Just a headnote, if you are using LFS on Linux with Wine, you'll need to install the vcrun2019 package.
Error code:

Is that possible, that I cannot turn off the radar?

Quote from vinibali :

Does anybody using it with Wine(Linux)?

It's working fine, the requirements for now:
* 64bit WINEPREFIX
* execute the binary while using the WINEPREFIX variable (if multiple can be found), because it's searching for the LFS process.

Do we need the last one? If using the insim port, this check could useless or just a hint at the startup. Currently it throws an exception.

Does anybody using it with Wine(Linux)?

Thanks, that's pretty obvious
We have two passwords now, but both needs to be the same in order to login and download mods at the same time.

Thanks!
I did this, however we just have there two passwords for some reason

Hi there!

My web and in-game passwords are different and I need to change the account password, while browsing mods inside the game.
Maybe I missed the explanation for this behavior, but if there's no chance to use the in-game password to download mods, we could have another value for the account inside the game.

Thanks

I would like to also subscribe.
Driver: \//RS BALI (vinibali)
Number: 90
Team: \//RS

Last edited by vinibali, Sun 31 May 2020, 14:52.

Quote from lfsrm :

Well if someone is fully ready to pay for LFS and at the same time being reluctant to invest into some cheap 40/50$ used GTX 660 or HD 7850 to play the game, I find that a little bit ironic...

Scawen is here to take advantage of the new features and enhancement to make his life easier, and at the same times delivering those features faster, but you guys are dragging him down with stone age hardware just because why not...

Simracing games are probably one of the most expensive genra to play properly, you need to understand that at least.

don't get me wrong, i'm so pleased with the progress. i just made a note, about the gaming opportunities for Linux gamers.

Quote from Bose321 :

That's what you get when you want to play games on Unix. Just go for the DX11 implementation and save yourself the trouble of maintaining it for the DX10 level. Like lfsrm said, you can buy a super cheap second hand GPU that supports DX11.

If you're still using hardware or an OS that doesn't support DX10 of DX11 in 2020 it's time for an upgrade

you've got lost in my short story Linux gamers will need a DX12/Vulkan capable GPU which should be at least an AMD GCN graphics processor with open-source AMDGPU driver

Quote from Scawen :

I don't have any plans to set up a Linux computer with a super graphics card to test DXVK - we don't have those kind of resources! As in space, time, money, etc. But hearing all those other games run fine, I would be surprised if LFS didn't.

you can count on us, i'm thinking about to set up topic here for racers you want to play on Linux.

Quote from Degats :

Vulkan supposedly works on cards supporting OpenGL 4.x and up, which means Radeon HD 5000 (2009), though still Fermi.

the first GCN (HD7730+) cards has the driver to run on Vulkan, maybe older architectures could be able to, but we don't have any closed/open driver for them. (sorry, i can talk about only the AMD part)

Last edited by vinibali, Tue 21 Apr 2020, 04:23.

very impressive!

i remember the time when the Linux community just had the opportunity to speed up gaming with Gallium nine. it's nearly a native graphics layer to play DX9 games on *nix with less GPU call transition compared to the classic DX -> OpenGL -> GPU driver conversion.
back in 2014Q4, Gallium nine was just merged to the mainline open-source GPU driver (Mesa3D) a few months after the DX9 based, 0.6F version was released.
i helped the community with a bugreport and the graphical glitches were sorted out in a month. finally the Wine performance was great, i didn't have to boot up W7.
i just noted this, because with DX9 - as Scawen mentioned - nearly anybody can play the game.
sadly if the DX10+ version will in place, people will need the DXVK implementation which works for DX9, DX10 and DX11, but in this scenario Linux gamers will need a DX12/Vulkan GPU, because this implementation transforms the DX calls to Vulkan.
i know it's pretty hard to maintain two DX versions, but can't we fallback the graphical engine/DX implementation to a lower level? effects, objects might miss couple of feature and quality, but might be still renderable.

for me the 24bit Z-buffer is fine. I dont know what do you find incorrect. unfortunately with wine nine and with my Ryzen 3400G 24bit is the highest option, what I can use.
isnt DXVK is for DX11->Vulkan conversion?

Dear devs!

I play LFS on my Arch Linux installation with Wine. This might be a known bug, but if not I'd like to fill a bug at freedesktop.org. It doesn't matter if Gallium nine is switched on or not, but 24bit Z-buffer is fine anyway.
Screenshots attached,
thanks

I'd like to say thank you guys for the comments, even if you said that I'm guilty.
Specially Draggo, who helped a lot with creating videos, analyzing the replay etc...
To be clear, I always race with the lowest amount of fuel. I experienced the lower weight might be better, I don't even loose so much time if I have to spare some fuel during the race with earlier upshifts and sailing.

Simon, thank you for the unban. But,
- your comments are malicious and full of personal remarks in general, but especially when you said in the MRc forums "Oh dear that did not go as you planned Vini LOL" (http://masterracecar.forumacti ... t-got-banned-for-230-days),
- you were paranoid, when you asked me about a potential DDOS attack.
IMHO if you are an admin you should behave differently, but first of all please be bit more patient.

See you at the race track, however in the past days I enjoyed so much racing at B2R Rotate.
PS: your last turn was nothing special

Thanks Draggo.
Seems like the structure has changed in the 0.6Q version.
0.6H, 0.6H4, 0.6J was really similar.
From the version J the third bunch of data is about the car and driver textures which might be a little behind in the earlier versions.
Btw I think the needed replay were handled by the trim command or the garbage collector.

