I still check in from time to time. I haven't done any racing in a couple of years. The old teams fell apart and since I really only did league racing, there hasn't been much reason to go back. I still have my rig and will probably give things a go once the physics updates come out.
Interesting thread. It also goes to show that tire physics are extremely complicated.
I know Tristan has way more real-world racing experience than most and I am sure he's correct, but it's pretty obvious the reason why the best lap comes after a few laps is tire pressure and temperature. For a real race, a crew chief will set the tire pressure to be less than optimum to allow for expansion when heated. That means that there is an optimum temperature/pressure where the tires will work best and it will take a couple of laps to hit that. Past that point, usually enough tread has worn off, and with the heat build-up, you'll usually never hit that optimum point again with that set of tires. (dependent on setup and other factors)
For real-world qualifying, the tire pressure is often set higher than for the race so the tires come to optimum temperature and pressure right away and they hit that sweet spot for a lap, maybe two, before the tires are done. They'll often reuse those tires for the race as "scrubbed" tires and they'll work just fine as long as the pressure is set properly for a long stint.
All of this is dependent on the setup of the car and the conditions, obviously. In endurance racing, you often hear drivers say that the tires were great all the way through the stint. This means the tires are good quality and the setup on the car was correct for the driving style and conditions. But so many other factors do come into play. I would suspect that the tire diameter has little overall affect on the results.
In LFS, I've only a few times noticed that I'm driving better laps near the end of a stint than the beginning. I suspect it has more to do with car setup and weight (less fuel) than it does the tires "gaining" grip. But that's my opinion.
I know Tristan has way more real-world racing experience than most and I am sure he's correct, but it's pretty obvious the reason why the best lap comes after a few laps is tire pressure and temperature. For a real race, a crew chief will set the tire pressure to be less than optimum to allow for expansion when heated. That means that there is an optimum temperature/pressure where the tires will work best and it will take a couple of laps to hit that. Past that point, usually enough tread has worn off, and with the heat build-up, you'll usually never hit that optimum point again with that set of tires. (dependent on setup and other factors)
For real-world qualifying, the tire pressure is often set higher than for the race so the tires come to optimum temperature and pressure right away and they hit that sweet spot for a lap, maybe two, before the tires are done. They'll often reuse those tires for the race as "scrubbed" tires and they'll work just fine as long as the pressure is set properly for a long stint.
All of this is dependent on the setup of the car and the conditions, obviously. In endurance racing, you often hear drivers say that the tires were great all the way through the stint. This means the tires are good quality and the setup on the car was correct for the driving style and conditions. But so many other factors do come into play. I would suspect that the tire diameter has little overall affect on the results.
In LFS, I've only a few times noticed that I'm driving better laps near the end of a stint than the beginning. I suspect it has more to do with car setup and weight (less fuel) than it does the tires "gaining" grip. But that's my opinion.
Yes, but...
The way the game was configured, wider track is always better so the setting is somewhat pointless. There's also other issues that aren't modeled by the sim like having to accommodate fender flairs and the additional stresses on the suspension system caused by a wider track. Yes, you can marginally change track on a production car, but not enough to have a huge benefit. It was better to just remove that setting than keep it as it really didn't add to the realism or fun. Everybody just bumped track out to the widest setting and left it there. There was no trade-off like in the real world.
Thanks for the link. And just so we're on the same page, all the User Account Control stuff that can be turned off is turned off on my machine. It's still best to put programs like LFS outside of the Program Files directory.
It is strange with regards to other versions of Windows though. That's why I used that word.
Last edited by Hallen, .
Windows 7 is very strange about what processes it allows to write to either of the Program Files directories. It seems to only allow Setup programs of some type to do that. After that, it disallows anything else to write there programatically. iRacing has massive problems with this with regards to its updates and their tech support is clueless about how to fix it.
For most Windows 7 users, it is best if you create your LFS Program directory outside of the Program Files directory. Then, you'll have control and things like the unlock and updates can write to the directory with no trouble and you won't have to "run as Administrator".
Ah, very good. I had put a start position in thinking it was needed for an open layout. Thanks for answering.
Unless said race driver has extensive sim experience -- like Tristan does. I don't trust drive-by opinions by pro racing drivers most of the time, because most of them aren't sim drivers. They give it a try, or maybe use it to learn the layout of a new track, but they don't seriously try to compete in sim racing.
Quite correct.
