PDA

View Full Version : Pcars 2 Aerodynamic Model



Zeratall
06-09-2018, 10:10
Been doing characterization testing for all the gt3 cars, and thought it was pretty cool the aero model for Pcars 2 actually shows the stall of a the diffuser.

259696


As you raise the rake (increasing rear ride height with fixed front ride height) you will see that the DF keeps increasing. That is until 85 mm of rear ride height. At that point the DF begins to fall off because the flow is beginning to separate because that angle of attack is too great. Pretty cool stuff.

Just thought I'd share!

Bealdor
06-09-2018, 10:56
Coming soon to ACC... :rolleyes:

cluck
06-09-2018, 11:10
To quote the wookie


Totally arcade:D

And yet some people still persist in suggesting that pCARS2 isn't a sim :rolleyes:

Zeratall
06-09-2018, 11:31
To quote the wookie

:D

And yet some people still persist in suggesting that pCARS2 isn't a sim :rolleyes:

259697

Sankyo
06-09-2018, 11:31
To quote the wookie

:D

And yet some people still persist in suggesting that pCARS2 isn't a sim :rolleyes:
While proclaiming that 'some other sim' is a real hardcore sim because of getting stuff implemented that pCARS2 already has since the start.

Darn, it's difficult restraining myself from jumping into this... :mad:

Bealdor
06-09-2018, 11:40
While proclaiming that 'some other sim' is a real hardcore sim because of getting stuff implemented that pCARS2 already has since the start.

The only gripe I have with PCARS 2 is that all those features weren't communicated enough (if at all).
I mean, who knows that lower air density at high altitude tracks is taken into account by the aero calculations making your aerodynamics less effective?
The only source for this is a post by Casey that's buried deep in a thread about tyre temperatures (http://forum.projectcarsgame.com/showthread.php?55303-Tire-temps-pcars-2&p=1401733&viewfull=1#post1401733)...

Zeratall
06-09-2018, 11:41
The only gripe I have with PCARS 2 is that all those features weren't communicated enough (if at all).
I mean, who knows that lower air density at high altitude tracks is taken into account by the aero calculations making your aerodynamics less effective?
The only source for this is a post by Casey that's buried deep in a thread about tyre temperatures (http://forum.projectcarsgame.com/showthread.php?55303-Tire-temps-pcars-2&p=1401733&viewfull=1#post1401733)...

That's why I've been trying to share all the cool insights I have, I love this game because it completely allows me to nerd out lol.

morpwr
06-09-2018, 11:46
While proclaiming that 'some other sim' is a real hardcore sim because of getting stuff implemented that pCARS2 already has since the start.

Darn, it's difficult restraining myself from jumping into this... :mad:


Whats going to happen if said sim suddenly feels more like pcars now? Does it suck?lol

cluck
06-09-2018, 11:46
I wonder if Casey could be persuaded to sit down for a couple of hours and do a really in-depth interview with somebody, someone who knows the sort of questions to ask. It would be fascinating to get it on record just how much stuff is simulated "under the hood", things we as players have simply "taken for granted" :)

Bealdor
06-09-2018, 11:46
That's why I've been trying to share all the cool insights I have, I love this game because it completely allows me to nerd out lol.

Keep doing that mate. Maybe we can create an overview thread to show all this stuff off then (although admittedly it'll be a bit late, one year after release).

Invincible
06-09-2018, 11:47
The only gripe I have with PCARS 2 is that all those features weren't communicated enough (if at all).
I mean, who knows that lower air density at high altitude tracks is taken into account by the aero calculations making your aerodynamics less effective?
The only source for this is a post by Casey that's buried deep in a thread about tyre temperatures (http://forum.projectcarsgame.com/showthread.php?55303-Tire-temps-pcars-2&p=1401733&viewfull=1#post1401733)...

Yeah. They really should have made some sort of booklet or even an in-game "tour" with all the awesome features and stuff that goes on behind the scenes and isn't directly apparent to the average user.

cluck
06-09-2018, 11:49
Whats going to happen if said sim suddenly feels more like pcars now? Does it suck?lolFunny really because, back when pCARS1 was in development and the world and it's wife was accusing it of being "arcade" because the cars were easy to drive, iRacing introduced their new tyre model which …….. made the cars a lot easier to drive.

morpwr
06-09-2018, 11:49
Keep doing that mate. Maybe we can create an overview thread to show all this stuff off then (although admittedly it'll be a bit late, one year after release).

Its not too late. Look at how many of us are still learning things about it.

cpcdem
06-09-2018, 12:05
Well, every sim out there proclaims that it has a very realistic model for almost everything. But then again, each and every one of them gets at some point an update where the devs say that they have completely rewritten the model for xyz physics aspect which is now extremely realistic. So then what was it before, it wasn't very realistic we say now?

I think all this talk about sims simulating real life with so much extreme precision is a little bit just hype. OK, things get modeled to great detail, but this does not mean they also work very accurate in practice, too. And how can it be with physics calculations taking place only 500 or 1000 times per second? And even for stuff that we do not even really know how exactly they work in real life, like the tires thing?

And if the modelling was very accurate, then we wouldn't have such glaring mismatches with reality, like in PCARS2 the kerbs sucking every RWD car like there's no tomorrow, fuel weight making practically almost zero difference and things like that. I'm mentioning PCARS2 issues because practically this is the only sim I play, but that holds true for every sim out there.

