• Welcome to PiBoSo Official Forum. Please login or sign up.
 
March 28, 2024, 02:40:06 PM

News:

GP Bikes beta21c available! :)


Jump start/Black flag/Stop&go.

Started by Grooveski, August 11, 2017, 11:28:07 AM

Previous topic - Next topic

Grooveski

Quote from: PeterV on August 14, 2017, 07:51:47 AM
The pitin and pitout area are not defined/setup properly on most tracks.
Thats the reason the ride throughs often dont work.

That'll be it.  ;)  There appears to be a short stretch with the wrong surface inbetween PIT and PITOUT.



...or what I'm presuming are PIT and PITOUT.  Hard to tell with the three sufraces being the same colour in TrackEd.  There's a break in the blue bit though which can't be good.
Weird thing is I don't recall the pit speed indicator going off and on again, which you'd imagine it would.

Hawk

Looking at that surface map may not be telling all how it is, if you know what I mean...... In that the red surface you see at PITOUT maybe actually underneath the PITOUT surface but is showing above it for some reason.... probably a bug in the exporter not being able to define the separate surfaces when very close together(probably the same reason we get flickering signs?), but also because some tracks have surfaces underneath others that really shouldn't be there.... Lazy modelling in my opinion..... Probably as a result of BTB track creation?  :)

Just my thoughts and by no means a statement of fact.  ;) 8)

Hawk.

Grooveski

August 14, 2017, 02:16:45 PM #17 Last Edit: August 14, 2017, 02:48:28 PM by Grooveski
[Edit - removed overly harsh comment about the model  - Thinking about it the 'duplicate polys' are probably for decals and are just flickering because of the flickering bug]

Hope I haven't derailed the thread too much  :-[ - that's one bug that's probably not a bug but there are still another half dozen listed that are.

Hawk

Good point Groove, about that red surface in the PITOUT section could be a decal surface layer. I think your right mate. ;)

Hawk.

PeterV

Quote from: Hawk on August 11, 2017, 02:51:38 PM
I noticed that on some tracks when going to the grid to start a race the bike literally drops down onto the track surface and if your the last bike onto the grid it will either give you a jump start straight away or a ride-thru penalty.

This seems to be a problem yeah, last bike too "drop"on the track is always moving a bit.


Just too show what we mean, its a big drop, and with a reasonable full server this causes trouble like DSQ


https://www.youtube.com/v/6re1cNoWnhA

PeterV

Here is another example.
Server set with Bandwidth setting 3 and player pings reading 30 and 60 in game.

Note: we also tried going to track with the brakes applied, at that time we did not get a DSQ
but the bikes still dropped onto the track like shown in the video.


https://www.youtube.com/v/kybMjun0Rs4

Napalm Nick

Good vids for Pib to see the problem  8)
"The post you are writing has been written at least ten times already in the last 15ish years. Its already been reported, suggested, discussed, ignored or archived (but mostly ignored). Why are you doing it again?"

PeterV

September 12, 2017, 06:55:29 PM #22 Last Edit: September 12, 2017, 07:36:04 PM by PeterV
Server:
CAWS Special #1
maxclient = 24
bandwidth = 3
Location Germany, provided by Dibu
report from Dibu, bandwidth when 16-19 players were online. The upload was 1,5 - 2,1 Mbit/s. Very good values. Memory usage was ~550 MB, also OK.


Connection:
Here is an example of the connection lag happening, the ping goes way up and then goes down in steps it seems.
Have a look at player Manu and player Matty connection ping.

After Matty you see the start of our race with 18-19 riders.
Tchemi in the bottom left disappears for very short times, i observed that more times with him.
Close too the start we see one or more riders (Tonygas, Tusrat ?) fooling around on the front row.
So the jump start there is to be expected, but Uberslug in the right bottom never moved and got a jumpstart.


Session(s):
Dont know if this is the correct place but i also observed lag / stuttering in the 30 min. practice session.
they where periods of 2 to 5 sec. in lenght, and riders where in server at the time.
The server was restarted 15 min. prior the start of the event.
I mention this because i did not experience the same behaviour in 15 min Qualyfing or in the 16 lap race.


Core:
3x Napalm Nick One in practice when i went to Settings, One on my Qualifying out lap, One going back to pits after race
1x Tchemi I had not installed the CBR, The game stuck on 'syncing' and I had a core.
2x Tommaso Levato I had one just before qualifying and one after coming back to the pits after the race.
2x finpower I had 2 times core-exe yesterday evening. both times in the practice session.
1x PeterV   in the practice session.

https://www.youtube.com/v/5mBs_zL8G3s

Napalm Nick

Another good vid Peter.

As it goes hand in hand...
Reminder for everyone here
"The post you are writing has been written at least ten times already in the last 15ish years. Its already been reported, suggested, discussed, ignored or archived (but mostly ignored). Why are you doing it again?"

BOBR6 84

Wow  :o iv not seen this happen yet on the grid.. but i see this happens when people spawn in the pits..

dibu

Thank you for the report Peter, it was interesting for me to see what happens at the other side of the server.

To complete your report:
- the replay size was 512
- the dedicated server didn't core during the whole session

In general the bandwidth usage of beta12(b) is much lower than it used to be in previous betas. Older betas used 2,5 - 3,5  Mbit/s for 16+ players and went nuts after a short while.
The stability of the dedicated servers seems to have improved too.

Hopefully PiBoSo finds the reason for the high pings during connects and the jumps when entering the track.

Did you also experience such a high number of cores at the first test race (Victoria /default bikes)?


PeterV

Quote from: dibu on September 13, 2017, 06:22:18 AM
Did you also experience such a high number of cores at the first test race (Victoria /default bikes)?
As far as i know we did not have any cores or stuttering in that race, syncing was better as well.