The online racing simulator
I did some small testing and I have a question. I had (according to LFS) ping 80, other player had something around 150-170. Is that enough to make his car fly after we collide with each other? (which we did). Just wondering how it works...

And also another thing. After joining attempt to a full host, I get "Did not receive track info", which never displayed in previos versions. Is that normal now or a bug?
'Did not receive track info' message now appears in chat field after failing to connect to a full host - B11 client and host.
Quote from baartlomiej :Host is full is as expected, but you get "Did not received track info" in the chat box

http://i.imgur.com/2tkrE.jpg

OK, I've reproduced it now.

This message comes up if there are admin slots.

It seems to be like this :

1) maxguests=2 / adminslots=1 : 1 guest joins - next guest gets "Did not receive track info" message.

2) maxguests=1 / adminslots=0 : 1 guest joins - next guest does not get the track info message.

EDIT : Doesn't seem to cause any problems. Do let me know if there is anything else.
Quote from DANIEL-CRO :about ping while connected localy, maybe its a bit lower ( minimal 10 ms ) but still maximal of 110 ms seems a bit too high

Just tested this on my machine - localhost dedicated server, same version as client, Win7 64bit

B9: 0.01-0.02s; mostly 0.02 about 80% of the time
B11: 10-20ms; mostly 0.01 >90% of the time

I haven't seen more than 20ms to a localhost server, although I did see 30ms (B9, Wine) or so to a LAN server sometimes, which is a little strange as real ping time is ~400µs.
B11 LAN server seems fairly solid at 10ms (>80% of the time), never above 20ms while on track.

Reported latency to Internet servers seems to be about OK in B11, mostly 40-50ms (max 60) to a server with real 36±1ms ping.
Quote from Degats :I haven't seen more than 20ms to a localhost server, although I did see 30ms (B9, Wine) or so to a LAN server sometimes, which is a little strange as real ping time is ~400µs.

That's gotta be game engine lag time when it comes to getting around to processing the packet. Like DANIEL-CRO said, the game engine's resolution is centiseconds. Really I think that the LFS engine is just more honest about it then other game engines.
Making the ping numbers in the connection list appear right indented would make them more easily distinguishable, like in main menu for example.
[RIGHT][INDENT]60
120
330
70
210[/INDENT][/RIGHT]

Quote from Dygear :That's gotta be game engine lag time when it comes to getting around to processing the packet.

Yeah, that's what I figured for the 10-20ms readings, although I thought the 30ms was a bit high. I guess it's just an obscure combination of timing and maybe CPU load.
Quote from Nilex :Making the ping numbers in the connection list appear right indented

That's a really good point, and would be a great idea to implement.

Quote from Degats :Yeah, that's what I figured for the 10-20ms readings, although I thought the 30ms was a bit high. I guess it's just an obscure combination of timing and maybe CPU load.

Well the over-under for ping time is anywhere from 0ms to 40ms in it's most extreme case.
I finally understood why my local ping was so high. Its because my FPS was limited to 10 (LFS instance only for testing on laptop). Without FPS limit (about 40 fps) I got maximum ping of 30 ms
Also other thing I noticed is that server list loads like 5 times faster in case of unlimited FPS, is it possible to make load time independable of FPS ?

edit: also in case of FPS limit you will see pings to servers only like 100,200,300, ... ms
Quote from DANIEL-CRO :I finally understood why my local ping was so high. Its because my FPS was limited to 10 (LFS instance only for testing on laptop). Without FPS limit (about 40 fps) I got maximum ping of 30 ms

This reminds me about Scawen's comment (http://www.lfsforum.net/showthread.php?p=1755308#post1755308) about decoupling the physics & redraws from the packet handling in B11. But that should have meant that you wouldn't have seen this high ping behaviour since B11 came out, and the way I read your posts, you did see it with B11. Maybe this is a clue that there is still a problem lurking.

