What I found out from analyzing a bunch of RAF files in LRA is the AutoClutch performs at 0.21s for each shift, for an average gain of ~0.07s over a clutch. As a comparison, the off-throttle time for the FBM is around the same.

I didn't test specifically for the button clutch as I've never bothered to cook the script up.

The difference is explained mainly by the impulsive nature of the 'attack' but the 'sustain' phase is shorter as well (no need to wait for the stick to move).

FWIW I remember quite clearly that some reputable users stated they could match the AutoClutch but at some cost, i.e. not being able to keep that pace for more than some time. (I'll try and see if I can dig the posts up)

This makes sense to me because the clutch time is not just a matter of pressing and releasing a pedal as fast as possible but also synchronizing the stick movement with the pedal, so the chance of a misshift greatly increases.

It happens only on some of the race cars apparently. I've successfully reproduced the 'issue' on the FXR, XRR, FZR, all the single seaters with the sole exception of the MRT.

I haven't tested all of the road cars, but picked some at random and none of them exhibited the same behavior. If you max out the spring stiffness settings, and then apply the intake restriction, the max stays unchanged.

If it's a bug, it's a very weird one

Back to the point raised by the OP, the relationship between forum regulars and 'overexcited', so to speak, newbies could not certainly change only through an increase of the number of moderators.

OTOH if such new moderators played the role of 'helpers', addressing the inflow of new users and taking on their shoulders the grinding which comes from the very same questions being asked over and over again, then the suggestion would make sense. They would also end up setting a standard for behavior on the forum.

TBH I've seen such a model in place on some boards, but those had different social structure and purposes from LFSforum, I'm not sure how it would work here

Quote from RIP2004 :

All cars would feel quite different in a lot of situations. Not only the start.

Sure, I'm quite positive an all-car hotlap reset would be in order and also most if not all sets would have to modified accordingly. It would raise a big wave but I'm still thinking 0.6A, despite the version bump, won't be that revolutionary.

Certainly the ESP is harder to code than the relatively simple ABS, I'm now wondering whether Scawen is busy getting it to work right: he certainly has a reputation for not taking the easy route around a problem, I would expect ESP to be coded quite realistically despite the issue with longitudinal traction, but then the default set would have to use unrealistic values to get close to the behavior of the real thing ?

Maybe the postponement has just given him the chance to do some more code restructuring and optimizations that were originally planned for a later patch, and I'm completely off the track.

In a previous post you wrote:

Quote from AjRose :

The dev's have to compromise between a slight advantage for cheaters and possibly losing a lot of their users.

What you would be compromising here is realism, hence my remark.

While we are back on this point, slowing down the driving helps would not make the users flee: there are a lot of G25s out there, but they're only being used as a souped up DFPs. This was quite evident in one recent poll about the choice of device.

In the most unlikely case that my suggestions were to be picked up, those users would either feel the need to 'upgrade' their driving habits or accept the fact their laptimes would not be as good as they were.

Quote from AjRose :

So now its because I was born in the USA? For a fully grown "euro guy" you seem to be very immature.

My point was that most road cars in LFS are european ones so I happen to have driven at least a couple of the real life cars after which the ones in LFS are modeled

Quote from Chrisuu01 :

This goes againt players as myself who use whell with no clutch pedal and have no other option becaus we have extremely smal amount of money

And some, despite having the money, must take the dreaded WAF into account but anyway...

TBH I can't see why anyone not willing to get a G25 (or comparable wheel) or use it fully should be rewarded with better laptimes

I was too quick to dismiss the reasons in your original post, indeed it is a relevant point since the VWS has been delayed exactly because of those same electronic assistance programs which you were talking about.

The TC on the VWS may actually be useless (or worse?) at the starting line, but since it's supposed to have an open diff, it would also affect performance in corners when the driver is pushing too hard, if I get this right.

So your limited informal testing would suggest there's currently little difference between locking the tyres hard and cadence braking in LFS, as long as you only take braking distance into account?

But ABS will also be better because, despite many claims of tracks in LFS being 'as flat as glass', it will ensure full driveability over bumps and on sloped sections of the track.

As most set have their max braking force tuned for the hardest part of the track, there will be consequences on the average laptimes. How much of a consequence will depend on the actual track/configuration, I guess.

Dear AjRose,

I see what you're trying to do. Let me sum this up, you are:

1) basically saying my suggestions are crap because you don't value realism that highly.

