POST A COMMENT

52 Comments

Back to Article

  • zephon85 - Wednesday, August 10, 2011 - link

    Any words on the impact of the different wifi adapters on notebook battery life? Would be quite good to know how much (if any) time you gain by using different wireless cards... Reply
  • Gigantopithecus - Wednesday, August 10, 2011 - link

    I could be wrong but given the upper limit on mini-PCI and mini-PCIe power capacity, I'd assume differences between cards on the same interface would be negligible in the real world. You might be able to demonstrate small statistically significant differences between cards using large sample sizes and very rigorous, controlled testing, but that's an enormous amount of time and effort to show that card X yields 5-6 more minutes of battery runtime than card Y. Reply
  • JarredWalton - Wednesday, August 10, 2011 - link

    I'll cover the battery life question in the laptop review; I didn't actually take the time to perform those tests yet. I don't expect much of a difference, as WiFi adapters are usually pulling less than 250mW, but we'll see. Reply
  • Souka - Wednesday, August 10, 2011 - link

    I'll put my $.02 in.

    Different cards have different drivers....each with their own defaul power settings.

    Unless all the various options are taken into account, it can be really hard to get a apples-2-apples comparison of power needs.

    But I agree to a point, the wi-fi power draw is really low compared to drive, memory, cpu, gpu, and LCD power needs.

    Kinda think of driving a car with the antenna up vs down.... yes it does make a difference, but not much.

    My $.02
    :)
    Reply
  • SquattingDog - Wednesday, August 10, 2011 - link

    Just to post on this - Wireless being on or off can make a huge difference to battery life on notebooks - so some testing between cards could definitely be good. Eg on a Netbook I have (Asus n10jc), I typically get around 30 - 45min more battery life with wifi OFF.

    In saying that, since upgrading from a 54mbps wifi network to a 300mbps wifi network, I have seen no difference in battery usage on the machine (connects at 300Mbps now).
    Reply
  • JarredWalton - Wednesday, August 10, 2011 - link

    True, but you're looking at a netbook that idles at around 7W. A reduction in power draw of .25W would be a lot more noticeable on that than on a full laptop that's drawing 12-15W minimum. Reply
  • Souka - Wednesday, August 10, 2011 - link

    with wi-fi on you may have a fair amount of network activity going on...which also increases the draw by the cpu/drive/mb/etc

    anyhow...good points all! :)
    Reply
  • philosofool - Wednesday, August 10, 2011 - link

    Wireless networking isn't something I really keep up on, so I don't know much past the various 802.11 names. What, forexample, does 3x3:3 mean and why might I like that more than some other configuration? Reply
  • A5 - Wednesday, August 10, 2011 - link

    From Page 1 of the review:

    "While we’re on the subject, let’s clarify what the MIMO numbers mean. When we’re talking about a 2x2:2 part, the first digit is the number of transmit chains, the second is the total number of receive chains, and the third is the total number of data streams supported. It’s possible to have a 3x3:2 device, for example, which would use the extra transmit and receive chains to improve SNR (Signal to Noise Ratio), but the number of streams cannot be more than the larger of the transmit/receive chains (so 2x2:3 isn’t possible, but 2x3:3 is)."

    Theoretically, a 3x3:3 device offers 3x the bandwidth of a 1x1:1 device.
    Reply
  • GeorgeH - Wednesday, August 10, 2011 - link

    Thanks for the review, it looks like my bias against everything "killer" will have to be adjusted a bit. While I'm still not sure that the performance difference is terribly meaningful, neither is $20 in most laptops. Reply
  • Aikouka - Wednesday, August 10, 2011 - link

    Jarred, if you asked me about a year ago... I never thought I'd recommend a Netgear product. My past encounters with their networking devices has been less than stellar. Even with that, I decided to take the plunge, and I purchased a WNDR3700 a few months back. It has treated me rather well so far and that's on stock firmware. It is DD-WRT compatible; so if you prefer their firmware, you're good to flash.

    Compared to the Airport Extreme, it is a bit weaker on the wireless front as it only supports MIMO 2x2, but its newer sibling, the WNDR4000, supports MIMO 3x3.