I don't understand why we need to go the route of this x sim simulated this, while y doesn't etc etc, while clearly none is perfect, I mean even real life F1 drivers do not bother running in their multi million dollar professional simulators, because even them are not simulating real car behavior very well. In my opinion, it's enough for us that all sims out there feel very good and are plausible enough, each with its advantages and disadvantages, but are all good enough so we can enjoy them, have a lot of fun and fool ourselves we are driving a real racing car. How I wish all that all that "our sim is better than yours" things (like many other similar "debates" in life) stopped sometime...

Konan
06-09-2018, 12:10
I don't understand why we need to go the route of this x sim simulated this, while y doesn't etc etc,

Good one...has it ever been otherwise?

cpcdem
06-09-2018, 12:18
Good one...has it ever been otherwise?

Probably not I guess :)

alonsomoslie
06-09-2018, 12:18
Sorry for interrupting you, because Im a green here, but you are true about talking and chatting with each other, learning things together is the best way of improving a game. Many game makers, has just a little developper team, including programmers, engineers, designers, etc.. A friend of mine, who lives in England, said, that Codemasters only has a team of 40 attendants, and they were the main developpers of the oldern days Grid Series. But Project CARS is a game, that has (one of the biggest) a forum, which can be everybody's home. Many users, who play AC or iracing, they just surf the websites to find setups for their car, and them go with it, without wanting to know whats behind it. But it is the majority of users. Being a developper is a hard work, you do a brilliant tool and the others just download it, and follow the instructions you made, and enjoy it.
Thank YOU all the hard work to create this game, I love it:)
Alonso:)

Zeratall
06-09-2018, 12:20
I think the fact that you can read a book on vehicle dynamics and then apply the same concepts to the game speaks for itself. That is the whole reason I started PC2Tuner, I originally only made it for myself because I love reading about things and then seeing if I can measure/optimize them. The level of detail in this game is insane, not only on the big things like suspension modeling, tire modeling, etc but even small details that set it apart from other sims, look at the tuning page for example, your presented with actual units, not some BS slider that says "soft" on one end and "hard" on the other like so many sims do today.

The sim community as a whole has a huge disconnect. People crying simcade without any evidence other than subjective feeling. My stance on the matter has always been, If something is a simulation of something physical than it should be measurable. And if something is not realistic than it should differ from real world by a fair amount.

Regarding Project Cars 2 being realistic or not, I think the data that we've (RST) collected speaks for itself.

cpcdem
06-09-2018, 12:34
I did not say PCARS is not realistic, I just said it is not realistic to the extend we like to believe it is. Like any other sim out there is not as realistic as their devs believe it is. Also PCARS1 was supposed to be very realistic, yet it was always faster using absolutely zero camber in every car in it. In my time with RRE, it was hard to go unnoticed that there's even no tire pressure adjustment allowed. In AC I remember I had been adjusting downforce and that was having a big effect on the cars balance even at speeds around 50km/h. I am not trying to bash PCARS, just making a general comment. But I would love to see a real life's GT3 car engineer face if somebody told him "Well, in my experience with x sim, in order to fix this understeer in y corner, you need to change this setup item to z value and it will be fine, trust me"...

Edit: And yes, of course comments like "PCARS is simcade" that we see in other forums are absolutely cr*p. But I've at least been in this forum for some months so I can be relatively confident that people will not think I'm trolling around by doubting some things, so I am writing it here. In other forums I wouldn't dare saying such a thing, at least not yet.

cpcdem
06-09-2018, 12:36
Sorry for interrupting you, because Im a green here, but you are true about talking and chatting with each other, learning things together is the best way of improving a game. Many game makers, has just a little developper team, including programmers, engineers, designers, etc.. A friend of mine, who lives in England, said, that Codemasters only has a team of 40 attendants, and they were the main developpers of the oldern days Grid Series. But Project CARS is a game, that has (one of the biggest) a forum, which can be everybody's home. Many users, who play AC or iracing, they just surf the websites to find setups for their car, and them go with it, without wanting to know whats behind it. But it is the majority of users. Being a developper is a hard work, you do a brilliant tool and the others just download it, and follow the instructions you made, and enjoy it.
Thank YOU all the hard work to create this game, I love it:)
Alonso:)

Well, saying that PCARS2 has by far a lot more features than any other sim out there is not an opinion, it's an absolutely obvious fact! :)

alonsomoslie
06-09-2018, 12:45
Once I played with AC, at a shopping mall, there was a motion sim, and you could try It out for 10euros for 5 mins. I got stomach ache after it. I really dislike the phisics of AC, Its annoying. But when I play Pcars2 at home, the car stays on the asphalt shoothly, and not jump up after every bump. Maybe it was just the motion sim setup, to show people, what it would feel like if you sit in a Ferrari 458 Italia.

Bealdor
06-09-2018, 13:16
I did not say PCARS is not realistic, I just said it is not realistic to the extend we like to believe it is. Like any other sim out there is not as realistic as their devs believe it is. Also PCARS1 was supposed to be very realistic, yet it was always faster using absolutely zero camber in every car in it. In my time with RRE, it was hard to go unnoticed that there's even no tire pressure adjustment allowed. In AC I remember I had been adjusting downforce and that was having a big effect on the cars balance even at speeds around 50km/h. I am not trying to bash PCARS, just making a general comment. But I would love to see a real life's GT3 car engineer face if somebody told him "Well, in my experience with x sim, in order to fix this understeer in y corner, you need to change this setup item to z value and it will be fine, trust me"...

I'm with you on that. All sims still have a long way to go but I also think it is crucial to show of the progress every one of them is doing.
IMO you have to communicate all this stuff, (mainly) not to pat yourself on the back for what you achieved but to improve them sim genre in general.
Showing off your features to the public puts pressure on the competition (and yourself if you want to release a sequel) leading to better games in the future.

