Friday Variety: Resolving Bluetooth Anxiety (Plus The Rest Of The Cellphone Summary)
Speaking of cellular reception, I had a happy breakthrough on Wednesday. As I’ve mentioned several times in the past, most recently on Monday afternoon, Apple computers running OS 10.5 ‘Leopard’ (whether from the factory or user-upgraded) and containing Broadcom transceivers have imperfect Bluetooth implementations. Specifically, the problem is with Bluetooth’s PAN (personal area network) profile; a wirelessly tethered computer connects to the phone using it and even gets a valid DHCP IP address assignment, but subsequent Internet access using the phone as a wireless modem is so slow as to be unusable (practically speaking).
‘Leopard’-based Macs with non-Broadcom (i.e. Cambridge Silicon Radio, etc) transceivers don’t seem to exhibit the issue, nor do Broadcom-inclusive Macs running prior OS 10.4. I spoke at length with Broadcom about the issue at CES, and although company representatives were aware of the problem and had repeatedly communicated their concerns to Apple, they couldn’t fix the issue themselves since Apple reportedly develops its own drivers. My multiple attempts to contact various individuals at Apple have to date been similarly unsuccessful. And last weekend, I finally (successfully, it seems so far…phew!) upgraded to OS 10.5.6, but as discussion group posts had already prepared me, this latest update released in mid-December didn’t fix the problem, either.
My job, as you can probably imagine, begs for robust online connectivity no matter where I am. Historically, tethering a laptop to my Dash or a prior-generation T-Mobile phone has been a reliable means of accomplishing this objective, for times when Wi-Fi was unavailable or I didn’t want to pay for it (the T-Mobile Dash’s EDGE bandwidth is generally faster than that delivered by my alternative, archaic, USB-tethered Sprint 1xRTT-based mobile phone). As such, the Bluetooth PAN problem was the only significant issue I stumbled across in migrating from an OS 10.4-based MacBook to my current OS 10.5-based MacBook Air that I wasn’t able to work around…until now
Microsoft removed alternative Bluetooth DUN (dialup networking) support (which reportedly still works fine in OS 10.5) from the version of Windows Mobile 5 running in my T-Mobile Dash. A hack reportedly restores Bluetooth DUN to the device, but only with Windows Mobile 6 versions, and if I were to go that route I’d also be forced to upgrade beyond Outlook 2000. Windows Mobile’s Internet Sharing applet alternatively supports phone-to-computer connections over USB, wherein it implements RNDIS (serialized Remote NDIS)…but OS 10.5 doesn’t support RNDIS, either.
I even took a stab at trying WMWiFiRouter (which enables the phone to act as a miniature Wi-Fi access point, with the cellular connection as its WAN port), but the program refused to install. Judging from other users’ feedback, I was reasonably confident that I could get Bluetooth tethering working within Windows virtual machines running on the MacBook Air, but I ideally wanted to get the host O/S connected, too.
My next idea was to use a non-Broadcom-based USB Bluetooth adapter as a means of superseding the internal transceiver. I’d stumbled across an online discussion thread which described a means of disabling integrated Bluetooth via a developer utility, so in order to get the 2.3 MByte Bluetooth Explorer program on my system, I shoehorned the entire 2+ GByte Xcode suite onto my miniscule-HDD machine. Believe it or not, the MacBook Air’s built-in Bluetooth drivers repeatedly crashed Bluetooth Explorer. And the first USB Bluetooth adapter I tried, Zonet’s ZUB6101C, was also Broadcom-derived (and completely disabled the computer’s Bluetooth capabilities when I plugged it in, to boot).
