EVGA Killer Xeno Pro: The Impact of Network Offloading
by Derek Wilson on July 3, 2009 4:20 AM EST- Posted in
- Networking
Mostly Deterministic Testing
Designing tests to determine the real world benefit of the Killer Xeno Pro has proven quite difficult. Even though frame rate testing with single player games isn't strictly deterministic, proper tests can produce results that are fairly consistent and have low variance. We haven't included many MMOs or multiplayer games that don't utilize timedemo functionality in our graphics hardware tests specifically because they are very hard to appropriately benchmark. We can get ideas about performance from play testing, but graphs and charts have a certain finality and authority to them that we just don't want to lend to tests that we can't stand behind are representative of relative performance.
We did come up with one test that is highly reliable, however. This test is a side by side comparison of framerate when playing EVE online. We ran two different computers side by side with exactly the same hardware and software setup except that we installed the Killer Xeno Pro in one box. Both instances of EVE undocked characters in Jita (a system that typically hosts about 1000 players at a time) and flew to nearly the same spot. Because EVE allows players to choose something to "look at" and centers the camera on that object, were were able to have two instances of the game running with players very near each other (requiring very similar network data) and with exactly the same graphical load (because they were looking at the same thing).
Our EVE test is in a place where there were a very high number of other players and we were able to eliminate as many other factors as possible from testing. This test showed no difference in performance with or without the Killer Xeno Pro:
| EVE Test | Average FPS | Min FPS | Max FPS |
| Killer Xeno Pro | 84.3 | 67 | 99 |
| On-board NIC | 84.5 | 68 | 98 |
We attempted testing in other multiplayer environments like Team Fortress 2 and World of Warcraft, but we couldn't eliminate graphics as a factor when side by side testing with different players like we could in EVE. If we did sequential testing, one run to the next had very high variability even on the same hardware (due to the influence of other players).
We did run some tests in not very highly populated areas of WoW and found that framerate and ping seemed to show no difference. This might be different for highly populated areas, but again we couldn't be very deterministic in testing this.
In trying to do the similar testing with Team Fortress 2, the Killer Xeno Pro would be faster in once instance and slower in the next. There was no real consistency to our data in this case.
Bigfoot claims that there is benefit from the hardware in games like WoW, Team Fortress 2, Counter Strike: Source, and other games with high volumes of network traffic. We really do not doubt the capability of the hardware to provide some sort of difference, but our tests just are not deterministic enough to appropriately compare the hardware. But in a way this does tell us something very important: factors other than client side networking (like the performance of the network itself, other players, servers, and potentially graphics) have a much higher impact on performance.
The Killer Xeno Pro does suggest another advantage: bandwidth prioritization and throttling. The hardware is capable of Quality of Service (QoS) like prioritization on a per application basis, and every application can have upload and download bandwidth caps. This could potentially help out when multiple network heavy applications are vying for bandwidth. We decided to test this with both EVE (for framerate and download speed) and WoW (for framerate and latency).
In our EVE test, we used uTorrent to download a 650 MB file while we played EVE. Because we had to do this test sequentially rather than side by side (the bandwidth demand from on torrenting computer would negatively impact the bandwidth available to both PCs -- a point we'll come back to later), our frame rates aren't directly comparable because of all the other player activity. Please keep in mind that fluctuations in the multiplayer environment make this a non-deterministic test despite the fact that framerates are similar.
| EVE Test + Torrent | Average FPS |
| Control (no download) | 98.7 |
| Killer Xeno Pro w/ Prioritization | 98.4 |
| On-board NIC | 98.5 |
We did, however, see a very large difference in the time it took to download our torrent.
| Torrent Time + EVE Test | Download Time in Minutes |
| Control (no game) | 27 Minutes |
| Killer Xeno Pro w/ Prioritization | 69 Minutes |
| On-board NIC | 30 Minutes |
Since we can't get an assessment of ping times in EVE, we did some testing on WoW in the same unpopulated area. Normalized to the average latency we experienced while not downloading a torrent, here's the latency incurred by downloading a torrent:
| WoW Test + Torrent | Increase in average Latency |
| Killer Xeno Pro w/ Prioritization | 15ms |
| On-board NIC | 25ms |
Even with these latency differences, our framerates were very constant at about 54 FPS with 0.4% difference between the three different setups.
Again, this might have a larger impact in a more highly populated area in WoW. But the hardware does show a ping time advantage over our on-board NIC when downloading a torrent while gaming.