This concept works quite well wrt obviously visible features like dynamic weather.
It's no coincidence that Kunos are pushing their dynamic track and weather stuff now and they probably wouldn't if there was no PCARS 2. Who knows if they'd even added it.

On the other hand this concept doesn't work well with those "under the hood" features.
There's a reason why RF2 is still considered the holy grail of (tyre) physics. And it's not because it's the most realistic one or the one with the most features.
It's because it's meeting the sim racing community's expectations perfectly. Now I know that RF2 has evolved significantly over the years (and the sim racing community has too, to a degree) but to me it's not the holy grail because IMO there isn't one. All sims have strengths and weaknesses, just as you said, but I'm always (trying to) prefer the one that tries to move the genre forward with new features.

The other reason why I think those features should be clearly communicated is simply to make the players aware of them.
I lost count on how many times I've seen reports on this forum about features of the game that were considered bugs simply because PCARS 2 was the first one to ever model them.
Be it tyre pressures rising while standing still because the brakes are heating up the air inside or hard and soft slicks not behaving like players are used to (soft = fast + short life and vice versa).
Not to mention the use of FEA for the tyre model (http://forum.projectcarsgame.com/showthread.php?49850-GTR-3&p=1536366&viewfull=1#post1536366) which is a huge thing for a consumer grade sim, that's also future proof.
Too many features/nuances are lost on the average player unfortunately because they don't know about them.

TL;DR: No matter the viewpoint, IMO it's always important to communicate this kind of "nerdy" stuff, no matter which sim you're talking about and from what perspective you're looking at it.


Well, saying that PCARS2 has by far a lot more features than any other sim out there is not an opinion, it's an absolutely obvious fact! :)

Unfortunately it's not obvious enough to many.

cpcdem
06-09-2018, 13:24
Yeah, 100% agreed on that Bealdor. And about the features aspect, if some people don't see it that PCARS2 is the most future rich by far (to date at least) sim, that can only be down to blind prejudice or wanting to believe that's not the case...

Sessionerror
06-09-2018, 14:30
@zeratall to come back to your original post for a second (great test btw!).

I guess the rake displayed is the static rake, right? I wonder at which dynamic rake you observed the maximum downforce. When driving top speed, your rake probably decreases - did you observe maximum downforce at zero dynamic rake (flat bottom of the car) or when there was still a little rake left even at highspeed?

Sankyo
06-09-2018, 14:37
...When driving top speed, your rake probably decreased
Is this what normally happens with 'normal' car set-ups? I imagine it's depending mainly on the ratio of downforce and suspension stiffness at the front and at the back of the car that ultimately determines whether the front is pushed down less or more than the rear of the car?

Sessionerror
06-09-2018, 14:43
Is this what normally happens with 'normal' car set-ups? I imagine it's depending mainly on the ratio of downforce and suspension stiffness at the front and at the back of the car that ultimately determines whether the front is pushed down less or more than the rear of the car?

Yes, downforce and spring stiffness will affect whether the front or the back will be pushed down further. But in most race cars, you'll run softer springs at the back of the car and also most cars do have much bigger rear spoilers - both results in a lower back. At least this is what telemetry tells me when driving faster and I always set up my car that I don't end up in negative rake. But I wonder whether a positive dynamic rake or a rake of zero is what I should be aiming for at high speed?

Invincible
06-09-2018, 14:44
Is this what normally happens with 'normal' car set-ups? I imagine it's depending mainly on the ratio of downforce and suspension stiffness at the front and at the back of the car that ultimately determines whether the front is pushed down less or more than the rear of the car?

Which would then also be affected by the engine layout of the car, because an FR car will most like have stiffer springs on the front than a RR car.

repsol23
06-09-2018, 15:13
@zeratall to come back to your original post for a second (great test btw!).

I guess the rake displayed is the static rake, right? I wonder at which dynamic rake you observed the maximum downforce. When driving top speed, your rake probably decreases - did you observe maximum downforce at zero dynamic rake (flat bottom of the car) or when there was still a little rake left even at highspeed?

I am not sure how he got his observations, but as a user of the tuner app I would assume it was at the "flat bottom" you referred to. I say this because part of the analysis of the app shows the ride height of rear/front vs time. I use this to adjust my downforce and heights such that I have an absolutely level car flying down the straights. It has many features which allow you to squeeze out every tenth, assuming you can push the car to the actual limits without driving off the track!!

hkraft300
06-09-2018, 15:17
This is from pcars1 (!) and I think very relevant:

http://forum.projectcarsgame.com/showthread.php?43295-The-Physics-of-Renault-Sport-Car-Pack&highlight=Physics

Sims are doing really cool, nerdy, accurate and incredible things. Yes they have oddities and things they need to polish but with each iteration it's a big jump ahead.


Which would then also be affected by the engine layout of the car, because an FR car will most like have stiffer springs on the front than a RR car.

It's mainly got to do with CoP, spring bias, static rake. Spring bias isn't often dictated by weight distribution either. You'll almost always have stiffer front springs even in a 911 I'm pretty sure.

Invincible
06-09-2018, 15:27
It's mainly got to do with CoP, spring bias, static rake. Spring bias isn't often dictated by weight distribution either. You'll almost always have stiffer front springs even in a 911 I'm pretty sure.

I know. I wasn't saying that a RR car has stiffer springs on the rear than on the front, but that it will most likely have stiffer springs on the rear than a FR or FF car.

