dslreports logo
site
spacer

spacer
 
   
spc
story category
OnLive CEO May Have Just Revolutionized Wireless
Developed New, Simpler Radios That Laugh at Shannon's Law
by Karl Bode 10:30AM Thursday Jun 30 2011 Tipped by sonicmerlin See Profile
A few years ago, Steve Perlman, the CEO of broadband streaming game company OnLive gave an interesting talk about the compression, latency and other delivery challenges faced by pushing that much streaming data to subscribers. A few years later (and after a successful launch of the service) Perlman returned to give a follow up talk at Columbia University, and the video of his presentation is again interesting to watch -- albeit this time for somewhat different reasons.

Click for full size
At around fifty-five minutes, Perlman drops something particularly interesting: he notes that another of his companies (Rearden) has managed to find away around Shannon's Law -- which lays the groundrules for limitations on networks and wireless spectrum. According to Perlman, they've developed new radios capable of getting ten times the potential usage Shannon's Law insists is possible, and 100 times what current wireless networks are doing using the exact-same spectrum.

Of particular interest is Perlman claims the radios he's using are actually simpler than current radios. They're also currently capable of 30 mile transmissions, and Perlman believes the devices may be capable of amazing 250 mile wireless broadband transmissions. That means that his company may have not only seriously delivered a killing blow to congestion issues, but created a potential for much more robust networks that could aid this country's coverage and competition issues. If Perlman's capable of following through on this and his findings can be verified, his company may have just revolutionized wireless.

view:
topics flat nest 

Mizzat
Will post for thumbs
Premium
join:2003-05-03
Atlanta, GA
kudos:1

Maybe why AT&T is an investor


Maybe why AT&T is an investor, or perhaps just coiencidence?
--
-M
mogamer

join:2011-04-20
Royal Oak, MI

Needed competition

Don't worry. The incumbants will find a way to prevent this. The same way they've neutered community based broadband. I mean why should they have competition? I really wish the big players on the net, Google, Amazon, Netflix, Apple, MS band together to start up a competing service to the current deadbeat isps.

Bellbubbalou
Premium
join:2002-04-26
Atlanta, GA

Re: Needed competition

The "incumbents" would actually be the equipment vendors that sell the radios to the wireless providers, and if they were able to offer this kind of spectrum efficiency, you bet the wireless providers would be interested!
--
There are 10 types of people in the world -- those who know binary, and those who don't.
decifal

join:2007-03-10
Bon Aqua, TN
kudos:1
If they ever want their cloud computing to take off. They are gonna have to...
Kamus

join:2011-01-27
El Paso, TX
said by mogamer:I really wish the big players on the net, Google, Amazon, Netflix, Apple, MS band together to start up a competing service to the current deadbeat isps.
[/BQUOTE :

Well, for argument's sake lets just say his claims are true and that this would make the tower issue go away, back haul a non issue and latency and bandwidth problems a thing of the past.

You could make wifi routers that would no longer just be access points for your home, but connect you directly to the internet.

So companies like the ones you mentioned could provide the software needed for such a product to come to life, but there would also be a huge incentive for current router manufacturers to embrace this technology, this would be a much hotter product than any WiFi router on the market, because it wouldn't be just for your house, but an internet IP device. This would be like hardware bitorrent.

So while i see why Telcos would never like something like this to come to life (it would basically render them obsolete, at least until even that kind of bandwidth wasn't enough) this would be just too big of an opportunity for hardware vendors.

In other words, if this turns out to be true, i don't think it could get stopped.


Corehhi

join:2002-01-28
Bluffton, SC
said by mogamer:

Don't worry. The incumbants will find a way to prevent this. The same way they've neutered community based broadband. I mean why should they have competition? I really wish the big players on the net, Google, Amazon, Netflix, Apple, MS band together to start up a competing service to the current deadbeat isps.

Wrong in this case the incumbants would be beating down the door.

Vamp9190
Premium
join:2002-02-11
Chantilly, VA
kudos:1

1 recommendation

Enough presentations

Less talking.....more doing

N3OGH
Yo Soy Col. "Bat" Guano
Premium
join:2003-11-11
Philly burbs
kudos:2

1 recommendation

Re: Enough presentations

Agreed.

Until there is independent verification of these claims, I call Vaporwear...
--
Petty people are disproportionally corrupted by petty power

Grethor
Today Is A Good Day To Format

join:2004-04-14
Tacoma, WA

Re: Enough presentations

said by N3OGH:

Agreed.

Until there is independent verification of these claims, I call Vaporwear...

BINGO!
openbox9
Premium
join:2004-01-26
Germany
kudos:2
It'd be nice, but often times these development companies need to "sell" their ideas to gain investor capital to actually build and deploy the products. That's what a lot of these dog and pony shows tend to be about. Vaporware tends to come about when there's not enough funding, or the technology was never viable in the first place and the venture capital is squandered (hello tech bubble).
axus

join:2001-06-18
Washington, DC

1 recommendation

Yeah, I'll side with Claude Shannon on this one.