A person can be enthusiastic and passionate without being a "fanboy". That label is generally used by people who disagree with others on a matter of opinion. It's best left out of the conversation. Just because you think you're a better critical thinker than others doesn't give you the right to label them.
I agree. The lower FOV does make the corners look a heck of a lot more tight though -- which does enhance your sense of speed when making the corner. I find it hard to drive as fast through corners at lower FOV's than at higher -- partially because I can see more of the corner at higher FOV than at lower -- but mostly because the corner looks way more sharp at lower FOV. I think this is another factor in LFS newbies thinking that there isn't enough grip in some of the cars. They use a high FOV like 95+ on a single monitor, and then wonder why they understeer off a corner at a speed that *seems* reasonable to them. Or, they spin while applying power when coming out of a corner. They don't realize how tightly they are turning and the speed they are going and they have no real life experience with a car that powerful.
Anyway, we've probably seen a million posts like this now from people who make a snap judgment based on minimal experience. There's tons of good advice in this thread. Hopefully, other newbies will read it.
When a layout is assigned to an open configuration, the number of laps seem to be controlled by the layout file settings. The laps can't be changed in the race settings screen. Is there a way around this limitation? I don't run a server, so maybe this is something that has been figured out a long time ago.
Yes, but the point is that they have to update cars independently which means they don't have universal tyre physics -- they fudge their calculations to work better with each individual car. When Scawen finishes the physics for the tyres, all cars will be impacted.
Sorry for the off-topic. I'm a bit pedantic about that kind of thing.
I'm looking forward to the new layout possibilities at all the tracks.
And yet the funny part about their physics update is that it only applies to certain tires on certain cars (unless they've expanded the scope since the last time I checked). What that implies is that they are fudging a lot of stuff with the tires to make them work approximately right for certain situations on certain cars. The current iRacing tire model is nothing to really crow about. It's better than LFS in a couple of areas, but it's also lacking in areas where LFS is very good. So, we'll see. But I expect this update to be very good for their NASCAR style cars, but pretty much ignore other cars.
When I used the word 'quality', I was referring to the simulator quality, not the driver quality. Also, I would definitely agree that current sim technology is a benefit to race drivers on many levels. I expect that my 'race craft' would be at a much higher level than most beginners if I were to actually race a real car.
What I was trying to say above is that because a sim has a real car on a real track and drivers post similar times to the real times, is not a good measure of how "realistic" a sim is. Some think it's the end-all-be-all, but I don't. There's many ways to skin a cat, some more elegant than others. It's that elegance that I suspect Scawen is shooting for.
All else remaining equal, if the physics of a sim were perfect, I would expect that an experienced 'alien' sim driver could always post a faster time than a top class race driver in the real car on a real track. But we know the sim will never be perfect, so there will always be exploits. Anyway, that's my thinking on why using real tracks and real cars isn't necessarily a great measure of how "real" a simulator is.
I do agree that you can more easily get to the limit and hold a car there when you are in the real car because of the better sensory input that you get. But the 'aliens' amongst us have ways of doing that in the sim that I personally think defies explanation.
While I agree with most of that, saying "direct comparison with other sims" is basically meaningless from a "realism" standpoint. Other sims do sometimes provide lap times consistent with their real-world counterparts, yet one has to wonder how that was achieved. In the sims I have used, I suspect that effect was achieved via tweaking of different aspects of the car and tire models rather than a physics model that was really representative. I know that's splitting hairs for most folks, but it's important to me.
People accustomed to driving sims should always be able to lay down a faster lap than a car on a real track with a real driver because of infinite practice and 0 consequences (no risk of death or injury). Therefore, making a sim that presents representative times isn't always a good measure of quality. I think Scawen understands this. He's working towards something that does a good job of representing reality. The real car with the real track did highlight the existing flaws which I think did drive the timing of this physics work. That's OK. Better is generally better.
That place at the zoo with them animals with longgnecks
All blow3d up
We do, we do. We sends money to feed them nasty looking mush.
That map is wrong. It forgot the place that does all our tech support.
Back off topic again: I can't let injustice stand.
The US Schludmuler type beer is awful. But we make some very fine brews now that rival anything I've had to drink in Germany so far. That's not disparaging German beers, they're great, but what you find up here in the NW US is awesome now. McMenamin's Terminator Stout, Rouge Brewery, Bridgeport India Pale Ale, and a host of others like Moose Drool, Fat Tire, etc, are very fine beers. Get with the 21st century and stop guzzling the High Life just because their commercials are cool.