Zeratall
06-09-2018, 16:44
@zeratall to come back to your original post for a second (great test btw!).

I guess the rake displayed is the static rake, right? I wonder at which dynamic rake you observed the maximum downforce. When driving top speed, your rake probably decreases - did you observe maximum downforce at zero dynamic rake (flat bottom of the car) or when there was still a little rake left even at highspeed?

So the line you see in gray titled rake is essentially the dynamic rake. The x axis is the static rake via varying the rear ride height. The whole point of this test was to find the ideal rake that is to say, the most under body downforce.

My test method: I found the wing ratio that gave me balance (remeber 0 front wing and 0 rear wing are not equal). I then kept the front ride height constant at 65, and drove a lap at each rear ride height. I then took a downforce measurement at the same spot for every track, with whst my ride height and downforce for each axle (using pc2tuner). I then calculate the rake using the dynamic ride heights from the tool. Now that I know what the angle is for this car. For any track and any wing configuration I know my (target dynamic rake). Which is awesome info to have.

The reason this is so important to know is becuase of something we in the aerospace world called "aerodynamic effecient" also know as Cl/CD, which is your lift coefficient over your drag coefficient. Typically under body downforce is a lot more aerodynamic effecient than wings. Which means you get more downforce for less drag which is everything for a race car. If we have two cars that have the same total downforce but one is using primarily wing while the other is primarily using under body, the car with the under body downforce will be faster on the straights.

Tar Heel
06-09-2018, 17:17
I'm just here to say I would love to see a conversation between Zeratall and Casey on the physics of PC2

ShneebnaMRR108
06-09-2018, 17:30
Here is another example (like above post) taken from RST Notes: Credit to Zeratall

If the rear ride height is too low, or too high, we will generate less downforce than if the rake is at its ideal angle. When you need more downforce, you can find at what rake you get the highest peak of downforce.

If you reference it to numbers, it might look something like this.
Say you experiment with four different rake ride heights and you arrive at these numbers.
40/50 = 5000 lbs. of downforce
40/51 = 5200 lbs. of downforce
40/53 = 5300 lbs. of downforce
40/54 = 5200 lbs. of downforce

If you are trying to generate more downforce, (in the above example) the rake and should be 40/53. (5300 lbs. peak)
If the rake is too low, our angle of attack or rake is too low, and you generate low downforce.
If the rake is too high, you stall the diffuser, which also results in less downforce.
Essentially, you are trying to find the “peak” downforce value.

repsol23
06-09-2018, 18:28
Wow, feels like I am back in school. Lol. Thanks Professor Z.

MuddyPaws73
06-09-2018, 18:42
How do you correct for the change in down force from the wings as the angle of their attack changes with the rake of the vehicle?

Zeratall
06-09-2018, 18:59
How do you correct for the change in down force from the wings as the angle of their attack changes with the rake of the vehicle?

So now that I know what the ideal rake is for each car let's say for a car under test that is 7 degrees (fake number). Then when I go to add wing or need to change the balance of the wing, I change the ride height with the new wing changes such that my dynamic rake angle with the new configuration is the same as what I measured as my ideal from before .

alonsomoslie
06-09-2018, 19:22
Sorry to interrupt Zeratall, but what car did you used at the post below, which has 65mm ride height on the front? You also said in the first post, that the 85mm is an average good ride height for GT3, for maximum under body downforce. An hour ago I tested at Silverstone with the AMG-GT3, setted rear ride height to 85 and frnt to its lowest, 77mm, and used medium level downforce. My car became snappier in fast corners (gained 0.3 secs.) but at accelerating out from slow corners, made my car drifting, lost 0.5 secs. What could be wrong with it, can you help me? Maybe its a wrong ARB setup.
Alonso:)

MJP
06-09-2018, 19:26
Coming soon to ACC... :rolleyes:

Not quite sure how to take this statement, are you implying that ACC doesn't or won't model this sort of stuff? Seems rather unlikely to me that they'd remove stuff their previous (current) sim and other sims already do.

Zeratall
06-09-2018, 19:43
Sorry to interrupt Zeratall, but what car did you used at the post below, which has 65mm ride height on the front? You also said in the first post, that the 85mm is an average good ride height for GT3, for maximum under body downforce. An hour ago I tested at Silverstone with the AMG-GT3, setted rear ride height to 85 and frnt to its lowest, 77mm, and used medium level downforce. My car became snappier in fast corners (gained 0.3 secs.) but at accelerating out from slow corners, made my car drifting, lost 0.5 secs. What could be wrong with it, can you help me? Maybe its a wrong ARB setup.
Alonso:)

That data was for the Lambo, I'll have to do some testing on the Amg and get back to you.

alonsomoslie
06-09-2018, 19:50
Thanks. Im also interested in Huracan GT3, Im an Italian:) I know that most of Italian cars hasn't got power, like the 488 GT3 (but I like its design very much), thats why I drive mostly german masterpieces.

Bealdor
06-09-2018, 20:01
Not quite sure how to take this statement, are you implying that ACC doesn't or won't model this sort of stuff? Seems rather unlikely to me that they'd remove stuff their previous (current) sim and other sims already do.