Facebook
Twitter
RSS
121 Comments
View All Comments
yyrkoon - Saturday, July 4, 2009 - link
Not to mention that ToE is not likely to make much of a difference in gaming ( if any ). Unless that machine is a server that has many high bandwidth connections, or the person gaming is moving files around from one machine to another while gaming (not very likely if they're pro gamers). I remember reading the original cards review, and claims from people saying something like how it improved their gaming experience immensely. I also remember wondering how full of **** these people were, or how much they were getting paid to make such claims. You're never going to see a difference that big, when you're pinging sub 1ms on a local connection, or much more than 10ms to a connection thats a state or two away. Even if there was, that difference will never be noticeable( assuming you're not dropping packets left and right, in which case you've got bigger problems ).People who game from home will see very little if any difference as their DSL, or cable service is much, much slower. Even if there was a difference, that difference would not be perceivable.
yyrkoon - Saturday, July 4, 2009 - link
Derek, all of Intel pro series cards should have ToE(TCP/IP offload Engine). Microsoft even supports a lot of these cards with special code for server applications. That is, I mean in their Server OSes. The Pro cards also have Link Aggregation, Fail Over, and high availability capabilities through software downloads from Intel.Now, I forget which model, but there is model from Intel that is basically two cards in one(two ports) that costs just as much as the Killer NiC here . . . and way out classes it minus the (oh joy) sound processing.
Anyways, I would expect a person doing a review on a NiC to know all of this already . . .
Qi - Friday, July 3, 2009 - link
Product page:http://www.intel.com/products/desktop/adapters/gig...">http://www.intel.com/products/desktop/a...ers/giga...
Product brief PDF:
http://www.intel.com/Assets/PDF/prodbrief/319831.p...">http://www.intel.com/Assets/PDF/prodbrief/319831.p...
And it's also somewhat aimed at gamers:
The Intel Gigabit CT Desktop Adapter allows you to take advantage of this dedicated I/O by combining Gigabit Ethernet with PCI Express to provide high-performance network connectivity for desktops with PCI Express slots. Make the Intel Gigabit CT Desktop Adapter your choice for applications utilizing rich media content such as video streaming, web applications, music, and gaming.
DerekWilson - Friday, July 3, 2009 - link
Again, nothing about network stack offloading ... just checksums and something about interrupts.mindless1 - Saturday, July 4, 2009 - link
The faster the system a nic is placed in, the less significant network stack offloading should be. IMO the article has it backwards, those benefitting the most would not have a certain level of higher end system, but at the same time it has to be remembered that networking in general is not an especially high latency scenario on the client side relative to wan latency.Spoelie - Monday, July 6, 2009 - link
Actually the article has it right..While the impact of the killer NIC might be more significant on a old/slow entry system, spending that $120 on other parts will still net you way more performance.
A killer NIC can only help during heavy network traffic, while a faster CPU is useful in almost any situation. So recommending this NIC to a gamer as an upgrade over other parts of his system is not a good idea IMO, only when there's nothing else to upgrade does this card become an option.
@Derek:
Maybe some other interesting tidbit is the impact on power consumption, since this is essentially a pc on a chip..
Qi - Friday, July 3, 2009 - link
The Intel Gigabit CT is a desktop NIC that offloads some of the work to my knowledge. This is on the Intel page:'Delivers increased performance while significantly reducing CPU usage'
And in the product brief PDF we can read this:
TCP checksum offload – transition control protocol (TCP), user diagram protocol (UDP), Internet protocol (IP)
DerekWilson - Friday, July 3, 2009 - link
This card does not offload the networks stack ...It looks like at most it offloads checksum generation / verification (though it is unspecific) and implements interrupts for some thing (though, again, it isn't very specific).
That card doesn't do nearly as much as the Killer from what I can see...
But, as per our article, anything as good or better than an on-board NIC is going to work just fine.
lyeoh - Monday, July 6, 2009 - link
You said:"That card doesn't do nearly as much as the Killer from what I can see..."
And:
"and with no significant difference between the baseline option and the Killer NIC, certainly no other add-in board is going to be worse ... and they aren't capable of being better. "
But you have provided NO PROOF that an add in card does worse or better.
We anandtech readers can make such claims as well. Forgive some of us for expecting Anandtech to do reasonably decent benchmarks - just like you do for other stuff.
If there's a very expensive video card that doesn't do much better than an integrated video, that does not automatically mean that a different add-in video card would do about the same.
At least with the SSD stuff Anandtech told the manufacturer "hey guys - this sucks", and the manufacturer eventually did something about it.
It would be good to find out which scenarios this card makes a diff compared to integrated NICs and other add in NICs (e.g. Intel, Broadcom, Realtek, Marvel). After all this card costs about 3 to 10 times the price!
GokieKS - Friday, July 3, 2009 - link
How many people do you know of that another add-in card NIC for their gaming system?For the vast majority, if not all, of the people who would consider buying this thing, the alternative is the integrated NIC on their motherboard.