• Welcome to PiBoSo Official Forum. Please login or sign up.
 
March 29, 2024, 06:11:35 AM

News:

GP Bikes beta21c available! :)


Dedicated Server Bandwidth Usage

Started by dibu, August 03, 2014, 08:40:54 PM

Previous topic - Next topic

dibu

August 03, 2014, 08:40:54 PM Last Edit: August 03, 2014, 08:56:32 PM by dibu
Peter's OldSchool race was the first event I could observe with beta5b and I'm impressed PiBoSo.  :)
The maximum bandwidth usage per user was less than 7.000 Bytes/s (upload). In previous betas it was up to 14.000 Bytes/s. Great improvement, thanks. :)

PeterV

And it was noticeable also, we had two players with 250 pings, with no troubles at all.
Server was very stable, not one core.exe from the 9 players that started.
This was with No Dynamic Surface, i am curious about with Dynamic now.

dibu

The interesting thing was also that the two 250ms players had a bandwidth usage of less than 3.000 Bytes/s while the other seven players used +6.000 Bytes/s.

BOBR6 84

normally high ping players bounce around and flicker on track.. everybody was rock solid on the track in that race  :)

Alone

Sounds gooooooood :D
I think that could be cause the Dynamic surface off, is the only way that high ping players could be in

Ian

Quote from: BOBR6 84 on August 03, 2014, 08:54:35 PM
normally high ping players bounce around and flicker on track.. everybody was rock solid on the track in that race  :)
For the first few laps someone was lagging around me especially noticeable on the start finish straight but I had the next highest ping at 120

HornetMaX

Quote from: dibu on August 03, 2014, 08:40:54 PM
Peter's OldSchool race was the first event I could observe with beta5b and I'm impressed PiBoSo.  :)
The maximum bandwidth usage per user was less than 7.000 Bytes/s (upload). In previous betas it was up to 14.000 Bytes/s. Great improvement, thanks. :)

Quote from: PeterV on August 03, 2014, 08:44:03 PM
And it was noticeable also, we had two players with 250 pings, with no troubles at all.
Server was very stable, not one core.exe from the 9 players that started.
This was with No Dynamic Surface, i am curious about with Dynamic now.

Iteresting indeed, even if Dyn Surf should be responsible of at least part of the bandwidth reduction.
It would be VERY VERY interesting to do the same race (same server, same track and bikes, same players as much as possible) and see what happens.
However, it is not sure that the instability issues associated with Dyn Surf do come from the increased bandwidth: if I recall correctly, Piboso wa smore angling towards a memory corruption (or something along that) occurring when Dyn Surf is used.

Thx PeterV and dibu !

MaX.

Vini

Hmm, the beta5b races I observed were very, very buggy.
A lot of core.exe and jumpstarts.