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
Exar3342 - Monday, July 6, 2009 - link
It is interesting how they don't note any of this information on the box, in the marketing information, or on the EVGA website for the card. This sounds like one of those "bad suprises" you get AFTER you waste your money on this and find out it doesn't do anything. The theme of the review isn't that the product doesn't do what it says it does (it does that well) but that it doesn't make an appreciable difference in the real world. It appears clear this product has paper benefits only, and no real gains anywhere.hooflung - Monday, July 6, 2009 - link
This is not a problem. The linux networking stack is more efficient than the XP stack. Even though the NPU isn't doing it in hardware its still not being done on XP.This is why the whole 'offloading' wording is confusing. The card is bypassing the OS stack, thus 'offloading' it to the card. The card then 'offloads' the UDP and 'some' but not 'all' TCP to hardware routines.
While EVE might use a bit of UDP for non critical things such as polling the market but it doesn't help speed up combat which needs the TCP/IP to poll where your ship is in space, what gun you fired, where your enemy is etc etc etc. If you do not have a constant TCP/IP connection, not UDP becuase UDP doesn't require you to have an active connection, you will be booted off the EVE server.
Its as simple as that. Derek should know that as he said he played EVE for 4 years.
The Card does help somewhat on older PC's since it will bypass the Windows stack but not on newer ones. It also DOES help ping on DSL because of how you can manage the network bandwidth and how it gives UDP priority in game mode.
Moreover, you need to be on a DMZ any time you use this card or the router with screw you over and take any performance you would gain and toss it out the door.
I've been a bigfoot customer and have the M1 in my Phenom II 940.
Rigan - Sunday, July 5, 2009 - link
Just for the record, the USB slot can be seen from the inside Linux install. So yes, it does have a use. And, yes, the thing does allow for the running of Linux apps on the inside Linux box. Works just fine.But, I'd be hard pressed to recommend this thing to anybody. In the world of modern multi-core cpu's the basic premise is rather silly.
DerekWilson - Sunday, July 5, 2009 - link
I might like to try that ... it could be fun to just play with. I mean from an ubernerd standpoint anyway.Is that available via their SDK, or is there some other hacking that needs to be done?
Per Hansson - Monday, July 6, 2009 - link
Why not just buy a decent router that supports OpenWRT and has a USB port instead?Then you have a great little device that supports QoS and whith which you can also download torrents etc while your computer is offline
The only use for the Killer NIC is if the system is really pushing 1gbps of traffic, think file server of very demanding webserver.
But that of course requires that the thing actually does offload the entire TCP/IP stack, and not just UDP (which I'm not 100% convinced it even does yet)
A cheaper NIC by any of the major players supporting ToE would probably be a better choice for the file or webserver of course (Due to more testing being done by their driver development departments)
http://en.wikipedia.org/wiki/TCP_Offload_Engine">http://en.wikipedia.org/wiki/TCP_Offload_Engine
Saying that there may be any benefit at all to doing TCP offload for MMO's and other games which work just fine on dialup still to this day (i.e. less than 5KB in bandwidth requirements) is just plain fraud IMO
But then again all reviews on the top sites including this one has come to the same consulsion so ;-)
titan7 - Sunday, July 5, 2009 - link
Read about why Linux considers these a very bad idea and explicitly won't support them:http://www.linuxfoundation.org/en/Net:TOE">http://www.linuxfoundation.org/en/Net:TOE
A friend bought the previous version. Their lousy drivers would bring down his entire system when doing torrents or even some games (e.g. Company of Heroes: Tales of Valor). No noticeable benefit, but it makes his system unstable.
Stay away from bigfoot (and nvidia) NICs!
DerekWilson - Sunday, July 5, 2009 - link
really, if bigfoot opened up their hardware the objections would fade away...honestly, bigfoot is probably using the linux network stack itself on the killer...
if linux devs could program straight to the hardware, it might really be something they would have more interest in.
davecason - Saturday, July 4, 2009 - link
I suspect that you will find that this sort of card would help a really old system that had PCI more than a new one. Think of it this way: if the task of the integrated network interface card is a burden for the CPU, this thing might actually help. Instead of an i7 chipset and compatible processor, think nForce 4 with an old 1.8GHz Athlon.We use Endace DAG cards at work and they work on basically the same principle: offload the Network work to the card.
http://www.endace.com/dag-network-monitoring-cards...">http://www.endace.com/dag-network-monitoring-cards...
Essentially the card is a computer withing your computer, dedicated to the task. The vendor recently let us know that we could get the same work done with standard gigabit ethernet cards in a more modern server... which supports my theory: not for high-end systems.
aadder - Sunday, July 5, 2009 - link
Hmmm those cards seem rather nice. Any idea where I might be able to buy the Endace DAG card?has407 - Sunday, July 5, 2009 - link
The Endace DAG cards are intended for special-purpose applications. Unless you need high speed capture and analysis, you can do better with lower-cost Intel, Broadcomm, Alactitech, etc. NICs.