No, I was just taking a lighthearted jab at them (or maybe more at their community) because their first physics dev blog made it sound like their new aero model and improved brake and tire heating stuff was something revolutionary while PCARS 2 already does all of that since release, but according to many it's not "sim" enough.

hkraft300
07-09-2018, 00:50
Stop screwing around with the GT3 cars and take the Indy for a spin if you want to see where pc2 Aero is at.

cpcdem
07-09-2018, 00:57
I did a few quick tests, took the Lambo GT3 in Daytona 3-oval and checked top speed:

- With default setup, so ride heights 67mm/79mm, the top speed was 280km/h
- Put maximum front ride height (85mm) and minimum rear (65mm), top speed was 278km/h
- Reversed, put minimum front ride height (65mm) and maximum rear (95mm), top speed was 280km/h

Maximum difference between the one extreme, other extreme, default was 2km/h, which is around 1mph.

Zeratall
07-09-2018, 01:10
I did a few quick tests, took the Lambo GT3 in Daytona 3-oval and checked top speed:

- With default setup, so ride heights 67mm/79mm, the top speed was 280km/h
- Put maximum front ride height (85mm) and minimum rear (65mm), top speed was 278km/h
- Reversed, put minimum front ride height (65mm) and maximum rear (95mm), top speed was 280km/h

Maximum difference between the one extreme, other extreme, default was 2km/h, which is around 1mph.

Makes sense your going from an extreme angle to the next, probally starving the diffuser in the first configuration, and stalling it in the second. That's the whole point of finding ideal.

cpcdem
07-09-2018, 01:16
Makes sense your going from an extreme angle to the next, probally starving the diffuser in the first configuration, and stalling it in the second. That's the whole point of finding ideal.

In all three cases, the top speed is practically identical...

Zeratall
07-09-2018, 01:24
In all three cases, the top speed is practically identical...

In your test you literally didn't keep anything constant, you changed literally every variable. You need to take a scientific approach, change one thing at a time iterate through that variable and measure the result.

Zeratall
07-09-2018, 01:25
Stop screwing around with the GT3 cars and take the Indy for a spin if you want to see where pc2 Aero is at.

Was one of the first things I did, this is what DRS looks like when you activate it (blue line is the rear wing). 259710

cpcdem
07-09-2018, 01:30
In your test you literally didn't keep anything constant, you changed literally every variable. You need to take a scientific approach, change one thing at a time iterate through that variable and measure the result.

I used default first. Then changed the rake to maximum, without touching anything else, top speed was the same. Then reversed rake, without touching anything else, top speed was again identical.

Zeratall
07-09-2018, 01:44
I used default first. Then changed the rake to maximum, without touching anything else, top speed was the same. Then reversed rake, without touching anything else, top speed was again identical.

You changed the front ride height along with the rear, your not only changing your rake angle, but your also changing the height, which also affects downforce. Read this article, although incredible simple will serve as a good reference to what your actually doing. http://www.formula1-dictionary.net/rake_angle.html

hkraft300
07-09-2018, 01:55
I used default first. Then changed the rake to maximum, without touching anything else, top speed was the same. Then reversed rake, without touching anything else, top speed was again identical.

Where are you testing the top speed? Rake can affect corner exit speed and therefore top speed.
30mm rake can increase drag and downforce to the point the car slows down. You can't say just based on this result that Aero effect is void.
Try: set front + rear wing to minimum, minimum front and minimum rear, do a run, then increase the rear ride height in 5mm increments for each run.
How quickly did you reach that top speed? Does the car have a high enough top speed and acceleration for the straight you're testing on?
If you're on redline at 280 or past the power curve of the car it's another variable affecting your Aero test. Your springs are also compressed and your rake has changed at 280kmh so you're actually at a different rake than what you had set in the tuning menu.
Better to test Aero at, say, 250kph and compare your power output + throttle %. This will indicate drag.
See how many variables there are?
Also I'd recommend a more Aero dependant car than a GT3, or at least set front + rear spring to the same stiffness (using a zeratall or jussi's calculator). Yet another variable.

cpcdem
07-09-2018, 02:22
OK, so I did a few more tests. Now kept front height at default and put rear first to minimum, then to maximum. Then put rear height at default, changed front to minimum, then maximum. In all 4 cases and the other 3 from my previous tests, the top speed is practically identical, 279kmh +/1 kmh.

Top speed was measured at the end of a straight, after being in it's top value for at least a couple seconds. The car was not redlining, it is actually easy by touching other setup items to make it go 290+ kmh.

And I did not say that aero effects are void, just said that as in every home sim, they are way less close to real life than some people hope they are. Do you guys really think that a personal PC can make so much sophisticated analysis of aerodynamics in realtime, that much to reliably detect that each given diffuser is stalled? And do that in addition to calculating a very complex tire model for 4 tires, manage the differential, suspension, AI, graphics and more? And all that in real time? Big organizations/racing teams invest millions of dollars in super computers to run cfd simulations (not in realtime!) and still their results are usually questionable to say the least, as the performance of certain F1 teams clearly shows...

Zeratall
07-09-2018, 03:45
OK, so I did a few more tests. Now kept front height at default and put rear first to minimum, then to maximum. Then put rear height at default, changed front to minimum, then maximum. In all 4 cases and the other 3 from my previous tests, the top speed is practically identical, 279kmh +/1 kmh.

Top speed was measured at the end of a straight, after being in it's top value for at least a couple seconds. The car was not redlining, it is actually easy by touching other setup items to make it go 290+ kmh.

And I did not say that aero effects are void, just said that as in every home sim, they are way less close to real life than some people hope they are. Do you guys really think that a personal PC can make so much sophisticated analysis of aerodynamics in realtime, that much to reliably detect that each given diffuser is stalled? And do that in addition to calculating a very complex tire model for 4 tires, manage the differential, suspension, AI, graphics and more? And all that in real time? Big organizations/racing teams invest millions of dollars in super computers to run cfd simulations (not in realtime!) and still their results are usually questionable to say the least, as the performance of certain F1 teams clearly shows...