Guess what? I do, that's the only reason I'm into LFS. The choice of a title for the thread might have been a giveaway, I know, but can't help it

The button clutch cheat and issues with fake sequential are Evil and go against the trend that LFS has displayed over the years, i.e. increasing realism by removing or toning down those driving helps which are too good. It happened with auto-blip, it happened with keyboard stabilised.

2) trying to persuade a fully grown up euro guy (me) that he doesn't know how euro cars are supposed to work.

Seriously, I've been driving cars with manuals already when you were still in kindergarten, do you really think I'm going to trust your posts over my own experience? :nol2:

3) trying to make me feel guilty for having a shifter (?!)

This is the icing on the cake

Quote from AjRose :

Also if you shift properly without flatshifting it will take very near the same amount of time as an H-shifter.

You're missing the point. The button clutch is a known cheat which is commonly used by some people to gain an unfair speed advantage. As things stand today, the fastest way to shift is use the button clutch cheat, followed by sequential mode, and last is the manual.

So the button clutch cheat should be eliminated because it is an overcomplicated and unrealistic way to seize an advantage over other drivers. And the driving help should be tamed as it shifts way too fast to be a driving help. It's just a shortcut to lower laptimes which has no place in LFS, a simulator that strives to achieve as much realism as possible.

Quote from AjRose :

Sure most cars wont be able to recover from 200 rpms but some will. And being in that situation is so unlikely most people will not ever notice it.

Engines with heavy flywheels would be able to recover. But that is definitely something you don't want to have on any car which you race with, as is the case for most road cars and race cars in LFS.

Back to the point, what happens in LFS is that in some spins you're able to avoid a stall even if you're slow on the clutch. I mean in cars like the XFG which is certainly not high displacement.

From a race perspective, this means you lose less time on a spin than you should. It may be a minor point, but it still is one nonetheless.

^ Riel this one is a reply to you too.

Quote from AjRose :

The gear indicator... Really? Is it that big of a deal for you to ignore it? Better yet why dont you buy everyone that plays LFS an H-shifter. There problem solved the Dev's will be able to get rid of it.

What you're saying here is there's no need for improvements because it's not a big deal to ignore the flaws in LFS. This is an improvement suggestions board, so it'd be kind of pointless to suggest not to make suggestions, am I wrong?

As for getting a shifter, if I was able to buy one than anyone can, unless they come from a country whose exchange rate is severely impaired.

It isn's some distinguished, luxurious mark used by rich people to recognize their peers. They are very affordable now and the main reason why some people can't be bothered to get one is simply because there's no speed to be gained from that.

Relying on a too-good-to-be-true driving help is much more comfortable than having to operate your own shifter.

Quote from AjRose :

I have never had the Rev's drop Instantly to 0 upon turning off a car Even from idle.

In LFS the engine takes too much time to reach 0, much more than needed IRL. That was what I was trying to say and I'll see if I can change the OP to make this more clear.

Quote from imadude10 :

I just think, as has been said before many times, that the engines need reworking.

Quote from S14 DRIFT :

Good points but most of them have been suggested before and are almost certainly being worked on. But these are not major problems and should be prioritized over new content and other major updates.

Quote from AjRose :

All of these have been suggested before.

Nothing that I wrote here was in Hyperactive's log and to search for all of those point would have required a lot of time.

If you like this better, think of this thread like a collection of suggestions that have already made before, but now they're all in the same place, so it has a value albeit it may seem small to you.

Besides, small fixes shouldn't be pushed forward unless there's a good reason for that, e.g. if it would mean fix something that would need to be rewritten anyway in a very short time.

New content is probably not going to come until S3 and all that bonus content which we received (or are hoping to receive), is just that: bonus content, we should be thankful for what we've got - but not expect more as all of those extras came for unplanned things that happened along the way (like V1 challenge, the VW deal etc)

Besides, one could use the 'content/major things' argument to counter all the suggestions for improvements, because they will either be minor and as such infinitely delayable, or too big and as such just too much to ask.

Quote from RIP2004 :

So this increases the water film and helps aqua planing to happen at a lower speed as with rotating tyres.

Makes sense, I stand corrected.

Quote from RIP2004 :

Of course aqua planing can appear everytime. But a locked wheel does help it to appear, because it makes the water film in front of the locked tyres thicker ...

It's easy to see that could happen with things like mud or snow, but water? I don't know

Quote from RIP2004 :

As you can read again on wikipedia and other sites : locked wheels push the water in front and the tyres begin to swim. Aqua planing.

