suspension gets killed, but rarely do you ever spin from such contacts. In iRacing a little tap often leads to a spin that you can't recover. If they can reduce the impact of these collisions, there'd be less carnage and more realistic racing.
I think it's because the body dosen't flex properly. IRL, collisions are greatly inelastic. Meaning, that a lot of energy will be dissipated by the car flexing on contact. In iracing, the collisions are like between billiard balls where all of the energy gets transmitted to the chassis.
When I came to iRacing I was really surprised on how easily someone spins when they hit you and you mostly stay on line. I don't think this is realistic but I do think it's by design and punishes poor driving habits. I have grown to like it.
@@frankovercrest2317 Good description. But i think that feeling is much more there in ACC. Less in iRacing, but still sometimes that "billiardy" behaviour.
@@huludical1 Sometimes i also feel after contact you spin too easily. Same ball park as the weired feeling of the tyres the first corner after a single lock up. But i also had a lot of 0x rubbing without bouncing effect. I think a lot of the spinny stuff is in combination with cold tyres. With cold tyres you spin quite quickly anyway. In that situation even little touches knock the grip away.
One thing we do with race simulator physics is bullshido, bullshido is used to fill the gaps that maths don't cover, all about reaching that perfect touchy feely feel. That's why some vehicles feel more correct than others and why some sims feel better than others in different situations. There will always be a level of artistic interpretation.
I would add that even now, in 2024-2025 we can't simulate tires with 99% even in F1. And even tire manufacturers themselves have to rely on a good old trial and error. Pirelli did a good job with current C1-C5 compounds, but even they fail miserably when predicting how long will they last, and so on. So there is still some black magic involved. That's why sim studios can't get the tires "right," and why every sim has different tires / way of modeling tires.
@@jamesmccaul2945 I do currently similar things for an aerospace company. What the guy means with artistic should be fairly obvious here but I explain for you. It’s a mixture of interpolation between data points and adding weight factors that are calculated based on a data set of a few, but not all situations possible. Thus, due to calculation limits the estimations need to be aligned to fit many different situations (grip level, temperatures and temperature gradients over the whole tire, environment temperature and radiation heat from the brake disc, load on the tire etc). So it should be obvious that not every situation can be calculated before it occurs in the simulator. That means that the racing simulator needs to do either ljve calculations, which can basically do only half the math needed, or it needs to use „look up tables“ of precalculated situations, or, and this is the case, a mixture of both. Now, for the lookup tables, there is as I said not enough data possible to fit in it. Thus, in the live calculations part, interpolation and using weight factors plays a role. But, as I explained, the weight factors themselves can not be accurate for any situation. This is what the first commentator meant by „things feel good and bad on one or the other simulator on different situations“. That is due to differences in weight factors and interpolation methods. For different feelings these weight factor can be adjusted. Let me make one example: 80deg tyre outside, 300deg inside, 15C environment and a bit heat from the brake disc is calculated. Car has a certain load on one tyre and it drives a specific radius with a specific friction coefficient on the track on that specific part where the tyre touches the track. In one simulator, this exakt situation is by coincidence precalculated, the other one interpolated and less accurate. Now you can add that on one simulator wants to have a more fun feeling and gives the tyre at the limit 5% more grip when the brake pressed, the other one makes it loose grip instantly due to the different loads on braking. Tadaa, magic is reached. Funny feeing. Ur welcome.
@@jamesmccaul2945I wanna add that the first commentator with „funny stuff“ wasn’t even wrong here. We literally have no clue what really happenes between interpolated data points and try to match our weihht factors on the set of precalculated situiation, which is indeed funny and random, because we literally need to extimate what situations to precalculate. And this is some real complex shit because there is hundreds of parameters and only limited storage for those datasets. So it is funny and random and estimated even tho we try our best. Nature is hard and complex my friend
One thing about tyres... Theyre not consistent even in real world. I can run a Avon tyre one week and a Hancook tyre the next and they're completely different. Shit even one set to the next of the same brand can be different. Its why you need to be adaptable
Every rl race driver i spoke about the iracing tiremodel, they all said something simular, that the tire midturn doesn't feel right, so this data makes sense to me. I hope they can find a sweatspot, where the car still feels responsive and the tire isn't overheating.
I have seen too many promises of tyre model updates in iRacing to believe that this will be a fundamental refresh which can fix the core of the problem. Having been lucky enough to drive a lot of things in real life, including LMP3 cars, I can say that the iRacing tyre model is extremely sensitive and unfortunately not realistic, especially with such a narrow peak. I do hope I am proven wrong, but this "fix" has been touted many times before, only to not really make a fundamental change to the tyre model issue.
Double precision means how floating point numbers are stored / calculated at. Single precision as in IEEE-754 means 32 bit / 4 bytes store a number. One bit for sign, 8 bits for exponent, and 23 bits for mantisaa. Double uses 64 bit / 8 bytes, so more bits for everything. Typically serious simulations use double precision.
Saying the term "double precision" used in the article means double precision floating points is a good guess but still only a guess. They could be doubling the physics engine tick rate and a marketing person could perfectly explain that as doubling the precision of the physics engine since a higher tick rate would mean exactly that: a more precise physics output. So good guess, but I wouldn't make that statement with so much confidence.
@@pixeljetstream they are testing higher rates ffb with no interpolation. And they anounced some time ago that the rework of the core physics system was underway. Their physics extern loop works at 60Hz, so probably they are working on it since they cannot upgrade FFB tickrate with no interpolation without uograding engine’s tickrate.