I never ever used any kind of cheat in racing (quali, sprint, endurance, etc...) LFS servers. (I hope this sentence is enough strait and well defined to not to be ambiguous like in politics).
This is why I was spent nearly all of my free time in the past 3 days after the work to get the mpr back from the filesystem level. Unfortunately I couldn't restore it and this is why the rest of the effort worth the price.
- If you look at the video from my perspective until the end of the scene(0:30-1:57), you can clearly see the throttle and even the steering was not so smooth from the last corner, like in was before (after the long straight until the last corner).
- The average distance was about the same up the hill until the finish line. He used a little bit of brake before the corner (I didn't) and we were on full throttle about the same time.
Who can I ask to judge this?
Do I have to ask Scawen to look at the replay?
I always race fair and try to be clear, this whole thing is just ridiculous.

Hello guys!

I was banned from a server. The admin said that I was cheating, but unfortunately the replay is only available from the server which is located in the UK. I think there was some latency issue, because the end of the race is so laggy.
Related topics:
https://www.lfs.net/forum/post/1926929#post1926929
http://masterracecar.forumacti ... t-got-banned-for-230-days
Here is the video:

And the replay file:
https://www.lfs.net/attachment/171116

I seek trough the SSD multiple times with different tools, I found mprs even from the 0.6H version, but not the local one from Sunday.
The first few lines contain the following information in this order:
filetype(LFSMPR), version, layout, trackname, fancy name - racer1, licence plate - racer1, car - racer1, username - racer1, fancy name - racer2, licence plate - racer2, car - racer2, username - racer2, ...... fancy name - racer*, licence plate - racer*, car - racer*, username - racer*.

Are the first few lines the same in the local and in the remote versions?

There was one race between the particular one and the ban, so the temporary file was not an option. However I was able to find the file on the filesystem. But unfortunately it's absolutely the same, so the remote and local mprs are identical.
So my question is more likely developer related, are the replays coming from the server? (even if the client is able to create his own copy)
Can we invite Scawen to answer this question?

Hello guys!

I was banned from a server. The admin said that I was cheating, but unfortunately the replay is only available from the server which is located in the UK. I think there was some latency issue, because the replay is so laggy, but this is just the background.

So I just want to get some information about the replay file's structure.
I'm using an experimental FS in my Linux environment, so I can't just restore the deleted files.
I seek trough the block device searching for the mpr-s.
I've found about 90 files which has the word "LFSMPR" in the first row, but I don't know what is the closing character or hexcode in the file, because of the "grep" search saved the files which was started with the word "LFSMPR" and cropped 2000 lines after the word. I've found a constant "FF 00" at the end of a few MPRs, but this combination appears many other times in the same file.
So the question is:
- is there any way to guess the end of the replays?

Thanks

The parameter needs to be placed in brackets, otherwise the bash/terminal wants to translate it.
So it works (in one line) this way:

Quote :

/usr/share/playonlinux/playonlinux --run "LFS" "Z:\home\balazs\.PlayOnLinux\wineprefix\LFS_nine\drive_c\LFS\data\spr\temp.spr"

or

Quote :

wine "/home/balazs/.PlayOnLinux/wineprefix/LFS_nine/drive_c/LFS/LFS.exe" "Z:\home\balazs\.PlayOnLinux\wineprefix\LFS_nine\drive_c\LFS\data\spr\temp.spr"

GALLIUM_HUD_DUMP_DIR=/media/ramdisk GALLIUM_HUD=cpu timeout 60 wine /media/ramdisk/LFS/LFS.exe /media/ramdisk/LFS/data/spr/temp.spr
says "Parameter 1 is invalid".

Hello there!

I want to create a script which tests different sw/hw settings and systems.
Just for your info, here are the opions:
- 2nd/4th gen AMD APU
- AMDGPU/Radeon kernel driver
- Linux/Linux-CK packages
- Wine compiled with O2/O3
- w/wo TearFree
- Gallium-Nine (w/wo CSMT)
- CSMT
This is a lot of work, so I just want to do the test with a script which minimizes the human-error ratio and helps me to do nothing
Is there any parameter or setting which just starts a replay? If I remember correctly the replay was a built-in feature in the previous releases but it's not working with 0.6R.
I've found an /spr parameter, but can't use it.
I can stop the benchmark with a timeout and forward all the relevant data with the GALLIUM_HUD_DUMP_DIR variable.

Thank you

did you installed the d3d9_43.dll and D3DCompiler_43.dll at lfs's virtual machine? where did you downloaded the wine and mesa packages?

what kind of? like in the #7 post?

looks like David updated the master branch!
https://bugs.freedesktop.org/show_bug.cgi?id=86635

"AMD A8-3870K (gpu OC at 720Mhz)
with Nine: avg 80 fps
without Nine: avg 50 fps

Please try Mesa from https://github.com/iXit/Mesa-3D/commits/master repository (fix is not merged in upstream yet).

It should work now, little bit blinking background of steering wheel, otherwise no issues."

UPDATE: IT WORKS LIKE A CHARM NOW!
A8-3850 stock
w/o nine: 45fps
with nine: 65fps

no graphical issues or artifacts

Last edited by vinibali, Fri 26 Dec 2014, 20:34.