Aquaplaning has nothing to do with braking, it happens whenever the water film on the road is so thick that the tyre cannot move enough of it to the sides to get a significant contact patch.

When that happens, besides praying to get outside the waterpool unscathed, you can't do much more. No ABS or anything will be able to save your valuable ass because the only things that kept you sticking to the road, i.e. tyres, are floating happily in water like kids in a pool during the summer break.

Back to your original point, improved tyre physics would certainly be a Good Thing(tm), but not for the reasons you've stated.

Quote from blackbird04217 :

Mis-shifts are very possible

That bit is under manual-in-seq mode, the driving help, aka when a car with a gated shifter is driven in sequential mode.

Quote from blackbird04217 :

As far as mis-shifting on a sequential, why does that need to be simulated?

Because shifts on a manual in sequential mode are too fast. They are as quick as a true sequential, which does not make any sense.

If manual-in-seq mode is an ambiguous definition can you suggest a better one please?

Quote from DrBen :

Just look at all those cheapos - like me - sometimes using mouse-steer or sporting one of the simpler "two-pedal and no h-shift" - wheels in lfs. You would definitely exclude these fellow racers when dissing the automated clutch mode.

I find any innuendos of elitism quite out of place, I certainly never have criticized anyone for whatever device they use or said I am better than people who don't happen to have a shifter.

If you call them "cheapos" like it was an insult that's your choice, because I've never used that word against you or anyone else, for the record.

Have you ever seen me online bragging about my wheel or vainly display a [G25] tag? I believe you would have a hard time proving that.

One has the choice of using whatever input device he/she fancies most, OTOH he/she should also face the consequences of whatever shortcomings his/her choice of device carries along.

You can certainly use a mouse but if you complained about the lack of analog control for the pedals, people would tell you to get a wheel and shut up.

But at the same time, non-shifter setups are actively promoted through the excessive skill of those driving helps which were devised to help people cope with limitations of the device, at a time when shifters were much more expensive and uncommon than they are now.

We have a brake help, too, but very few use it because modulating the pedal is regarded as faster. That's the way things should be: driving helps are made available but they do not to make you faster for no reason.

So I say it again, let's remove the button clutch cheat (you're not going to defend cheats, are you?) and then slow down the driving helps so that they become what they were intended to be.

Quote from Riel :

An engine gets started by starter at around 80 rpm. So it shouldn't stall at 200

The world is choke full of things that shouldn't do things we don't expect, but frankly, a petrol car revving happily a long at 200 RPM is not one of them.

I'm certainly not Gordon Murray, but I've seen and driven a few cars and none of them could do that.

Since you appear to believe otherwise, I'd kindly invite you to post proof of what you're saying.

Quote from wheel4hummer :

Most updates don't require a restart except for graphics drivers and the like

Actually, the only updates which require a restart are those concerning the kernel. My linux box has been up for 85 days as of now.

Quote from sam93 :

What are the advantages that Linux has over Windows exactly. I have heard that Linux is a lot better than Windows but I do not yet have the knowledge of what Linux can do to know why it is better. Could someone please tell me what are the advantages and disadvantages if you don't mind.

Just some random thoughts

Pros: distros with good package managers are way easier to keep up to date than Windows (one centralized resource rather than windows update + a zillion different websites/programs to maintain); no need for an antivirus/antimalware/whatever, no real time check means snappier operation and no time wasted with routinary scans; no need for defragmenting (I love this one ); safer, security holes are fixed much more quickly than Windows (with MS it takes a couple of months if you're lucky). 64-bit support is almost perfect which means some programs are faster. For some type of applications, there's actually more choice than under Windows (e.g. free non-linear video editors). Extremely customizable. Wine allows you to run many common Windows programs with little inconvenience and more or less the same performance. Infinitely cheaper than Windows.

Cons, GNU/Linux being a patchwork of different software, user interfaces are not always coherent or as polished as you would like; there are several different standards for storing options, managing configuration files etc. This situation has been steadily worked on and has been improving over the years, and one could argue Windows is not completely free from this kind of issues anyway. Stumbling in the wrong distro can be quite a disappointing/frustrating experience. Fixing some problems, sometimes, can take quite some time and effort, but tends to run smoothly forever afterwards. For some type of applications, you simply don't have as much choice than Windows. Hardware support is not as complete and up-to-date as Windows, so linux users still have the habit to check compatibility and quality of drivers/support before getting new stuff.

That's certainly far from complete. Just what I could come up with in about 5 minutes.