Hey Nils. iRacing has been busy lately with their tire (specially since they signed Terence Groening from RF2 to work alongside Dave Kaemmer). This is part from February 2023 development blog: "The tires themselves are also being worked on constantly which will improve all forms of racing. Of course, rain created a tremendous amount of new tire and track surface work, or as our engineers call the rain tire work in their technical design document, “Tires on a Film of Fluid”. We’re going to get a little technical here, and I’m borrowing some details from Dave Kaemmer‘s various technical tire documents/designs. The tire model is a collection of dozens of models, properties, data and recipes (rubber, etc.) that all work together. Our tire model is the most sophisticated and in-depth model in the industry, but back to being self-critical, it can and will be better. Most of the sub-models within the overall model work amazingly well. However, a few of these sub-models could be better and you feel those issues at times when you drive. Improving those last pieces of the model will make a difference in the feel and the behavior of the tires. Just for example, we think we do the following quite well: Modeling of forces and tire deflections from stopped to high speed for car tires; Carcass foundational stiffnesses from construction parameters; Slip curves from tread rubber properties, including dynamically, as temps and speeds change; Transient thermal modeling near and at tire surface and effect of temperature on grip; Relaxation modulus from basic rubber recipe, and given conditioning level; Model for grip, including track thermal properties. Some examples of sub-models within the tire model that we want to improve include: Improvements to the contact patch model; Rolling Drag; Speed of tread conditioning; Heat buildup and temperature gradient through the tread; Camber modeling. A major project over 2022 in regard to tires was developing a finite element model (FEM) specific to vehicle racing tires. The intention of this separate model was to build a tool to accurately simulate the behavior of tires under different operating conditions, especially to analyze the loads and forces/moments occurring in the contact patch. With this tool available, we should be able to extend and improve iRacing’s current tire model to respond more accurately to these loads, forces and moments." Since then, they finished the rain tire and they did it amazing, noticeably better than their dry tire. I think they will deliver with V10 version. Currently, the newer iterations of tires are also better (Porsche Cup and TCR were recently reworked, and their tires feel like a clear step forward regarding exactly what you say). It's very difficult to build a physical tire, specially if you don't want to allow any 'driftsim' behaviour. My opinion is that they were not ready yet to make that step without becoming a driftsim like RF2 did.
Hmm, maybe this update will get me back on iRacing to try it out, the tire model made me give up on the sim after waiting years for improvement, honestly don't expect much from this with this old engine.
11:30 double precision means that they will go from using Floats (4 bytes) to Doubles (8 bytes) hence the name double. This means that the calculations can be much more precise since they can have larger numbers and / or more decimals
yeah cheers, was mentioned somewhere else in the meantime. though, something like that will hardly change precision apart from tiny rounding errors. at least i doubt its gonna be anywhere near noticable really
Forza updated the frequency of physics calculations (number of iterations per second). Double precision in programming terms means you're using doubles instead of floats. Those are floating point number representations using respectively 64 and 32 bits. Not sure this is what they meant, but sure sounds like. In any way, two separate things (frequency of calculations vs precision of a single calculation).
Every time you look at inside/mid/outside tyre temps in iracing in motec, they are totally SCREWED! Rather than fix it, they double down with update after update when the philosophy of the tyre model is wrong to begin with. We have seen novels written by iracing stating that this is how tyres behave even though every man and his dog knows the tyres are flawed but they keep doubling down.
I'm so fed up of driving in iRacing. The tyres are atrocious. GT3's are the worst. To me, GTE's are the best, then I guess the GT4's. Porsche Cup is better but it's like all of their tyres are skinny bicycle tyres, just barely touching the surface. I can't stand it anymore lol
7:58 somewhat sounds like what I feel in LMU, and to some extent rF2. LMU’s initial bite has improved over rF2, but the cars feel more snappy and finicky like what I constantly hear about iRacing. I noticed all cars in LMU also use the lowest tyre pressure too, 135-140 kpa which is around 19.5-20.5 psi or 1.35-1.40 bar and that sounds very low even for cold pressure.
@@lorenzorubino2195 I did some further googling and indeed 140kpa is the minimum tyre pressure, at least according to one of the many WEC files I found. I then found one that specifies LMP2 tyres for the 2023 6h of Spa, there it mentions that the minimum stabilized and static relative pressures should be around 1.90-1.95 bar. Honestly, I don't quite understand what stabilized and static really means but 1.9 bar cold pressure is definitely too high. I decided to run a few laps in Spa at 22 air temp, 35 track, on mediums. After 7 laps of just driving how I would in a race (2:07-2:08 lap times) the tyres never went over 170kpa and their temps floated around 80-83 degrees, a bit under ideal temps but not far. Unless I got something wrong with these weird stabilized and static terms, these tyre pressures are way too low for driving 7 laps of race pace on a warm track. Genuinely don't know what to think, I could never get to grips with LMU and I'm just trying to understand why.
@ Good research, I am a GT3 engineer in GTWC, not WEC, but I can tell you this: minimum tyre pressure cold is usually a suggestion, let’s say that teams have always a way to go lower. The 1.90 bar stabilised means that after one run, when the car comes back to the pit stops, with the hot tyres, the pressure reset should be 1.90 (basically bleeding out air until the tyre reaches that pressure). I don’t know about the Goodyear tyres, just to give some info: with Pirelli (with the tyre heater) we go as low as 1.10 bar (even lower) when it’s hot, without the tyre heater, 1.30 bar (obviously depending on track temp, tyre temp, run type etc..).
@@utkarshchaurasia2233 Well, irl there isn't really a low as possible. When there is a pressure min like in WEC, there are always ways to go lower (by naturally heating up the tyre) so at the end the pressures are always being adjusted based on conditions. In game, if the min is 1.40, and the tyre is modeled correctly, in hot weather one should have tyre overheating issues. But to be honest, it is very rare that a commercial tyre model gets the temp and pressure correct, it is a level of detail that nobody cares about, and that it complicates things quite a lot.
Every low downforce car on slicks feels like shit to drive on the limit because of the "flash heat" issue, so lets update the tires on the high downforce cars whose high downforce masks this issue. Maybe we'll get tires that behave like IRL this time next year
it shows character trying to get help. but maybe iracing set off to try do it too complicated, as 16 years after launch the tire does not feel or behave correct. would be great if they'd eventually succeed, but it would almost be a black swan event by now
@@SimracingPopometerI'm so fed up of driving on their version of tyres in most cars. The game has been out for over a decade, how the hell is this still a problem. GT3's are the most frustrating thing to drive.
