Super-fast Wi-Fi coming: 802.11ac-2013
zdnet.com
zdnet.com
Pretty much all this guarantees is that wireless n routers will stay $50, AC will stay $100 - 150, and AC-2013 will be $200+, due to complexity. Maybe in a few years n routers will spin down demand and AC will have market effects drop the price, but then I worry about spectrum limitations. We already see huge issues with n routers congesting on the 2.4 band, and making channels even wider for AC means congestion leads to even worse bandwidth and range loss.
Sure, 2.4Ghz is hilariously congested, buy as 5Ghz devices are sold, it will be less so and what I half expect is when people start trying to beam 5Ghz through three office walls (and fail), they'll want to know why and eventually find themselves back on a 2.4Ghz solution, simply because it's not so sensitive to solid objects.
I don't either, which is part of the reason I've spent a couple dollars on ethernet cables that connect some of the obvious places. A year ago I wrote this comment (https://news.ycombinator.com/item?id=5192451):
"Ethernet was developed in the context of the internet with its seven levels of the ISO reference model," he said. "So we had a job to do at level one and two, and we didn't burden Ethernet with all the other stuff that we knew would be above us. We didn't do security, we didn't do encryption, we didn't even do acknowledgements."
Very interesting story—I wish it had been more detailed in places, but I love this middle section, about how simplicity sometimes beats complexity.
It's also amazing how prevalent Ethernet still is, even when wireless is a competitor. The other day I left this comment: http://news.ycombinator.com/item?id=5052448 on HN, because in some circumstances running a cheap ethernet cable from a router to a desk, couch, or other work station can still be a real win, especially given how inexpensive even very long ethernet cables are from Monoprice.com.
They last forever, aren't subject to the level of interference wireless is, and, in many conditions, have faster data transfer speeds. Ethernet is still great.
- there is security at the Ethernet level with 802.1x. Mostly used in professional settings, to prevent anyone with access to an area to plug into an Ethernet jack and sniff traffic with ARP poisoning or access anything;
- the ACK support for WiFi is perfectly fine at L2. The reason Ethernet doesn't do ACK is that the PER is very low so there's really no need for it. The reason why all wireless standards do some form of ACK is that wireless is very lossy and relying on higher layers for retransmission would give poor performance (retransmission at L2 is faster, and transparent to TCP). There's ACK for WiFi but in wireless LAN (e.g.: 4G/LTE) that are also lossy and with a relatively high latency you even have two levels of ACK: at the L1/L2 with HARQ, and at L2 with ARQ. Different contexts, different mechanisms.
Anyway, can't beat 1GbE for low latencies and high speeds, it's still faster than an AC router, unless maybe your computer is right next to it.
Go ahead, do whatever you want with the laptop on the wireless, the roku won't care because its wired.
Buying cables and hooking everything up that you can, is by far the cheapest, fastest, and easiest way to upgrade your wireless experience.
Or they will put a single CAT6 to each room and put a low power 5GHz AP on it. If you are the only person in a few thousand feet 2.4GHz is fine. You cannot expect to have the ability to penetrate walls and have numerous neighbors doing the same.
My main snag with WiFi right now is its half-duplex nature and dropped packets. 100Mbps WLAN is horrendously worse than 100Mbps LAN for networked storage
I personally ended up using power-line ethernet to replace a lot of the wireless throughout my apartment. While the wireless connection was much faster in theory (and in practice when everything was perfect) than the powerline ethernet speeds, trying to stream HD movies and such was always an exercise in frustration with the wifi no matter how good the routers and/or repeaters were on the wifi connection and how well they were set up. With the powerline ethernet adapters, OTOH, I just plugged them in and everything Just Worked, much slower peak speed, but nice and consistent and more than enough bandwidth for HD movie streaming.
In any case the horrible state of US ISPs is the bottleneck for networking at home for me right now regardless of how I wire myself internally. My two options are Time Warner Cable (which cuts out between the modem and their headend for minutes multiple times a day and was never fixed despite half a dozen on-site technician visits) or AT&T DSL which has been much more reliable but is annoyingly slow for big downloads and such. I'm currently using the AT&T connection, but both options are terrible in their own way and nothing else usable is available to me.
Fuck you, US ISPs.
Powerline networking is still a shared medium. You're sharing the available bandwidth with all of the neighboring houses on the same utility transformer.
