The new forums will be named Coin Return (based on the most recent vote)! You can check on the status and timeline of the transition to the new forums here.
The Guiding Principles and New Rules document is now in effect.
[PATV] Wednesday, May 30, 2012 - Extra Credits Season 4, Ep. 15: Spectrum Crunch
Dog
Registered User, Administrator, Vanilla Staff admin
Registered User, Administrator, Vanilla Staff admin
This week, we discuss a major bandwidth problem we're about to run into.<br />Come discuss this topic in the <a href="http://extra-credits.net/episodes/spectrum-crunch/#discuss" target="_blank">forums</a>!
Dog on
0
Posts
While I think that the FCC should seek to allocate spectrum as efficiently as possible, I'll hazard that over the air TV use is (temporarily) on the rise as people cut cable in favor of Internet streaming. And the cellular companies have an onus to make the most efficient use of the spectrum they have been given: I'd like to see them roll out true 4G services and upgrade their customer's phones before they go begging for spectrum from FCC/NASA. It might even be more economical for them to use their existing spectrum x% more efficiently than to buy x% more spectrum, and it would generate massive goodwill with their customer base.
OTA TV isn't legacy. Hook up a UHF antenna to your TV and you'll see a picture better than what cable offers.
http://www.wired.com/epicenter/2011/06/perlman-holy-grail-wireless/
You can watch his talk at Columbia Engineering School where he talked about his wireless tech first, if you are interested in hearing the words from the horses mouth. He starts talking about his new wireless tech about 55 minutes in if you want to skip past his presentation of Mova and OnLive.
And here is a PDF I just found, that describes DIDO (Steve Perlman's magical wireless tech).
http://www.reardenwireless.com/110727-DIDO-A New Approach to Multiuser Wireless.pdf
It sounds like it might not be that far away. So he'll probably come along next year or so and save the world from the "Spectrum Crunch."
Now we're headed face-first into a bandwidth crunch in 2014 that would be completely unnecessary if AT&T were allowed to make their acquisitions. Not to mention this whole bandwidth auction mess was started by the government with its whole FCC-in-charge process of auctioning off bandwidth
I was reading about this very issue at least a year ago in the Wall Street Journal when communication companies openly spoke about the writing that was on the wall. If you want I can try to dig up some articles from when I read them (Sadly the hard copies have long been recycled) or you can do some digging yourself
The angle here that the government and FCC are going to be the ones that resolve this issue is naive at best. There is no way that the government can better allocate a resource such as wireless spectrum than a free market with a healthy free media
You want to see some crazy numbers, look at fiber optics layed down since 2007 and the increase in the amount of bandwidth there. Compared to the increase in traffic. Through out the history of the internet bandwidth has always out paced our use and continues to do so now. There is no reason to believe this trend has changed. The people screaming "the sky is falling" are the same ones who said we wouldn't be streaming 1080p EVER.
The "bandwidth crunch" has a ton of solutions, many of which predicted back in the 1990s. But most of them aren't implemented because this is a fake problem. 2014, lol. I hope these guys revisit this post when 2014 comes around and dish out some serious humble pie for their fearmongering.
This set of slides was some serious hokum. Go look at the wireless data growth leads to spectrum deficit slide. lulz, stop trolling, guys.
There are two basic problems that can create loss of service quality. The first problem is bandwidth saturation, as mention here. But that only happens in crowded places, like say grand central station. The trick here to alleviate the problem is to put up more antennas with less power. Less power means two antennas that use the same frequency can be places closer together without interfering. Here the real problem is investment costs for the antennas. The important point here is that is has nothing to do with overall data usage, you can use as much data as you like, you just get it slower in a crowded space.
The second problem is a lack if foresight on the part of the telcos. When you had voice communication your average connection was local. The advantage is that you do not have to maintain much infrastructure for regional, national and international networks. But the internet changes everything. Now most of the time you are connecting to some server which is on the other end of the world. Suddenly most of your connections are at a national level. The wired telcos has to battle that problem in the 80s and 90s and are mostly done with it. But the smartphone revolution hit the wireless operators right in the face and they where not prepared. Now they are making up a big fuss about it. And THAT is the reason you won't get a unlimited data plan; has nothing to do with bandwidth saturation.
No we are going to hit a hard limit soon, since there is a bunch of unused RF band. (Listened to AM radio recently?) And don't think telcos can just buy the problem away, unless they bring up a new standard and you get a new phone that frequency band is as useless as it is now. Congestion happens in very crowded areas, but it is not the massive problem we are lead to think it is.
I mean, really, Draw Something is going to have problems? Might as well say the entire cellular network isn't going to work. Oh noes, OnLive won't work on your phone. This doesn't mean it's doomed, it just means you'll need to be at home.