FFH
Premium
join:2002-03-03
Tavistock NJ
kudos:5
Yes, I'll believe it when it happens. This just sounds like some new data compression scheme than a real wireless technical breakthrough.
--
Record your speedtest.net results in DSLReports SpeedWave
»www.speedtest.net/wave/afe201cb84d45c88
PerfectCode

join:2009-06-12
Portland, OR
said by Vamp9190:

Less talking.....more doing

Is that the power of the Home Depot?

NOCMan
MacChatter
Premium
join:2004-09-30
Colorado Springs, CO

Companies Invent New Batteries All The Time

And each states oh it'll come out in 10 years and have batteries really gotten better?

I'm not debunking this guys claims, but we hear about all these things all the time and maybe 1% of them ever become commercialized.
--
Mac Chatter www.MacChatter.com

HappyAnarchy

@iauq.com

Re: Companies Invent New Batteries All The Time

I know the batteries in my cel phones continue to get better despite increasing demands placed on them.

Same thing for rechargable battery packs in my Xbox controllers vs the last generation of batter packs I used. Even my Gameboys batteries seem to keep improving.

ctceo
Premium
join:2001-04-26
South Bend, IN
Reviews:
·Virgin Mobile Br..

Vaporware

As long as metered billing is on the up and up, and fractal monetary systems are controlling the growth of said systems, this new "revolution" would be no more than the discovery that solar energy alone would make your electric bill all but disappear aside from some simple yearly maintenance to make sure your panels are free from debris.

I'll believe it when I see it.
--
»were.boldlygoingnowhere.org if we don't change out ways!

pnjunction
Teksavvy Extreme
Premium
join:2008-01-24
Toronto, ON
kudos:1

Yeah revolutionized...

with the force of a single powerpoint slide.

It's going to take a bit more than a slide with "10x Shannon! 100x! 1000x!!1!omgwtfbbq" to convince me that he is doing anything but talking shit. Classic CEO hype.

Someone said compression, that's probably it. Like how v.42 or v.44 or whatever could speed my dial-up modem up by 5x, but only for certain types of data.
Kamus

join:2011-01-27
El Paso, TX

The internet the way it's meant to be?

Well, if this is anything like he claims, this would be the first step to turn the internet into a product rather than a service. (why rent access when you can just buy a router that will connect to many others?)

He claims there would be no need to run a back haul cable, and you could just piggy back these things.
He also claims much better latency and capacity.
Oh and let's not forget he's also saying that there would be no need for towers anymore.

With devices using this, you could simply chain these things all over the world and have ubiquitous internet access, then the cloud would stop being a collection of privately owned corporate networks, and instead would be a collection of user owned devices.

So, it's basically promising what the internet should end up being.

But like many of you, i think that until this is verified by his peers. and ends up in a router you can buy, i consider it just vaporware.

If his claims are even half true, this would not only solve the last mile problem, but would also make back haul a non issue anymore.

rchandra
Stargate Universe fan
Premium
join:2000-11-09
14225-2105

Re: The internet the way it's meant to be?

no towers, so then it's a(nother) proposed mesh network?

If so, for administrative reasons, I don't think it will ever work, simply because of the current attitude. End users are typically made responsible for everything which transits their connection. When so, I don't want to take the chance someone might find anyone else's traffic somehow objectionable. I'll take responsibility for the traffic I generate, but I probably don't want to be responsible for someone else's.

And I would argue that's not really how the Internet (if you meant that with an uppercase "I") is aupposed to be. The original architecture might have been intended to have end-to-end connectivity (before v4 address scarcity started to become a problem, thus launching things like NAT). It does not specify what topology that should take, which your post implies to me the Internet should be a mesh.

As soon as a good chunk of the Internet participants go IPv6, that end-to-end connectivity should be restored for the foreseeable future.

If this tech extends range to 250 miles, the great boon there will be the necessity for fewer towers, and if it's architected as a mesh as you imply, it's possible not to have to have separate backhaul to the Internet at large. And it will conceivably supplement or replace satellite-based Internet, especially for those in rural areas. Hopefully this will put downward pressure on rural Internet broadband access prices. (Hmmm.....I wonder what costs more....sticking up the necessary towers, or building and launching a geosynchronous satellite.)
--
English is a difficult enough language to interpret correctly when its rules are followed, let alone when a writer chooses not to follow those rules.

Jeopardy! replies and randomcaps REALLY suck!
axus

join:2001-06-18
Washington, DC

Re: The internet the way it's meant to be?

Yeah, I think people on a mesh network would need to respect network neutrality of packets traversing their way. That could be pretty tough if they are a node for a bunch of endpoints.
Kamus

join:2011-01-27
El Paso, TX
said by rchandra:

no towers, so then it's a(nother) proposed mesh network?

