Previous 1 48 49 50 51 52 Next 2,266 Replies Latest reply: Mar 28, 2014 2:35 AM by WSR Go to original post Branched to a new discussion.
  • moskovit Level 1 Level 1

    lupunus wrote:


    moskovit wrote:


    2. A question for Lupunus: [..] I haven't seen you post anything about the kext downgrade solution (perhaps I missed it).[...] What you think about the kext downgrade solution?  Do you reject/ignore it because you are convinced that the wi-fi problems people are experiencing have nothing to do with Lion?

    I did not ignore it and there may be eventually situations on which a downgrade of a Kernel extension could be a solution, indeed.


    However, it changes the design of the OS


    How precisely does downgrading the kext change the design of the OS?  I, for one, am happy to change the design of the OS, if it will allow me to use my wi-fi without having to wait when I wake my computer from sleep.


    and may cause unforeseeable problems with other functions of the OS


    Of course.  If that happens, one can easily revert the downgrade.


    it did not resolve the reason for the issue as such.


    There's only one way to find out: by trying it.  It looks like others have had success using this solution where other potential solutions have failed.


    As I could see on the last developer version of 10.7.2 there where NO changes in the wireless stack.

    That is confirmed by reports of people having installed the 10.7.2, hoping that will solve there wireless problems. Guess what, it did not.


    That does not mean that Lion is not to blame.  It could mean that Apple has not yet solved the problem.


    Imagine that there are several millions of Lion users out there. I guess at least 90% of them are connected to a network, probably the majority on a wireless. If there where a major issue with Lion on wireless, we definitively would know about.


    By the way ... there where the exact same discussions on Tiger, Leopard and Snow Leopard about wireless problems.

    Take a look by copying that line into Google:

    apple 10.6. wireless problem


    You will find 82.900 hits only for this community.


    Consider this analogy: A car manufacturer makes a car.  It tests it.  The car works great.  The manufacturer releases it.  The car works great, except for people who operate it in extremely low temperatures.  These people complain.  The manufacturer ignores their complaints.  An auto mechanic named Lupunus tells them that their problems are not the fault of the model or the manufactuer, because the car works great for almost everyone else, and Lupunus offers dozens of different tweaks for different drivers, depending on how their cars have been customized.  Eventually, the manufacturer realizes that the car was not properly designed for extremely cold conditions, but does not want to admit its mistake, so it quietly improves the design.  But then, before it releases next year's model, it again fails to test the car in extremely cold conditions, and the same cycle continues, year after year after year.


    Do you see my point?  Just because not everyone experiences this problem, and just because wi-fi problems recur with every major version upgrade, does not mean that the version upgrades are not to blame for the problems.


    Of course, our situation is probably a lot more complex than the car analogy, but this only strengthens my point.  It is not clear exactly what the conditions are that cause the wi-fi problems, and the conditions that cause the problems may be multiple and may change from upgrade to upgrade.


    Also, you have neglected to address the point I made in my last message about so many of us claiming that this problem started right when we upgraded to Lion.  Either that is a HUGE coincidence, or you think we've all misremembered what happened.  How do you explain this?


    Would anyone else care to chime in here?

  • moskovit Level 1 Level 1

    Okay, so downgrading the kext seemed to help—or, at least, didn't hurt—for half a day, and now my wi-fi is back to stalling before reconnecting after waking from sleep.


    As I wrote before (see below), the kext downgrade I tried was supposed to work only for Atheros chipsets.  I have a Broadcom chipset.  Is there a different kext (or a different version of the same kext) that I might try downgrading that might help those who have Broadcom chipsets?


    I no longer have any files from Snow Leopard, so if there is a file to replace, I would appreciate someone posting the file online and pasting a link here, so that I (and others) can try downloading and installing it.

    moskovit wrote:


    Looks like I spoke too soon in praising Lupunus's "2 BSSID" solution.  Like many solutions I've read on this thread, it seemed to work for awhile—a week, in this case—and then it stopped working.  Once again, my MacBook Pro became slow to re-connect to wi-fi from sleep.