It does have the benefit that signal strength drops off with linear wiring distance. Thus, you're farther from your neighbors than you would be with Wi-Fi.
However, as with 5 GHz, you're getting good results right now because powerline networking isn't too popular. But if every house were to get powerline networking, then you would start encountering interfering networks.
Yes, but not enough attenuation to block powerline communications.
This is not a theoretical concern. People have tested HomePlug units and successfully connected to neighboring houses. There's even explicit provision in the spec for neighboring networks, so that they share the bandwidth instead of stomping all over each other.
Ideal wifi: 1) should be fast 2) should be reliable
I would much rather see optimizations on #2 than #1 at this point. I agree that once speeds hit 100mbps, reliability is a much more important engineering goal to focus on.
Wireless is and can be very reliable. Just go get a FCC spectrum licence for a dedicated and unshared frequency and watch your reliability issues disappear. It's just expensive.
Reliability on open spectrums is always going to be an issue, and one you should never depend fully on. There will always be old equipment running that doesn't work well with the new standards and tosses it's new found gains out the window. Unless you can convince each of your neighbors to throw away that ancient 2.4GHz phone handset and old B router.
Reliability on telco wireless is a different issue. They would rather run more customers on a segment, reducing reliability, but increasing profits.
A drop in quality corresponding to lower bandwidth--delivered reliably--is acceptable. Freezing on a perfectly-detailed closeup of a stupid expression because someone came as the internet went (out) is not.
Reliability >> speed.
I had tough luck getting even 802.11n speeds on Linux. Are there any USB WiFi cards which actually work? I got D-Link DWA-160, but it never connects on the n band, only on b/g which limits it to approx. 50 Mbit/s. (I'm using rt2800usb driver).
Especially when we know machines are routinely compromised to circumvent user freedom, why on earth would you want to help give a foothold to others in your personal space computationally?
I don't think that Internet of Things is that big of a practical gain, but I do see the potential there for some pretty cool applications. Imagine if every power switch in your house could be controlled from your phone. Or if you could do inventory in a warehouse by asking each crate what its contents was. Or a farm where every tomato plant can tell you if it needs more water.
The privacy issues are separate from the possibilities that the technology provides. Surely we can figure out a way to build a system that's secure. Or choose not to introduce these things into our daily lives. I think just the industrial uses are very interesting here.
Let us draw a difference between private and industrial IoT. IIoT (your tomato example, maybe shipping, etc.) is done for practical logistical reasons, and likely as not will end up on a private network or embedded system. There is little likelihood that a government acquiring the tomato files is going to ever truly hurt the consumer, because there is an airgap of commerce between that information and the consumer--I buy the tomato from a supermarket with cash, and the trail goes cold.
PIoT, though, means that I have dozens of little spies all reporting on every aspect of my life, and that eventually that data is collected and correlated in one place--as it must, if it is to be of maximum utility. And that data, once concentrated, is something worthy of scrutiny by third parties. It suddenly makes it a lot more useful to spend a little effort to break into a management system or throw the book at a service provider, because the return is so great.
Moreover, even without extracting the data, the devices required afford computational capabilities, and the only truly trustworthy computer systems that exist are those that are powered off and disassembled--anything else is suspect (for the truly paranoid).
I just don't think that the personal internet of things is a good idea.
I would be curious how you think that open source will make PIoT any less risky or a compromise any less dangerous.
See eg. http://www.oreillynet.com/pub/a/wireless/2003/08/08/wireless...
Then I would hope Router could work on their software and not crash every month or so. It would be great if they could restart itself every week during at night when there is Zero activity.
Added with a decent implementation of USB 3.0 / 3.1, so i could use the external USB HDD as an NAS instead or buying a separate unit.
I have plenty of 100mW APs, and AFAIK they use that power all the time. Sure would be considerate if they'd scale down to use the appropriate amount of power required to reach each particular node.
802.11ad also has engineering work being poured into protocol adaption layers, so interconnects such as DisplayPort or PCIe can run atop it and presumably the 802.11 MAC that's supposed to be core to 802.11ad.
Does that mean you need 8 antennas to max out the range and bandwidth? Cool!
Network backup is another potentially time consuming task, but this is typically done late at night. Enterprise WiFi adoption is unlikely because this makes a company vulnerable to hacking and DOS (via interference or jamming) attacks.