If you tackle this issue in the future Extra Credits, please take the time to be more specific with what you are describing, and what exactly is associated with this particular issue. To be honest you came off as fear-mongering at one point, and you're not even rallying people behind a cause, so it is completely non-constructive.
Looking over the pdf, I will say it's a novel approach, and could be promising, but good god in heaven, I cannot stand how it's presented. This is one of the most annoying papers I've ever read. And the degree of "Oh, all the scientists we talked to either told us it was impossible, or were just baffled at how it could be possible," in the press about it? Ugh. If someone said it was impossible, you probably just did a bad job explaining it, because it's really not that complicated. It's clever and novel, but it's not that complicated.
Honestly, if the guy would just present his ideas and be like "I think it could work. If my numbers are right, we should see <result x>." But he doesn't. He's like "Yeah, this thing just totally blows away everyone who's ever seen it, maaaaan. It's gonna completely revolutionize everything, maaaaaaan. It's so cutting edge, we're seeing shit that we don't even fully understand, and we're the ones making it, maaaaan. It's gonna fix all our problems, and then candy's gonna shoot out, and it'll even be magic candy that gives you sweet abs."
your = belonging to you
their = belonging to them
there = not here
they're = they are
1) There is no practical or even theoretical limit to the maximum amount of terrestrial bandwidth available. Basically, you can run fiber optics all over the planet and get more bandwidth to any given point then we could ever conceivably use. It's not even that expensive.
2) Once you have terrestrial bandwidth near to where you need it, you put up a wireless access node. There may be practical limits to how many nodes you can have in a single location, but we'd be talking about not being able to stream 1080p to a stadium worth of cell phones, not your typical coffee shop or commute area.
3) In short, the answer is just more cells, which means more money. As soon as customers start demanding more bandwidth and are willing to pay for it, there will be more cells (or micro cells) built and the problem will go away.
--
Note also that bandwidth AVAILABILITY is what is driving the usage, not the other way around. Netflix would not have been conceivable in 1995-2000 when everyone used a dial-up modem to connect to the internet. Now that everyone who wants it had high speed cable, DSL or even Microwave links, it's totally doable. Right now no one would write an FPS for tablets because of bandwidth limitations. But they won't always be there. There is a LOT of money to be made in offering more and faster connections.
Case in point:
Do you remember back when cable modems used to be the WORST, because each neighborhood shared the same node and if your neighbor was a huge Napster user your experience was suck? Notice how that is no longer the case? That's because the cable companies can charge $50/month for internet access and can therefore pay to roll out infrastructure to service those customers with the bandwidth they want/need.
Yes, it looks like a BIG NASTY PROBLEM now . . . but it is eminently solvable. It just takes investment.
But on this issue i must agree on mmecha, ash7, rioki and all the others, there is not that big of a crunch, maybe for wifi but not for wireless data. So yeah, you might need to change your phone before 2014 to be able to use the new wireless broadbands on different frequencies. Considering that in some country, most people keep their smartphone less than 2 years. There is no reason to panic.
But I am suprised how technologicaly aware your public is, reacting intelligently in the comments and everything to point out your mistakes. You should feel lucky to have such a great audiance with as much critical sense.
I also agree with mmecha on the blury distinction between bandwith and wireless bandwith capacity.
Anyway, i am affraid this episode lacks a bit of research.
I still love the show though, i'll be there next week to watch next episode. Love you guys, keep on the good work. Good job community for not believing everything you're told and offering your mind.
Think of those frequencies as bandwidth and the maximum number of times they can be subdivided as throughput: the number that determines if bandwidth is running short in the first place. With a wire, that would be the amount of electricity that can be passed through versus how much electricity the wire can hold. Or how many bands of light can be passed through a fiber optic cable versus the maximum number of bands of light that can be passed in the first place.
The good news though, again, as they mistook 'cellular network' with 'Internet', this applies mostly to wireless internet and any wireless device like your smartphone. The LAN connection in your house wouldn't suffer from this.
Course, are they talking about memory address space? A different issue all together, but still important, as the combinations being used up in the TCP/IP protocols are running out and have been for a while.
Sadly, these guys are not tech experts and they wouldn't know what to look for, so when they did their homework, they had to quote experts and pray they were in fact experts.
Though seriously, unless you can prove that the FCC is on the take or using lax hiring practices or something, stop claiming they don't know jack about anything they deal with.