    I was reluctant to try downgrading the kext, explained at because I read that that solution is appropriate only for Atheros chipsets, as opposed to Broadcom chipsets, the latter of which my MacBook Pro has.  But out of desperation, I tried implementing the kext downgrade solution and de-implementing Lupunus's "2 BSSID" solution (quoted below).


    To my surprise, the kext downgrade solution didn't mess anything up, and, in fact, it may have solved the problem.  It's too early to tell, because it's only been an hour or so, but so far, wi-fi is re-connecting instantly on wake from sleep.  I'll try to remember to give an update after a week or two to see whether the solution sticks.


    A couple of questions:


    1. A question for anyone: It was previously reported that the kext downgrade works only for Atheros chipsets (note that I installed the kext that I downloaded from the link above, about which the author writes, "only have a go if your Mac has Atheros Wi-Fi hardware!").  I have a Broadcom chipset, and the kext downgrade seems to have worked for me (at the very least, it didn't mess anything up).  What's the deal?  Is the author wrong that the kext is appropriate only for Atheros chipsets?


    2. A question for Lupunus: Though you refuse the title, I still think you're a genius (or at least very smart), as well as very generous (though admittedly harsh).  But I haven't seen you post anything about the kext downgrade solution (perhaps I missed it).  Instead, you provide solutions that are tailored to each user's configuration, without reference to the kext.  I'm curious: What you think about the kext downgrade solution?  Do you reject/ignore it because you are convinced that the wi-fi problems people are experiencing have nothing to do with Lion?


    And now, Lupunus, a reply to your last reply to me:

    lupunus wrote:


    As a matter of fact it's also clear, that in 99.9% of the reported problems neither Lion

    nor Apple or Windows is responsible for the problem.


    If one integrate a correct configured Mac in a correct designed and configured wireless network, the system will run seamlessly in that environment.

    if this where not fact, the several millions of Lion users had ALL experienced a wireless disaster.


    I'm trying to avoid getting personal here, Lupunus, but I'm concerned that your personal biases might be leading people astray with many of your suggestions.  As I said, I get the sense that you're extremely bright, knowledgeable, and generous with your suggestions.  But it strikes me as incredibly presumptuous for you to claim you know it to be a fact that "99.9% of the reported problems neither Lion nor Apple or Windows is responsible for the problem."  Just because the problem doesn't affect all Lion users doesn't mean that Lion isn't the problem.  And just because lots of people have wi-fi problems with each OS version upgrade doesn't mean that those OS versions aren't responsible for their problems.  I myself saw a very clear correlation in time between when I upgraded to Lion and when the wi-fi problems began.  So did many others in this discussion.  For you to imply that we're all wrong or that these are all just coincidences strikes me as incredibly arrogant.


    I would let it go, except that I and many others in this discussion are spending quite a bit of time reading, implementing, and replying to your suggestions.  If you were more humble and open-minded to the possibility that Lion might, at least in some cases, be the problem, then I would put a lot more credence in your suggestions.  But given your apparent arrogance and closed-mindedness, I am concerned that you may be wasting many people's time.  And I know I'm not alone, because I'm far from the first person to point out your haughtiness in this and other Apple discussions. 


    My suggestion, for what it's worth, is that you ask a few people you trust to be honest with you whether they can understand how others might perceive you as arrogant—and do a bit of soul-searching—before continuing to participate in these discussions.


    moskovit wrote:


    Wow! Lupunus, you are the only person on this forum who has provided a solution that's worked for me, and I tried many, many of the solutions offered on this forum.   My wi-fi has worked perfectly for about a week since I implemented your solution.  You are a genuis, and a saint


    (That said, I must say, Lupunus, that I do, at times, feel that you are needlessly harsh with people who are understandably frustrated with Apple and come here to vent their frustration.  Your solution did solve my problem, but that doesn't change the fact that the problem began precisely when I upgraded to Lion.  I would much prefer that Apple fix these problems quickly than me having to spend hours on this forum and tinker with esoteric settings, and so I commend people for voicing their frustration, in an effort to get Apple to improve their software [though it's not clear to me whether Apple employees actually read these forums].  So I appreciate your kindness in offering me a solution, and I would also appreciate you expressing a kinder attitude towards others on this forum.)