The issue of outside of the tire cooling down slower is pretty simple really. Heat transfers better into tarmac than air. The outside gets hot during cornering like the other parts of the tire, but on the straights, it's not touching the tarmac and thus cools down slower. And when traveling on a straight line, there is of course not enough load to create more heat than is getting dissipated, so the entire tire gets cooler, at different rates depending what part is touching the tarmac and what is not. iRacing has some major issues with tires that are heat related though, but the above part is just thermal dynamics.
ruclips.net/video/BspEp2dVsZk/видео.html in short, no. while yes tarmac better transfers heat, the tire's contact patch is only in contact with the ground for not even 1/10th of the time it is exposed to air. plus, somehow various commenters forgot that the tire is producing heat as well as it is going straight and being flexed whenever the specific part of the tire becomes the contact patch for a split second. you can see this on pretty much any infrared recording of tires. the inside on cambered tires will remain warmer than outside. plus, iracing somehow is the only game that shows this phenomenon. no other sim has that pattern
@@SimracingPopometer Comparing to that video the issue seen in your iRacing telemetry is more that the outside of the tire gets so hot in the first place. If you look at the video you linked after the carousel, you can see that inside gets the hottest but it also cools down the quickest as that part is touching the tarmac the most. As far as the straightline part, yes tire produces heat while going straight but not enough to prevent it from cooling down on the straights after cornering. So even if the straight was long enough, the tire would of course never cool down to the same temp as the road is. I think iRacing has some major issues with how heat affects the tires though, pretty much always has. And the fact that more rubbered track means less grip has always been weird to me too. I really hope this new update improves things, we've waited that for a long time 😄
@@IlkkaHaapala because this is a thing in Nascar tires. The road tires still have probably this behaviour from that. A couple of real Nascar engineer confirmed this in the old forum (they did not work for iRacing). Anyway, it is a PLAUSIBLE thing to happen. It may differ from real life in the amount and frequency it happens; but it does.
Maybe you did not see tire temp changes outside and inside on Oultons straight since real-time telemetry is unavailable to third party apps. I have read they do this to avoid cheating.
1:56 where did you get this information from? I expect that the M2 gets a rookie series, just like the MX5 cup. it would be completely useless otherwise.
think that might've been my first time skim reading, but i can't find that part anymore. "This car will be used at times in our series and will be selectable for you all in AI and Hosted iRacing." maybe the "at times" part. so probably wrong here on my part
@@SimracingPopometer It seems to be as following: Due to the high negative camber, of course on straights the outside of the tire don’t have contact to the asphalt. But the tire cools down exactly through the asphalt, that’s why the outside doesn’t = no conductive cooling -> air is an insulator. That’s physics but I do understand that it looks weird on the first look.
Therefore that’s not an argument to underline the stiff carcass construction. This disappears at higher downforce cars but I’m not sure if you can see this also on other low down force cars with high negative cambers. Has to be verified.
It's unclear from their update whether the lmp2/GTP tire update is the same as the v10 tire. If it is it's a little odd that they didn't just call it v10 in the first part of the update.
@SimracingPopometer from the article "The 499P will join our now 16-car-deep Endurance Series, as well as the BMW M Hybrid V8, Cadillac V-Series.R GTP, Acura ARX-06 GTP, and Porsche 963 GTP in a reconfigured/reconstituted Prototype Challenge Series. " aka replaces lmp2 fixed in a gtp fixed series and also in the 6h series the issue is iracing imsa is licensed so they can't use 499p
How do you manage to check tyre temps in real time? 😮 All channels I found in simhub updates temps and wear only after pit stop, same like in Black box.
this is from the telemetry file, not shared memory, and only available after the lap. check and download popometer and have a look. even though temps are part of the advanced channels with the sub
God i hope they fix the fps drops. Going from 90fps in vr to 50 in one corner has costed me multiple races. Ive messed with the settings and have everything so low its ridiculous but nothing seems to have any effect
Is the tire just getting insulated by the air between it and the track stopping the brake temp from dispersing throughout the tire and then to the track? So rather than the tyre getting cool because it has no friction it gets hot because it cannot transfer the heat to anywhere? I have no idea about how this works in sims I'm just curious.
I saw on iracing forums it's as you say - both tarmac and air have lower temperature than tyre surface but tarmac has higher thermal conductivity so inside and middle of a tyre cool down quicker thanks to touching tarmac and transferring heat to it. I don't know if the temperature differences should be like they are currently or maybe lower or higher but that's their explanation to outside not losing as much temperature on straights as inside and middle of a tyre.
why is it only on some cars then, and why is it the only sim that does that. and why does a tire on the road not already produce temperature from being flexed while driving? we're talking minimum pressures here. no matter how they put it, its odd at least!
At the end of the day iracing is the same as every other sim. Some cars feel good. Some don't. Whatever the data tells you, iracing feels just as realistic as any other sim. At least they go forward with development and don't make objectively worse games like kunos. People trash iracing but somehow ACC doesn't get crap but it feels nothing like driving a car in any way.
@@SimracingPopometer My wild guess is that some property is over or undertuned like heat transfer between tyres and ground or heat buildup from friction, especially that most race cars run non zero toe. It also might be that they got those things right but something else is wrong that affects inside vs outside temps that gets more noticable with specific suspension geometry.
Typical case of "let's make it different because we want to be unique and special" and they just made the worst tyre model (that I've tried at least). Made the switch to PC, and my driving style might not be optimal (I have a very Porsche tailored driving style) but AC, ACC, rFactor 2, Rennsport, even on the AMS2 demo I can drive the same. iRacing? I'm scared of turning the wheel, I have no idea when the tyres are gonna go past that ultra thin grip-no grip treshold. I got a year for the price for one month and it's been like a month since I last touched it. Between the tyres and the price I'd rather hotlap on ACC tbh.
