17 Replies Latest reply: Dec 9, 2011 10:07 AM by AEon_Xao RSS

Bandwidth usage on 360 - Does it explain away the supposed lag comp issue?

Some of you may find this information useful for reasons other than the intent of this thread, as some ISP's around the world still have max donwload caps per month or whatever.(I read somewhere that this is a big problem in Australia..?).  Those discussion threads are what led me to inquire about Mw3's bandwidth usage.

 

Many people have been blaming more lag in each successive CoD on the game being more complicated, more stuff going on, and therefore more network traffic.  It kind of makes sense, but without data it IS still only guessing.

 

As it turns out, my router has a monitor built in for how much data transfer there is.

 

In 30 minutes of hard gaming(backing out every lobby and re-searching, I had almost all mid-game joins, minimizing down time between data exchange periods during 1 minute lobby waits), I ended up with 7.5mb or so Downloaded, and 6.5mb or so uploaded.  Sure, I could test for a longer time, but in the end that leads to innaccuracies of it's own.  hosts rage quitting(leading to more wait lobbies and searching), the need to get up to pee, drink, smoke, etc.

 

I am assuming that I was not host.

 

This is on par with what I could google on Mw2's 15mb/hour rates.  I'll even be generous and up my numbers to an even 20mb/hour rate to account for games where everyone is dropping bombs and other air support, and to account for the lobby and search time.  Keep in mind some of this is post game as well, uploading of stats and such.

 

Lets get into some numbers on what that means to you and your connection.(here's the part that people with bad ISP's will want to pay attention to)

 

A low end DSL connection is 1mb(Or is it labeled Mb?  whichever).  In reality, when that user downloads a file, the best he can expect is about 100k/sec, and that adjusts on scale.  3Mb connections will download at about 300k/sec(assuming a strong source).(that's what I have for a connection)

 

Now, lets split up our generous 20mb an hour.  Divide by 60 for the per minute rate, then again by 60 for the per second rate.

5.5(repeating)k/sec.  HOLY CRAP!!! NO WONDER WE LAG.....wait a minute..

 

That's a little over what amounts to max....for dial-up 56k users.

 

Host will obviously see more traffic than these numbers as they have to sent out the information to everyone.  I rarely pull host at home so I cannot provide those numbers.   I'm sure the community would be grateful for a known host to contribute his numbers.  but again, we could be generous and make an informed guess. 

 

TDM is 6v6, and so host has to send info to 11 clients.  So at a maximum, we could multiply our results by 11, coming up to 60.5k/sec for an average rate host would need to utilize.  Still well under the typical 1Mb download max of 100k/s.

 

I don't see any support for the pro-lag comp argument that says "The game has more events, and therefore needs more bandwidth, and that's why it's laggy" as far as being a client participant goes(maybe host has issues with this, but I have doubts)

Any into-it gamer geek won't be suprised by those low low numbers.  FPS games are constant traffic, but never bandwidth heavy.  The aspects of connection that are far more important are ping and line quality(dropped packets and high delay screw with gaming).

 

/tests were done without using party chat and playing solo

//It's rumored(and I've seen in other games) that party chatting with the built in xbox feature can screw up a game's performance.  It's done that since Halo 3(all the way through BO), you see someone teleporting(bad lag in previous games would correct a players position with an instant teleport) and 8 times out of ten they were in a voice party.

///I dislike that they artificially made this game smooth.  You don't see much of that auto-correct of player location, only in really really bad lag spikes.  It was a useful tool in previous games to see who was lagging, if EVERY one was teleporting around, it was a bad host, otherwise it was just those that were teleporting.  It was a convenient tool to see if you or others were suffering a lag.

////pardon any mislabelings of k/s, mb, Mb.  etc.  I can never keep it straight which gets capitalized and which not.