Final Words

The approach and design of the Killer Xeno Pro are more efficient than standard network cards. Bypassing the windows networking stack will reduce load on the operating system and the CPU. Bypassing the CPU and OS when sending and receiving audio using supported voice chat software is a cool thing as well. Built in hardware prioritization (QoS) and bandwidth throttling are also interesting features.

But the bottom line is that none of this makes a significant difference in the gaming experience on modern PCs when paired with current games, nor does it offer an advantage over alternatives.

The biggest advantage the Killer Xeno Pro showed was in it's ability to prioritize games over other applications. At the same time, this only works for the one PC that is doing both downloading and gaming. If there are other computers on the network at all, it would be much more cost effective to purchase a router that can handle QoS and bandwidth control on a per application (or per port) basis. Using a router to handle this means that I can download huge massive files on one PC and my wife can play Team Fortress 2 on another without experiencing problems.

I could even play a game on the computer that I'm downloading with in that case, but it remains our recommendation to simply not download in the background while playing a game. More than just networking is affected by downloading in the background, as the harddrive is constantly being hit and this can be a significant source of loading pauses and hitching in and of itself.

If you don't want to spend any money, most torrent and other downloading applications come with built in (or add on) bandwidth controls that can be employed to achieve the same end as hardware QoS. Hardware QoS and bandwidth control are nice features to have, but they are not worth $120.

The voice chat acceleration could be beneficial when gaming while chatting, but currently most applications are not supported. Teamspeak, Ventrillo, and Skype all need out of the box support at the very least. At best we would want all games with built in voice chat to support this as well, but that isn't likely unless and until the hardware becomes more popular. In addition to application support, voice chat doesn't take up a significant amount of CPU time and the most significant impact on latency is still going to be the network as a whole.

TCP/IP offload is a better way to do things, but the benefit to the gamer just isn't there. Network load just isn't high enough to really take advantage of the hardware in modern games. But it isn't like the potential benefit of an NPU can never be realized: it starts to matter in the server space where technology like this was originally targeted. Offloading the CPU of a heavily loaded database server can definitely leave more CPU time for processing tasks and can increase network responsiveness. This just isn't what the Killer Xeno Pro is targeted towards.

So, when you've already got an on-board network card, is the Killer Xeno Pro worth $120-$130 USD? When that money can be put into either CPU or graphics, the answer just has to be no. At the same price as a Radeon HD 4850, there is just no reason not to look toward upgrading older graphics solutions. If you've already got something on the level of the 4850, then that money should be saved for your next graphics hardware upgrade where it will still have a higher impact on performance and experience.

For professional gamers and those obsessed with twitch shooters, for the gamers running 1280x800 on a 30" panel with most of the settings turned down on the highest end hardware money can buy, for those who are always after whatever option might give them the slightest edge: the Killer Xeno Pro might be for you. But even then, this hardware is the icing on the cake rather than a core ingredient.

What the geek inside me really wants to see is more general access to the hardware. This is, after all, a PC on a PCIe card. If Bigfoot gave us deeper access to the hardware, we might find more (even if equally niche) uses for an extra PowerPC processor in our computers. Additionally, to satisfy our intellectual curiosity, we would like to get our hands on a couple more of these cards in order to do some LAN testing using combinations of standard and Killer network cards to see how overall network performance is changed (if at all) especially with respect to voice chat performance.

Beyond this, there is a caveat. Perhaps, as broadband becomes more pervasive, game developers might want to push networking. At some point in time, games may need the PCs they run on to handle a much larger volume of network traffic in order to function well. Right now, game developers are targeting current bandwidths using current commodity network hardware. Games can't be designed to require higher performance networking gear because consumers either don't have access to high speed internet or they don't have a network card that does TCP/IP offload (among other things).

At some point down the line, something like the Killer Xeno Pro might become a significant requirement. But right now, for the vast majority of gamers out there, our advice is to save your money.

Experience Testing
Comments Locked

121 Comments

View All Comments

  • Pirks - Friday, July 3, 2009 - link

    "offload TCP/IP work from the CPU by bypassing the Windows TCP/IP stack" <<<--- THIS my friends if the proper fix for the infamous Vista MP3 network throttling issue. I wish this card were around when Mark Russinovich made his famous blog post.
  • Zolcos - Friday, July 3, 2009 - link

    Of course, internet latency is mostly dependent on the internet. Maybe I read this product wrong, but I always thought it was more for LAN gamers. After all, the client-side processing makes up a larger percentage of total latency on a fast LAN than over the internet. I'd like to see some game latency tests with a few computers on a gigabit LAN and no internet connection. Maybe even do a test with just a crossover cable between 2 Killer Xeno Pro cards to see "how low can you go".
  • DerekWilson - Friday, July 3, 2009 - link

    I'd like to do that sort of testing as well. We've only got one card in our labs right now though.
  • andylawcc - Friday, July 3, 2009 - link

    how much did they pay you guys to review this?
  • james jwb - Friday, July 3, 2009 - link

    the "Anandtech has gone down hill" troll remarks aren't going to reappear again, are they? The last few months it's been nice here...

    There was nothing wrong with this review. Bigfoot marketed it at gamers, it was reviewed with this in mind, and it's failed in that respect. The review was solid, if anything blame Bigfoot Networks. Got it?
  • DerekWilson - Friday, July 3, 2009 - link

    Really?

    I was actually interested in finding out if it was any good.

    And we did recommend that people not buy it ... so ... there's that.
  • crimson117 - Friday, July 3, 2009 - link

    Okay then... um... how much did their competitors pay you to review it? Yeah!
  • HerrK - Friday, July 3, 2009 - link

    It would be nice to see how good the ping reduction for torrent+WOW is in comparisson to a software solution like CfosSpeed, which does the trick at my PC for yeras now. And I would like to see CPU-utilization compared to onboard NIC´s for GBit file transfers in an internal network. I'm aware that you wanted to show the real performance benefits for a gamer, but you know...


    Last, since this is my first post here, I would like to thank you all for your great work at Anandtech, it is one of the sites I check almost every day, and be delighted.
  • DerekWilson - Friday, July 3, 2009 - link

    This could definitely be interesting and is something we would like to look into.

    we do really want to test with more than one card to see how it changes overall network performance.
  • hyc - Monday, July 6, 2009 - link

    In my experience, having done a lot of heavy load testing on servers, you're only going to see any difference when you're near saturation of the network fabric. I.e., you need to be pushing enough packets to be at over 60% of the network's packet-per-second limit before you'll see any performance difference from any offload engine. For gigabit ethernet the maximum frame rate (at minimum frame size) is about 1.488M packets/sec. At anything less than 10% network utilization I doubt you'll even be able to measure the CPU overhead of network processing.

Log in

Don't have an account? Sign up now