I race irl and I like the tire feel. Yes they are sensitive but it forces you to focus on the fundamentals of driving which actually improves you in all other games due to the focus on technique
how is vr performance? am tempted to give iracing a shot but will a 5700x3d be enough if i turn shadows off.. am upgrading gpu but am worried about a cpu bottle neck
It's been a very long journey for iRacing with their physical tire model approach where most sim racing titles have been turning knobs on the old emprical models. I hope this update is going to be the breakthrough
iRacing has literally tried to strong arm people before with legal action for trashing there tire model publicly. Now they admit that they were doing it wrong all along... lol!
im not sure that simply mixing people that produce 0 slip tires and people that produce infinite slim tires will magically result in somewhat medium slip tires :P but of course we remain hopeful
@@SimracingPopometer I’m sure these guys know something about tires since that had been there life previously and now they have Iracing way of doing it and come to a better place.
@@SimracingPopometer It's 'too fast'. The game feels like it's speed up. FOV is broken and the best way to understand is to watch replays and check at the objects in the background. They are unrealistically large. The whole 3D world is f**** up. Try other games, like ACC, and everything will be much more real (like in the real world). This then affects handling too.
@@SimracingPopometerfor 4 yrs on iRacing I was never happy with the FOV. It always felt odd, as if I was sitting on the headrest in the car. The ONLY thing that finally fixed it for me was to follow Sense of Speed's long 45 minute video of FOV. Now I finally feel like I'm sitting IN the car.
@@SimracingPopometer it's a long video but I skipped through it until he posted #'s on the screen indicating a change being made. You'd have to watch the video but you'll end up changing WorldNearPlane distance & the "point size" which I think has to do with how the pixels are drawn. Honestly, it's a much watch. Driving the Lamborghini, I can finally see more out of the windshield because the larger monitor isn't such a big issue anymore. Whatever iRacing setup in the ini's files for FOV is all wrong like many other things I'm noticing with iRacing (especially the tyres)
iracing and it's tires are religion among the users, they are used to it to the point they think it's correct. And there is nothing bad about it, I tougth that partly because user base is so used to it it's hard to change. I usually hope only good things to everybody, but Iam not so keen to see iracing developing theyr weakness. if they really get it rigth they migth get users from other platforms that allready have few users. That way i could force us that don't wan't to pay ridicilous amounts to join iracing if we want to race online. What do I know physics update migth also deter old hard core iracing users..
iRacing V10 tyre model seems not simulate deformable mesh of the tyre. If has this they will show picture in development blog. Maybe just some improvement on contact patch brushes model. Not high level simulation as Le Mans Ultimate and rFactor2.
@SimracingPopometer I do agreed with you that nothing is perfect but it's iracing is worst simulator in terms of tire model and ffb . By the way if you wanted try 60hz you need to open config file and disable simucube 360 api
The correct tire in the digital environment with 0 risks produces incorrect driver behavior. Like too aggressive close driving and driving always on the grip limit mentality. I don't know if by design or by happy accident but iracing sketchy grip limit cut off kinda fixed those common sim problems. It seams like the incorrect tire is producing the most correct racing. I hope they will keep it in mind while fixing the tire.
My personal conspiracy theory is that iRacing intentionally fudge the tyres to make them less driveable so that eSports aliens can't drift around IRL drivers and drive them away from the platform.
I cant take anyone seriously that thinks ACC is anything like driving an actual car. That game is objectively the worst sim on the maket. The game they made twn years prior is much much much more realistic.
James Baldwin said himself that ACC simulated GT3s better than iRacing and he drives in the Fanatec World Challenge. What weight does your opinion hold compared to his irl experience?
@@TheOfficialRandomGuy Max and Lando and every other f2 driver is on Iracing and they can KICK James ass in a car. James is like any other karting sim racer who is quick. Opinions are opinions and not the 12 commandments.
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
My biggest issue with iRacing and a lot of other sims is actually not how a single car drives but how they interact. Netcode aside, the cars spin way too easily after contact. If you look at real life motorsport, the start of a GT3 race is full of contact, little bumps here and there, maybe a bit of bodywork flies off, sometimes you're unlucky and the suspension gets killed, but rarely do you ever spin from such contacts. In iRacing a little tap often leads to a spin that you can't recover. If they can reduce the impact of these collisions, there'd be less carnage and more realistic racing.
I think it's because the body dosen't flex properly.
IRL, collisions are greatly inelastic. Meaning, that a lot of energy will be dissipated by the car flexing on contact. In iracing, the collisions are like between billiard balls where all of the energy gets transmitted to the chassis.
When I came to iRacing I was really surprised on how easily someone spins when they hit you and you mostly stay on line. I don't think this is realistic but I do think it's by design and punishes poor driving habits. I have grown to like it.
@@huludical1 well, it depends on the car, but it goes both ways, often both cars spin
@@frankovercrest2317 Good description. But i think that feeling is much more there in ACC. Less in iRacing, but still sometimes that "billiardy" behaviour.
@@huludical1 Sometimes i also feel after contact you spin too easily. Same ball park as the weired feeling of the tyres the first corner after a single lock up. But i also had a lot of 0x rubbing without bouncing effect. I think a lot of the spinny stuff is in combination with cold tyres. With cold tyres you spin quite quickly anyway. In that situation even little touches knock the grip away.
One thing we do with race simulator physics is bullshido, bullshido is used to fill the gaps that maths don't cover, all about reaching that perfect touchy feely feel. That's why some vehicles feel more correct than others and why some sims feel better than others in different situations. There will always be a level of artistic interpretation.
I would add that even now, in 2024-2025 we can't simulate tires with 99% even in F1. And even tire manufacturers themselves have to rely on a good old trial and error. Pirelli did a good job with current C1-C5 compounds, but even they fail miserably when predicting how long will they last, and so on. So there is still some black magic involved. That's why sim studios can't get the tires "right," and why every sim has different tires / way of modeling tires.