Quote from Bob Smith :

Yep, a little toe in for high speeds, counteracted at low speeds by using slightly more Ackermann.

If I get this right, your suggestion would be even better than pure Rev Ackermann.

It looks too good, why would they apply Rev Ackermann then?

You can't alter a set via scripts.

I know the problem well since I used to race with pad too. If you will accept a free piece of advice, buy a wheel ASAP.

You'll be starting again from scratch as soon as you switch controller; that's the reason why some aliens keep racing with the mouse even if they have a proper wheel.

Quote from Ikaponthus :

Secondly, I think if any air did get under an F1 car it probably would fly (haven't watched the youtube video btw). It's wings would act like aeroplane wings.

It's quite the opposite.

The wings on planes are supposed to create a high pressure zone on the underside. The pressure difference results in a upwards force.

Cars OTOH like to stick to the ground, so the wing creates a high pressure zone on the upperside, that's why it's called downforce.

The spectacular accidents you're talking about are the consequence of the ground effect which creates hideous amounts of downforce as long as there's a certain clearance below the car and not more.

If the car is lifted beyond a certain limit, the ground effect immediately disappears and the car sorts of bounce up.

That said the videos of GT5P which have been posted, look really unrealistic: as usual cars appear to have no weight.

Quote from dawesdust_12 :

Metapackages for Ubuntu that install the other window managers, however they are known to sometimes reak havoc on machines, so be careful.

Really? How weird, I'm a Debian/lenny user and switched DE several times with no ill consequences.

Quote from Ikaponthus :

I really get the impression that a lot of the people in this thread don't even have their driver's licence yet. Maybe I'm too old for this.

That's how the forum works. Though the average age of licensed drivers is around 27 IIRC.

OTOH I find the kids more bearable than the adepts of the Church of Kaemmer that use to disguise as hardcore simmers, those older chaps are more pathetic too, you would expect them to have learned something over the years

Last edited by NightShift, Thu 15 Jan 2009, 19:50.

Quote from Flame CZE :

I think that it's implemented.... I can't play LFS now so it maybe is only a Back button...

There's an undo button but it only works as long as you haven't saved the set yet.

Quote from zeugnimod :

+1, learn it right the first time.

Yep, and anyway damage is far too little as it is now. I'd also like to have better suspension damage so finally chicanes would have a real meaning in LFS.

Things which I feel are wrong in LFS and bug me. This list is mostly about small things that I imagine could be fixed without major alterations to the code. It is also more about removing or tweaking existing things, rather than adding new features.

AFAICT none of the following is present in the SI log by Hyperactive as of now.

Transmission Simulation

• Fake sequential (the sequential driving help on gated shifters) is too good and needs to be tweaked, despite being a driving help it makes people faster for no good reason, and as such is every bit as bad as keyboard stabilized was once (when it used to catch slides for you better than a person could ever hope to); similarly, fake sequential on gated shifters is able to shift as fast as a true sequential and usually better than a person with a shifter, also without any risks of misshifts. So:
  • Shift speed should be toned down, so the fake sequential becomes slower than an average driver, thereby introducing an incentive to use a real shifter in hotlaps and online races.
    
  • The risk of random misshifts could be introduced, to enhance realism.
    
• Button clutch exploit. At the moment it's possible to gain an unfair advantage by exploiting the button clutch, a possible solution is:
  • Make the button clutch slower than a proper analog clutch pedal and fake sequential. This will constitute an incentive to learn to use an analog clutch properly and promptly get rid of all the cheating.
    
• Clutch heats up too quickly on road cars. IRL one can keep a car stationary against a slight slope, by playing with the clutch and applying a little gas. An unexperienced driver will often perform the maneuver revving as high as 2000-2500 RPM, without any short-term effects on the clutch. In LFS this would probably induce a noticeable slipping under power.
  
• The Gear Indicator on road cars is unrealistic, but also useful for those users who don't drive with a gated shifter. Two ideas here:
  • Enable the gear display in road cars only when transmission is set to manual-in-seq or auto mode.
    
  • Add an option to enable/disable it manually.
    

Engine Simulation

• Revs should drop much quicker than now, when throttle is closed - I suppose this also could be affecting engine brake to a degree.
  
• Engine should stop more or less instantly when ignition is turned off: instead it takes quite some time (seconds?) to go from idle to zero.
  
• The engine on most cars should not be able to rev as low as 100-200 RPM and then come back without stalling.
  

Last edited by NightShift, Wed 14 Jan 2009, 18:03. Reason : Tried to improve readability of the suggestions