Quote from glyphon :

how about the mazda 787b, which is the car that won le mans. it was a NA 4 rotor (which is 2.67L, not 2.4. each rotor is 0.67L ), and produced 700hp (that's 262hp/L)

That's for an engine designed to survive a 24 hour enduro, NOT a firecracker F1 engine. If it was tuned for F-1 levels of endurance...

For those who already have power, all you'll want is more power and your only significant fear is the loss of power. Dominance is power, and anything that threatened the dominance of conventional 4 cycle petrol engines spends a lot of heads rolling. I haven't bothered to check yet, but I'll not be surprised if the ban on rotaries for Lemans was preceded by pleas of other teams to ban the "unfair technology"

"Unfair technology"? What is this? Racing is supposed to the purest form of meritocracy and as far as I'm concerned, it's SUPPOSED to be unfair! Remember the heydays of downforce when people were trying to exploit this unfair advantage to the max? Racing was much more exciting then as the public anticipated technically fascinating and groundbreaking cars to show up every now and then and set new benchmarks to force the other teams out of mediocrity. Not to mention the heroic drivers that tested and this new technology, boldly going where no one has gone before.

Do discerning racing enthusiasts like me want a bunch of mediocre cars with even more mediocre drivers bang and crash their way towards victory? No! I want the best drivers teamed up with the best cars. Contrary to popular belief, unfair advantages are exactly what racing needs to improve the show. I want to see the top teams to absolute dust the mediocre off the track and fight tight battles amongst themselves on the track to prove their superiority, not a just some "random" engine blowups to "add some drama".

The real advantages of rotaries are not just the huge power to mass ratios and specific power output. The greatest advantage is its sheer lack of volume, allowing packaging that others can only dream of. Front or mid engined, it's very easy to get desirable mass distribution and moment of inertia with this engine.

Quote from Scawen :

Well, I've said many, many, many, many times, that I'd do this :

Fully Compatible Patches <- DONE
Multiplayer Incompatible Patches <- NEARLY DONE
Physically Incompatible Patches <- SOON STARTING

So, it's all gone exactly as planned so far, only there were a lot more things added in each stage. That's a good sign, isn't it?

Well, THE MAN has spoken and as far as I'm concerned, he has given us his personal assurance that LFS will only get better. Patch X is obviously more netcode related than physics related, though clutch diffs have done a lot to solve the "why is my clutch diffed RWD car so ridiculously unstable off throttle and mid corner whilst being so ineffective on power?" problem. I remember in another thread that he told us that everything physics related will get more emphasis especially after patch X, so why all this whining?

Quote from Mike85 :

It shouldn't be random. Because it would kill the competition. But the "randomnes" would come later when the engine model is complete. Then it would be exciting.

Its already exciting racing wheel to wheel in LFS. Atleast that thing works in LFS.

There is no such thing as random engine failure. It's always down to human errors such as driver mistreatment, engineering/technical faults, unforseen environmental factors, etc.

Until the day engines are properly simulated along with environmental factors e.g. dust and LFS becomes a true racing simulator that covers both the driving and engineering side, there shouldn't be "random" engine failures.

On the flipside, if a car hits a wall at any speed over say 0-30km/h it should suffer coolant leaks, engine misalignment, etc and effectively DNFed. Same for silly kerbing, where it should result in broken suspension and of course another DNF. These are what I agree should be there because these cases have real causes and aren't random.

Last time I checked Subarus used multilink rear suspension, so no matter how much you calibrate the suspension, the car will never behave like the real thing. Dynamic toe, camber recovery rate, antisquat etc all have effects on handling and grip that are very significant, yet many still fail to appreciate.

My point is, the real thing actually toes in on rear compression.

Last edited by Jamexing, Tue 5 Jun 2007, 11:07.

The ultimate test of how good a simulator is is to simulate a real car, with all the relevant specifications as closely matched to the real thing as possible. This is the only test that can truly make or break a sim. If the simulated real car handles very similarly to the real thing is the only absolute proof of realism in terms of physics simulation.

Quote from tristancliffe :

Ensures that you input and output shafts of the gearbox are matched, so that the gears can engage quickly, smoothly and without damage. It is vitally important on non-syncro engines, but gearbox designers invented the syncro ring as most people are too LAZY to LEARN the simple movement required.

The fact that the majority of modern manual transmission cars come with TINY pedals don't help. Nothing a larger pedal surface won't fix.