    The one feature I'd definitely argue for though is simultaneous dual-band. I don't think most people have homes with only 5Ghz devices. So with just a dual-band router, you'd be limited to 2.4Ghz anyway.
    Reply
  • xand42 - Wednesday, August 10, 2011 - link

    Any idea why the (default parameter) netperf tcp transmit results are so horrible across the board? I just ran netperf on my Advanced-N 6200 notebook and got 190Mbit/s to my Cisco E3000 router, basically the same value as with iperf and netcat and every other benchmark. Reply
  • Rick83 - Wednesday, August 10, 2011 - link

    It would be fairly interesting to see the difference between the Killer NIC, server NICs, laptop NICs and desktop NIC's from different vendors (and in different implementations).
    As I run only a SSD in my desktop, and have my user profile residing on a NAS (Intel i5 powered NAS, though, not limited by the available cycles and RAM) on a RAID 5, with reasonable linear throughput, network performance is quite important for me, as it accelerates such tasks as generating thumbnails of files, listing large directories, local decompression (that's probably the worst offender, and with large archives I do this on the NAS directly) and many tasks that access my user profile that may be latency or throughput limited.

    For this reason I recently specifically got a mainboard with Intel network adapter, hoping for enhanced performance, but only realizing, that I'd have to hack the .inf to get driver support under my OS.

    On the NAS, I am currently using the RTL 8111D chips, one connected to the switch and the other to the modem. If using a decent chip increases my samba/nfs performance, I'd put down the 50 euros for the intel pro chip in a blink.

    Also you could test supported cable length, jumbo frame support, documentation (with some cards the maximum MTU is not properly documented, and it takes ages of non-fragmented pinging to discover the correct MTU) and performance of teaming and fail-over mechanisms.

    Also, 5Ghz was something that I tried in my parent's home, but we couldn't even connect a machine that was one ceiling and wall away, at less than 10 meters distance. Might have been just a spectacularly bad router (Linksys 320N) or maybe 5GHz is really just for line-of-sight applications.
    Reply
  • pityme - Wednesday, August 10, 2011 - link

    Jarred,

    I think it would be nice to run tests of multiple wireless network interference like the kind observed in apartment/townhouse/condo situations. This is a big problem that no one seems to ever test/talk about.
    Reply
  • ckryan - Wednesday, August 10, 2011 - link

    True that. I live in a condo building in uptown Charlotte, NC and every unit has a wireless router of one kind or another. To make matters worse, the elevators seem to interfere somehow -- along with every wireless phone and microwave, the acres of glass and the steel structure. To top it off, there's an office building right next door which has many wireless systems on its own. Reply
  • ckryan - Wednesday, August 10, 2011 - link

    Sometimes it seems like someone has cast some manner of evil voodoo on WiFi as a whole.

    I bought a Killer 2100 NIC because I thought it was a good idea -- and it was on sale. It is better than Intel's on-board Ethernet (and better still than Realtek) if only just barely. In reality, I just liked the idea of dedicated hardware for my networking needs. It really did make a difference, and really is some good stuff, but probably not worth the substantial premium over garden variety PCIe NICs (for most people). I'm glad I bought it... but only because it was on sale.

    The Killer product line is in the position of having a product family which does perform mostly as advertised, but becomes a tough sell when you already have onboard ethernet or wifi. It has to be difficult to sell a consumer NIC that costs as much as a 80GB SSD or mid range video card. For $65 or $70 it's great. At $140, not so much. As such, I'm glad to see them trying to put their hardware inside other hardware, like networking cordon bleu. Shoving their "NPU" onto the PCBs of video and sound cards, motherboards, and laptops is a good idea and if nothing else add a little variety to the mix. It would easily be worth an extra $25 -$35 premium for a motherboard in any price range to get their NIC over the bog standard Realtek. It can't possibly cost more than the NF200 or Hydra chips on mid and high end mainboards and is surely worth an extra $20 on top of a laptop as well. Whether they can sell this particular device sans lappy at a reasonable price or even at all remains to be seen.
    Reply
  • honvl - Wednesday, August 10, 2011 - link