@@jamesmccaul2945 I do currently similar things for an aerospace company. What the guy means with artistic should be fairly obvious here but I explain for you. It’s a mixture of interpolation between data points and adding weight factors that are calculated based on a data set of a few, but not all situations possible. Thus, due to calculation limits the estimations need to be aligned to fit many different situations (grip level, temperatures and temperature gradients over the whole tire, environment temperature and radiation heat from the brake disc, load on the tire etc). So it should be obvious that not every situation can be calculated before it occurs in the simulator. That means that the racing simulator needs to do either ljve calculations, which can basically do only half the math needed, or it needs to use „look up tables“ of precalculated situations, or, and this is the case, a mixture of both. Now, for the lookup tables, there is as I said not enough data possible to fit in it. Thus, in the live calculations part, interpolation and using weight factors plays a role. But, as I explained, the weight factors themselves can not be accurate for any situation. This is what the first commentator meant by „things feel good and bad on one or the other simulator on different situations“. That is due to differences in weight factors and interpolation methods. For different feelings these weight factor can be adjusted.
Let me make one example: 80deg tyre outside, 300deg inside, 15C environment and a bit heat from the brake disc is calculated. Car has a certain load on one tyre and it drives a specific radius with a specific friction coefficient on the track on that specific part where the tyre touches the track. In one simulator, this exakt situation is by coincidence precalculated, the other one interpolated and less accurate. Now you can add that on one simulator wants to have a more fun feeling and gives the tyre at the limit 5% more grip when the brake pressed, the other one makes it loose grip instantly due to the different loads on braking. Tadaa, magic is reached. Funny feeing. Ur welcome.
@@jamesmccaul2945I wanna add that the first commentator with „funny stuff“ wasn’t even wrong here. We literally have no clue what really happenes between interpolated data points and try to match our weihht factors on the set of precalculated situiation, which is indeed funny and random, because we literally need to extimate what situations to precalculate. And this is some real complex shit because there is hundreds of parameters and only limited storage for those datasets. So it is funny and random and estimated even tho we try our best. Nature is hard and complex my friend
One thing about tyres... Theyre not consistent even in real world.
I can run a Avon tyre one week and a Hancook tyre the next and they're completely different. Shit even one set to the next of the same brand can be different. Its why you need to be adaptable
Every rl race driver i spoke about the iracing tiremodel, they all said something simular, that the tire midturn doesn't feel right, so this data makes sense to me. I hope they can find a sweatspot, where the car still feels responsive and the tire isn't overheating.
I have seen too many promises of tyre model updates in iRacing to believe that this will be a fundamental refresh which can fix the core of the problem. Having been lucky enough to drive a lot of things in real life, including LMP3 cars, I can say that the iRacing tyre model is extremely sensitive and unfortunately not realistic, especially with such a narrow peak. I do hope I am proven wrong, but this "fix" has been touted many times before, only to not really make a fundamental change to the tyre model issue.
Double precision means how floating point numbers are stored / calculated at. Single precision as in IEEE-754 means 32 bit / 4 bytes store a number. One bit for sign, 8 bits for exponent, and 23 bits for mantisaa. Double uses 64 bit / 8 bytes, so more bits for everything. Typically serious simulations use double precision.
confirmed: iracing not a serious sim for the past 13 years
😂😂
Saying the term "double precision" used in the article means double precision floating points is a good guess but still only a guess. They could be doubling the physics engine tick rate and a marketing person could perfectly explain that as doubling the precision of the physics engine since a higher tick rate would mean exactly that: a more precise physics output. So good guess, but I wouldn't make that statement with so much confidence.
@@pixeljetstream they are testing higher rates ffb with no interpolation. And they anounced some time ago that the rework of the core physics system was underway.
Their physics extern loop works at 60Hz, so probably they are working on it since they cannot upgrade FFB tickrate with no interpolation without uograding engine’s tickrate.
@@f1_onboardsAC okay seems a bit odd choice of words coming from programmers if it was just the rate, but possible.
Hey Nils. iRacing has been busy lately with their tire (specially since they signed Terence Groening from RF2 to work alongside Dave Kaemmer). This is part from February 2023 development blog:
"The tires themselves are also being worked on constantly which will improve all forms of racing. Of course, rain created a tremendous amount of new tire and track surface work, or as our engineers call the rain tire work in their technical design document, “Tires on a Film of Fluid”.
We’re going to get a little technical here, and I’m borrowing some details from Dave Kaemmer‘s various technical tire documents/designs. The tire model is a collection of dozens of models, properties, data and recipes (rubber, etc.) that all work together. Our tire model is the most sophisticated and in-depth model in the industry, but back to being self-critical, it can and will be better. Most of the sub-models within the overall model work amazingly well. However, a few of these sub-models could be better and you feel those issues at times when you drive. Improving those last pieces of the model will make a difference in the feel and the behavior of the tires. Just for example, we think we do the following quite well: Modeling of forces and tire deflections from stopped to high speed for car tires; Carcass foundational stiffnesses from construction parameters; Slip curves from tread rubber properties, including dynamically, as temps and speeds change; Transient thermal modeling near and at tire surface and effect of temperature on grip; Relaxation modulus from basic rubber recipe, and given conditioning level; Model for grip, including track thermal properties. Some examples of sub-models within the tire model that we want to improve include: Improvements to the contact patch model; Rolling Drag; Speed of tread conditioning; Heat buildup and temperature gradient through the tread; Camber modeling.
A major project over 2022 in regard to tires was developing a finite element model (FEM) specific to vehicle racing tires. The intention of this separate model was to build a tool to accurately simulate the behavior of tires under different operating conditions, especially to analyze the loads and forces/moments occurring in the contact patch. With this tool available, we should be able to extend and improve iRacing’s current tire model to respond more accurately to these loads, forces and moments."
Since then, they finished the rain tire and they did it amazing, noticeably better than their dry tire. I think they will deliver with V10 version. Currently, the newer iterations of tires are also better (Porsche Cup and TCR were recently reworked, and their tires feel like a clear step forward regarding exactly what you say).
It's very difficult to build a physical tire, specially if you don't want to allow any 'driftsim' behaviour. My opinion is that they were not ready yet to make that step without becoming a driftsim like RF2 did.
Hmm, maybe this update will get me back on iRacing to try it out, the tire model made me give up on the sim after waiting years for improvement, honestly don't expect much from this with this old engine.