So you have a flaw in your thinking your measuring top speed to indicate downforce, when in fact your actually measuring drag. In my original statement I said that their is an ideal rake angle for downforce, its measurable, I did it and have done it for other cars with the same result. They may not actually be reproducing the affects of aerodynamic efficiency in diffuser vs wing drag. Which would indicate your top speed theory.

That being said, your statement about CFD is void, actually since about early 2000's, check out X-Plane, they are not only using Finite Element to calculate coeffecients of drag, they are also taking in vicious affects, and yes in RealTime here's the link if you still don't believe me. https://www.x-plane.com/desktop/how-x-plane-works/

Big corporations like F1 teams are spending a ton of money for super detailed analysis because they are making design decesions based on the data which is why they want a super accurate model, its cheap, quick, and easy to get a down and dirty estimation when you don't consider things like viscous affects, etc. We are not making design decisions, we are literally just enjoying the sim for fun, so they could cheat without even needing to calculate any aerodynamic forces, for all we know they could have an if statement, that wouldn't take much code now would it?

cpcdem
07-09-2018, 04:14
OK, let's leave it here, let's just agree to disagree. I probably should not had made my first post in this thread at all, but sometimes I can't help it...

Zeratall
07-09-2018, 04:52
OK, let's leave it here, let's just agree to disagree. I probably should not had made my first post in this thread at all, but sometimes I can't help it...

I think that's wise.

hkraft300
07-09-2018, 07:39
Was one of the first things I did, this is what DRS looks like when you activate it (blue line is the rear wing). 259710

Is that the Renault 3.5?
Indy doesn't have DRS.

The Renault And Indy are interesting how they behave with rake. Indy is a little less sensitive I think but I could be wrong.

Wing drag on Indy is huge.

Have you analysed the TS040/50?

Zeratall
07-09-2018, 07:46
Is that the Renault 3.5?
Indy doesn't have DRS.

The Renault And Indy are interesting how they behave with rake. Indy is a little less sensitive I think but I could be wrong.

Wing drag on Indy is huge.

Have you analysed the TS040/50?

Actually now that you mention it I do think that was the renault 3.5. It's my favorite formula car in game, makes me sad can never find a lobby for it. I haven't tested those yet, there's so much content in this game I could literally have years of things or cars to look at hahahah I love it!! I really wanna do some testing on some of the lmp cars as I feel like they will be very interesting to analyze.

alonsomoslie
07-09-2018, 11:55
Well I tried LMP cars, and as a rookie, I would say, these are the most complicated racecars, ever built by human engineers. I know that F1 is great, but LMS is the most cuttting edge aero based racecars. Some say, that F1 cars suffer over 6Gs at high speed corners, but I think these Porsches, Toyotas, even Audis can do more that that.

TexasTyme214
07-09-2018, 21:50
Is that the Renault 3.5?
Indy doesn't have DRS.

The Renault And Indy are interesting how they behave with rake. Indy is a little less sensitive I think but I could be wrong.

Wing drag on Indy is huge.

Have you analysed the TS040/50?

I gave the 919 a shot, and I saw similar behaviors. You can't run the car at absolute minimum ride height or you'll choke the diffuser. You also have to ensure the rear doesn't sit lower than the front at speed. Besides that, I haven't gotten far into rake angles. I'll do more of that this weekend. For anyone curious, Monza seems to be a strong test track for this.

alonsomoslie
07-09-2018, 22:13
Hi TexasTyme 214:)
You maybe found my problem of high speed sitting down of my rear.:)
I posted my problem at Car Setup Talking thread. I totally forgotten this effect.
Alonso:)

Urban Chaos 2.0
07-09-2018, 22:43
Coming soon to ACC... :rolleyes:

Hahahaha. In Stefano's and all the AC fanboys' juvenile dreams :highly_amused:

To quote the wookie

:D

And yet some people still persist in suggesting that pCARS2 isn't a sim :rolleyes:

These are the same people who persist in the belief of Santa Clause, the tooth fairy, and the official 9/11 narrative. They also seem to think FFB = physics. Go figure...

Zeratall
07-09-2018, 23:10
FFB = physics. Go figure...

Ive see so many people make that confusion all the time, makes me sad......

Zeratall
07-09-2018, 23:12
I gave the 919 a shot, and I saw similar behaviors. You can't run the car at absolute minimum ride height or you'll choke the diffuser. You also have to ensure the rear doesn't sit lower than the front at speed. Besides that, I haven't gotten far into rake angles. I'll do more of that this weekend. For anyone curious, Monza seems to be a strong test track for this.

Not only can you choke the diffuser but if you increase the overall ride height aka move front and rear ride heights up, you actually get slightly less df due to side bleeding.

bmanic
08-09-2018, 05:19
OK, let's leave it here, let's just agree to disagree. I probably should not had made my first post in this thread at all, but sometimes I can't help it...

.. so when your flawed methology and conclusions are pointed out, this is the conclusion, in stead of actually learning from the feedback? I actually learned a lot from your post (well, not yours but the counter feedback).

cpcdem
08-09-2018, 12:55
.. so when your flawed methology and conclusions are pointed out, this is the conclusion, in stead of actually learning from the feedback? I actually learned a lot from your post (well, not yours but the counter feedback).