And if you spend the time to watch the show you'll note they make it clear they're talking about wireless the entire time. The issues they express are distinctly applied to cellular devices throughout the narrative. They close by saying the issue is solvable but at a higher level than the normal grassroots actions they advocate. We're looking at an issue of municipal infrastructure colliding with corporate interest and the broader market. Suggesting that we can just 'do a thing' many thousands of times in order to fix the issue, and that that plan is both obvious and easy, misses the larger point, e.g. that we live in the real world, not an EE 101 word problem.
http://www.nab.org/documents/newsRoom/pdfs/042511_Solving_the_Capacity_Crunch.pdf
Data growth - there's only so much data that a human being can digest, and while we're not at the limit by any means, there's a strong chance we're past the biggest boom. Also, what we're losing is not bandwidth, it's spectrum. So no matter how big games get, or how high the definition of our streaming videos is, it's not going to be a problem. The issue is how many devices can be in the same space. Sharing link space is one of the biggest things preventing us from maximizing our potential over the airwaves. The major problem being that wireless radios have a lot of trouble detecting collisions (Multiple transmitters transmitting at once, therefore destroying the data from both) - chiefly because of the way signal propagates. See the law of inverse squares. At distance D from the source, the signal's strength is reduced to 1/D^2 of it's initial strength. This means they're transmitting very powerfully nearby their sensors, and picking up on much weaker signals. As mentioned earlier, this can really be solved by adding more infrastructure.
Technology - People can be so blind about the growth rate of technology, even today. Trust me, the space is subdividable. Frequency is directly tied to wavelength, and wavelength could theoretically be distinguishable down to the Plank Length. We just don't have the technology to further subdivide it. But we will. Believe me, we will. Once upon a time, streaming live video at 1080p seemed like the fever dream of a madman. But today, people do it every day. Just be patient, and the smartest minds on the planet (Not me, of course) will sort it out.
Innovation is at the core of the Human condition. We are unique among all the critters because of our incredible ability to surmount seemingly superhuman tasks. I'm sure someone will figure this one out, too.
Or they could go the RIAA route and try to render the new model so inconvenient and/or legally questionably to cling on to what gives them profit.
Why?
All of our terrestrial TV signals have been turned off already. I think there's a very differant system of broadcast licencing laws which allowed for that one.
The problem with gaming only comes in when there are patches, updates or with digital distribution. I'm sorry but if you try to download a full game off of a cellular data connection, you either live out in the middle of nowhere or you're an idiot. Though to be fair if you're in the middle of nowhere cellular network infrastructure is most likely crap. Or severely underutilized. The same thing goes for these services threatening to send you video feed of their cloud servers running Crysis for you maxed. Not that the game makes any impact on how much bandwidth the video feed consumes, it is still a damn video feed. It could be Plants vs. Zombies for all I care. That is the kind of service I would block with extreme prejudice at work. I understand your work computer sucks but I'm not buying a larger bandwidth line for you to stream video games on.
I understand that some people have to use cellular data to work/etc. I understand that some people think "Screw you all I am a consumer I'm going to eat bandwidth like Cheetos at a LAN party!" Trust me I have to tether at work since most locations I visit have secured LANs that I can't access. Instead I suggest that people be intelligent about their bandwidth usage while this spectrum problem is solved. Try not to stream video or download large files like full games/large patches if you are on a cellular data connection. LANs are still king when it comes to this stuff. Instead try to just play a game or read. That's the wonderful thing about smart phones, most have the ability to connect to wireless LANs and use that data instead. I know that ISPs are being lazy and imposing data caps, but that is caused by a problem that we can punch through. Force them to lay more fiber and improve the backbone network infrastructure. That should solve some of our cell data crunch too since cell towers tie into said infrastructure.
TLDR version: Video streaming and downloading games on your cell phone is retarded. Play what you have downloaded until you get to a LAN.
Also, to piggyback on a comment in the duplicate thread, switching to landlines is almost as difficult an option as freeing up the over the air bandwidth. Even in the most connected countries there are a large number of areas where a high-speed landline connection is unavailable, not to mention unstable.
OnLive is an example of Bad Programming. This kind of thing needs to stop. The calculation for the the 40 hour HD video needed to play a typical game will far out weigh me having a system that can just run it itself.
Games that require Broadband in order to play with only 4 Players are examples of Bad Programming. When I could play Phantasy Star Online 4 Players No Lag over dial-up that's an example of good programming. Me being unable to play a game like Diablo III just by myself over Dial-up Lag Free is an example of Bad Programming and wasted bandwidth.
You want full 3D games running on your cell phone or iPad then get programmers up to speed on parallel programming. The only way it will happen is properly utilizing multicore technology, and currently they are very bad at multicore programming.
http://gizmodo.com/5889900/the-pasta+shaped-radio-waves-that-promise-infinite-bandwidth
http://iopscience.iop.org/1367-2630/14/3/033001/article
this talk is about one of the possible solutions to the bandwidth problem