i remember some discussion a long long time ago that race clutches are like that. basically they are made to grip 100% as fast as possible which makes sense. i guess even on the slow cars in LFS they have been given a race clutch.
your non-race clutch in your car lets you slip it quite a bit so the poor old lady next to you at the green light doesn't think you want to race when you only want to make sure you don't stall
back in october there was an interesting discussion about framerate vs steering lag. i was sure the two were totally independent in LFS but of course that is wrong as you can feel sometimes the laggy steering in a full grid start.
so i got to thinking would FFB really be better at 144hz? i had seen some 144hz lcd monitors a while back but it was too expensive. but if FFB in LFS would be better...
so i bought one and i can tell you, it really is much different. not only is the scenery moving by smoother, but more importantly the suspension movements feel much clearer and the traction breaking and regaining is much more immediate. as to whether this matters to you, that depends
so is it true "real men only need 20 fps"? well if someone can get a WR at 20 fps then that is a real man lol
the box says windows home premium 7 sp1 64bit. it's an oem version so that may have something to do with it?
i've disabled the service that controls which items get executed so i can't go back and find the survey launcher. it was buried deep in a (WMI or WBEM?) configuration tool and was set to run something like 10am of the first wednesday at least 6 months after installation. this from memory from almost 2 years ago.
Because you can't get things to work, Windows 7 is crap?
right after installing windows 7, i downloaded process explorer and noticed the OS was performing about 30,000 operations per second at idle.
most of these were registry enumeration/open/read/close cycles and many were related with windows management instrumentation which is supposedly meant for administrators to manage a collection of machines in an office or whatever.
while attacking this problem, some of the things i disabled were pretty annoying, like at pre-programmed times a survey will pop up asking you questions on behalf of Microsoft.
after a few weeks of research and disabling many things the number of operations per second at idle was less than 100. i was satisfied with it then until after the recent windows update problems.
my windows 7 always wants to do an automatic update but fails. all steps Microsoft suggests to correct this fail. i was at a friend's house a few days ago and her windows 7 laptop also has an chronic automatic update failure.
imho windows 7 isn't crap, but it is only a minor incremental change to windows XP and i think this tactic of revving software with imperceptible mostly cosmetic changes is crap. like for example the various file-incompatible Micosoft Office updates like 95, 98 etc.
i will say in Microsoft's defense that the apparent misery they put programmers through is through incompetence rather than malice. the programmers are capable enough but the mega-corporation culture unfortunately works against them in many cases
Scawen, there is a tool called Rufus which can be very useful when installing windows XP. a lot of newer computers don't use IDE anymore for the DVD drive so the installation may fail looking for a SCSI driver.
Rufus uses your existing XP installation media to create a bootable USB stick that can get the installation done.
of course this is only possible due to negligence and/or incompetence on the part of the engineers responsible for the internet's design. (to be gentle it could be said as lack of foresight.)
regardless, in a world where as young children we are made masters of some world - even imaginary - through computer games, we become addicted to this control and this idea of being king.
the logical conclusion is that the engineering problems will be fixed and we will talk about this like the strange dark ages of child-king internet ****s.
And what would really be awesome is if it would keep data for each track environment for a server instance, so you'd have to deal with the rubber and stuff laid down from the other combos as well
unless i'm mistaken, your car in lfs is not at an x,y,z point but rather at position p along a 3d curve. it's another way of saying the same thing, except even if two curves cross each other and because of that the car can be at the same point on two combos, i don't believe the lfs engine sees it as the same point for the purposes of grip evaluation.
i believe this is part of the simplification the lfs physics engine uses and because of that it is impossible to share rubber paths between combos. (at least more than visually)
- dynamic wind, including gusts etc
already implemented?? maybe the gustiness could be configurable
In here I'd like you to respond not only with your favorite car/track combo, but also explain why it is your favorite.
FZ5 @ AS5 - because making the tires last is a challenge, it's hard to get the best pace through the uphill section with this heavy car and turn 19 is a very rare corner in lfs where you gradually lift to keep the front glued. (turn 19: corner with aston north hairpin). good match of track for this car.
FXO @ AS11 (central) - very high speed and difficult braking sections.
good for close races.
XFG @ FE2R - for such a slow car there is always something happening at every moment in this combo. to get a quick pace you have to flirt with flipping or going off.
FZR @ BL1R - a good combo for endurance racing because of the easy flatspots braking for the chicane. it's so hard to make the tires last a full tank while going for pace! love the blind t1/t2 crest too.
How do you move so smoothly with the mouse? Mine's all weird.
along with the 1/2 or lower speed replay trick, i believe setting the "movement speed" also affects how fast the mouse can pan the camera. try playing with that and seeing if it changes anything.
(movement speed, the slider that changes how fast the camera moves with the arrow keys)
Scawen, you said yourself that you didn't get much done this whole year. Well, developing other (at this point) unnecessary things won't speed things up as far as I can see.
actually, achieving something in the scope of a stalled project can be reinvigorating. even if that something isn't on the critical path.
