Quote from ColGaddafi :

You don't read, you don't answer. And read above.

There is no rule about it.
But I love how you're trying to dodge my statement.

Quote from ColGaddafi :

what the hell are you on about? how about reading the thread. I've been s2 licensed far longer than you have just as a side note.

I did most of it. But I cba to read 246565462451435 post of the same XRG != XRT. Can drive one nuts, you know?
And your profile says otherwise. So if you're S2 licensed, why post this using a demo account?

Quote from ColGaddafi :

200cc isn't the only thing. as I said already, xrt has bigger wheels and stiffer suspension and more settings to it.

Like I said and you confirmed, it's like comparing m10b18 engined e30's to e30 s14 engined ones.

There is a reason that XRG is slower, has smaller wheels and "softer suspension" you know. It's the thing that gives new guys a RWD car to practice instead of putting them straight into FO8 and watch them spin along the track all day long.

If you want to drive a turboed version of XR GT and want to "roll on your big wheelz, dawg", just buy S2 and you'll have XRT(which is not the same, so you wont troll the same thing all over again)
Or if XRT is "gay" for you, there is always FZ5, RAC, LX4/6, XRR, FZR or any of the open wheel cars. Yes, some don't have turbo, but I'm sure that GTRs or OWC will be man enough for you.

Quote from FPVaaron :

The reason for two turbos is to reduce the lag

Yes, I am a turbo noob. But 2 turbos on 1.8L engine?

Quote from Nick7 :

So, you think if you add two turbo's it would have twice the power than with 1 turbo, right?

Wrong.

It would have "twice" the lag. :P

Orly nao? Haven't noticed those in the past 10 years working with Linux.

Or yes, you could make symlinks into /usr/(s)bin

If the source software package was made by a moron or the software it self was written by a moron, who will use a common name for his own libs i.e., it would be risky yes. Because if he named his one of his libs "libglib.so", you'd have a problem if you installed this and overwrite the original lib.
But AFAIK, wine is not developed by morons and their lib names are quite different from other software, so there is no danger in manually compiling and installing wine.

If you are still afraid you'll overwrite something or that you'll have to spend much time hunting down files to delete should you decide to remove the software, you could pass "--preffix=/opt/mysoftware" to install it in that dir. You would then need to alter your $PATH or pass "--bindir=/usr/bin --sbindir=/usr/sbin" to have the binaries installed in a place that is already included in $PATH.

Compiling is bad?
If you compile software on your own machine it is better than download a compiled package. File conflicts? I've never had those and I've been compiling software from source since like, forever.

And again, dpkg. Another distro specific software. make is cross-distro, so it will work anywhere, as long as you have the right libs and compilers installed.

I do agree that "make install" files are harder to hunt down, but like you said, you can have the source compiled and packaged in a distro specific package and install that. I just didn't want to include it in the above "guide", because it would again be distro specific.
But if you want it, you can use src2pkg on slackware. Maybe even on some others, but I cba to try out other distros.

Quote from misiek08 :

Why not apt-get/aptitude install wine ?
Then wine LFS.exe /cfg=setup.cfg and done.

Not everyone is using Ubuntu/Debian(thank god) and I don't even know if it will install under Ubuntu that way. X11 might come up as dep. packet. And if you don't want X11 on your server(which you probably don't if you have everything in place "up there") then your best shot is to either find a precompiled package of wine that has been compiled with above options or you compile if your self. Which I personally prefer.

Quote from Guthix :

I can see that Scawen have posted another post, where? It doesnt show up on the "look for moar posts by dis user" button/section/thingy under his username.

Stalker.

Would anyone be so kind to render my skin:
http://www.shrani.si/f/2h/sJ/t7ySFgY/xrrxfgreeny.jpg

And I'd have one small "request". Can the tyres please stay slicks?

Fine, if you like it, I'll make a new thread. In 5 years time then we'll have 50 threads explaining the same thing.

And if I friggin knew how to field 32 cars I would have already done it.

Quote from the_angry_angel :

Yes it will run on that. Don't forget that the problem is bandwidth, not processing power.

Which is correct now. I've been searching for some requirements and I found this:
http://forums.overclockers.co.uk/showthread.php?p=7979961

It completely contradicts your claim on CPU usage. So which is correct now?

I am runnin a 32 slot server on a 700MHz Duron for testing purposes. The connection is FTTH 20Mbps in both ways, so connection should be enough for so many slots, but I'm worried about the CPU speed. Will it handle it?
Anyone want to try it out?

Yeah well, that is an option, unfortunately I found it too late as I thought our local TV will be streaming full 24h online again like it did in 2009.

If they wont do it this year, Imma write a hate mail to them. They can stream all those MotoGP, F1 and other road sports just fine, but when it comes to the queen of all racing they kinda back down. Not to mention all those football/basketball/etcball matches in xyz league.

Quote from Tomrius :

Hosting LFS behind a company firewall will usually be impossible...

If you don't have access to the firewall it is impossible yes. But for the rest, there still is "ESTABLISHED,RELATED" state of connection.