    I'm on a wireless card and want to test my latency. Where can I find this GaNE tool? Reply
  • Flunk - Wednesday, August 10, 2011 - link

    This result really surprised me, I was expecting results like their wired products (no difference). My guess is that the biggest factor is the support for more channels but the performance really does seem markably better. I'm gearing up to by a new Alienware laptop near the end of the year and I may very well check the box for a Killer-N upgrade. Reply
  • StormyParis - Wednesday, August 10, 2011 - link

    Isn't there a wide gap between what tech sites review, and what users use ? I know it's part of the purview of a tech site to go for the 0.001%-of-users, bleeding-edge stuff, but... has anyone ever seen a review of mainstream Wifi adapters/laptop anywhere ? As one of the 99.999%, I'd be interested ! Reply
  • JarredWalton - Wednesday, August 10, 2011 - link

    Bigfoot offered, so I thought, "Hey, this could be fun." Well, it wasn't. WiFi testing is a real pain in the butt I've decided. Still, it would be good to look at some other products. We might come back to the subject in the near future; I'm going to see about soliciting a bunch of different WiFi adapters and see what turns up. Obviously, I have at least the 6230 and 6300 from Intel. I want to add the Killer 1103, some Broadcom stuff, etc. Really, I'd want to focus on just dual-band cards, though -- anything that can't do 5GHz becomes less interesting after playing with the Linksys E4200. Reply
  • DesktopMan - Wednesday, August 10, 2011 - link

    What's the reason for the big difference with these results: http://www.smallnetbuilder.com/wireless/wireless-r...

    Anyone know?
    Reply
  • Reflex - Wednesday, August 10, 2011 - link

    Probably different laptops. This review is unfortunatly not very good because if I'm reading the first couple pages correctly, he used different laptops for each card. Contrary to his earlier experience, most laptops will accept any wifi card you wish. I swapped in a 6300 in my Dell a year ago and it works great.

    They need to establish a baseline testing platform to isolate the perf between the cards. Testing them all on different laptops invalidates the test. Hard drives, CPU's, memory speed, etc can have a *huge* impact on wifi performance, especially for file copy type operations. And the range test is completely irrelevant as everyone has their own way of routing the antennas up through the lid.
    Reply
  • JarredWalton - Wednesday, August 10, 2011 - link

    For wireless, the storage actually matters almost not at all. I swapped in an HDD to one of the laptops and ran the two file copy tests. The HDD was withing 1 second of the SSD for the large file, and within 3 seconds on the small files. On GbE, HDD vs. SSD is a huge disparity, but with WiFi topping out at <30MBps it really doesn't matter much. The WiFi latency appears to be almost as bad as the HDD latency for seek operations.

    But you're right: the different laptops all make it hard to to apples-to-apples, and depending on vendor swapping in a different WiFi card may or may not work. The real issue for me was lack of time; I kept going back and forth between devices as I discovered a potential issue with one of the results. Now that I'm more comfortable with what WiFi testing entails, I'm hoping (not right now, but maybe in a couple months) to go through and test a bunch of cards in a single laptop, as well as in a PCI-E x1 desktop adapter.
    Reply
  • endrebjorsvik - Sunday, August 14, 2011 - link

    This puzzles me as well. The last couple of days I have been struggling with getting decent performance from my own setup. I have a Netgear WNDR3700v2 and a Lenovo X220 fitted with i5-2520M and Intel 6300 3x3 and running W7. A HP ProLiant ML110 G6 with GbE and 4x2TB RAID-Z is serving the test-files.
    According to smallnetbuilder.com, the WNDR3700v2 ( http://www.smallnetbuilder.com/wireless/wireless-r... ) should be faster than WNR3500L ( http://www.smallnetbuilder.com/wireless/wireless-r... ), so my setup should at least be as fast as Jarred's Netgear-Intel6300-Ideal-result (154 Mb/s).
    I have tried both 2,4 and 5 GHz with both 20 and 40 MHz BW and with both stock and open firmware (dd-wrt), but I don't even get to 90 Mb/s (Windows file transfer tops out at 11 MB/s = 88 Mb/s, and usually stays below 10 MB/s). The distance between the router and laptop is ~6 feet, and I have tried every possible position of the router (different antenna directions). The laptop lid is open (~90 degrees).

