The online racing simulator
Searching in All forums
(21 results)
wollschaf
S3 licensed
Most probably the resolution and / or refresh rate you selected in LFS were too high for your CRT. Newer CRTs shut off in that case, older ones display multiple images and can get damaged by that.
wollschaf
S3 licensed
I'll never buy Microsoft gaming hardware again after having had the unique experience of discontinued driver support. Sure, basic drivers exist, but there is nothing to change the pedals from combined to separate axis besides copying windows 2000 and XP driver files back and forth.
wollschaf
S3 licensed
I know these problems... All you have to do is replace your GcKernel.sys from XP (59136 bytes) with an older version from either Windows 2000 or the SideWinder 4.0 driver package - I don't remember where I finally got it from. The size is 100276 bytes and the date is Nov 18, 1999. The older version will only allow you to use separate brake and throttle axis for the pedals.

If you backup the original file from xp it is indeed possible to change the behaviour of the wheel by unplugging, exchanging the file, and plugging it back in (in another usb port). Two little batch files that copy the files around is easy to write, one as example here and the other one as exercise:

cd %WINDIR%\system32\drivers
del GcKernel.sys
copy GcKernel-separate.sys GcKernel.sys

This of course assumes that you have two GcKernel-*.sys prepared in the system32\drivers directory and are okay with your old GcKernel.sys to be deleted. Do make a backup if you're not sure what above commands do.

I have had calibration issuses with the throttle pedal, which occasionally would behave half-binary: zero until 50% throttle, then normal behaviour up to 100% - and occasionally not falling back down to zero, but remaining at 50%, imagine that while racing
I think that I solved it by pressing down the pedal a little bit while plugging the wheel in; In my case the calibration button in the controller applet is doing no good at all... the wheel (especially the pedals) calibrate themselves - more or less. Especially if you open the controller applet, press down the pedals, but resist from clicking that button. Luckily, calibration has not been an issue for sevreal months now, so I've had no problems at all in LFS and RBR
wollschaf
S3 licensed
Quote from Electrik Kar :If it's night-time and raining softly at BL for everybody everywhere, it helps cement the notion that Blackwood's a real location somewhere, naturally operating under it's own laws of environmental and temporal variability.

I strongly agree. I'd love to see a global weather system that simply cannot be turned off. Drivers would have to adapt their driving style, and league races definitely were a lot more interesting. It's a simulation, after all

Track and vehicle usage is also likely to change with the weather, and that's a good thing.

I see a problem with global time zones, though. Some tracks could just be raced on at night time for certain users, which is unfortunate. Splitting the global weather and time system into separate worlds (LFS World #1, 2 and 3...) could solve this; Users with certain track and time preferences are then able to choose something close to their preferences. The user base has to be big enough to ensure every world is populated enough, though.

I can also imagine that many lfs players would not like such a system at all.
wollschaf
S3 licensed
Quote from KMSpeed :I 'd like what Victor proposed combined:

The first line for the user name, WR and date
and the second line for the splits.

That solution looks best to me. It's still easy to see the WR and completely ignore the splits for a quick glance, should one choose to.
wollschaf
S3 licensed
Quality control is not easy for a sim open to mods. It' not possible yet to restrict user modeled cars to real physical behaviour. In a perfectly simulated environment, bad mods would behave badly and unrealistic ones could not be used at all. I don't see that kind of simulation being feasible on home computers for the next ten years.

Maybe somwhen inbetween, premodeled parts could be made avaiable to build cars, but then it's more of a building simulation. After S3 is finished, the developers may start creating "Live to build" and then start an online engineering community that builds cars for S4.
Even with premanufactured parts, it's hard to avoid bugusing... this would be rather hard to achieve.

It's a shame that so many people out there would certainly build unrealistic, overpowered cars that will ruin the whole LFS experience to others. It's also a shame that there are people out there who can build better cars than the ones included in LFS (more time, no money considerations, ...) - that effectively can't be used because of the more or less crappy other mods.

One possibility to solve this is to allow only approved mods to race online. But then again the approval process would have to be very strict, leading to frustration on both sides and endless flamewars on the forums. Doesn't sound too convincing to me.

I don't think that real cars should be included in the game, if that costs money. That money should be used to improve the simulation. I bet that when S3 comes out, some more (small) manufacturers actually want to be included in the game

No, not more real cars, but more realistic (and detailed) models of the ones we have now. If they appear to be real, they become real. Then people wish to have those cars in front of their house, and not only in the computer.

All in time.
wollschaf
S3 licensed
Quote from tristancliffe :How accurate is GPS these days? +-40 feet? +- 20 feet? Either way, hardly good enough to make realistic roads!

Accuracy of GPS is given in reference to the real position on a map, thus there are huge absolute errors. With good visibility of the sky, the GPS itself is much more accurate in a relative way: You get a smooth line, your position does not jerk around, and you can claarly see where you changed a lane, for example. Even though the absolute position on the map could be terribly wrong, the shape of the track itself would be mostly accurate.

Recording the track twice on different days will produce a shifted version within the range of absolute GPS accuracy, and perhaps it is shifted a bit more from N to S than from E to W. For height data the same applies, but the variations are much bigger.

It all depends on satellite visibility and positions (which change), but the raw data is good enough as a basis to work on. Not for exact replica, but for something being very similar... for fun.

Attached is a tracklog from driving around a motorway interchange (1sec logging interval, as fast as it gets on consumer units) and a screenshot of a tracklog visualisation - including height data. Driving onto a mountain on narrow roads, a different way back, and around the area in general.
wollschaf
S3 licensed
Quote from al heeley :Little arcade-like HUD graphic indicators would really downgrade this sim, imo.

That's right. Exact distance measurements should be avoided. But perhaps it is possible to create a system that just enhances the sim driver's senses, which does not at all have to be accurate. There are certain things you notice while driving a real car that are not visible in LFS.

I think about some kind of shadow (transparent bar, or something) that appears at the sides of the screen if a nearby / overtaking car would be visible in the corner of the eye in a real car. Just that you know that there is something, but not exactly how far it is away.

A feature like this would enhance the driving experience - and not be at all unrealistic. Things you don't notice in a real car without turning your head should require turning your head in LFS, too.
wollschaf
S3 licensed
The problem with interpolation is that it generates a lag in realtime applications. For a perfectly smooth transition this will be the delay between two packets.

To decrease this lag, some kind of predictive data will have to be sent (or made up from older packets received) so the interpolation can assume the wheel will turn further with a certain speed, and display that assumption. This is probably bad, as wheel motion can't be predicted that well and overshooting would look very weird.

As wheel motion data is not very complex or big, it could be sent separately and more frequently to improve the multiplayer experience.

Interpolating from a saved replay should not be a problem to implement if the current code fits. That would be a great improvement!
wollschaf
S3 licensed
The issue reminds me of this...

http://www.escapistmagazine.com/issue/17/27

So... we either need more girls playing LFS, ore more guys pretending to be girls so everybody is afraid to ask players what they really are

Some manners wouldn't hurt, either.
wollschaf
S3 licensed
I had this forum in mind, and the website I thought of was LFS World

Voting at different locations would create a huge development and management effort - if it was an official feature, the user database could be used to prevent multiple votes per user.
wollschaf
S3 licensed
As there are quite a lot of people who like to experiment with different combos, it should be possible to get those together for some racing events. Having all these people join the same server is a serious problem.

Perhaps it would be best to set up a 24/7 server or two that offer
- different selected (not random) combos for 12/24/xx hours
- a website / forum to vote for the next combination
- not-so popular combinations, and never one of the popular ones
- statistics per day and / or driver to make it even more attractive (best race, fastest lap, ...)

Something like a one-day racing event, every day

An official lfs world racing server that stayed on top of the list would be perfect
wollschaf
S3 licensed
There's also a possibility to enforce certain lap or split times within a reasonable range. Most wreckers I've seen (not many) were standing around somewhere most of the time. Not meeting the time requirements could result in pitting and not being able to join again.

The time limit should of course take account of slower racers and allow a t1 pileup and some spins on the track... It should not be noticed by anybody who tries to complete the track.

On the other hand, there are still plenty of ways to wreck others.

Better stick with the LFS principle... better physics and engine damage would certainly not allow more than one heavy crash.

I assume it is relatively easy to make the car's engine simply stop working after a certain amount of deformation on the front (or wherever the engine is located). Blocking the wheels after bending them over a certain angle would also help (and increase realism). And maybe just kill the driver after experiencing more than, say 80G acceleration. Just for really, really bad crashes. This would include having to fix the pit entry of South City (Club?) where flying around for touching the walls slightly (i.e not frontal) is not so rare for me...

This, of course, would also hurt the one being wrecked. But only once...

The number of deaths could in some way used to restrict joining servers and in general improve more careful driving. It could be automaticaly decreased over time... 3 unlocks per Month, 30 lives for online racing

The basic assumtion is that wrecking people need far more lives than normal driving ones, including new drivers who sometimes hit the wall...

So all we need is another (crash) physics update
wollschaf
S3 licensed
We just need the Ariel Atom in LFS. It's the perfect car, somehow even the philosophy of it fits perfectly. I bet it would be so much fun to drive...

Please? :boing:
wollschaf
S3 licensed
Disallowing Shift-S and even resetting the car during a race would be neat and indeed realistic. Wreckers should have a harder time then

A problem could be if a completely demolished car stays on the track and isn't able to move by itself. So perhaps an option could be implemented to set a timeout, a "pit after n seconds"-button. n as an adjustable value, of course.
wollschaf
S3 licensed
A great game!

wollschaf
S3 licensed
Encryption seems like the way to go.

Every player could be assigned a secret n-bit key. Skins uploaded to LFS World that are marked as private could be encrypted using this key.

Upon trying to display the encrypted skin LFS connects to the Master / LFSWorld server to get the username and player's key. Decryption is then possible... but only if the skin is used by the right player.

A default skin displaying "I tried stealing skins from other users" could be displayed if decryption fails

Of course this system can be beaten, too. Not with a simple offline-tool, though. Something that grabs the transferred keys or reads the texture from memory can.

But it's not at all importand to make it perfectly secure...

Although... I'd rather see physics updates instead of that feature.
wollschaf
S3 licensed
:sheep:
wollschaf
S3 licensed
You should ban him from your server if he keeps crashing...

Oh no, wait.


Maybe your computer is infected with ad / spyware -> try downloading AdAware and see what it finds.
wollschaf
S3 licensed
My SideWinder Precision Wheel (non ff) has the pedals connected through a 4-Pin telephone plug. The part Number of the pedals is X04-74735.

I am pretty sure that opening the pedals might help to get the spring back in place.

I need my pedals, though. Sorry.
wollschaf
S3 licensed
There is a possibility to view those videos from germany, too. Just use google language tools and translate the URL of the video

I'm eagerly awaiting the big physics update
FGED GREDG RDFGDR GSFDG