Let me explain.
When someone makes a connection to your server which you are hosting on port 69332 i.e. he makes a connection from a random port(usually) ranging from 1024-65535. Why not from 1? Because 1-1024 ports can only be opened up from users with special rights(root on *nix systems).
You could either:
a) open up the port range to any host, not a good idea
b) allow the connection out if it was already previously established or is related.
With b) no one can make the connection to the outside(except for allowed ports, i.e. 53, 80, 110, 143,...if you allow them) but connections that were first established "to the inside", to your LFS server will not get denied on their way back out.

I think this needs not to be explained further, because if you have access to company firewall, you probably already know this. If you still need explanation on this matter, please tell your boss to hire qualified sysadmins.



And TCP does not use "handshake" for each packet. Three-way handshake is the TCP connection establish greeting(which has a huge security hole in it btw) it goes like:
Client establishing connection sends SYN packet.
Server receives it and returns SYN,ACK packet.
Client receives that and returns ACK packet.
ACK being acknowledge SYN being sync(I think). This way a connection is established and ensures that the two sides are communicating with each other.
The "thingy" that TCP uses to ensure no loss of data and corruption of data are actually two:
Sequence numbers(ISN) which is agreed in the initial three-way handshake and each packet sent in any direction increments that number by 1. This ensures that duplicates are discarded, lost packets retransmitted etc.
And then there is error detection. TCP uses a weak 16bit checksum for it's header and data. Sender calculates the checksum of the packet that it is about to send and includes it in the actual sent packet. Receiver gets the packet and calculates it's checksum and then compares the two checksums, the one calculated and the one transmitted, if this is correct it almost assures data correctness. But even if this is a weak check we still have CRC or other integrity checks on layer two of OSI model.

If you want to:
And the security hole. One system can recieve only X number of SYN packets and it must wait for Y seconds to get the second ACK packet back. Not including any hosts or anything.
So let's say the system can hold only 10 SYN packets at a time(the number is much higher in reality) and it must wait at least 5 minutes(lower in reality).
I send 10 SYN packets to you in 1 second, I recieve 10 SYN,ACK packets from you back but refuse to return the final ACK packet to you. Now you need to wait 5 minutes before you can drop those 10 SYN packets and as a result no one can connect to you until those SYN packets are dropped or I decide to send you the final ACK packets.

I derailed a thread.

And, and, and, they should totally add a "light a cigarette" action/button, for those really cool race endings.

Oh great! Thanks guys!

Thanks everyone for replying.
Only one more thing, where can one find aprox. times of other classes?
Mostly I'm interested in AS7 GT3 class times. I love that track and I love throwing my car down through that "blind" chicane.

I've managed to get 2.02.07 time with XR3 there, although it was with full tank and only after like 10-15 laps. Is this on the good side, or is there much to improve?

Hi,

I know some sites with the setups, but they are all for default cars, but I haven't found any that would include setups for cars with added restrictions like:
FX2,FX3,FZ2,FZ3,XR2,XR3,.....
Do any sites like this exist?

Is it ok to download an XRR set for an XR3 car? What would need to be changed? Wing adjustment? Height? Susp.?

Thanks

Quote from DEVIL 007 :

Because your CPU is not able to count all physics for full grid. I had AMD K8 which was running at 2.8Ghz which is much faster processor for LFS then your P4 at 3.2Ghz. LFS is not GPU bound but heavily CPU bound and P4 was never great for LFS.

Ok, but why does it not lag when I turn down graphics then?
I did a test yesterday.
I put AA up to 8 and I got 15-20FPS on a server with around 15-20 cars on the grid. On the next round I turned AA back down to 2 and I got 50-60FPS.
The amount of physics to process stayed the same.

I know P4 is not the best CPU around, but hell, I'm not going to use my quad laptop to play LFS.

Quote from bishtop :

nice skin xfirestorm

Thank you!

And here is the skin file:
http://www.shrani.si/f/2h/sJ/t7ySFgY/xrrxfgreeny.jpg

I didn't want to upload it before, as it wasn't finished, the credits were missing.

Save it as XRR_XFGREENY.jpg as it's uploaded to lfsworld.net as such already.
I know the number plate doesn't have a number. I didn't want it to have it.
Please do not edit, except for the number. Thanks.

Last edited by xfirestorm, Fri, 18 Feb 2011 19:05.

Take a look at this one:
http://www.youtube.com/watch?v ... 0Yh3k&feature=related
Then yours. See the difference?

I'm not saying it's complete utter cr**, but there is room for improvement. Lot's of it.

Use the path tool, it should make things easier for you.
Make four points on the skin about where you want the line(or 3 points), but don't position them perfectly just yet, as you will move them around.
Between the dots, you should have lines. If you click the dot and drag it, you will move the dot(edge), but if you click the line, it should start bending.
When you're happy with the look of it, click on "Path to Selection" button in the toolbox, which will then select your path and you can fill it with color.
It takes time and practice tho.
If you don't want to do it you can search for some vinils on google, like this:
http://images.google.com/image ... +vinyl+vector&spell=1

And open it as layer on your skin.
To remove the white stuff from the vinyl(if it's not transparent) select all the white stuff and right click on the layer in layer window and click "Add alpha channel", once done, just press delete and white stuff should disappear and you should see the rest of your skin behind your new vinyl. Just rotate/position/resize it to fit your skin.

My first skin that I really made an effort in making it.

http://www.shrani.si/f/A/116/U9s5Ruj/newskin.png
http://www.shrani.si/f/2w/xb/38b6qKoI/newskin2.png