I had much better success with my next candidate, a Cambridge Silicon Radio-based D-Link DBT-120 whose integrated firmware I’d previously upgraded for use with (among other things) a PowerPC G4-based Power Mac running OS 10.3. When I plugged the DBT-120 into the MacBook Air, OS 10.5 prompted me to create a new Bluetooth PAN profile for it. For good measure, I re-paired the T-Mobile Dash to the MacBook Air via the DBT-120, and I also re-ran Bluetooth Explorer to confirm that the DBT-120 was now the primary transceiver in use (it was by default, as another Internet discussion thread had already tipped me off would likely be the case).
Then I crossed my fingers, launched Internet Sharing on the phone and hit ‘connect’ to activate its Bluetooth PAN facilities. Switching to the MacBook Air, I connected it to the T-Mobile Dash-generated Bluetooth network, and saw that it’d (correctly) gotten a different DHCP-provided IP address assignment than before when I’d tried this maneuver via the laptop’s integrated Broadcom resources. The final step, a ping of Yahoo’s URL, was successful wherein it’d consistently failed before (in case you were wondering, this wasn’t just a successfully surmounted DHCP server problem…I’d tried hard-coded IP addresses in the past without success, as well), and email, web browsing and other Internet-based tasks now worked, too.
On paper, at least, the DBT-120 seems like a step backwards from a performance standpoint, since my version B4 unit only supports Bluetooth v1.1 speeds whereas the Broadcom transceiver inside the MacBook handles Bluetooth 2.1+EDR. And so far, EDGE-based MacBook Air Internet sessions via the T-Mobile Dash do seem slower than they used to be with my Broadcom-inclusive and OS 10.4-based MacBook. But my testing isn’t yet extensive enough to be statistically significant, and anyway the Dash’s Bluetooth 2.0 support doesn’t include EDR’s speed boost. Plus, we’re talking about EDGE, not UMTS HSDPA. Bottom line: at the end of the day, slow connections are better than no connections, right?
So there you have it, Apple users, a workaround until the company decides to release a patch. Perhaps obviously, tethering is most relevant to laptop computers, so owners of all MacBook Air iterations along with recent-generation MacBooks and MacBook Pros are candidates. Cynically speaking, I doubt Apple will be sufficiently motivated to implement a fix until the company’s cellular service partners start offering iPhone tethering-over-Bluetooth data plans. Why, after all, would Apple want to do anything that’d encourage you to stick with your existing cellphone? (any sarcasm you may be sensing is intentional)
Speaking of slow connections, T-Mobile’s taken its sweet time in rolling out 3G data services compared to both its GSM and CDMA competitors, although at least part of the blame might lie elsewhere. As such, I got excited when nearly two years ago I started getting word about EDGE speed upgrades that were compatible with existing network and handset hardware. I haven’t heard anything more since last March, though, and with UMTS rollouts now well underway in major markets (and LTE trials in progress) I doubt Ericsson and Nokia Simens’ pronouncements will amount to much. Still, I dream…
Wrapping up, I’ll pass along several other mobile telephony-related links I found interesting and suspect you also might enjoy:
- What Do Cell Phone Reception Bars Mean?
- Four bars? The disconnect between bars and cell signal
- How Mobile Phones Work Behind the Scenes
- The Definitive Coast-to-Coast 3G Data Test
Happy weekend, all.
wrwhifzzydt commented:
4hyrDa , [url=http://biwlslvsckli.com/]biwlslvsckli[/url], [link=http://hjhoupxnhspf.com/]hjhoupxnhspf[/link], http://zqspqruplfas.com/
rtafhu commented:
fvf8CF sgsfqnoohxrx
kxbqcqozewk commented:
KH1py5 , [url=http://qqumgnnaspoe.com/]qqumgnnaspoe[/url], [link=http://htartjglfpbt.com/]htartjglfpbt[/link], http://iwangezmxxez.com/
nmathnaf commented:
RVWkQS melcgvacjafg
Jonalyn commented:
Why do I bother calling up poeple when I can just read this!
Trevion commented:
Articles like this are an eapmxle of quick, helpful answers.
Brian Dipert commented:
iDavid commented:















