Bandwidth Study
A while back I mentioned wanting to analyze different games bandwidth usage on Xbox Live. Not long after I made that post, I actually found some software and did the analysis. I've had these graph images on my laptop now for a good 6 months and am just now getting around to posting it.
Anyways, in the graphs, download is marked in green and upload in red. First, lets look at a graph for the game Gears of War.
This game isn't known for it's great network code. There are instances where the game lags quite a bit (one of the reasons I sold it). It doesn't look like the bandwidth usage is very high for the game, but it is very inconsistent.
Next is a graph from playing the arcade game Catan.
It may look very inconsistent at first, but if you look closely, you'll see a very consistent line for upload and download near the bottom. The spikes in bandwidth are likely due to voice communication in the game. It appears the game uses a very set amount of bandwidth that doesn't vary at all during the game.
Here is a graph from playing Halo 3 in the playlist Big Team Battle (12 players).
As you can see, the first part is during the pre-game lobby when not much game data is being transmitted, and the latter part is from during gameplay. Halo 3 is known to have some of the best network code out there for Xbox Live games, handling a myriad of issues that can come up during a game with grace. The bandwidth isn't that much, when compared to other games. A lot of grief players experience during a Halo 3 match isn't due to bandwidth, but more due to latency between different players.
You may not think the bandwidth has been very much so far. Here is a graph from playing Call of Duty 4.
The graph illustrates the different bandwidth requirements between when you are hosting a game or not. The first half of the graph is from a Team Deathmatch game where I did not have host. The middle is during the pre-game lobby and the latter half is from when I did have host.
As you can see, there's a big difference in bandwidth requirements when you're host or not. On the other hand, most broadband connections can easily handle that amount of bandwidth with no problem whatsoever. So, you may ask, why don't you get host more often? The game host is likely determined (in most games) not by bandwidth available, but by latency. So, you may have a 50 Mbps connection, but if you're out in the middle of nowhere and the latency is too high, good luck ever getting host.
Anyways, in the graphs, download is marked in green and upload in red. First, lets look at a graph for the game Gears of War.
Next is a graph from playing the arcade game Catan.
Here is a graph from playing Halo 3 in the playlist Big Team Battle (12 players).
You may not think the bandwidth has been very much so far. Here is a graph from playing Call of Duty 4.
As you can see, there's a big difference in bandwidth requirements when you're host or not. On the other hand, most broadband connections can easily handle that amount of bandwidth with no problem whatsoever. So, you may ask, why don't you get host more often? The game host is likely determined (in most games) not by bandwidth available, but by latency. So, you may have a 50 Mbps connection, but if you're out in the middle of nowhere and the latency is too high, good luck ever getting host.
Labels: Call of Duty 4, Catan, Gears of War, General Geekiness, Halo 3, Xbox Live

0 Comments:
Post a Comment
<< Home