11:30 double precision means that they will go from using Floats (4 bytes) to Doubles (8 bytes) hence the name double. This means that the calculations can be much more precise since they can have larger numbers and / or more decimals
yeah cheers, was mentioned somewhere else in the meantime. though, something like that will hardly change precision apart from tiny rounding errors. at least i doubt its gonna be anywhere near noticable really
Forza updated the frequency of physics calculations (number of iterations per second).
Double precision in programming terms means you're using doubles instead of floats. Those are floating point number representations using respectively 64 and 32 bits. Not sure this is what they meant, but sure sounds like.
In any way, two separate things (frequency of calculations vs precision of a single calculation).
Every time you look at inside/mid/outside tyre temps in iracing in motec, they are totally SCREWED! Rather than fix it, they double down with update after update when the philosophy of the tyre model is wrong to begin with. We have seen novels written by iracing stating that this is how tyres behave even though every man and his dog knows the tyres are flawed but they keep doubling down.
I'm so fed up of driving in iRacing. The tyres are atrocious. GT3's are the worst. To me, GTE's are the best, then I guess the GT4's. Porsche Cup is better but it's like all of their tyres are skinny bicycle tyres, just barely touching the surface. I can't stand it anymore lol
Appreciate the video Nils, this was the video breaking down some of the physics elements that I was hoping someone would make.
7:58 somewhat sounds like what I feel in LMU, and to some extent rF2.
LMU’s initial bite has improved over rF2, but the cars feel more snappy and finicky like what I constantly hear about iRacing.
I noticed all cars in LMU also use the lowest tyre pressure too, 135-140 kpa which is around 19.5-20.5 psi or 1.35-1.40 bar and that sounds very low even for cold pressure.
Actually for cold tyre pressures, depending on the temperature and track, those values are pretty spot on.
@@lorenzorubino2195 I did some further googling and indeed 140kpa is the minimum tyre pressure, at least according to one of the many WEC files I found.
I then found one that specifies LMP2 tyres for the 2023 6h of Spa, there it mentions that the minimum stabilized and static relative pressures should be around 1.90-1.95 bar. Honestly, I don't quite understand what stabilized and static really means but 1.9 bar cold pressure is definitely too high.
I decided to run a few laps in Spa at 22 air temp, 35 track, on mediums. After 7 laps of just driving how I would in a race (2:07-2:08 lap times) the tyres never went over 170kpa and their temps floated around 80-83 degrees, a bit under ideal temps but not far.
Unless I got something wrong with these weird stabilized and static terms, these tyre pressures are way too low for driving 7 laps of race pace on a warm track.
Genuinely don't know what to think, I could never get to grips with LMU and I'm just trying to understand why.
@ Good research, I am a GT3 engineer in GTWC, not WEC, but I can tell you this: minimum tyre pressure cold is usually a suggestion, let’s say that teams have always a way to go lower. The 1.90 bar stabilised means that after one run, when the car comes back to the pit stops, with the hot tyres, the pressure reset should be 1.90 (basically bleeding out air until the tyre reaches that pressure). I don’t know about the Goodyear tyres, just to give some info: with Pirelli (with the tyre heater) we go as low as 1.10 bar (even lower) when it’s hot, without the tyre heater, 1.30 bar (obviously depending on track temp, tyre temp, run type etc..).
@@lorenzorubino2195 So do you try to run the tyre pressures as low as possible in real life too? Or does it depend on the conditions
@@utkarshchaurasia2233 Well, irl there isn't really a low as possible. When there is a pressure min like in WEC, there are always ways to go lower (by naturally heating up the tyre) so at the end the pressures are always being adjusted based on conditions. In game, if the min is 1.40, and the tyre is modeled correctly, in hot weather one should have tyre overheating issues. But to be honest, it is very rare that a commercial tyre model gets the temp and pressure correct, it is a level of detail that nobody cares about, and that it complicates things quite a lot.
Every low downforce car on slicks feels like shit to drive on the limit because of the "flash heat" issue, so lets update the tires on the high downforce cars whose high downforce masks this issue. Maybe we'll get tires that behave like IRL this time next year
Good break down, I appreciate your assessment.
Dave K hired the tire guru that developed tires at rf2. Very surprised as Dave K has been on Alice mission to make his “own” tire.
it shows character trying to get help.
but maybe iracing set off to try do it too complicated, as 16 years after launch the tire does not feel or behave correct. would be great if they'd eventually succeed, but it would almost be a black swan event by now
@@SimracingPopometerI'm so fed up of driving on their version of tyres in most cars. The game has been out for over a decade, how the hell is this still a problem. GT3's are the most frustrating thing to drive.
@@SimracingPopometer iRacing was released in 2008.
@@slowanddeliberate6893 corrected to 16 years ;)
The issue of outside of the tire cooling down slower is pretty simple really. Heat transfers better into tarmac than air. The outside gets hot during cornering like the other parts of the tire, but on the straights, it's not touching the tarmac and thus cools down slower. And when traveling on a straight line, there is of course not enough load to create more heat than is getting dissipated, so the entire tire gets cooler, at different rates depending what part is touching the tarmac and what is not.
iRacing has some major issues with tires that are heat related though, but the above part is just thermal dynamics.
ruclips.net/video/BspEp2dVsZk/видео.html
in short, no.
while yes tarmac better transfers heat, the tire's contact patch is only in contact with the ground for not even 1/10th of the time it is exposed to air.
plus, somehow various commenters forgot that the tire is producing heat as well as it is going straight and being flexed whenever the specific part of the tire becomes the contact patch for a split second.
you can see this on pretty much any infrared recording of tires. the inside on cambered tires will remain warmer than outside.
plus, iracing somehow is the only game that shows this phenomenon. no other sim has that pattern
@@SimracingPopometer Comparing to that video the issue seen in your iRacing telemetry is more that the outside of the tire gets so hot in the first place. If you look at the video you linked after the carousel, you can see that inside gets the hottest but it also cools down the quickest as that part is touching the tarmac the most.
