What am I doing wrong ?

Discussion in 'Wireless Internet' started by Mike Jones Super Hero, Jul 18, 2004.

  1. I have a D-Link DI-614+ the one with twin antennas (rev A ? ) which
    appears to work well , I can connect to it via my Sony Vaio laptop using a
    D-Link AirPlus DWL-650+ without any problems at all , but I have just been
    given an Acer laptop and a D-Link AirPlus Extreme G+ DWL-G650+ card and I
    cannot get it to connect to the router no matter what I do . If I run
    netstumbler the G+ card can see the router as well as another router locally
    (on a different channel) . I have tried the non G card in the Acer and that
    works . Am I right in assuming that the G+ card should be backward
    compatible with the b router or is this the problem ?

    Any ideas gratefully received


    Cheers

    Mike

    --
     
    Mike Jones Super Hero, Jul 18, 2004
    #1
    1. Advertisements

  2. Mike Jones Super Hero

    JM Guest

    compatibility between hardware is not garaunteed, stick with the same
    manufacture with wi-fi hardware.
     
    JM, Jul 18, 2004
    #2
    1. Advertisements

  3. Its all D-Link kit just mixed between b and g technology

    --
     
    Mike Jones Super Hero, Jul 18, 2004
    #3
  4. Mike Jones Super Hero

    JM Guest

    It has been noted by some D-Link users that the backwards compatibility is
    not that great.
     
    JM, Jul 18, 2004
    #4
  5. Mike Jones Super Hero

    JM Guest

    JM, Jul 18, 2004
    #5
  6. Mike Jones Super Hero, Jul 18, 2004
    #6
  7. It "should" work. An 802.11g card should be downward compatible with
    802.11b. The fact that NetStumbler can see the 802.11b only access
    point underscores that it should be compatible. Therefore, it has to
    be something to make it incompatible. My guess(tm) is:

    1. Your WEP keys are out of sync. There have been a few reports and
    comments from DLink that ASCII WEP keys are "unreliable" and that one
    should use Hexadecimal keys. I've never seen this, but it's possible.
    This might help:
    http://www.warewolflabs.com/portfolio/programming/wepskg/wepskg.html
    If this is the case, I would be interested in the firmware versions.
    I have a DI-614+ (with the two antennas) in my office and a DWL-G650+
    card in stock. If it really is a WEP incompatibility issue, I wanna
    try it.

    2. You might wanna disable the "Wireless Zero Config" service and use
    the DLink utility instead.

    3. You have something set in the DWL-G650+ that says something like
    "use only G or high data rate speeds" or perhaps a fixed speed. This
    is common on some 802.11g access points. I dunno if it exists in the
    DWL-G650+ setup.
     
    Jeff Liebermann, Jul 18, 2004
    #7
  8. Mike Jones Super Hero

    gary Guest

    And if the DI-614+ is using 22Mbps PBCC, I wouldn't necessarily assume it
    will interoperate with all the other adapters. The DWL-G650+ documentation
    claims back-compatibility with "802.11b+", but I'm not sure how inclusive
    that is. If there's a way to disable 22 Mbps, I suggest trying that.
     
    gary, Jul 18, 2004
    #8
  9. Good idea. Despite owning a DI-614+, I never even noticed that it
    would do the 22Mbit/sec pre-802.11g data rate. I had lots of trouble
    getting a DWL-900AP+ with the 22Mbit/sec rate enabled to connect to an
    802.11b only access point. Not exactly the same problem, but
    definately a possibility.

    See:
    http://support.dlink.com/techtool/di614+/emulator/adv_perform.htm
    and move the Tx Rates down to 1-11Mbits/sec.

    Drivel: I've had this DI-614+ for about 2 months and never even
    noticed that it can do 22Mbits/sec. Ooops.
     