    So I wonder if you (Jarred) came across any mindblowing tricks that increased the throughput dramatically? Or was the Netgear-Intel6300-combo just plug'n'play?
    Reply
  • JarredWalton - Sunday, August 21, 2011 - link

    What are you copying from? 11MB/s max sounds like you've got the Ethernet side hooked up to a 100Mb port, or else you're doing a transfer from one wireless PC to another? In that case, you'd be doing 22MB/s of wireless traffic, which would be pretty good considering collisions and such. Reply
  • ss284 - Wednesday, August 10, 2011 - link

    It would have been really great if a recent macbook's wireless throughput was tested. I believe all the recent refreshes have the same broadcom based wireless adapter. Reply
  • xdrol - Wednesday, August 10, 2011 - link

    "the number of streams cannot be more than the larger of the transmit/receive chains (so 2x2:3 isn’t possible, but 2x3:3 is)"

    No it is not. It cannot be more than the SMALLER of the two. But the transmit and receive antennas are on a different device, so a given device could support more than it's Tx/Rx antennas, but only in the other direction (where it does have more antennas).

    As for specifying what 1 given device can do, then there are actually 4 different numbers, 3 are not enough:
    - The number of Tx antennas (a)
    - The number of spatial streams to be received (<=a)
    - The number of Rx antennas (b)
    - The number of spatial streams to be transmitted (<=b)

    As WiFi is a symmetrical system, the Tx and Rx features of a device are usually the same (read: I'm yet to see any that differ) - unlike e.g. LTE, where the usual MIMO currently is only downlink, but not uplink (it has different PHY for uplink anyway).

    In the example, the 2x3:3 is valid only if you meant it has 2 Tx antennas, 3 Rx antennas, and it can RECEIVE 3 spatial streams. As it has only 2 antennas, the maximum outbound spatial streams is 2.
    Reply
  • Brian Klug - Wednesday, August 10, 2011 - link

    That's technically right, and we do mention that the Intel 1030 can do two streams on Rx and one on Tx, but I've seen very few routers actually support an asymmetrical MIMO scheme like that. Even the intel card for example always only shows 1 stream being used for Tx and Rx, so in practice really it should be symmetrical.

    -Brian
    Reply
  • James5mith - Wednesday, August 10, 2011 - link

    Maybe I'm used to living in smaller places, but 60 feet from your front door to the router? That seems a bit extreme. Is the router in the attic, and the front door in the basement corner of the house or something? Reply
  • James5mith - Wednesday, August 10, 2011 - link

    Actually, more to the point, if it's 60 ft to your front door, then your google maps view shows the Intel 6300 making it nearly 500 ft from the router in the Cisco 2.4GHz test. You stated it was 200 ft. Reply
  • JarredWalton - Wednesday, August 10, 2011 - link

    I was estimating. I just paced it off, and walking (around corners) it's about 50 feet. In a direct line, it's more like 30 feet. I'll update the distances, though it's all very rough. (I don't have blueprints for the house, but it's about 2300 sqft with an upstairs and downstairs; corner to corner is about 50 feet I'm guessing. I'm revising the distances to be as accurate as possible, but for reference that red vehicle in the driveway is almost 20 feet long if that helps. Reply
  • theqat - Wednesday, August 10, 2011 - link

    Any idea when we can expect the Pollux review? I'm trying to make a decision on a laptop fairly soon and I'd love to see what you guys think about it before I do so. Reply
  • Hrel - Wednesday, August 10, 2011 - link

    I kept looking for a range test but didn't see it. Having an issue with my Clevo P150HM, had to send it back to cyberpowerpc. Got no wifi signal at all at public places like panera and starbucks. I kind need internet in those places. Huge part of the reason I even have a laptop. It has the Intel 6230 in it cause I need bluetooth. Reply
  • ggathagan - Wednesday, August 10, 2011 - link

    Page 7? Entitled "Testing Signal Range"? Reply
  • Focher - Wednesday, August 10, 2011 - link