Back on topic -- wait, there is no real topic to this thread.
It's most likely both just as in LFS. LFS has the tire loading grip problem and the lack of some suspension dynamics that Scawen is working on now. You can't really bad mouth either sim much for these problems because it is monstrously complex.
The way iRacing behaves with the changes leads me to believe that there is more fudging going on with the suspensions and tires than there is in LFS, but that's just my impression. In other words, you have little hacks in to adjust for known problems and every time you make changes, those hacks have to be modified to try and balance things out again.
I've done a bit of racing in iRacing. I see the same things I see in LFS, just not quite as extreme when it comes to driving technique. I haven't messed with setups much other than to see that the tire pressures work as expected where LFS is backwards on that.
Contrary to BBT, I see understeer being more of a problem in iRacing than LFS. You pretty much have to turn in and continue to crank in large amounts of steering angle to get a car to rotate through a corner. Yes, in some cars the throttle steer works really well, almost too well. If you build a loose setup, you'll have a major, unpredictable handful. On the good side, once you start a slide, the tank-slapper effect seems pretty darned reasonable to me.
Ignorance is bliss. Seriously.
Against what datum are they measuring their "accuracy". I doubt the guy who does sound recordings, etc, knows one way or another. Unless you are a professional who does this kind of stuff regularly, you aren't going to know.
I doubt they are lying, they're just not stating all the facts. Listen, I'm not making this stuff up. That link I put a few posts ago... that's who I work for. I don't sell the scanning stuff, but I do know precision field measurements. If the instrument says plus or minus 2mm distances, which is reasonable, it doesn't mean squat if you vertical is out or if your calibration is off, or you measure at too oblique of an angle or... the list goes on. I can also guarantee they will not move that instrument and do more setups than is absolutely required to get the minimum data to do the job. It's just a fact of economics and time.
I do know a few things about this kind of work. Believe what you want. I could be wrong and they could be the most perfect field surveyors in the world.
Yep, but like I said, you can quote the accuracy of the equipment on a single, perfect measurement, but that certainly does not indicate what you are actually going to get. There's a lot more that goes into it, I can assure you of that.
I'm not trying to dis iRacing tracks. I'm just saying that they will not capture everything, even with a scan.
Scanning is awesome and there is no better way to get a track into a game, given the current technology. I'm all for it. I'm just saying it isn't perfect, it has to be done right, and you can easily over-state the accuracy.
Even at 2cm, you are going to get a very good representation of the track. The edge of the pavement will be blended a bit. Curbs will probably need some touching up, etc. But by saying you scanned it simply does not mean it is a perfect representation. It's close, but some fudging will have to be done.
Only if you expect gimmicky and unrealistic shaker effects.
The scanning process is very complex. The machines used are very expensive, usually in the $70k-$80k range. They have the ability to shoot thousands of points per second. Their accuracy will vary with range and angle to the object being measured. The technology is basically time of flight where the amount of time it takes the light to leave the instrument, hit the target, and bounce back to the instrument is measured and then the distance can be calculated. Or, it can be a phase measurement system that basically counts the whole number of waves that it takes to reach the point.
Depending on the process, special targets are placed in the area to be measured. Those targets are shot form at least two different locations. Then the targets can be used to stitch together the measurements from various locations.
An example of the type of instrument that could have been used is here:
http://www.trimble.com/trimblegx.shtml
People can claim that the process will produce measurements within 2cm, and that is reasonable. However, the instruments themselves can measure more close than that, down into the mm range, but because of variances in setups, temperature, length of shot, visibility conditions, etc, getting that level of accuracy across an entire job site is difficult. If the proper care is not taken, you can end up with points that are off by quite a bit. Lets say your stated accuracy is plus or minus 2cm. That means that any given point could be 2cm too high and the point right next to it could be 2cm too low giving a pretty big bump. There are techniques to improve this of course, but it is extrapolation, not actual.
So, even the vaunted iRacing scanned tracks are not going to represent every bump accurately. Any trend, like a slope, or bigger bumps like you would find at Sebring, can be modeled pretty well. But small bumps, or sharper bumps will most likely get lost. So, it's true when they say they are accurate within 2cm, but realistically, they are not really that good. A lot of additional measurements would have to be taken using different instruments to get things really close, and that kind of thing would probably be prohibitively expensive.