    Jeff Liebermann, Jul 18, 2004
    #9
  10. ---- Original Message ----
    From: "Jeff Liebermann" <-cruz.ca.us>
    Newsgroups: alt.internet.wireless
    Sent: Sunday, July 18, 2004 6:11 PM
    Subject: Re: What am I doing wrong ?
    This is my Firmware Version: 2.10 , Fri, 22 Nov 2002



    Just did that didn't make any difference
    Set to basic rates to 1-2-5.5-11(Mbps) and TX rates to
    1-2-5.5-11-22(Mbps) and whoo hoo ......success we have a connection


    Nice One Jeff
     
    Mike Jones Super Hero, Jul 18, 2004
    #10
  11. D-Link has a bad habit of re-using model names/numbers when the
    products themselves are entirely different.

    This is true for certain of the products in this case: the DWL-650,
    DWL-650+, DWL-G650, DWL-G650+ are *entirely* different products.

    You have the correct products fro maximum compatibility with your
    router: the clue is in the + which in the case of D-Link means they
    are based on a wireless chipset from Texas Intruments.

    The so-called 802.11b+ products (DI-614+, DWL-650+) use the TI
    ACX100, the DWL-G650+ uses the TI TNETW1130. All these products
    include PBCC modulation (an option in 802.11g) and support the 22
    Mbit/s speed (also an option in 802.11g). This is the highest speed
    you will get with your setup.

    The earlier post about compatibility with SuperG etc is not relevant
    - this is a proprietary extension using the Atheros wireless chipset.
    and your D-Link products use a TI wireless chipset.

    Hope this helps

    --

    Richard Perkin
    To email me, change the AT in the address below
    richard.perkinATmyrealbox.com

    It's is not, it isn't ain't, and it's it's, not its, if you mean it
    is. If you don't, it's its. Then too, it's hers. It isn't her's.
    It isn't our's either. It's ours, and likewise yours and theirs.
    -- Oxford University Press, Edpress News
     
    Richard Perkin, Jul 18, 2004
    #11
  12. The plot thickens. I logged into my office DI-614+ remotely and
    noticed that there was this box on the bottom of the wireless page
    http://support.dlink.com/techtool/di614+/emulator/adv_perform.htm
    inscribed:
    4x mode [ ]enable [ ]disable
    I'd never noticed that. So, I dig into the support pile and find:
    http://support.dlink.com/faq/view.asp?prod_id=1245
    which declares:
    "If you are using non-4X adapters in your wireless network,
    performance will degrade if 4X is enabled."
    Ummmm... perhaps we should turn it off?
     
    Jeff Liebermann, Jul 19, 2004
    #12
  13. Mike Jones Super Hero

    gary Guest

    Good question. My guess would be that this is akin to saying that if you mix
    802.11b and 802.11g devices, performance will degrade (but not significantly
    below 802.11b highest bitrates). My uneducated guess would be that a mix of
    D-Link 4x and non-4x adapters is expected (by D-Link) to downgrade to
    standard 802.11b bitrates. D-Link has said that it only recommends using 4x
    mode if all of the network elements are Airplus, so they expect it might
    break if you have other vendors in the mix.

    I also found a reference:

    http://www.tomsnetworking.com/FAQ-18-DLink+Wireless-0.php#475

    that indicates that early revisions of the DI-614+ firmware can cause
    Centrino laptops to crash during association, with or without 4x mode. Can't
    tell from the article if there's a fix for the 614+ (there is one for the
    900-AP+), but this suggests some flakiness experienced by clients trying to
    associate with DI-614+.
     
    gary, Jul 19, 2004
    #13
  14. As I understand it, the so-called "4x" mode is not a higher bitrate
    (which remains at 22 Mbit/s) - but uses higher frame packing rates to
    reduce overheads.

    22 Mbit/s using PBCC modulation is fine IMO, since it's now part
    (albeit optional) of 802.11g.

    But I'm none too keen on proprietary extensions of whatever kind,
    whether they're called marketingspeak things like SuperG, Nitro,
    Turbo or 2x or 4x or whatever. It all affects interoperability - the
    very reason for standardisation in the first place...