@DrBen's thoughts: good point about evolution vs revolution. LFS can be moved forward although exactly how is very debatable. many people want fancy extras that i fear are not at the core of what LFS means. this sim is supposed to enable people who want to compete to get on a virtual track in a virtual car and go at it.
adding shaders.. meh. yes dynamic lighting would rock and i can see a challenge aspect where pilots can be better than each other based on how well they adapt to low light or sun glaring in the windshield, but is that really the best step to bring LFS closer to perfect?
someone mentioned improving collision detection. to be honest, except when a car is squeezed by two others or there is a lot of lag, the current collision system is very very good. most importantly, i really believe moving this system up one notch would be as huge a math puzzle as the tire model. can we all agree that the net code in LFS is top notch and that the internet is what needs to improve? and besides physics limits us very much in what is possible even in theory, with the speed of light imposing a minimum lag when competitors are far away from each other.
when we talk about the damage model, breaking off pieces, etc. it's true the current damage model is very light. but think about it, that is just common sense. it is so rare for people who want to race online to be in a race situation with someone about equal to their skill that you don't want to just end that too fast with a damage model that is too punishing. the current one does a good job: if someone is careless and hits things too hard or too often, they will fall behind in the race. but if something unpredictable happens, some lag spike or some idiot screwing up, the race is not necessarily over right away. also this model has depth, so toe damage front and rear does what you would expect, camber problems as well, and so on.
we must ask ourselves not what all the things are we dream of seeing in the future of LFS but what is the next thing that is stopping us from having this online competition we want so badly. or if not stopping us then slowing us down or making the competition less good in some way.
sorry, wall of text, but in that perspective, adding shaders to LFS does a tiny bit to make the competition have more depth, but at the same time stops many racers from even getting into the race due to their hardware. making the damage more severe cuts short a nice battle between two racers because some lap-down kid makes a mistake while being passed.
if we get a better tire model, we get a better platform to test our skill against each other. if we get more realistic setups, we get races where there is less space between the racers, another good thing.
if we got some way to level the playing field with the clutch hax, racing would get more competitive too. right now there are a few ways competitions is being degraded in LFS and that's where the development focus should be seeing what the sim's priorities are.
Strange that so many write VR off so easily, "not interested"? I mean it has to be the next big thing
when it can be done right it will be worth it.
i was an early adopter of the 3d shutter glasses. at the time i played a lot of counter-strike and the 3d was really cool but it was much harder to be competitive in 3d than in 2d.
i don't think Scawen said he was starting the tire model over from scratch. if i read his message right, he was thinking of releasing the results he got to so far, without quite reaching his intended goals of being close to reality.
For example, for a 720p content (resolution of 1280 x 720), the frame for the left eye will be 640 x 720 and the frame for the right eye will also be at a 640×720 resolution.
a step down from the nvidia shutter glass solution then. they use a d3d/opengl driver shim to make each frame change the camera position from left eye to right eye (the game/app doesn't do anything). your 3d glasses time LCD shutters so each eye only sees the correct frame.
still, shutter glasses have problems:
even when the "shutter" is closed you can still see through a bit. this "ghosting" really hurts the effect. lowering the brightness helps the ghosting but the glasses are like sunglasses so even with the shutter open they already make the screen a bit dim.
you need a monitor that has 150hz or more so each eye gets at least 75hz. i had a crt that could do this speed of refresh and it weighed 50 pounds and screeched like a banshee.
honestly for the money, a good projector with a curved screen will probably give you much more immersion without the headache and eye strain.
just wait till 1M minecraft kiddies (edit: and CoDtards) get their parents (or their 30 year old selves ) to buy them a 1st gen oculus rift, then the 2nd gen will get special attention from screen manufacturers and we will get enough resolution.
it is not always the best to simulate the reality as close as possible, because there are still some aspects missing that could not be simulated as well. in that case it is best practice to simulate the "best immersion of reality"
agreed. LFS is great because everything inside the LFS world works as expected. this world might not be exactly like the real world but the important thing is that this doesn't mean a race in the LFS world is not exactly like a race in the real world.
in comparison, iRacing surely has some results that are closer to real life. if a track, tire and physics model gives a resulting lap time that is 0.01s away from a real life lap time, you can claim it is the most accurate simulation available. in a sense that is correct, even if the same model fails to simulate inertia.
LFS's #1 priority is to enable online racing. personally i think a physics model that makes sense under all conditions is better one that is the closest to real life. having a perfect reproduction of reality as priority #1 leads to compromises: the simulated world has some parts that are almost exactly like real life at the expense of having holes in the simulation.
it looks like Scawen tried to add new restrictions forcing his simulation to be closer to real life while maintaining LFS's unified physics model and eventually decided that can't work for the moment. probably mainstream hardware isn't powerful enough yet.
it also looks like he decided to make some update to his unified physics model without imposing the real-world restriction. no doubt this will give us a better immersion factor!