Or... maybe the high number you saw was purely "local" info and wouldn't be reflected in what other people see for you? Scawen will know I guess. (Does 10 FPS mean the physics loop really only runs at 10Hz too? Surely not.)

And as for loading more slowly with low FPS, that sounds like an accidental (and pretty surprising too! ) side-effect. Funny though.
Quote from Neilser : (Does 10 FPS mean the physics loop really only runs at 10Hz too? Surely not.)

By some short test I've done it seems physics calculates only for each frame, which generaly is fine for CPU time. But thing that FPS affects ping to servers, etc. could be maybe improved. We'll wait to see Scawen thoughts about this "issue".
Quote from DANIEL-CRO :By some short test I've done it seems physics calculates only for each frame, which generaly is fine for CPU time. But thing that FPS affects ping to servers, etc. could be maybe improved. We'll wait to see Scawen thoughts about this "issue".

Isn't this just a limitation of the FPS limiter implementation rather than the game engine itself? If the entire game loop recalculates every 1000/limit milliseconds when the FPS limiter is set, it's logical that you see pings only as multiples of the delay between two iterations of the loop. For instance, if you set the limiter to 35 FPS, the only ping values you'll get are 28, 56, 84, 112 ms.
Quote from MadCatX :Isn't this just a limitation of the FPS limiter implementation rather than the game engine itself? If the entire game loop recalculates every 1000/limit milliseconds when the FPS limiter is set, it's logical that you see pings only as multiples of the delay between two iterations of the loop. For instance, if you set the limiter to 35 FPS, the only ping values you'll get are 28, 56, 84, 112 ms.

I agree, but I'm trying to suggest to remove FPS limit while loading server list, or to make like separate thread which will ping every server and when comes time to render new frame to show all server which replied untill this moment.


One bug I noticed in B11 hotlaping:
Start hotlaping on AU4 (you will see start lights), change to BL1 and you'll have lights on BL1, and you can get penalty for starting earlier :P . Maybe to completly remove lights from hotlaping
Daniel, you should get a job as a tester!!!

Say you'll find all bugs, now matter how obscure.

Demand very high pay!
Quote from Scawen :Daniel, you should get a job as a tester!!!

Say you'll find all bugs, now matter how obscure.

Demand very high pay!

And make sure you recommend him
Yes, as LFS tester
Scawen, as B12 and all next will be incompatible, could be possible to add in InSim, a field that indicate what language is using the driver? I think that having something like eng spa fre... 3 letters should be enough..

That could help a lot servers that runs differents InSims programs, so that InSim program read that info, and show to final user his language, or any default one..
I suggested that before but then all InSims will be incompatible. => no option!
Quote from cargame.nl :I suggested that before but then all InSims will be incompatible. => no option!

That's a stupid excuse not to add something new...

The InSim systems can be easily updated - I don't see a problem here!
Excuse? Why didn't you have such a big mouth last week?

There are a LOT of InSims out there, some are very old but still in use, like TV director, Airio and who knows what. All relying on dll's. Its stupid to break compatibility for such a small thing. You are selfish :thumbsdow .

It can be done but for a bigger release / fresh cycle of patches.

And like I said, I requested it already nearly two months ago at the beginning of this cycle. It didn't make it. Probably because of the reason (not excuse, reason) I just mentioned. End of story.

.
Yeee, no need for anymore LFS updates!!! Current LFSWorld stats will be wiped and we need to keep 'em!
Breaking InSim binary compatibility without a really good reason is obviously not the way to go. I guess that supplying the locale info in a new optional packet would be the best solution. I for once agree that it could be quite useful.
Quote from MadCatX :I for once agree that it could be quite useful.

I agree but well.. Adding NCN with IP/language info is nicest but would render all InSims useless.. New packet with user info and announcing that after a NCN or requesting it with a TINY request? It's a solution but there is some overhead involved and I believe Scawen is allergic to overhead
This thread is closed

FGED GREDG RDFGDR GSFDG