Round 6 - Grosser Preis von Deutschland [POSTPONED]

Has the server been closed?

Yes, we will announce here when we will be running it again.

I'm ready to test, but there is no server available

Alberto Ibañez wrote:

WE ARE GOING TO RACE ANOTHER DAY, STAY TUNED NEXT DAYS IN THE FORUM.

Bummer, strange issue that.

The group C server Czesarius, we closed the Procar one.

Bueno , una pena no haber corrido . Llevo varios años como organizador y administrando servidores , así que se que con más de 20 pilotos se complica , y en este caso de distintos países . En mi caso siempre tengo ping muy alto , espero no haber molestado , y en esa situación sólo tienen que avisarme si se torna molesto .
Se que aquí hay gente con mucha experiencia en esto , pero pregunto como está el servidor de espacio (GB disponibles) , ya que esa suele ser una causa de mal funcionamiento . Un recurso que aprendí hace poco es eliminar todos los archivos .dds de los autos y las carpetas Sounds , el servidor no las necesita y ocupan un espacio realmente importante .
Igual no está demás revisar si no hay algún proceso extraño consumiendo recursos . Siempre suponiendo que hay algo más además de problemas de red .

We have been informed that another league suffered the same problems when racing at this track. We raced at Brands Hatch with the same mod, server, etc, and everything was fine, so it is likely a track issue. We will be adding an alternate version of Hockenheim (Same layout but converted rFactor track) over the week to act as reserve and to check against the vanilla one in the DRM race (Gold Star) next sunday. Once we are sure that one works, we will reschedule the Procar race.

Sorry for all the inconveniencies, if there was one track we never thought could cause us trouble, that was one of the vanilla content ones

Quote :

Se que aquí hay gente con mucha experiencia en esto , pero pregunto como está el servidor de espacio (GB disponibles) , ya que esa suele ser una causa de mal funcionamiento . Un recurso que aprendí hace poco es eliminar todos los archivos .dds de los autos y las carpetas Sounds , el servidor no las necesita y ocupan un espacio realmente importante .

También se pueden eliminar los archivos de sonido, pero el caso es que tenemos espacio de sobra, y también ancho de banda, procesador y memoria. La verdad es que el fallo de ayer no tiene sentido si no es por causa de la propia pista.

So, I am a bit confused now. Should I count Österreichring as Round 6, since the original round 6 was postponed somewhere behind?

I don't know as of now if the Hockenheim race will take place before the Osterreichring one or if the whole schedule will be pushed backwards. We will decide that later.

I just remembered that we also ran the 1971 F1-mod at this track for the Jochen Rindt Memorial race.

I do not think we had these issues then. So maybe it was just a freak night?

Timo

Maybe so. Yesterday Richard had to restart the server because of the problems you pointed out and which I confirmed after checking, so it might be that it had been autoupdating or doing a regular backup or God knows what during our race. But still, I got word that the track has caused issues in other leagues.

Alberto Ibañez wrote:

We have been informed that another league suffered the same problems when racing at this track. We raced at Brands Hatch with the same mod, server, etc, and everything was fine, so it is likely a track issue. We will be adding an alternate version of Hockenheim (Same layout but converted rFactor track) over the week to act as reserve and to check against the vanilla one in the DRM race (Gold Star) next sunday. Once we are sure that one works, we will reschedule the Procar race.

Sorry for all the inconveniencies, if there was one track we never thought could cause us trouble, that was one of the vanilla content ones

Glad the problem was found!
Great news that it wasn't an HSO or league member issue!
Would be good to know the specific issue with track!! And why it didnt seem to effect the 88 version later that evening with GroupC cars? Just less cars maybe?
Anyway, thanks for the efforts!

The ping increased noticeably with the Group C cars, but we never had more than 17 at the same time, and I think when it goes past 20 or so it's when it shoots through the roof. 

But we used the track with F1 71 and had no trouble, so it's still not clear to me what the mod has to have in order to cause those huge pings.

Alberto Ibañez wrote:

The ping increased noticeably with the Group C cars, but we never had more than 17 at the same time, and I think when it goes past 20 or so it's when it shoots through the roof. 