Well, let's be honest here.
This all sounds too good to be true. (and you know how the saying goes)
It sounds like a scam to get VC investing on this.
Then again, to give the guy some credit, this is what i thought of Onlive as well, and while it's far from perfect it is now a reality.

Of course the implications of something like this being real are tremendous, it would turn the world on it's head, and would take care of our bandwidth needs for quite some time.

His comments on the tech were really vague though, he did say it was a "different beast" but of course that's not saying much at all.

Even so, i can't help but feel excited about this, and i would love to know a lot more about this, and just how and when will this be used in the real world.
sonicmerlin

join:2009-05-24
Cleveland, OH
kudos:1

Re: The internet the way it's meant to be?

Here are two of his applied for patents related to this technology:

»www.freepatentsonline.com/y2011/0002410.html
»www.freepatentsonline.com/y2007/0025464.html

Alexq

@qwest.net
Just to clarify, I don't think AT&T has any investment in Rearden.

Also , it doesn't sound like mere compression when he is talking about simpler hardware, it sounds more the stuff used to describe mesh technologies.

This is Pearlman. I am sure he has had his share of failure but he doesn't seem to do BS and apparently doesn't need to.

He did WebTV which MS bought which turned into Ultimate TV which turned into Xbox.

He did Quick Time for Apple.

He and someone else did Android for Google or he at least had some substantial involvement.

He did Mova a motion capture firm which along with some other similar firms has helped glitz up Hollywood lately and prefigured (to an extent) MS's Kinect and changes interfaces.

He did OnLive when so may people were insisting it couldn't be done.
Not only that he is pushing it into every device and seems to have no viable competition at the moment.

Also in partnership with Juniper he is promising 10gb connections when even Google in its Kansas city project is only doing 1gb. Further he is promising to do the browser at this speed and on the server side in a way that bypasses silly caps.

This guy is real and I'd take this with as much credibility as if a Gates or a Jobs said it. Further this is not Gates talking about say Terapower and what might be, while cautioning that everything under sun be tried. This is Perlman saying they have working units now.

Also we have precedent for this kind of leap in bandwidth in cellular in particular. There was a discovery in the late 80s or early 90s that through open what was previously though possible.

If you are wondering why he is making noise about it now, it may have been as a pretense for his attack on a IP law. Its clear patents aren't working, they reward corporations and stifle innovation too much, but it seems Perlman is dead set against one the bills to address patent gridlock or rather corporations getting sued for IP infringement.
Cobra11M

join:2010-12-23
Mineral Wells, TX

ok

so will AT&T and verizon ever use this kind of tech??, idk AT&T prob not verizon maybe, it sounds pretty promissing i mean 30 miles to what ever, if you think outside of traditional thats a heck of lot of space covered by one tower.., and if really able to handle more data..then thats good to...i personally dont understand how thought and im sure im not the only one..., haha cause wifi and cell towers can only hold so many people and then they slow down of course.. hmm we will see i guess, anyone else inlighten me??
rradina

join:2000-08-08
Chesterfield, MO

Re: ok

If there's sufficient download bandwidth, the 30 mile thing could work for the receiving side but the sheer number of clients transmitting TO THE TOWER in the almost 3,000 square miles such a tower covers is nuts! How could that ever be coordinated to provide any meaningful throughput? Even if the radio scheme can support it, what kind of hardware is necessary to sort through that many routing requests at once? We could be talking millions of devices. To put in perspective, if one million devices wanted one packet of data in the same second, that's 1,000 requests per millisecond to achieve a rather pathetic 1,000ms latency.

Then there's the 250 mile claim which incinerates the 30 mile claim.

slader

@amis.net

Rearden already plans to spin-out a company on this tech

More here: »onlivespot.blogspot.com/2011/06/···sed.html

slader

@amis.net

Steve Perlman Wireless patent

Here's an article about the underlying patent and the link to the patent: »onlivespot.blogspot.com/2011/06/···eve.html

slader

@amis.net

Steve Perlman names his tech DIDO

DIDO => Distributed-Input-Distributed-Output technology.

More here: »onlivespot.blogspot.com/2011/07/dido.html

tcubed_bus

@sbcglobal.net

Perlman follows Cooper and Reed

Marty Cooper founded ArrayComm because he was (and remains) convinced that 10x RF spectrum capacity improvements were doable. He's certainly got the chops to make such a claim. Today, Cooper blames a combination of telco naysaying and corporate profiteering (i.e. incumbents make money by perpetuating artificial shortages) for his business problems. The technology is fine.

David P Reed has spent years arguing for a more nuanced understanding of spectrum capacity. In particular, his thoughts on RF interference, S/N ratios, and Shannon are worth reading (reed.com/dpr). A sample:
Information loss is a systems design and architectural issue, not a physical inevitability!
If Perlman is talking space/time diversity, MIMO, phased arrays, and coding, he could be on the money, and well within understood technology, with his claims.