    Once again, Bigfoot's product doesn't really offer anything of extra value to alternative products (personally, I think the UDP test results are wonky and it's unbelievable that a current wifi device would best wired gigabit Ethernet, let alone dominate to the extent that the test showed). You're better off getting a true 3x3:3 card, even if the current performance only matches the 1102. At worst, performance will stay the same. But there's at least a chance that the addition transmit and receive channels will enable improved performance with a firmware update. Sure, if the 1102 is the default card in a particular laptop then fine (although I'd still take the Intel 6300 over it) but I sure wouldn't pay any price premium for it unless it's an upgrade option to replaces some default low end device. Reply
  • neothe0ne - Wednesday, August 10, 2011 - link

    yeah, Lenovo and HP are horrible companies. HP is even worse because all their consumer products force you on WiFi Link 1000 (their Bluetooth is offered in a separate chip). Most of the Probooks and Elitebooks and the Envies use Centrino 62xx cards, thankfully... but they're still whitelisted. Reply
  • Peroxyde - Wednesday, August 10, 2011 - link

    Hi experts,

    Sorry for off topic question. Can you please recommend a good wireless router? I have heard of TP-Link 1043ND as being well rated. Is it really better than Linksys or DLink? Thanks in advance
    Reply
  • jigglywiggly - Wednesday, August 10, 2011 - link

    A LIL LATE FOR THIS ARTICLE
    I HAVE A CLEVO WITH A 6230
    AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
    Reply
  • iamkyle - Thursday, August 11, 2011 - link

    The majority of laptops can't even use this card thanks to the legalized monopolization of laptops via whitelists. That crap should be illegal. Reply
  • The_Laughing_Man - Thursday, August 11, 2011 - link

    I think Bigfoot should sue those tier one laptop suppliers for anti-competitive practices.
    Whitelist should be illegal.

    I have a brand new HP laptop, I can replace the WiFi card after booting and it works just fine. It is even in the same WiFi family that comes with the laptop and Windows uses the same set of drivers for both cards. But the notebook will not boot with newer dual-band card installed during start-up due to HP whit-list in their BIOS.

    On top of that HP encrypts their BIOS such that is is very difficult to hack their BIOS to allow valid WiFi cards, not to mention other things.

    HP doesn't even offer an upgrade to a dual-band WiFi card for this very expensive high-end notebook the DV7T, even if I was willing to pay them for it.

    And Dell, Asus, Acer, and Sony all do the same thing.

    They all need to be sued. And until they remove their whitelists, Bigfoot will be forced to have a very small market to sell their mini-pcie killer-n card to.
    Reply
  • neothe0ne - Sunday, August 14, 2011 - link

    "And Dell, Asus, Acer, and Sony all do the same thing."

    Are you sure about that? I was under the impression HP and Lenovo were alone in the industry with the WLAN whitelist. And anyway, Dell does offer the Intel Centrino 6230 on the XPS 15 now, unlike HP's dv6 which is stuck in budget-tier Intel WiFi Link 1000 land.
    Reply
  • cjl - Tuesday, August 16, 2011 - link

    Dell, at least in their Alienware products, definitely does not whitelist. After reading this article, I got one of the Killer 1102 cards for my M11xR2 (which comes with a rather terrible card by default, and there were no upgrade options offered), and it works just fine. I popped it in, installed the drivers, and everything has been working great since. Reply
  • Musafir_86 - Thursday, August 11, 2011 - link

    Hello,

    -Thanks for the article, but did you tested those adapters with or without any security/encryption/password protection scheme? I mean WEP or WPA/WPA2 - I think encryption put some overhead in the throughput.

    Thanks.
    Reply
  • JarredWalton - Thursday, August 11, 2011 - link

    All testing was done with WPA2 AES. Most modern cards do fine with that, though a few years back it was sometimes slower IIRC. Reply
  • Musafir_86 - Thursday, August 11, 2011 - link

    -Okay, thanks for the clarification. :) Reply
  • Yummer72 - Thursday, August 11, 2011 - link

    Thanks for the informative review.