But we used the track with F1 71 and had no trouble, so it's still not clear to me what the mod has to have in order to cause those huge pings.

was thinking about this too, went back and double checked, grid was 19 cars. so unfortunately, can't really tell what would of happened it we had more than 20 back then too.

David Jaques wrote:

Alberto Ibañez wrote:

The ping increased noticeably with the Group C cars, but we never had more than 17 at the same time, and I think when it goes past 20 or so it's when it shoots through the roof. 

But we used the track with F1 71 and had no trouble, so it's still not clear to me what the mod has to have in order to cause those huge pings.

was thinking about this too, went back and double checked, grid was 19 cars.  so unfortunately, can't really tell what would of happened it we had more than 20 back then too.

Is it possibly keeping the higher poly count car models visible at too great a distance compared to other circuits? Is this one of longest AMS Included vanilla circuits?
Cant research any theories, not home

high values of LODS are in most cases that causes high fps with a ful field of cars

icm with 10.000 lods objects on tracks , like buildings billboards etc
it can hurt alot on fps

a save lod is 5000 for buildings ( they dont pop up in distance )

objects on car lods like wings (0.400) is a good value

but this is i know from rfactor  , AMS is alot different and most of all icm with those damn d3d3 plugins
its a big bug fest that hit very hard online on the fps

basicly its to find the balance in between track AND and mod and that takes much more people get involved in HSO who are dedicated to this task, however AMS is a strange platform to tweak this despite its rfactor orgin which i know the tricks how to solve this

LODs matter in your own computer, but the server is not using any graphics power, for it the race is just a set of positions in 3 axis. It communicates with your PC to keep track of where you are, but your own PC is who calculates your physics and displays graphically the rest of cars and track where the server tells it they are. 

The server lag happens when it has problems receiving/sending the info or it's too busy with other background processes and can't run the game quick enough (As f.e. while autoupdating Windows or other programs). The track can contribute here if it is bugged and causes contradictory info inside the server, f.e. when people's position as relayed by their PCs is outside the track, in the wrong sector, jumping sectors, etc.

the one thing that stands out to me about this track is the styrofoam bails at the chicanes. when one gets hit it splits into several smaller parts that all persist. can't imagine this is good for lag.

David Jaques wrote:

the one thing that stands out to me about this track is the styrofoam bails at the chicanes.  when one gets hit it splits into several smaller parts that all persist.  can't imagine this is good for lag.

Yes and not only in chicanes. It's also at entry and exit of stadium section.

David Jaques wrote:

the one thing that stands out to me about this track is the styrofoam bails at the chicanes.  when one gets hit it splits into several smaller parts that all persist.  can't imagine this is good for lag.

Yeah Mega-POLYs for nothing.
But based on Albertos explaination, polycount shouldnt matter. I just struggle to believe it doesnt somehow.

Also, Remmen comment about d3d3 plugins makes one wonder if disabling them could be a cure?

Brian Janik wrote:

David Jaques wrote:

the one thing that stands out to me about this track is the styrofoam bails at the chicanes.  when one gets hit it splits into several smaller parts that all persist.  can't imagine this is good for lag.

Yeah Mega-POLYs for nothing.
But based on Albertos explaination, polycount shouldnt matter. I just struggle to believe it doesnt somehow.

Also, Remmen comment about d3d3 plugins makes one wonder if disabling them could be a cure?

im not talking graphically, but wouldn't each individual piece of broken styrofoam need to be reported between the server and all client PCs the same as each car position would need to be reported? once a block is broken its like adding 10+ cars to the track instantly.

No

And that's all I can say, without giving clues to cheating 

I'm sorry, but I won't be able to race on Friday 5th. I'm on a business trip, I'll drive 700km back home from Germany on Friday afternoon. But most importantly, if I came home after being away for one week and jumped straight into the cockpit, I guess my wife might ask for a divorce    Enjoy the race, guys.

EDIT: Oh, it's 5th MAY, not April. Okay then

HSO doesn't race on Fridays Petr. Not until we do a 96 Hour race around the Nürburgring anyway.