MSA TORA TCC Season 8 - General Discussion

Totally agree Chris, a good target size would be 12, with hard cap of 14, and then adjust as required depending on the numbers. Chances are it will fall around 12 or 13 for most lobbies, which in the past has worked out OK. It almost seems petty to argue about 14 or 15 cars, but the actual difference in data terms is quite large - 14% increase just for one extra car which is why it seems to push some connections over the limit.

Tweaking the last example then. What would you do with 29?

I like having a big field of cars, feels way more real that way and for me its peoples connections more than the amount of people in the race, just my view.  

CQR Aero wrote:Tweaking the last example then.  What would you do with 29?

29 cars in a TORA race, did you miss the 1 off the front  

Joking aside, the huge number of entries for big series gives that flexibility to run optimal lobby sizes that reduce lag and allow better racing.

To answer the question... if everyone had 100Mb fibre optic lines I'd happily race 16/15 cars in that scenario, but as things stand where >50% of people don't I'd go for a 10/10/9 split any day to avoid this (and ruin a championship):



And yes that was Chad in front...

Speed of the connection isn't really the issue. 100Mb fibre isn't any better than a good quality 8Mb ADSL line for XBL (ie no packet loss, good ping). It doesn't help that a lot of ISPs provide cruddy routers that can't reliably manage 200 concurrent connections ... co-incidentally around the 14 player unofficial cap.

It's the upload speed that's the issue TFR - many ADSL connections have slow upload (512kbit or 768kbit, 0.5Mb or 0.75Mb) which is marginal for more than 12 players (you need enough bandwidth to send data, voice etc to all the other players). Fibre connections typically have much higher upload speeds (5Mb, 10Mb etc), so can cope with more players more easily. Good point about the standard routers, some of them are awful, especially Sky's mandatory router!

Too many Magics round here.  

Fair point on the upload. I've used Forza on a Draytek router with traffic monitoring and it never appeared to exceed 400Kb up (or down) in-race with a 16 player lobby, but there were a high number of connections. That's on my 22Mb/1Mb Sky ADSL connection.

The old Sky white Netgear was ... problematic. I used to power cycle it before each race, and the uPnP was buggy. The new Sky Hub is a decent bit of kit in my experience and it's very stable, that said it's a one router fits all solution so it's designed for 80MB Fibre.

lol yeah Magic! Draytek are excellent routers, I'm sure if BT/Sky/Virgin supplied them as standard we'd have a lot less lag issues! I've seen some people who race here post there upload speeds of less than 0.2Mb, live a long way from the exchange so get terrible speeds. Roll on Forza 5 and dedicated servers, only uploading to 1 person (the server) then, so should make things so much better

Wish I did more of this in school rather than this  now.

LOL Warspite, learning some of this stuff would make you want to  ! There's so many factors that affect online gaming, especially P2P like Forza - connection speeds, routers, latency. It's not something you can ever solve, just try to make it as good as can be - TORA do a great job and we get great racing as a result.

I'm trying to follow it all best I can haha, but IT and Maths were never my strong point.

My idea of repairing a Technology hardware error, is to use a Hammer

I appreciate the time you've taken to explain it though, same to TFR Magic (Magic&Magic), it does help to see what the community thinks is best for lobby organising, and providing knowledge on pro's cons with such as Connections ect.

Chris does a Fantastic job to make sure every lobby is hosted and kept to a reasonable number considering Forza's limits, so all the honest opinions provided here is a huge help to such a difficult task.

If we can manage 14/15 player lobbies in TEC & that's with a streamer so why can't we for TCC?

I think the proximity of the cars increases lag - when cars are closer together (as they often are in TCC), data is sent more frequently to update car positions (same in shooters, where 'line-of-sight' affects the frequency of updates). It's why you often find there's more lag at the start of a race when everyone's bunched up together and it settles down after a lap or two as the cars become more spaced. 14 worked well in TEC and no reason 12-14 can't work well for TCC too!

@Warspite:
I'd highly recommend Microsoft Research's Donnybrook project, they looked into the effects of latency and connection speeds on online gaming and have used that information in building and improving Xbox Live:

http://research.microsoft.com/en-us/projects/donnybrook/

One of the main points raised in the "Maximizing total upload in latency-sensitive P2P applications" research was

"Utilization of the total upload bandwidth is the crucial bottleneck in distributed peer-to-peer games because in residential areas, broadband access is typically asymmetric with high download rates, but only small
upload rates. Thus, the higher the upload bandwidth utilization, the
more accurately the game play can be rendered."

And that's fundamentally what we see with Forza, we hit a limit where the slowest persons upload in the lobby starts to create a bottleneck of lag that affects the quality of the online game. I think some of this Donnybrook research has gone into the Xbox Live Cloud for dedicated servers to improve the gaming experience, I am impressed with the effort Microsoft puts into all this stuff, same with multiplayer matchmaking, Trueskill etc, there's a lot of detailed research goes into making our gaming work!

LMR WARSPITE wrote:Wish I did more of this :study:in school rather than this  now.

least you had computers I remember my old slate and chalk.......

S2M Ranger wrote:

LMR WARSPITE wrote:Wish I did more of this :study:in school rather than this  now.

least you had computers I remember my old slate and chalk.......

I used an abacus in grade school  

Don't lie Dewey you never went to school

S2M Ranger wrote:Don't lie Dewey you never went to school

Uh huh....... and surprisingly my old Amiga 500 computer from 1986 still works! Wished I wouldn't have lost the mouse to it. I could be playing Indy 500.  

DeweyTee wrote:

S2M Ranger wrote:Don't lie Dewey you never went to school

Uh huh....... and surprisingly my old Amiga 500 computer from 1986 still works! Wished I wouldn't have lost the mouse to it. I could be playing Indy 500.  

LOL. Found the PC version when clearing out the parents cellar the other week... disk wouldn't read.

DeweyTee wrote:

S2M Ranger wrote:

LMR WARSPITE wrote:Wish I did more of this :study:in school rather than this  now.

least you had computers I remember my old slate and chalk.......

I used an abacus in grade school  

Show off, I had to do my sums via the medium of cave paintings.  

CQR MAGiC wrote:

And yes that was Chad in front...

I was involved in this too (the Vodafone M3), i got bummed... royally.  

BG Beanz wrote:

CQR MAGiC wrote:

And yes that was Chad in front...

I was involved in this too (the Vodafone M3), i got bummed... royally.  

That's funny, I only remember winning that race  

BG Wumba wrote:That's funny, I only remember winning that race  

Gamertag change: BG Mr finish last in the first race, start the second race on pole and run away.

Whoops, longer than sixteen characters!  

After the stewards announcement about jump starts, was I given a penalty for races 1 or 2? If it was race 3 I completely understand as I did, by mistake go too early.

D4RR3N 3579 wrote:After the stewards announcement about jump starts, was I given a penalty for races 1 or 2? If it was race 3 I completely understand as I did, by mistake go too early.

Reading through the announcement, it seems that you have received a single penalty, so I would assume that to be for a single race. My guess is race 3.

Could I ask which race out of the 3 I jump started ?