I noticed with B10 that when you are in hotlap mode, if you pit in, when out of setup screen you do not restart the lap as it was the case before, you are in the pit and recording is no more active (normal behaviour).
And this is annoying. For example, I was trying to overtake a car, which turned right, directly into me. I did brake to avoid contact, and then I see him teleporting some meters in front, so he actually never turned into me. Just a braking can make you a loss of few seconds, or in many drivers race like cargame - you can loose some positions and even cause accident.
Do you have any figures for the numbers over the cars, as seen on a computer where you are seeing a lot of left to right glitching? That is the number that should describe how much glitching you see on your computer.
Obviously I am completely wrong sometimes, but I'll need some scientific convincing on this one, as I can't see any reason why glitching would be worse if the overhead car numbers are the same, unless I changed the physics, but all hotlaps still work so that means I didn't change the physics.
Remember that sometimes you can install a new patch, and the same day there are internet problems. So the person who installed the patch and has internet problems is then convinced that the problem is the patch. We do often get this at patch times.
I know, we had problems for 2 days on B9, then updated to B10 and the problem persisted. As many mates complained the problem is either at our hosting or in LFS itself, either way I hope to find the cause & solution.
I'll do some tests with network debug enabled to be able to tell you the exact ping figures.
Thank you Whiskey and Krammeh, for trying to examine this closely. I'm interested to hear what you find when you do the comparative tests. I'm not sure that a replay would help, because you might not see the jumping in the replay, as there is no latency in an MPR. The glitching in an MPR is only due to the time interval between position packets, without the transmission delay added to that. However, if a B10 MPR looked significantly worse than a pre-B9 MPR, that could indicate missing position packets, which is the only other thing I can think of that would cause cars to glitch worse than usual.
Hi Daniel. Don't worry, I check my mail several times a day. I have added your comment to some notes. It is a good find - thank you for for that. I'm going to see what I can do.
But I don't really want to reply to forum posts or emails just to say that I have read them. I just want to reply if I need more information or after investigating, when I have something more interesting to say.
I understand it may seem rude that I don't reply to posts that people have made, it may look like I'm ignoring them. But I'm reading and noting every post. If I did reply to every post then there would be loads of posts from me through the test patch thread saying "thanks, I've added it to my notes" and that would be a form of spam by me that would make the thread harder to read.
Shannon (failure) from NZ? He was just on our S2 server, used him as a test object and for me its really hard to tell the difference. Lag hits/crashes are just as bad as before I would say (0.5).
Also madcow was on, he is from AUS.. Went a little bit better, his line is a bit more stable (0.3) but hmm a <B9 versus B10 test would be ultimate I guess.
* Btw: those "TCP AV : InitTime += 1" messages spamming the chat, is that really necessary during shift-F8 mode? Its a bit distracting during driving and I wonder if somebody knows what its trying to say anyway. I have no idea.
I've made a small change which could possibly improve the apparent car lagging.
Although we don't have scientific results for it, and without thinking too hard about it, I just thought I'd try this small change, which is to send the cached multiplayer packets after the packet receive but before the physics loop and graphics draw. The idea is to send most packets, including the timing packets, more promptly.
It seems to result in lower apparent ping times, so it may keep the guests more accurately in time with the host. Though I have no real reason to think that car lagging will be improved by this, it is possible.
To implement some better cheat protection, we'll need to test an incompatible patch, but this version is still compatible so I think it would be a good idea to release a compatible B11 this evening.
Other changes :
Ping / lag time now shown in ms when less than 1 second
Ping / lag time is now also shown in game setup screen
In B10 hotlapping did not restart after leaving the pits
InSim button ids are not used for the internal LFS buttons, so there is no conflict. The lag numbers are only taking up space previous used by wider user name buttons so there is no conflict problem.
I don't think the shift+u interface should disappear when a certain id is used. I guess it disappears when an InSim button is placed outside the "recommended area".
Please try again with this evening's version.
I don't know when that will be changed, but definitely not in this round of patches. It is due to current limitations of the collision detection system. Not really a bug, just something that is not covered by the system.
What is the best way to debug each build you give us Scawen as far as the networking flow? Do you have special developer tools that are built into the game's exe that would allow us to see more information while we are on each server so that we can give you better feedback? As you've said yourself that MPRs are not really useful as they contain engine events from the server side, it does not give an exact representation of what each client was feeling during the online session. If we could get more information during our online sessions maybe we could give you better feedback if the packet / buffer system is working better then the previous versions.
Do you have any command line switches, or keyboard shortcuts that have not been disclosed that would give us more information then a normal client would normally set when playing?
Also, if you could please report ping time's in Milliseconds only, as this is standard notation I think that would generally help quite a lot with people's understanding of what is going on.
No, I don't have any more info, just the ping / lag info and the latency (over car).
The ping is displayed in milliseconds in B11, up to 1 second.
Then we are in the range of "lag" (time since the host last heard from the guest) rather than "ping" (round trip time). And it is displayed in red, in seconds that are steadily increasing. I think that "23" is more informative than 23000 and it fits on a small button. I think you'll find it ok, I think it looks quite good now.