As far as I can tell, most car manufacturers expect their cars to not survive beyond the warranty period to insure demand for newer, fancier and even more fire cracker like cars.

Last edited by Jamexing, Mon 4 Jun 2007, 10:57.

Quote from cardriverx :

rotarys rule, and the people who bash it are just plain mis-informed. Whats the big deal about a quart of oil every 3k miles? Its part of how a rotary works..

+1000!!

If you don't mind carrying spare oil around with you, it's not TOO bad. The real problem with the oil consumption is the fact that the oil is either burnt or exits to the exhaust valve as unburnt hydrocarbons, both scenarios which aren't exactly brilliant for emissions control. Modern exhaust after treatment and engine management along with superior mechanical design has done a lot to severely reduce its emissions and fuel efficiency issues (no surprise), though due to the way lubrication of rotaries are done, oil consumption still leaves a lot to be desired.

Unlike some here, I recognize both the strengths and weaknesses of all engine types and appreciate them in the context of their intended application. I remember a time when diesels were seen as nothing more than noisy, smoky and underpowered lumps that have furl economy as their only saving grace. Now take a look at BMW's 2.0L turbo with 100hp/L (excellent even in comparison to a lot of turbocharged petrol engines), all the power and torque with improved fuel efficiency (not just simple economy).

http://www.motorauthority.com/ ... r-banger-diesel-from-bmw/

http://www.motorauthority.com/ ... diesel07/BMW_diesel03.jpg

Note the beautiful power curve that comes along with the sheer power and torque. What really matters is that just 5 years ago, such a diesel engine that's road legal (i.e. passes latest and future emissions tests e.g. Euro 4,5) was beyond our wildest dreams.

So what do diesels have to do with rotaries? Well, both suffer a history of relative underdevelopment compared to petrol 4 cycle engines. I just wish that companies like Mazda push the envelop and come up with rotaries that are superior to their predecessors in every way with each new iteration of rotary engines.

Besides the clutch preload, patch X won't be anything too terrifyingly exciting. Well, that's if you're like me who thinks a 32 car T1 pileup = not exciting.

Quote from Mako. :

/\ win!

If a technology is so domineering that it's BANNED after its maiden race/victory, then it must be REALLY GOOD.

Reminds me of the old GTs that experimented with variable height rear wings which were determined to be so potentially superior to the competition that it was BANNED as continuous development brought them closer and closer to their true potential. Gordon Murray's fan car was even better, with maximum downforce coupled with minimal drag that dusted off the competition. Oh, it was BANNED after the first race/win after it went circles around the competition...

I remember someone in this forum said that racing is a meritocracy. The above 3 examples prove otherwise as racing tech is more along the lines of mediocrity after the good old yesteryears when the learning curve was vertical. Unreliability rate of 30% in 2006 even with nothing more than ultra refined dinosaur tech? If failures were mostly the result of new and radical tech that's one thing, but this?

Now THAT's mediocrity!

Quote from VTiRoj :

+1, it would be extremely impressive to see the LFS's physics engine being used to simulate different types of engines. Especially a rotary.

If it had to be used in one car though, it would have to be a LMP class car or something to challenge the GTR's. We need a high revving monster of a racer

Next thing you know every server's going to ban it on the grounds of "unfair technology"...

Quote from Bob Smith :

Yes, they are identical, hence the shared components. I agree the FZR has aero that is well suited to its mass distribution, whereas the others have big problems in this area. While the wing locations are defined by the 3D model, I'm not sure what method is used to determine the centre of the undertray. If these centres were re-adjusted, it would help balance the class for sure. Atm the FZR can use way more downforce on twistier circuits, as the other two cars are limited by the maximum front wing angle.

Not to mention the L/D factor. FZR uses as much downforce as it wants with way LESS penalty than the other 2.

Last edited by Jamexing, Sun 3 Jun 2007, 10:20.

Quote from Bob Smith :

Well if you want me to be really anal, that's not quite true. The aero dynamic devices themselves are shared between the three cars but their placement on the cars varies which alters the forces being applied to the wheels. This, coupled with varying mass distributions, gives the need for very different wing angles between the cars (needed to maintain a neutral aero balance).