    For those of you experiencing a slow re-connect to wi-fi from sleep, you might try implementing Lupunus's solution (reprinted below), especially if you own a Time Machine (or another type of Airport router) like me.


    Thanks again, Lupunus, for your help!


    lupunus wrote:


    moskovit wrote:


    The problem I was experiencing is my MacBook Pro taking awhile to connect to wi-fi upon waking from sleep.  The problem started after I upgraded to Lion.  Resetting the time capsule to factory settings (via Airport Utility->Base Station drop-down menu->Restore Default Settings) solved this problem for a couple of days, but then the problem came back.

    Looks like the "2 BSSID" problem. See other posts here.


    • On TimeCapsule set a dedicated name (SSID) for the 5GHz wireless.
    • If no iPone or iPod nano is connected to the TC, eventually set: 802.11n only (5GHz) - 802.11n only (2.4GHz).
    • If iPhone or older wireless system should get connected use: 802.11n only (5GHz) - 802.11b/g. You will reach that settings with holding option key (alt) while clicking on the dropdown menu.
    • Force the Mac to use the 5GHz SSID
    • Delete the other SSID's of your home wireless from the list of "known networks"
    • Eventually delete the keys pointing to them from keychain access
    • Uncheck "Ask to join new networks"
    • Reboot
    • Rejoin your wireless.


    moskovit wrote:

    1. Please don't post a solution unless you've tested it for at least a couple of days.  Otherwise, you may be giving false hope and inspiring people to waste time trying things that don't actually work.


    2. Please specify how many days you have been problem-free since implementing a solution.

    Tested on own wireless with Airport Extreme (same as TC but no hard disk) on 7.5.2, 2011 MBPro on 10.6.8, 2008 MacBook 10.6.8, WindowsXPPro SP3, 4 AP-Express (7.5.2), 2010 MBook (customer) on 10.7.1.


    Extreme up without problems now for 48D 1h 43m, MBPro uptime 23D 14h 23m, WinXP ocasionally online, Expresses uptime same as Router, Uptime customers MBook 11 days.



  • Sandpuppy Level 1 Level 1

    This from Apple support today:

    "So sorry for this inconvenience. I just emailed engineering last week and got a response that there is no update yet. I haven't forgotten you and will email engineering again at the end of the week. I wish I had more information for you. I know with the growing need for this to be fixed that engineering is working very hard to find a resolution."

  • mampo Level 1 Level 1



    isn't -20 better than -40 regarding signal strength??

    If you calculate the SNR then you would get 80 which is better than you example with -60 signal strength.

    The lower the better.


    I cannot understand you point that -20 should be worse than -60.

    Neither regarding SNR maths nor looking at the graph.





  • mampo Level 1 Level 1

    lupunus wrote:


    As I could see on the last developer version of 10.7.2 there where NO changes in the wireless stack.

    That is confirmed by reports of people having installed the 10.7.2, hoping that will solve there wireless problems. Guess what, it did not.

    This is not true. The stack changed.

    In the newer seeds, the kernel extension definitely as a different number.

    It also reduced my problem by far.

    I had VERY frequent drop outs.

    I don't have them like that any more.


    They are a lot less frequent now.

    The problem is not completely solved but seems better in my case.

    Cannot speak for all.

    I have a current MBP 15", and I tried all the suggestions mentioned here in the forum.


    All the best


  • lupunus Level 4 Level 4

    mampo wrote:


    isn't -20 better than -40 regarding signal strength??



    thanks for the hint!


