Last Active
  • FCC will probably restore Net Neutrality on April 25

    Net Neutrality is based on a bunch of big bad "what if's" that "might" be implemented by Big ISPs. It's a solution looking for a problem. There are only a few Big ISPs with the clout and marketshare to even try to do what they're claiming, while there are over 2000 smaller ISPs (like mine) that'll never be in a position to even consider such things without suffering.  Customers would drop them in a heartbeat.

    The problem is nobody thinks about the "what if's" on the government/regulatory side.  Sure, the current majority leadership of the FCC (who are appointed, by the way, not elected, and therefore face no ramifications for their actions from the people/entities within their stewardship--just a bunch of attaboys and pats on the back from the people who are in power) have signaled that they will cherry-pick only "relevant" rules from old Title II 1900's-era (AT&T days) regulation.  But what if a future administration decides to implement more of the existing (old) rules? Or come up with another loophole to add more?  

    Or worse: what if the government "Net-Neutralizes" one or more ISPs out of existence for not toeing the line when it comes to the ISP standing up for free speech or other constitutional issues, or trying to exercise its own rights?  The same administration that wants to "Neutral-ize" the Internet has used all kinds of means to try to keep Elon Musk from buying/holding onto Twitter, coerced all the social media platforms to manipulate speech during the pandemic, and they're questioning Apple's "monopolistic stronghold" on their own smaller fraction of the market.

    Congress has given so much rule-making power away to the Executive branch's agencies for so long that, before Congress can do anything to solve the problem (do they do that nowadays?), the damage is done.

    We should all be very wary of any government entity offering to step in and take more power (i.e. create more regulation) in the interest of the common good, especially based on "what if" scenarios that the free market itself can (and often does) correct. 
  • USB4 Version 2.0 to offer up to 80 Gbps data transfer

    urahara said:
    Here is a sequence for you to come down a little: 802.11 -> 802.11b -> 802.11a -> 802.11g -> 802.11n -> 802.11ac -> 802.11ax -> 802.11be ;)

    WiFi's more of a tree with branches. Some standards are band dependent.

    2.4GHz = 802.11b -> 802.11g -> 802.11n
    5GHz =  802.11a -> 802.11n -> 802.11ac (WiFi 5) -> 802.11ac Wave 2 -> 802.11ax (WiFi 6) -> WiFi 6E -> 802.11be (WiFi 7)
    6GHz = 802.11ax (WiFi 6E) -> 802.11be (WiFi 7)
    60GHz = 802.11ad -> 802.11ay

  • Elon Musk and T-Mobile try beating Apple with satellite vaporware

    I'm guessing a lot of people didn't watch the event.

    This will use a thin slice of existing T-Mobile mid-band (PCS, 1900-2100MHz) spectrum, which they have licenses for across the entire US.  This isn't necessarily 5G, and they said "the phone you already have."  I'm guessing that they'll only need to implement the 4G/LTE spec (or a lite version of it), enough to support voice and simple data.

    According to Musk, the antennas to support this will be roughly 5-7 meters wide and tall, and there will be serious electronics involved to counteract doppler effect etc. Because cell phones transmit omnidirectionally, cell antennas on towers are already designed to be really big ears to pick up the weakest of signals.  Licensed spectrum is much quieter than Part 15 (unlicensed) bands where WiFi and baby monitors and IoT devices live.  So they wouldn't have put out the "vaporware" if they didn't believe there was technical capability. 

    There's also only so much bandwidth available (2-4Mbps) for the whole spot beam, so they're starting with the easy stuff. SMS uses the SS7 signaling network (which runs on most telco phone switches across two 56K channels). They've got oodles of bandwidth to support texting and messaging, and with store-and-forward capabilities onboard the bird, that part's pretty much figured out.

    Since T-Mobile already owns the spectrum, the biggest challenge could be regulatory issues.
    just cruisinmuthuk_vanalingamhtbduryeigoh
  • Apple's M1 Ultra combines two M1 Max into a powerhouse chip

    The keynote also mentioned "the last" M1 chip in the lineup was Ultra.

    My bet is Mac Pro gets M2 (Max/Ultra) with each chip having 2 interconnects instead of one and perhaps 128GB per package, supporting quad/hex/octo configs within insane core counts.
  • T-Mobile will shut down Sprint's 3G network in March

    davgreg said:
    The only area rough around the edges is the millimeter wave 5G which is still problematic. I was in Nashville near Vanderbilt Thursday and the millimeter wave was up and running but the network not well optimized - crazy fast signals outside and nothing inside a simple brick restaurant. It was like this in Memphis recently as the same tech was rolled out but was much improved within a month or so.
    Physics.  The lower the frequency, the greater the likelihood it will travel far and go through objects.  Converse is true: the higher the frequency, the shorter the distance traveled and the more likely items are to interfere with that frequency's propagation.

    Lower bands = lower bandwidth channels; higher frequency bands = huge bandwidth channels.  Wider channels travel shorter distances; narrower channels can go further, generally speaking.

    The addition of 700MHz for LTE and 600MHz for 5G to T-Mobile's portfolio mean they will benefit tremendously from the lower frequencies, but you won't get crazy fast on those bands at great distance (I would argue that anything over 100Mbps to a cell phone > 1 mile away is pretty cool).

    Millimeter wave bands such as 28-47GHz, their Part 15 counterparts, 24GHz and 60GHz, and the lightly-licensed 70/80GHz E-band are amazing. I have deployed point-to-point links on 24-, 60-, and 70/80GHz, and they range in capacity from 1Gbps to 10Gbps.

    But those frequencies don't propagate through most solids. The antennas have to have line of sight to each other. At anything over 18GHz, rain begins to cause problems. 
    At 58-65GHz, oxygen absorbs the signal. Human bodies certainly have a blocking effect, and building walls most certainly will too.

    The 5G spec has all kinds of redudnancy/resiliency/aggregation built in to allow the phone's radios to switch back and forth between bands, based on what's available.