All true and valid points, and it is this serious mismatch of center of mass with center of aero pressure that has REALLY plagued the GTR class with ridiculous imbalance issues for all this time. Worst case scenario is the FXO, where to get a certain amount of usable (neutrally balanced) downforce it pays a with disproportionately large amount of drag compared to the FZR which incidentally has mass distribution that fits the aero balance perfectly. In simplistic terms, lets say if the FXO wants 5000N of useful downforce at a given speed to match that of an FXR, it pays with disproportionately larger amounts of drag since the front angle needs to be cranked more than the rear, easily placing it outside of the range of maximum L/D ratio. Same for the XXR, only not so serious. The FZR simply does not need to worry of such since it can easily have both front and rear angles set within their optimal L/D region with no penalty in aero balance.

Even if all 3 cars had identical front spoilers and rear wings, the actual downforce, L/D ratio etc will still be very different between 3 cars since air interacts with the aero devices and each cars unique bodywork in all sorts of complex ways. Yet amazingly they all generate the same L/D forces.

To validate my point, just go to the setup screens of all 3 GTR cars and take a look. Lift and drag forces vs. angle for all 3 cars on both front and rear spoilers/wings are practically identical!

Quote from ajp71 :

I'd love to get rid of the majority of the adjustments to the road cars in LFS, seeing as atm the cars people race have little relation to road cars.

Why on earth would any one in his/her right mind race fully stock road cars?

The TBOs are road legal TRACK cars! NOT road cars!

Quote from wheel4hummer :

I was riding with my friend, and he went around a sharp turn at 40mph, in a GMC Safari! It scared the hell out of me. First, the van almost rolled over. Second, the tires were sqealing like crazy.

Either:

1. Your friend knows his car and the road so well that it was pretty normal to him.

2. He's got balls of carbon nanotube reinforced titanium alloy.

3. He's pretty crazy.

Maybe a combination of the 3?

Quote from Bob Smith :

You must have copied and pasted it from somewhere.

Witness the power of COPY AND PASTE!

Anyway, I remember myslef giving a similar description in the Car Setups section some time ago when preloaded clutch diffs were new new to LFS racers. Only that mine actually includes a simple experiment to allow people to understand preload on a much more instinctive and fundamental level.

Guys, unlike some here I don't really have anything against rotaries, just pointing out their annoying characteristics that make them a bit more of a hassle to live with.

BTW, compression ratio of 11:1? Here's what I got from a Mazda's official websites:

http://www.mazda.com.au/rx8/specifications.aspx?ID=43

http://www.mazdausa.com/MusaWe ... Specs&vehicleCode=RX8

http://4a.mazda.com/product/rx8/spec.html

?!?!?!?

Good news: emissions and fuel efficiency has improved quite a bit thanks to more development. Bad news? It'll still drink oil at a rate piston engines in good condition don't. The new NA powerband is actually better than the old powerband from the tubo 13B, though it was more a problem of the fiendishly complex and unrelaible turbo system than a fundamental problem of the engine itself (in case people misunderstood my previous post).

I'm all for putting a good RX-7 lookalike and drivealike in LFS, but as I've said before they're already struggling to get the 4 pot turbos right, so something like a rotary could be an even bigger nightmare.

Testing bumps on the rally cross track? Too many variables as you have infinite possible lines to approach or completely avoid bumps.

There's only one way to test suspension bump response properly, and that is to run the car straight through a bump track with very well designed and defined bumps or to run it through a curve that is very well defined and bumps placed and ran over at the right places.

BTW, nothing can teach people about the effects of setup on steady state cornering grip better than a skidpad. Since this is a computer program, we should just make it as big as possible and allow people to test cornering at all sorts of radiuses. How is this relevant to racing? Well, it teaches you a lot about setup vs. apex speed, especially on tracks where there is a lot of time spent in mid corner.

Slalom? That teaches people a lot about how to set up cars to be agile and responsive in transient conditions without making them ridiculously unstable. A great example is a Mitsubishi Lancer Evolution, which did great in slalom tests and surprise surprise, it's extremely good on RL mountain passes where transient states are dominant. The same could be said of cars such as the Lotus Exige, which was great in slalom tests too which translated well to agile yet very controllable track behavior.

If test tracks were really half as useless as some here insist they are are, car manufacturers would have abandoned them a long time ago. They may not simulate a track exactly, but they allow a excellent way to come up with base setups that are fundamentally good.

Last edited by Jamexing, Sat 2 Jun 2007, 07:02.

Anyone with the slightest bit of sense would answer "don't know" to what new cars would S3 possibly have, or may just speculate for fun, but all these big "NOs" are getting silly and don't make any sense anymore.