I don't think anything was flawed and IMO the conclusions are very obvious that several things are way way off, but what's the point on continuing this? I will just be that guy that "complains and moans", almost everybody else will be arguing that the (any) simulation is almost 1:1 to real life, so let it be, it's not worth fighting this. Live believing this and let live :)

hkraft300
08-09-2018, 14:12
I don't think anything was flawed and IMO the conclusions are very obvious that several things are way way off, but what's the point on continuing this? I will just be that guy that "complains and moans", almost everybody else will be arguing that the (any) simulation is almost 1:1 to real life, so let it be, it's not worth fighting this. Live believing this and let live :)

Your methodology was completely off. Demonstrates your complete lack of understanding, not that the simulation has problems.

Maybe stick to setting lap records and leave the geek stuff to the geeks.

Konan
08-09-2018, 14:17
ok ok...he already said he was going to leave it at that so let's leave it at that before this turns into a discussion going back and forth with each claiming they're right (like so many well started discussions)

hkraft300
09-09-2018, 00:42
ok ok...he already said he was going to leave it at that so let's leave it at that...

Sure. Like a guy walking into a cloud formation discussion mentioning "chemtrails".

hkraft300
13-09-2018, 00:07
...
Extra wings and aero parts are added front and rear on the FXX-K, with the claimed gains being about 40% more downforce over the LaFerrari for as much as 520kg @ 200km/h. That is A LOT of downforce - greater than GT3 levels - and refers to the highest downforce mode of all the active flaps when at maximum braking effort. The official claims also note a number of 675kg downforce at maximum speed, which means more than 50% of the maximum downforce is removed when the active aero elements are in their low-drag positions. Our active aero model copies this by looking at brake pedal force, speed, and cornering effort to blend between those downforce targets with the bulk of the maximum downforce load going to the rear - roughly a 10% rearward shift in aero balance for stability when it detects you need it.



So pcars2 simulates active Aero...

Casey Ringley
13-09-2018, 00:40
Yeah, based on various inputs like speed, braking/steering input, G forces, etc. It is hooked into the visual model then, too, so the wing position is synced up with where in the setting range the active aero model puts things.

Casey Ringley
13-09-2018, 00:50
Stop screwing around with the GT3 cars and take the Indy for a spin if you want to see where pc2 Aero is at.

The power of really (really!) good reference material. Was very pleased that we managed to match our model within 2% to the real thing for downforce, CoP, and drag over the entire aero map (ride heights from 5-75mm front and rear) for the entire wing setting range. Having that correct really helps to emphasize how important getting in sync your aero and suspension setup is on these modern formula cars. There is massive performance to be found by running as low as possible, sometimes to the detriment of low-speed mechanical grip. Very different approach to something like the Lambo GT3 car which, as Zeratall has shown here, likes a certain amount of rake and gets choked of airflow if running too low.

This was a fun thread to read. One new tidbit since I've been doing some code review lately to learn if there's anything we didn't know the model already does: It turns out that our aero model already factors in the draft to things like brake and radiator cooling. It's not a huge effect, but it is there.

hkraft300
13-09-2018, 01:59
I'm studying mechanical engineering. My fluid dynamics lecturer is the head of the auto engineering department. Some of the concepts he's covered; even if they are implemented at a basic level can have massive consequences to the performance of the vehicles in the game. I'm starting to get a real appreciation of the depth and difficulty of simulating Aero effects, and the impact they have on the cars in the game.
Some Aero behaviour can, to the untrained mind, seem bugged because it's so extreme. What they don't realise is the scale of forces involved and how strong they can be at race track speeds. I think we're all old enough to remember Webber back flipping the Merc CLK at le Mans. 30 odd feet in the air.

Zeratall
13-09-2018, 02:53
I'm studying mechanical engineering. My fluid dynamics lecturer is the head of the auto engineering department. Some of the concepts he's covered; even if they are implemented at a basic level can have massive consequences to the performance of the vehicles in the game. I'm starting to get a real appreciation of the depth and difficulty of simulating Aero effects, and the impact they have on the cars in the game.
Some Aero behaviour can, to the untrained mind, seem bugged because it's so extreme. What they don't realise is the scale of forces involved and how strong they can be at race track speeds. I think we're all old enough to remember Webber back flipping the Merc CLK at le Mans. 30 odd feet in the air.

Hkraft300 I think your my long lost sim racing brother or something lol.

hkraft300
13-09-2018, 04:17
I have had many engineering nerdgasms playing pc2.

Bealdor
13-09-2018, 05:42
I have had many engineering nerdgasms playing pc2.

Yeah, me too. :cool:
Unfortunately most of this stuff is lost on too many people.

Konan
13-09-2018, 05:44
Yeah, me too. :cool:
Unfortunately most of this stuff is lost on too many people.

Like me...i'd like to get inTensor with it though...:rolleyes:

hkraft300
13-09-2018, 07:31
Like me...i'd like to get inTensor with it though...:rolleyes:

You should. It's easy for me because I'm getting it practically spoon fed to me in class, and I have things like project cars, the discussions here and motor racing to relate it to. I see some guys in my class struggling because when they leave the class they don't have anything to connect the education to in practical/real world terms.
You can mostly teach yourself. You can read about springs and watch engineering explained as he goes through the math. You have guys like Jussi and Casey at your disposal.
I'm yet to pick the brains of my fluids lecturer. He's also the man behind the success of my school in Formula SAE. He asked in the first class who would like to join, and I put my hand up enthusiastically. He looked me dead in the eye and said "well you'll have to become a better engineer first". Ego destroyed.