    I wonder if Bigfoot will continue to have an advantage if the "WLAN Optimizer" program was used with the other WiFi cards?

    http://www.martin-majowski.de/wlanoptimizer/

    I have personally seen significantly improved performance and the elimination of "lag spikes" (QuakeLive) with this software tweak.

    Any comments?
    Reply
  • JarredWalton - Thursday, August 11, 2011 - link

    I'll give that a try; it could very well remove the spikes, leaving the primary advantage as the lower base latency. Reply
  • bhima - Thursday, August 11, 2011 - link

    You should review that 95% color gamut matte screen in that Mythlogic ;) Reply
  • loopingz - Thursday, August 11, 2011 - link

    First of all thanks for highlighting that I can change my wifi adaptator on my laptop. Mine is always frozing during transfert in windows (linux is fine).

    Second thanks for helping me choosing the good one.

    I hesitate now between intel 6300 for range, correct performance and price, and the 110 2/3 for pure performance.
    May be best of two worlds intel 6300 in the eeepc that travel a lot and bigfoot in the main home laptop.

    Can I recycle a my old wifi card or a new one using an antenna and puting it in my desktop computer?

    I will give try to Wlanoptimizer too because watching movie from the raid5 nas still not perfect (router linksys e3k).

    Thanks for the good job.
    Reply
  • name99 - Thursday, August 11, 2011 - link

    "Wireless networking also tends to need more overhead for error checking and interference losses, and there’s a question of whether the streams are linearly independent enough to get higher throughput, orientation, directionality of signal, etc. Even though you might connect at 450Mbps or 300Mbps, you’ll never actually reach anywhere near that level of throughput. In our testing, the highest throughput we ever saw was around 75% utilization of the available bandwidth, and that was on a 300Mbps connection."

    This is not a useful description of the situation. The nominal speed of a connection (ie the MCS index) already includes error correction overhead --- that's why you see a range of bit-rates, with the same parameters (modulation, number of streams, bandwidth) --- these different bit-rates correspond to different levels of error correction, from the strongest (1/2 coding rate) to the weakest (5/6).

    It is also unlikely that corrupt packets and the retransmission (what you are calling "interference losses", though in your environment noise is likely more relevant than interference) are substantial --- both ends aggressively modify the MCS index to get the best throughput, and try to keep the number of corrupt packets low.

    The real issue is the MAC --- the negotiations over who next gets airtime. This used to be a big deal with wired ethernet as well, of course, but it went away with switches around the time we all moved to 100TX. The basic 802.11n MAC does not rely on any real co-ordination, just on timing windows and retries, and it wastes a phenomenal amount of time. 802.11e improves the situation somewhat (I expect all the systems that get 75% efficiency are using 802.11e, otherwise they'd see around 50% efficiency), but it's still not perfect.
    What one really wants is a central arbiter (like in a cell system) that hands out time slots, with very specific rules about who can talk when. For reasons I don't understand, 802.11 has been very resistant to adding such a MAC protocol (802.11e has elements of this, but does not go full-hog), but I would not be surprised if we finally see such as part of the 802.11n successor --- it's just such an obvious place to pick up some improvement. The real problem is that to do it right you have to give up backward compatibility, and no-one wants to do that. At least if we'd had it in 802.11n, then we'd be part way to a better world (people could switch it on once all their g equipment died, eg at home).
    Reply
  • ktwebb - Thursday, August 11, 2011 - link

    We didn't "start" at 11mb. For the consumer yes but I was putting in 900Mhz 2mb in schools in 97' and Aironet (bought by Cisco) by late 98. Reply
  • AEROSPIKE - Wednesday, August 24, 2011 - link

    802.11b WAS NOT THE FIRST MARKETED WiFi HARDWARE BUT THE FIRST MANUFACTURER'S BACKED STANDARD BY THE IEEE, 802.11 WAS THE FIRST WITH UP TO 6 CHANNELS AT 1-2 MEG BANDWIDTH BACK IN THE NINETY S. I WAS AN EARLY ADOPTER, THE SOFTWARE AND HARDWARE OF THOSE UNITS WERE VERY UNSTABLE. Reply

Log in

Don't have an account? Sign up now