Are people actually saying that they DON'T want new cars in S3?!?

The 2 points on physics modelling are mainly focused on things DIRECTLY relevant to the handling and performance of a car.

Reliability. Though properly model that all chassis and engine parts that could modified need to be accounted for. Even if we disregard the processing power needed and assume that we all run supercomputers, it'll necessitate freeing up many parameters to the player's manipulation. For instance, the engine must be tunable to allow compromises between outright performance and reliability along with almost all else. Yet so many simply dismiss any suggestion of making engines tunable. What a contradictory load of crap. Or are we just going to do something silly like artificially stating that a car that's revved till redline on every upshift will blow up earlier than the one that's driven at just half its useful revs over a race? Or even more silly stuff like applying randomness and cars blow up at the luck of the draw? IRL, cars fail both due to driver and also often due to engineering/technical error. If only people here have the slightest idea how hard it is to model these things.

And 3D suspension geometry not so important? Whoever says this must have been hiding under a rock for a pretty long time. Ever wonder why BMWs handle so well? Toe in on compression and vice versa from the rear multilink suspension has great influence on a cars ability to turn in and corner well whilst still allowing good powerdown on exit. If the multilink suspension I've just described is applied to the FZ50, you'll all be amazed by the difference it makes to overall balance.

Body flex not important? Oh dear. Last time I checked, any competent and skilled driver could easily spot the difference in handling between a seam welded and non seam welded car. The difference is even greater if additional chassis bracing and a roll cage is installed. Of course, LFS cars behave as if they have infinite stiffness, so when(if) chassis flex does get modeled, the difference between road cars and roll cage braced and seam welded race cars becomes very significant, especially in terms of handling precision.

Since LFS has attempted to simulate downforce cars as downforce dominant as an F1 car, aerodynamic modeling is of supreme importance. Yet we have silly things like 3 VERY different GTS with absolutely identical aero and an F1 car that has absolutely no pitch sensitivity. All these things are DIRECTLY relevant to racing and do more to improve realism and actual driving experience than anything else could.

And don't even get me started on turbo modeling, one of the most blaringly obvious yet neglected phyisics issues that received no improvement to this day.

Thank goodness that clutch LSD preload has finally been implemented properly. One blaring obvious physics problem that's finally addressed after a long time.

My point is, if the most important physics modeling remain as incomplete as it is, many things can really wait unless it's relatively simple and directly relates the potential performance cars or is directly relevant to improving the racing experience..

Tire simulation is one problem, though it's nowhere as horrible as before patch U.

2 things remain that make LFS not as realistic as possible. On is aero, but that's a complex issue I'll not try to discuss here.

The other problem is suspension. Real cars have 3D suspension geometry instead of LFSs oversimplified 2D models. There's a complete lack of antidive/squat multilink suspensions that move in complex paths don't exist in LFS either.

IRL, just try 2 similar cars similar in every way accept one: rear suspension. A car like the Toyota Corolla runs cheap rear beam suspension that will skittle over bumps and actually skip the rear end and throw the car sideways when pushed hard on very twisty and bumpy conditions. Next try the Mazda 3, with multilink rear suspension that generates desirable bump steer. Good high speed stability and a very controllable but still usable rear end that copes well with mid corner bumps is what one could expect.

More to the point:

The REAL trouble with rotaries besides the apex seals, lack of low end torque in N/A examples, etc is its fuel energy conversion efficiency.

The real problem with them is that their disproportionately high fuel consumption is a result of their inability to convert fuel to mechanical energy quite as effeciently as piston engines, compared to the current FSI petrols and especially turbodiesels with high pressure direct injection. Due to the inherent design of the Wankel, it is extremely difficult to design and implement rotor geometry that can yield the high level of compression piston engines could much more easily achieve.

The flatspot issue could be countered with VGT (Variable Geometry Turbo), a system much simpler and more effective than those horribly complicated and gremlin infested sequential turbos of old. We'll just have to wish for a new RX-7 and HOPEFULLY Mazda bothers to use VGTs...

When was the last time anyone saw a rotary with compression ratios along the lines of 13:1?

Quote from tristancliffe :

And forget it for any car that has to move at less than 40mph for any length of time - the vast lack or torque these engines produce mean they are useless and slow on the road. On the track where speeds are rarely less than 40mph and revs can be kept high they are okay I suppose.

But they sound silly. They eat seals. And barry-boys think they are the bees knees because of the Mazda adverts and the bodykit explosion for the RX-7.