As far as the straightline part, yes tire produces heat while going straight but not enough to prevent it from cooling down on the straights after cornering. So even if the straight was long enough, the tire would of course never cool down to the same temp as the road is.
I think iRacing has some major issues with how heat affects the tires though, pretty much always has. And the fact that more rubbered track means less grip has always been weird to me too. I really hope this new update improves things, we've waited that for a long time 😄
@@IlkkaHaapala you mean thermodynamics?
@@eatisonline Yes, yes I do 😅
@@IlkkaHaapala because this is a thing in Nascar tires. The road tires still have probably this behaviour from that. A couple of real Nascar engineer confirmed this in the old forum (they did not work for iRacing).
Anyway, it is a PLAUSIBLE thing to happen. It may differ from real life in the amount and frequency it happens; but it does.
Maybe you did not see tire temp changes outside and inside on Oultons straight since real-time telemetry is unavailable to third party apps. I have read they do this to avoid cheating.
this isnt real time from shared memory, but from iracing's telemetry file ".ibt" - so this is correct.
the track is oran park btw.
1:56 where did you get this information from? I expect that the M2 gets a rookie series, just like the MX5 cup. it would be completely useless otherwise.
think that might've been my first time skim reading, but i can't find that part anymore.
"This car will be used at times in our series and will be selectable for you all in AI and Hosted iRacing."
maybe the "at times" part.
so probably wrong here on my part
@@SimracingPopometer "we are thrilled to offer the BMW M2 CSR as the latest entry into our racing ladder at the Rookie level. "
@@MakikouZ Yup, this is covered as well in the iRacing DownShift podcast released yesterday.
cheers!
Cayman GT4 does that weird 'hot outside' thing too
i would think all gt4s
@@marcinpohl3264 hmmmm..what exactly do you mean with the hot outside thing? Outside tyres getting too hot?
@@SimracingPopometer It seems to be as following: Due to the high negative camber, of course on straights the outside of the tire don’t have contact to the asphalt. But the tire cools down exactly through the asphalt, that’s why the outside doesn’t = no conductive cooling -> air is an insulator.
That’s physics but I do understand that it looks weird on the first look.
Therefore that’s not an argument to underline the stiff carcass construction. This disappears at higher downforce cars but I’m not sure if you can see this also on other low down force cars with high negative cambers.
Has to be verified.
They all do, the ir18 is their best car/tire/feel combo in the sim, imo
It's unclear from their update whether the lmp2/GTP tire update is the same as the v10 tire. If it is it's a little odd that they didn't just call it v10 in the first part of the update.
they should have telemetry viewer built in...
just to say 499p won't be in imsa... it is only in a prototype series and 6h endurance events, according to this article.
No it doesn't. It doesn't specify either way.
sad if so
@SimracingPopometer from the article "The 499P will join our now 16-car-deep Endurance Series, as well as the BMW M Hybrid V8, Cadillac V-Series.R GTP, Acura ARX-06 GTP, and Porsche 963 GTP in a reconfigured/reconstituted Prototype Challenge Series. "
aka replaces lmp2 fixed in a gtp fixed series and also in the 6h series
the issue is iracing imsa is licensed so they can't use 499p
@@1xRacer I get it, but this license shit is annoying for simracing as fuck
@SimracingPopometer 100% agree
I'd suggest including a link to the update site in you vid description for easy access by those that want to find it without needing to search.
its there :D
2:04 they mentioned that the bmw will be a rookie series car
misread, cheers!
How do you manage to check tyre temps in real time? 😮 All channels I found in simhub updates temps and wear only after pit stop, same like in Black box.
this is from the telemetry file, not shared memory, and only available after the lap. check and download popometer and have a look. even though temps are part of the advanced channels with the sub
God i hope they fix the fps drops. Going from 90fps in vr to 50 in one corner has costed me multiple races. Ive messed with the settings and have everything so low its ridiculous but nothing seems to have any effect
Is the tire just getting insulated by the air between it and the track stopping the brake temp from dispersing throughout the tire and then to the track? So rather than the tyre getting cool because it has no friction it gets hot because it cannot transfer the heat to anywhere? I have no idea about how this works in sims I'm just curious.
I saw on iracing forums it's as you say - both tarmac and air have lower temperature than tyre surface but tarmac has higher thermal conductivity so inside and middle of a tyre cool down quicker thanks to touching tarmac and transferring heat to it.
I don't know if the temperature differences should be like they are currently or maybe lower or higher but that's their explanation to outside not losing as much temperature on straights as inside and middle of a tyre.
why is it only on some cars then, and why is it the only sim that does that. and why does a tire on the road not already produce temperature from being flexed while driving? we're talking minimum pressures here.
no matter how they put it, its odd at least!
At the end of the day iracing is the same as every other sim. Some cars feel good. Some don't. Whatever the data tells you, iracing feels just as realistic as any other sim. At least they go forward with development and don't make objectively worse games like kunos. People trash iracing but somehow ACC doesn't get crap but it feels nothing like driving a car in any way.
@@SimracingPopometer My wild guess is that some property is over or undertuned like heat transfer between tyres and ground or heat buildup from friction, especially that most race cars run non zero toe. It also might be that they got those things right but something else is wrong that affects inside vs outside temps that gets more noticable with specific suspension geometry.
@@jackmakackov7077i take it you have real world experience in a GT3 car to draw a comparison then?
Coolest thing would also be if they add oil / fluid spills
I manipulate the car mid corner all the time. Especially anything FWD.
What cars are you referring to?
@@Real28 probably talking about high downforce cars like a gtp
Typical case of "let's make it different because we want to be unique and special" and they just made the worst tyre model (that I've tried at least). Made the switch to PC, and my driving style might not be optimal (I have a very Porsche tailored driving style) but AC, ACC, rFactor 2, Rennsport, even on the AMS2 demo I can drive the same. iRacing? I'm scared of turning the wheel, I have no idea when the tyres are gonna go past that ultra thin grip-no grip treshold. I got a year for the price for one month and it's been like a month since I last touched it. Between the tyres and the price I'd rather hotlap on ACC tbh.