    Hope this helps

    --

    Richard Perkin
    To email me, change the AT in the address below
    richard.perkinATmyrealbox.com

    It's is not, it isn't ain't, and it's it's, not its, if you mean it
    is. If you don't, it's its. Then too, it's hers. It isn't her's.
    It isn't our's either. It's ours, and likewise yours and theirs.
    -- Oxford University Press, Edpress News
     
    Richard Perkin, Jul 19, 2004
    #14
  15. Mike Jones Super Hero

    gary Guest

    This agrees with what I have read. From what I understand, the claim is not
    that the max bitrate is 4x the 802.11b max bitrate, but that max real-world
    throughput is 4x. The underlying encoding method is 22 Mbps PBCC.
    I've read that one of the reasons TI started shipping 22 Mbps was that early
    drafts of 802.11g did not support PBCC. It was a way of forcing the issue.
    Don't know if this is true, but it sounds plausible.
    Yes, I completely agree. But my guess is that these tweaks are relatively
    inexpensive for both chipmaker and vendor, and the temptation to gain a
    product differentiator is almost irresistible.
     
    gary, Jul 19, 2004
    #15
  16. My guess is that since enabling the 4X mode radically chances the
    fragmentation threshold and "normal" data rate, methinks Dlink is
    doing something quite different. I suspect that they've:
    1. Increased the fragmentation threshold so it only sends big
    packets.
    2. Increased the CTS/RTS threshold so that it NEVER sends flow
    control frames. This will work if you don't have other nearby radios,
    but will cause chronic collisions in a crowded environment.
    3. Prevent the radio from slowing down to 802.11 (1-2Mbits/sec)
    rates. Never mind that some radios don't understand management frames
    at any speed other than 1Mbit/sec.
    4. Decrease the beacon rate to reduce wasted airtime.
    5. Crank up the DTIM message interval so that other radios will wait
    longer to listen for broadcasts leaving lots of airtime for the 4X
    radios.
    Any failure can be adequately explained by a suitable conspiracy
    theory. When DLink cranks up the size of the average packet, ignores
    flow control, and drops the inter-packet delay, it's assumed that the
    receiving radio is capable of buffering the incoming traffic.
    Centrino, where everything is done is software, was probably designed
    for typical 11Mbit/sec traffic rates. I probably overflowed some
    buffer trying to digest 4 times as much data as expected. This is
    pure speculation on my part but sorta assigns the blame to the
    appropriate party. Netgear also had to issue firmware updates to get
    their access points to work with Centrino. I had an MR814 that would
    lurch and hiccup its way through wireless transfers to/from a Toshiba
    Centrino laptop, that was magically fixed by a firmware update.

    I have 3 fairly new Centrino based laptops in the office complex
    mooching off my DI-614+ wireless connection. No problems after the
    usual Windoze XP updates with 4X turned on. I guess it's fixed. I
    just turned off 4X and will find out Monday if anyone complains.
     
    Jeff Liebermann, Jul 19, 2004
    #16
  17. This old story covers the situation as I vaguely remember it:
    http://www.wi-fiplanet.com/columns/article.php/1478441
    Since TI bought Alantro, the company that invented PBCC, for a nominal
    $300 million, they were obligated by their stockholders to push for
    its adoption. Meanwhile, Intersil (Harris) didn't like the idea of
    any other company trampling on its turf. Intersil owned CCK and
    wanted it based on CCK. The battle lines were drawn and the vote came
    to a fairly even split. To break the deadlock, the committee decided
    on a 3rd option. They picked OFDM straight from the 802.11a spec,
    with optional support for Intersil or TI's protocols.

    So it is written, and so we are stuck with it.
     
    Jeff Liebermann, Jul 19, 2004
    #17
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.