I know, more than you wanted to know, right?
I know you most likely won't see this post, but I have to add my thanks too. You've made a great product here and it's given me years of massive entertainment and thrill. I'm excited about the updates and look forward to them, but it doesn't drive my life like it seems to for some of these guys.
To some of you dissatisfied people: you really need to get a grip on life. It's OK to complain and hope for better stuff, but this "I want it now" (Willy Wonka style), attitude is disappointing. Scawen gives you an honest, realistic statement, and you flat blast him. And then you turn around and complain that you don't get any updates. Go figure.
You've never been lied to. Scawen tells it like it is. That's refreshing in this day and age.
I was referring to a spring attached to the pedal in some way to give a bit of resistance. The real sensor is of course the potentiometer that is attached to the pedal and reads how far it has traveled. Pedal travel becomes the measure of how much braking is happening on a commercial brake pedal set used for simulators.
The system here uses something that looks like a metal bar and it measures pressure, not motion. So yeah, you would need some kind of free travel, probably with spring resistance, and then contact with the pressure bar.
A truly accurate system would have some way of moving that pressure bar "back" so that if your brakes get hot or worn out (like in a 24h race), the "soft" part of the pedal throw would get longer and longer. It would take some kind of output from the game in order to do that. This, in my mind, would be the ultimate brake simulator setup.
A properly setup, properly bled brake system should only have about 2-4cm of travel. After that, it's all in how hard you press. The pedal won't move any more from there.
The first bit of travel provides very minimal braking force. The real force gets applied once you hit the point where the pedal isn't moving anymore.
A spring system must use pedal travel to gauge how much braking to apply. It may get harder to go further down on the pedal, but it's still the movement of the pedal that indicates how much braking to do. That's not how a real hydraulic brake system works.
It makes sense, and has been discussed a lot before. No simulation is perfect. It also makes a difference by what he means as being a "handful". If the GT5 cars are unstable at high speed, then what cars are we talking about here? Monstrously powerful 1960's vintage cars that don't have any downforce? Well, yeah, they'd be unstable. Also, not seeing GT5, I can't say much, but a lot of console games over do the gyrations of the car to give the impression of more speed. In other words, everything is bouncing around on the screen and the user thinks that makes things more unstable.
LFS does seem to have issues with how the tires break loose and how they regain grip, sometimes. It is dependent on the situation and the setup used.
You'll find everybody has an opinion on how accurate and realistic a particular sim is. It mostly depends on their experience with games and what they think is important. A lot of people say Forza is more realistic than LFS simply because the graphics look better. It's what they deem important.
lol
Wash! err, no, you're right, that's Alpha.
Not nearly as much as they would on something like, oh I don't know, a rallye car maybe?
But your point is correct. On a race track, especially the LFS tracks, you aren't going to find much in the way of paved surface bumps that would push a shock into the fast bump area. For a track where you are taking a lot of curb and the curb is fairly sharp, then yeah, you'll get into that range.
I would point out a track like Sebring, and just about any of the US street courses definitely would put you into the fast bump range. But that doesn't apply to LFS so the damper settings we have are OK for the majority of it. But it would be nice to have the progressive type dampers.
I think it is pretty critical that sooner, rather than later, LFS incorporates speed sensitivity (progressive rates) into the dampers. Again, I don't have a ton of experience here, but at least for production model cars, I think almost all dampers are going to be speed sensitive. I know the Bilsteins on my car are and every shock graph I have ever seen shows the same type of curve that yours shows. The linear rate is impossible to match up with a progressive rate damper across the normal spectrum.
Of course, it would also be very nice to have adjustable dampers for fast and slow bump too. But again, these have been covered in the suggestions threads.
I found all this out when trying to setup the XRG like my BMW E30 track car (thanks again Bob for the help with that). I got it close, but you can't really nail it because of the dampers. (BTW, with the upcoming changes with setup restrictions, it would really be nice to have the rebound setting back for the XRG.
)
Of course, it would also be very nice to have adjustable dampers for fast and slow bump too. But again, these have been covered in the suggestions threads.
I found all this out when trying to setup the XRG like my BMW E30 track car (thanks again Bob for the help with that). I got it close, but you can't really nail it because of the dampers. (BTW, with the upcoming changes with setup restrictions, it would really be nice to have the rebound setting back for the XRG.
)FGED GREDG RDFGDR GSFDG