A Network Card That Gives You the Power to DESTROY

A few months ago I blogged about add-on cards, and more specifically about how I think the technology behind them is the soul of the enthusiast industry. In that article I mentioned the KillerNIC, a product from Bigfoot Networks, as a great example of how smaller companies are key innovators in the market. The stars just hadn't aligned correctly for me to try this product personally until now, even though I've been dealing with Bigfoot directly and we've been selling these cards in systems for over a year. Well, over the past few weeks I've been rocking it out with this thing, and I've messed with it enough that I'm satisfied that they've got something here that's more than just a fancy heatsink... Although the bling is awesome. I've come to the conclusion that if you don't have a KillerNIC, you will be pwned by someone that does. It's that simple.

First, the quick once-over in case you haven't heard of this product: The KillerNIC is a Linux computer on a PCI card which bypasses the Windows subsystem and delivers network traffic directly to your game. By bypassing all of that Windows overhead with a dedicated 400 mhz processor and specialized software that prioritizes gaming traffic over everything else, you are ensured that your computer and/or your bloated OS aren't slowing down your online action. That's about as technical as I can get for you. If you understand Linux and networking deep enough to understand how this thing works much more than that, you've reached a level of geekdom that I haven't yet attained. Kudos to that! Myself, I like to call it magic.

There are a ton of reviews out there for this thing. I don't even know what to make of them all because the testing of this product seems to be hard to quantify. How do you measure such a situational, dynamic element as online game lag when every second on every server is constantly different? Apparently no one knows. Many people have spent far longer than I would taking manual ping measurements and what have you, and then finally come up with very little to show -- yet have eerie feelings of smoother gameplay. Apparently this very thing makes for the world's greatest hardware scandal on Slashdot. For more technical reading, there's a good article at Nordic Hardware that is very in-depth as well as a whole list of further reviews on Bigfoot's website. I have to say up front that my purpose here is not to write yet another one of those reviews. Those guys know what they're doing and I can't add anything to that conversation.

Besides, I don't believe in data. After all, everyone knows that science is the devil.

I have a better method. Here's what I did: I played with it and pwned with it. I jumped on a CS:Source server on a small map packed with as many people as I could find. These are typically the laggiest servers and I figured I'd put myself into a position where I was definitely going to see a difference if there was one. Well, there was. It was noticeable immediately. I experienced drastic accuracy increases, headshots were coming far more frequently, dodging was far more effective. In CounterStrike you can die with one bullet from the weakest pistol if someone gets you up close in the face, so when you come around a corner and a dude is standing right there, it frequently comes down to the faster shot. What I was feeling was that it was way easier to get that faster shot off. It seemed like I was somehow a split second ahead of my opponents on the server, and in a fast-paced FPS like CounterStrike, it's a sick advantage.

I've switched back and forth numerous times between the Killer and my onboard, trying to get an accurate feel for the difference. I go to the same servers playing the same maps with just minutes in between during the changes. I even try to stay to the same areas on the maps so I can get the closest comparison. It's actually funny that since I learned how much lag had to do with my effectiveness, I learned to avoid the types of situations where I knew the lag was screwing me up (close-up encounters, or sniper vs. sniper battles where the faster shot wins). On the flip side, with the Killer, I learned to exploit those situations. And moreover, I can even say that the card helped give me a higher level of immersion, because the game is working and looking like it should. It removes a lot of frustration that occurs when all of sudden you're dead when you get shot by a guy you didn't even see because of a split second of low frames.

I have been in touch with Harlan Beverly, the creator of this product and owner of Bigfoot Networks, through the review process. He explained to me that there are many different "types" of lag, and that the KillerNIC is specifically reducing client and network lag. It's reducing the computer's time and effort to read network packets. This is particularly true during moments of intense action, which just happens to be when you need it the most!


I also tried testing this on Frontlines and Battlefield 2142. The problem is that these games are a lot different and accuracy isn't as much of an issue (they are a different breed of FPS, and the tactical element is much, much less). Plus, both of these games are plagued with horrible server lag, constantly, and nothing on the client side can fix that. Regardless, a slight improvement was still noticeable, specifically with the reduction of the need to lead shots, both with automatic weapons and sniper rifles. I found that on these games, playing with less players on the server produced more of a difference from what I am assuming was overall less server lag.

A few more notes: I have this card in a machine that I'd class as a very capable, mainstream gaming machine. It's got 4GB of DDR2, an e6850, and an 8800GT. This machine handles these games like a champ, and these are the kind of specs that are specifically rated as able to benefit from the KillerNIC. If your machine isn't up-to-date and handling games well by itself, the KillerNIC isn't going to do squat for you. Furthermore, Bigfoot acknowledges that certain games benefit much more than others with the card. I obviously saw this in action. Whether it's coding issues or how certain games handle network traffic themselves, that has an impact on how much the KillerNIC can help.

Yes, I love this card, but is it an obvious choice for everyone? Well, I'd say there is certainly a targeted niche here. The price tag is considerable with models MSRP'ing for $199 and $279 (although you can find them for less), but if you fit the mold that this card addresses, I really think it's a no-brainer. You need to be a competitive, online gamer, playing the specific games this product helps with a capable machine to run it. The fact is that for myself, I wouldn't want to play FPS games without this card anymore. It would not be as much fun for me at all - and that's definitely worth something.