VelvetTorpedo
13-09-2018, 11:32
@hkraft300 you should really join our pc2tuner discord then, if you love geeking out about this stuff. The guys there are obsessed haha

https://discord.gg/njPdbEu

hkraft300
13-09-2018, 14:17
@hkraft300 you should really join our pc2tuner discord then, if you love geeking out about this stuff. The guys there are obsessed haha

https://discord.gg/njPdbEu

Thanks mate. Learning my way round the app now. I’ll catch up with the conversation soon enough.

Jussi Karjalainen
13-09-2018, 16:59
So pcars2 simulates active Aero...Eh, old news, active aero was already in pCARS 1. =)

And it's not just front end and back end either, we also do lateral active aero where applicable, though that's pretty much limited to the Huayra BC, I think. Not many cars doing such things.

Jussi Karjalainen
13-09-2018, 18:27
To be fair cpcdem's example was a valid point, and the answers aren't as obvious as you might think. I did a bit of testing on vintage Mulsanne as well as some analysis on a special test track, and here are some of the reasons why what he saw happens:

1. The Daytona oval isn't exactly the best place for these kinds of tests due to the corners, banking, etc. Ideally if you really wanted to accurately see what the aero is doing you'd use a pure straight, and one that's as flat as possible. Vintage Mulsanne isn't exactly the flattest, but averaging across a few runs with different settings I lost about 5 km/h on max negative rake and about 3 km/h on max positive rake compared to running stock (consistency can at times be a bit tough, gotta be careful that the wind isn't changing directions etc.).

2. The maximum negative rake (max height front, min height rear) doesn't just purely cause more drag, it also reduces downforce. Reducing downforce tends to reduce drag in itself, so if your DF goes down a lot with adjustments like this the actual drag might not change quite as much. On our test track and reading the exact corner forces I saw the drag increase by about 6% when running max negative rake vs. the default settings, but downforce wen't down by over 17%, losing about 100 kg of total DF. Losing this much DF would have a big effect on drag. And less downforce also means less rolling resistance, since the tyres are sitting light on the road. It just happens that in this car (and probably a fair few GT3 cars) running max negative rake doesn't add a humongous amount of extra total drag, but it does cost you a huge amount of DF. The total aero efficiency in the tests went from something like 1.85:1 L/D to about 1.44:1 L/D with the rake change.

3. Don't underestimate how significant the loss of a few km/h is at these speeds. After I got the ~5 km/h slower speed on Mulsanne with max negative rake, I put the default setup back and started reducing power (via air restrictor) until I got the same loss of ~5 km/h, and ended up going down by over 20 hp. So a speed difference of just a few km/h isn't an insignificant thing when you're going that fast.

Zeratall
13-09-2018, 20:18
To be fair cpcdem's example was a valid point, and the answers aren't as obvious as you might think. I did a bit of testing on vintage Mulsanne as well as some analysis on a special test track, and here are some of the reasons why what he saw happens:

1. The Daytona oval isn't exactly the best place for these kinds of tests due to the corners, banking, etc. Ideally if you really wanted to accurately see what the aero is doing you'd use a pure straight, and one that's as flat as possible. Vintage Mulsanne isn't exactly the flattest, but averaging across a few runs with different settings I lost about 5 km/h on max negative rake and about 3 km/h on max positive rake compared to running stock (consistency can at times be a bit tough, gotta be careful that the wind isn't changing directions etc.).

2. The maximum negative rake (max height front, min height rear) doesn't just purely cause more drag, it also reduces downforce. Reducing downforce tends to reduce drag in itself, so if your DF goes down a lot with adjustments like this the actual drag might not change quite as much. On our test track and reading the exact corner forces I saw the drag increase by about 6% when running max negative rake vs. the default settings, but downforce wen't down by over 17%, losing about 100 kg of total DF. Losing this much DF would have a big effect on drag. And less downforce also means less rolling resistance, since the tyres are sitting light on the road. It just happens that in this car (and probably a fair few GT3 cars) running max negative rake doesn't add a humongous amount of extra total drag, but it does cost you a huge amount of DF. The total aero efficiency in the tests went from something like 1.85:1 L/D to about 1.44:1 L/D with the rake change.

3. Don't underestimate how significant the loss of a few km/h is at these speeds. After I got the ~5 km/h slower speed on Mulsanne with max negative rake, I put the default setup back and started reducing power (via air restrictor) until I got the same loss of ~5 km/h, and ended up going down by over 20 hp. So a speed difference of just a few km/h isn't an insignificant thing when you're going that fast.

Thats the difference between parasitic drag and induced drag, and why it's not always good to relate downforce directly with drag ( you have to know where you live in the drag bucket) which is what I think people were trying to point out in cpcdem simplistic analysis, as well as the test methodology.

alonsomoslie
14-09-2018, 11:18
This conversations between you two are quite instructive for me, I love reading these high level discussions.
Alonso:)

Markspeed
21-09-2018, 18:39
Some Aero behaviour can, to the untrained mind, seem bugged because it's so extreme. What they don't realise is the scale of forces involved and how strong they can be at race track speeds. I think we're all old enough to remember Webber back flipping the Merc CLK at le Mans. 30 odd feet in the air.

Vividly! That was back when I could watch all 24 hour's of it without a wink of sleep. That was a good race brewing up till the moment Dumbreck went over the armco before Indy. Then the race settled into cruise mode till about 5:30 am when Kayama's TS020 found thunderous god of sunrise speed! But then all was lost when the left rear exploded at about 190 halfway towards Indy. For some strange reason Indy hated speed that year. All the drama during that race happened on the way to Indy. I believe one of the Rook Racing Porsche's got tripped over by a Viper on the way into Indy as well. If memory serves.