Bigfoot Network’s Network Interface Card (NIC) is a PCI-e based gigabit Ethernet card which contains its own proprietary network processing unit (NPU) and Bigfoot’s Game Networking DNA™ technology to identify, classify and accelerate online game traffic in your PC. The NPU takes the network information processing load off the CPU and the software suite allows you to tweak your computer’s network traffic and the applications it is running in the background. In doing so, it seeks to reduce stuttering, freezing, mob rubber-banding and other symptoms of lag, improving game performance and responsiveness.
Hardware
The card is a compact unit with a PCI-e 1x connector that can also be used in any other free PCI-e slot on your motherboard, be it 1x, 4x or 16x. The cage over it serves to protect the components on the board as well as display the Bigfoot Networks logo, looking much more high tech than its predecessor with the ghetto-looking letter K over it.
Software
The Bigfoot Networks Killer Network Manager is installed with the drivers and is the software control interface for the card. With it, you can view your system information, your bandwidth usage, monitor your PC by viewing graphically, your CPU and NPU usage, Ping and FPS (Frames per Second using FRAPS if it is installed and active), view all your active applications and throttle or block them completely if you wish and you can also configure your network settings and the Network Manager features. A very useful suite of tools as it allows you to tweak what’s going on in your PC and fine tune it so that it’s dedicated only to the game you are playing – and of course whatever else you want running in the background.
)
)
)
Testing Methodology
I may be set up to do all sorts of tests and measures where it comes to sound and music, but for actual technical testing of PC hardware components such as a NIC card or even the Matrox TripleHead I previously reviewed, I have to decide what I’m trying to test, review and show. At MMORPG, the obvious test for the Killer NIC was how well MMOs ran on it. It all sounds good on paper, but the proof as they like to say, is in the pudding. I have a middling gaming rig and a rotten DSL connection that is prone to stuttering. Downloading the latest driver from the Bigfoot website directly, I plugged the card into a PCI-e 4x slot (the 1x slot was too close to my graphics card) and was ready to take it for a whirl.
I played a few MMOs during the weeks I was actively reviewing the card. APB, EverQuest, EverQuest2, LoTRO and Final Fantasy XI as well as a few betas that shall remain un-named. The test was conducted simply by moving my LAN connector from the onboard NIC to the Killer NIC and noting how smoothly (or not) the games ran while on either connection. Now… remember the rotten DSL connection? My experience with every game seemed to be smoother with the Killer NIC with less stuttering, especially when there were many other people on screen.
To do actual frame rate measures with as few variables as possible such as number of players in zone and on screen at a time, I created new toons and looked for empty zones. These were easily found in EverQuest. Consistently, I was running about 10 frames per second faster with the Killer NIC in EverQuest (yes, people still play).
Now I looked for further measures that could be simply accomplished. I looked to Pingtest.net and did the tests over several days and times of the day to eliminate any unusual server/internet events and threw out the extremes – like the 20% packet loss I experienced before my network went down one day.
Pingtest.net looks at three components of internet latency. Packet Loss – that’s when packets of data fail to reach their destination. Ping – the time it takes for a packet of data to hit the host and come back, and Jitter – shaky pulses of data, the variance in ping.
The tests were inconclusive for Packet Loss as pingtest could not measure beyond single percentage point differences and nearly all measures were at 0% loss without using the Killer NIC.
Ping tests were consistently lower when using the Killer NIC. Depending on the distance of the servers, the improvements could be slight or substantive, but the telling component was jitter. The Killer NIC was smoothing out the variances. Below is an indicative sample:
Server – Newbury Park in California, Aerioconnect at 950miles away from me.
I also stream video and music – enjoying internet radio while I’m working on an article or working around the house. It was great to have much fewer drop outs and decibel variances in my music. Using only the onboard connection, it sounded sometimes like there was a kid constantly playing with the volume control, especially if I’m also downloading or patching games at the same time.
Conclusions:
MMO performance and responsiveness does not only depend on your internet connection, but also have to do with your computer specs, and this was obvious to me as frame rates could be considerably different in the same game, depending on the population and graphics in the zone. I am currently running the following:
CPU: AMD Athlon 64 X2 Dual Core 4600+ RAM: 2gig Graphics: ATI Radeon HD 5770 OS: Windows XP Pro 32 Connection: DSL 2.5 Mb/s download, 0.5 Mb/s upload (not near what I’m supposed to be getting).
Any component you upgrade in your computer could improve its performance and responsiveness. The decision to be made is figuring out where to find the most bang for your buck. How much improvement do you expect for your dollar, and also where is it important to upgrade. After a while, you also reach the point of diminishing returns, especially when you have a fully tricked out rig. The Killer NIC 2100 recently dropped its price from $129.99 to $89.99 – a far more palatable price point than what it debuted at (about half) several years ago, and it works. Consensus among other hardware reviewers is that improvements are more dramatic for gamers with PCs of middling specs. Now it is up to you, dear reader, to decide if the improvement is worth the cost. For me and my dodgy internet connection? Totally worth it.
