TORA MSA GT Meeting 8 - Road Atlanta (21st Oct)

AFAIK, rFactor uses dedicated servers (i.e. one player is the host), so as long as that host has a fast upload/download it can support 100's of players (as each player is only uploaded and downloading from the host). Xbox is totally different model, no central host, so with 16 players in the room it is in effect the same as a 225 player 'hosted' room in terms of data transfer if you see what I mean lol

yep defo we need dedicated servers which we could rent out

Yeah! That is number 1 on my wish list for Forza 5 - please please please let us rent a server for running league races. Even people with slow (<0.4Mb) upload speeds would be able to race in 16 player lobbies then. The rented Xbox servers work well on Battlefield!

So if you need 0.8Mbit (800kbit) "16 cars" for lag free racing why is TORA asking for "512K upload is considered the minimum for the lowest form of lag free gaming" in the introduction?

I think a problem also is that people are using wireless connections and those are not so stable as a wired one.

That's only from my own personal experience and testing, 512kbit will usually support 10-12 players, but 16 requires about 800kbit. As you say, there are other factors too such as wireless connections, but these typically show as cars 'shifting' around or jittering, whereas flying cars and massive lag bombs are usually upload saturation.

To be honest, if there was a crash involving 5-6 cars in real life, they would throw the red flag and stop the race.

If it was real life I'd be dead

Ianmr wrote:But he'll my race was wrecked yet you don't hear me whining like everyone else

Where you see whining, I see feedback.

CQR MAGiC wrote:If it was real life I'd be dead

You don't have much faith in your car's rollcage do you.

14 players max, drivers that consistent lag and don't improve their connection which jeopardizes other drivers may not be allowed to race if connection improvements cant be made.

SelectiveRogue wrote:drivers that consistent lag and don't improve their connection which jeopardizes other drivers may not be allowed to race if connection improvements cant be made.

Will be extremely difficult to implement this as different people lag for each person. eg for me some americans lag but you can bet that if 2 of those americans raced each other there would be no lag. Only way around that is to have separate championships for europe and the americas until broadband gets a hell of a lot better or just accept that lag incidents are our equivalent of mechanical issues.

As for last nights 16 player lobbies, I thought they went well, only had one incident with BloodyCain due to a minor lag bomb. For me theres no difference if you have 12 or 16 players, the trickiest part is getting the racers in the room.

CQR HABURi wrote:To be honest, if there was a crash involving 5-6 cars in real life, they would throw the red flag and stop the race.

I would Agree with this if most of the lobby wanted a re start. In the situation Lobby A had I would have done a 9 lap race with the rolling start order of the last full lap.

That's an interesting idea Chaddy, never though of that before - not something you can do in real-world racing because of damaged cars, but could be worth thinking about for the future for virtual racing? Was really looking forward to having a battle with you again after our battle in race 1 and then BANG!!!

BG Hainesy wrote:

SelectiveRogue wrote:drivers that consistent lag and don't improve their connection which jeopardizes other drivers may not be allowed to race if connection improvements cant be made.

Will be extremely difficult to implement this as different people lag for each person. eg for me some americans lag but you can bet that if 2 of those americans raced each other there would be no lag. Only way around that is to have separate championships for europe and the americas until broadband gets a hell of a lot better or just accept that lag incidents are our equivalent of mechanical issues.

As for last nights 16 player lobbies, I thought they went well, only had one incident with BloodyCain due to a minor lag bomb. For me theres no difference if you have 12 or 16 players, the trickiest part is getting the racers in the room.

if its an upload issue you can get an idea of some 1 by doing a speed test etc

Indeed, a mandatory speed test from www.speedtest.net as part of the sign-up process is a very good way of spotting potential lag issues (other series I've raced elsewhere use this system, and amazingly I've seen upload speeds as low as 0.18Mb )

e.g:

This would help if peoples connections were a constant but theres nothing to say that if you post up a test showing say 10m d/l and 1m u/l that your connection wont fluctuate. ADSL connections fluctuate all over the place, wireless is another cause of fluctuations. Something as simple as a microwave can cause a wireless connection to go ballistic so for me these tests are only valid at the time of the test.

Racert46 wrote:
I had rather been in C lobby with 12 drivers and have a proper race then this lag issue in B lobby with 16 drivers.

Actually mate, there was 16 in lobby C and some epic lag in race 2. I'm putting the replay on store front tonight...watch out for the (Bora/Porsche/Space hopper which started just behind me.

I have to say thanks to the guys for letting me host. Nerve-racking and I think that affected my performance...that and 100% aero / 70% transmission / 90% Engine / 70% Brake damage...and everything else was about 50% coming out of turn one...lap one...yeah...I'm a bit sour.

Oh, I'd like to thank the top 4 lads in Lobby C Race 1 for the clean racing though. Some good stuff and some good moves, was a good battle to be in the middle of.

BG Hainesy wrote:This would help if peoples connections were a constant but theres nothing to say that if you post up a test showing say 10m d/l and 1m u/l that your connection wont fluctuate. ADSL connections fluctuate all over the place, wireless is another cause of fluctuations. Something as simple as a microwave can cause a wireless connection to go ballistic so for me these tests are only valid at the time of the test.

ADSL upload speeds shouldn't vary by any great degree, they're typically fixed at specific profiles (384kbps, 512kbps or 768kbps), downloads do vary more but they're not really the issue that causes these lag problems.

@ Graftersroom,

What i mean is that i rather race with 12 drivers and less lag than full lobby with lag.

With yesterdays quali result normally I would have been in C lobby (29th or something)but because of the 16 people lobby i was just in the B lobby.

And it seems that in every lobby there was massive lag. Maybe because there where 16 players in the lobby.

End of the day, it's a game working through a server with everyone using their individual connection. Sometimes can't be helped. Like Magic said, least we're all not dead & we can all hope that Turn 10 will give TORA a free server for Forza 5.

Graftersroom wrote:End of the day, it's a game working through a server with everyone using their individual connection. Sometimes can't be helped. Like Magic said, least we're all not dead & we can all hope that Turn 10 will give TORA a free server for Forza 5.

exactly

Here is my test. Taken at 6 pm GMT when everybody is going online in Holland.....



And for your viewing pleasure the lag incident.

can i just say, its all well and fine posting speed tests to your closest server, but for a more accurate reading, put it to the LOBBY HOSTS nearest server, and you will see some very different results

Aside from the two lag spikes that both gave me slight damage I had two great races. I think Hainesy and I were within 200ft the entire two races. Had a great time moving up thru the field with these guys. Great driving in lobby B!

Really wish the races were longer. My replay was only 21 mintues long. Why are we not maximizing the length of races based on replay times? These races are too short. The second race if anytrhing should be longer than the first to give the reverse grid drivers the chance to catch the leaders. I've now seen two straight weeks where a driver finishes last in the first race and cruises to a win in the second. Something not right there.