Iracing's tyres, especially for the GT3's are absolutely attrocious.
I race irl and I like the tire feel. Yes they are sensitive but it forces you to focus on the fundamentals of driving which actually improves you in all other games due to the focus on technique
Then be happy with ACC. Others are happy with iRacing. So, no problem here. And, the tires are the same for all drivers.
Rain and night are currently VR FPS killers - would love to see this improved as I actually really enjoy racing in rain
its the same without VR :D
how is vr performance? am tempted to give iracing a shot but will a 5700x3d be enough if i turn shadows off.. am upgrading gpu but am worried about a cpu bottle neck
It's been a very long journey for iRacing with their physical tire model approach where most sim racing titles have been turning knobs on the old emprical models. I hope this update is going to be the breakthrough
Sounds like it only applies to GTP and LMP2
Perhaps iRacing should hire someone from the Pirelli or Michelin performance tire team(s) to help get the tire behavior as realistic as possible.
Good quality Data is all that is needed…… and powerful computers. These guys can do it but previous limitations of physics engine hold it back.
Ummmmm... 100% not Pirelli. I have more faith in Michelin based on the comments from drivers on Michelins in WEC
Hopefully nobody from Pirelli ever goes near iracing.. Michelin have always been the superior brand, would benefit us all a lot more than pirelli
iRacing has literally tried to strong arm people before with legal action for trashing there tire model publicly. Now they admit that they were doing it wrong all along... lol!
I agree. Iracing feels like wood. LMU i find way better
Off course they will improve their tyre model since they hired two persons form 395 studio they were working on tyre model of LMU and Rf2
im not sure that simply mixing people that produce 0 slip tires and people that produce infinite slim tires will magically result in somewhat medium slip tires :P
but of course we remain hopeful
@@SimracingPopometer
I’m sure these guys know something about tires since that had been there life previously and now they have Iracing way of doing it and come to a better place.
no absolutely. but its not that you get great people and they can simply put their knowledge into a working system in a game.
So for the last 20 years it didn't feel correct? /s How about we talk about elephant in the room? They should fix their stupid FOV first.
what's wrong with the fov though?
@@SimracingPopometer It's 'too fast'. The game feels like it's speed up. FOV is broken and the best way to understand is to watch replays and check at the objects in the background. They are unrealistically large. The whole 3D world is f**** up. Try other games, like ACC, and everything will be much more real (like in the real world). This then affects handling too.
@@SimracingPopometerfor 4 yrs on iRacing I was never happy with the FOV. It always felt odd, as if I was sitting on the headrest in the car. The ONLY thing that finally fixed it for me was to follow Sense of Speed's long 45 minute video of FOV. Now I finally feel like I'm sitting IN the car.
@@S2GUnit so what did you change in the end?
@@SimracingPopometer it's a long video but I skipped through it until he posted #'s on the screen indicating a change being made. You'd have to watch the video but you'll end up changing WorldNearPlane distance & the "point size" which I think has to do with how the pixels are drawn.
Honestly, it's a much watch. Driving the Lamborghini, I can finally see more out of the windshield because the larger monitor isn't such a big issue anymore.
Whatever iRacing setup in the ini's files for FOV is all wrong like many other things I'm noticing with iRacing (especially the tyres)
iracing and it's tires are religion among the users, they are used to it to the point they think it's correct. And there is nothing bad about it, I tougth that partly because user base is so used to it it's hard to change. I usually hope only good things to everybody, but Iam not so keen to see iracing developing theyr weakness. if they really get it rigth they migth get users from other platforms that allready have few users. That way i could force us that don't wan't to pay ridicilous amounts to join iracing if we want to race online. What do I know physics update migth also deter old hard core iracing users..
10 years ago "SIM ISN'T HELPFUL CUZ DEBRIS APPEARS IN REAL LIFE!"
now: hehehehe
Not until their new engine is out
it will be one big argument less against iracing
We don't know what it should feel like we're just a bunch of idiots.
16:35 🤔
old footage from winning lamborghini esports ;)
iRacing V10 tyre model seems not simulate deformable mesh of the tyre. If has this they will show picture in development blog. Maybe just some improvement on contact patch brushes model. Not high level simulation as Le Mans Ultimate and rFactor2.
true the tyre model is shit compare rfactor or lmu😂 also ffb is shit 😂
I still can't quite comprehend where rf2/LMU earned the praise of tyre model and/or physics. it's just as borked as any other game.
@SimracingPopometer I do agreed with you that nothing is perfect but it's iracing is worst simulator in terms of tire model and ffb . By the way if you wanted try 60hz you need to open config file and disable simucube 360 api
The correct tire in the digital environment with 0 risks produces incorrect driver behavior. Like too aggressive close driving and driving always on the grip limit mentality. I don't know if by design or by happy accident but iracing sketchy grip limit cut off kinda fixed those common sim problems. It seams like the incorrect tire is producing the most correct racing. I hope they will keep it in mind while fixing the tire.
I'm shocked reading the comments, lol. Reddit ppl are a different animals 😅
My personal conspiracy theory is that iRacing intentionally fudge the tyres to make them less driveable so that eSports aliens can't drift around IRL drivers and drive them away from the platform.
at least you called it a conspiracy yourself :D
I cant take anyone seriously that thinks ACC is anything like driving an actual car. That game is objectively the worst sim on the maket. The game they made twn years prior is much much much more realistic.
Same can be said for iRacing.
this video hasnt mentioned ACC :D
James Baldwin said himself that ACC simulated GT3s better than iRacing and he drives in the Fanatec World Challenge. What weight does your opinion hold compared to his irl experience?
@@TheOfficialRandomGuy
Max and Lando and every other f2 driver is on Iracing and they can KICK James ass in a car.
James is like any other karting sim racer who is quick. Opinions are opinions and not the 12 commandments.
@@Max__apex James is a top tier sim racer, and has beaten Max straight up, so let’s not stretch the goal post.