    I have to admit that I made a mistake on Sandypuppy's readings. She has indeed a very good signal.

    Shame on me and a note to myself: "Read more carefully"


    As we are talking in negatives here:

    • A noise value of -90dBm is better than -85dBm.
    • A signal of -65dBm is better (stronger) than -70dBm


    Received signal strength reads negative numbers because of the free space and path loss of the signal; e.g. from obstacles, reflections or interferences.


    Eventually the use of a tool like iStumbler will give one a better expression of the situation in his wireless, as iStumbler displays the signal and the noise as percentage of a maximum.

    For example, my strongest neighborhood wifi displays a receiving 20% signal on a noise floor of 13%.




  • moskovit Level 1 Level 1

    Good news (hopefully):


    Mac OS 10.7.2 (just released) may solve many of our wi-fi problems.  Contrary to Lupunus's insistence that the OS is not the problem, and his perception that there are no wi-fi changes in 10.7.2, the release notes say this:


    "Address an issue that may cause a delay in accessing the network after waking from sleep."


    Post here whether the update solves your problems, please, so we can see how successful it is.

  • lupunus Level 4 Level 4

    moskovit wrote:


    the release notes say this:

    We're sorry.


    We can't find the article you're looking for.




    Message was edited by: lupunus --> ...

  • togume Level 1 Level 1

    Bad news:


    I upgraded to 10.7.2 this afternoon (bridging the WiFi by tethering over USB), and the problem persists. Same issue is encountered: The connection is stable for 5-10 minutes before the BSSID changes, and the network connectivity (not wireless connection) drops.


    I'm very dissapointed in the management of this issue. I'm running the latest hardware in a corporate environment where Snow Leopard's networking was stable. Furthermore, I have a colleague that just purchased the same spec machine, which came with Lion from the factory, and has the same issue.


    This is not a hardware issue; this is a software issue where the Wi-Fi driver does not know how to handle changes in the wireless network (i.e. switching BSSIDs due to load balancing).

  • gkillmaster Level 1 Level 1

    I upgraded and now its worse. With my computer inches from the wireless hub, it doesn't detect it at all. In 10.7.1 at least it showed up. There is no menu in my wifi drop down now at all. Just says WiFi: No hardware installed. SO SO frustrating... I've spent many days on this problem now with no clues leading in the direction of a solution.

  • IzzyJG99 Level 1 Level 1

    I was downloading 10.7.2 to upgrade for the 10th time from 10.6.8 (Stable as ****, btw), but after reading I stopped the download. Apple's losing my business more and more each day. This kind of thing should've been addressed and fixed at the very least in the .2 update. At this point I am beginning to think they're doing this in purpose to make us buy new computers. Everyone I know who has a Post-2010 Mac or Airport built in 2010 or after doesn't have the problem. Everyone, like me, who has older hardware (Mid 2009 iMac and original Airport Express) have this problem.


    As I mentioned before I was contacted some months ago by an Apple Technician about this problem via phone. He said they had no idea what was going on. That the engineers saw no flaws in any code or hardware. He asked if he could install a special bug watcher on my iMac while it run 10.7.1. I flat out told him no, I had already moved back to 10.6.8 and wasn't going to waste any more of my time.


    It frankly smells of either bad engineering or something meant to cause people to grow annoyed and purchase a new computer or airport in the hope it'll fix the problem. And guess what? A friend of mine did that now works on that new machine. What a joke.

  • gkillmaster Level 1 Level 1

    I'd give anything to have snow leopard back. Is there any way to go back and keep my settings as they are now intact? I just imagine that may be a nightmare...

  • moonspots Level 1 Level 1

    I bought my iMac in May 2011, and I'm having the same wireless problem and have had no luck fixing it. It's not just older models.

  • benjikan Level 1 Level 1

    I have the July 2011 Quad Core iMac and the problem still exists.  This is the most recent model available.

Previous 1 48 49 50 51 52 Next