The sound part is highly subjective, but yes they aren't the most practical engines for real world use. On the track, the Mazda rotary was deemed so domineering after its debut victory that it was banned. Turbocharging could compensate for the lack of low and midrange torque, but the sequential turbo from the RX-7 was a complete PITA for both engineers and technicians who worked on them. Wasn't terribly reliable too and even if it worked perfectly it STILL had flatspots.

Oh, besides oil, do remember to stock up on apex seals too...

Great engine overall, but just don't forget to bring a bottle of motor along with you in the trunk on every drive...

Higher frequency = more responsive, less grip True, especially when the car skips over bumps instead of riding them.
Lower frequency = less responsive, more grip True until it rolls so much that the sidewalls scrape the ground and a whole lot of other possible reasons.

Wider track width = more responsive, less body roll, more grip due to reduced load transfer, reducing the effects of load sensitivity.
Shorter track width = less responsive, more body roll, less grip, increased effects of load sensitivity from increased load transfer.

Increased front track leads to understeer and vice versa due to change in the roll couple between front and rear as the wider end is effectively stiffer in roll relative to the narrow end.

Now, RWD cars seldom have a lack of "oversteer", thus it's not a bad idea to give it a bit of understeer by default, so to say. This means lower frequency at rear to give it more grip. True to quite an extent, a bit of understeer is required to allow earlier acceleration on exit.

Quote from srdsprinter :

calm down. He's refering to the F/R weight bias.

IRL, 911's are known punish mistakes like sudden mid-corner braking, not braking in a straight line, etc, with sudden snap uncorectable oversteer. For the most part, this isn't seen in the FZR. With a decent setup in LFS, you can drive the FZR with (mostly) disregard to its engine location.

True, the unforgiving nature R/R does show itself really well especially in the non GTR version of the FZ series. The same could be said for the FZR at speeds low enough to make aero downforce relatively insignificant.

But since the FZR has the benefit of WIDE rear slicks and moderately wide fronts to match, added with the fact that it's aero matches so well with its F/R weight distribution, has rendered the R/R flaws relatively insignificant whilst preserving all its wonderfully positive traits. In fact, this aero synergy alone makes it light years ahead of the other 2 GTRs that suffer from horribly mismatched aero, with the FXR being the most obvious and crippling example. A front heavy car with extremly REAR BIASED downforce?

If F/R downforce was much better matched for all 3 GTRs, there won't be such a huge need for weight and intake restriction to balance them. The fact that the FZR needs 80kg just to balance the class just goes to show how fundamentally superior its performance really is to the other 2 since its fundamental characteristics such as F/R downforce ratio and mass balance are just too well matched.

Look at it this way. If an FXR driver wants to match the overall grip of his car to the FZR, he would of course need very similar amounts of downforce since the 2 cars weigh similarly. To use the downforce properly, center of pressure must be closely matched to center of gravity. The FZR could easily achieve this whilst maintaining excellent overall L/D ratio. The FXR, on the other hand, magically possesses the same rear biased aero as the FZR with its VERY front biased mass distribution.

Though possible to achieve the same downforce overall as the FZR if the FZR was set on relatively low downforce, it still loses out since this can only be achieved with severely compromised overall L/D ratio as the front downforce is cranked REALLY high whilst the rear wing remains underworked. If the FZR was set up for maximum downforce, the other 2 cars would have no chance in hell of matching their amount of downforce and hence overall grip to it whilst still suffering from mismatched aero induced L/D inefficiencies. This is especially severe with the FXR. Everyone seems to blame its 4wd drivetrain efficiency losses as the main contributor to lower track performance, but this aero issue seems surprisingly neglected. Of course more 2 more differentials and more driveshafts and rotating parts do sap a bit of power, but that alone can't explain why the FZR needed so much ballast to keep things more reasonably fair. Even with 80kg of ballast in the FZR, the XRR still needed some reduction of ballast to keep up with it! Wow.

With the TBO class, realistic turbo and powerband modelling coupled with maybe some changes in tire sizes could easily bring much better balance. But with GTRs things aren't that simple. With aero still at this rudimentary stage, there's still a long way to go. I'm just glad that Scawen has wisely chosen to implement this quick, simple and effective handicap band aid solution so things won't just be one FZR display of absolute superiority and domination after another.

Last edited by Jamexing, Fri 1 Jun 2007, 05:58.