Researchers unlock fiber optic connection 1.2 million times faster than broadband

It's going to affect the internet backbone first, fiber optic does go all the way to homes in some neighborhoods but it will be awhile before you can buy a card that will take advantage of that Tech, that will fit into your PC ... AND ... be realistically affordable. It might even take a new motherboard or even a whole new PC, but until that time in the future, the bottle neck will be YOUR equipment.

I see it really benefiting Virtual Reality. Designers won't be limited to the data densities they have to work with now.

 

Probably not much. We have it at work. I have cable at home. I don't see any real difference in terms of the user end. It probably makes a difference at the server levels with so much stuff cloud based, real-time remote backed up, and huge amounts moving back and forth simultaneously that we never appreciate until it decides to stop working.

 

I agree with what you said, but I'll add this.

Comcast gives me 800 megs as part of the package I have. Hard wired I get around 750 megs. Now going wireless depends on which device I'm using and where I am in the house. My Androids get around 350 when I'm close to the modem, and about 130 megs just 30 feet away and around two corners. Right now I'm in my garage smoking a cigar and the computer (on wireless) is giving me 48 megs on a speed test I just ran, I'm 25 feet away from the modem and there are 3 walls in between. Keep in mind that the speeds I 'supposedly' have differ greatly depending on which speed testing website I use, and of course, Comcast's speed test measures the fastest speed.

As a former field/tech ops manager for several cable companies, I'm trying to figure out what kind of speeds can ultimately be seen using coax. Only a small percentage of homes in the USA are capable of having fiber ran all the way to their modem. What's happening now is a fiber is ran to a node that will feed about 150 homes (in a big city there can be thousands of nodes), the node converts the light to regular RF signal (be it analog or digital) and then a typical cable TV type distribution system using coax cable, amplifiers and line extenders takes that signal into the home.

 

So I was thinking, to take advantage of this Tech, you would run fiber to the room where your gaming or main consumption work is being done and go wireless in the rest of the house taking advantage of "mesh" technology. I'm, basically, doing that in my home now, only it's cat 6 from the pedestal to my "main consumption room" and wireless everywhere else, and I just have a high end router that seems to cover the whole house just fine. I can even take a laptop up on our roof deck and work fine. Coax is dead, I pulled it all out when we moved into this place 4 years ago; don't miss it'.

I'm retired IT (Systems Engineer) and am certainly not up to speed on this or any other technology. I have no clue what kind of connection it would take to actually be able to take advantage of it. PCI slots, even muxed together may not be capable of that kind of throughput; maybe it could, I don't know, But it could require a whole new leap in PC technology. Anyway, my intuition and experience (which admittedly is way out of date) tell me it's going to be awhile before you see those kind of speeds at a home desktop.

 

Not a critique; just expanding on this from a techy perspective. Fiber is considered difficult in the general consumer household because the cables are very finicky and get damaged easily. Additionally, the fiber based switches tend to be very expensive. That's part of why the NBASE-T Alliance was formed (as well as increasing throughput to POE wireless access points without having to pull new cables). These switches are more affordable, and you can get 2.5-10G over most existing cables people have in their homes already.

Wireless meshing is great, but something to keep in mind is that you will be eating up the channels with the backhaul tunnel between APs. Any time you can wire an AP, it's worth doing. The OP mentions they were tuning this with optical amplifiers and other means, which effectively means this will likely be optical carrier-grade technology and won't be feasible for consumer client use. It also seems as if this was probably just a test to see if they could get new wavelengths to work from a theory perspective and then creating products around it will take quite some time. Carrier-grade optical equipment is pretty specialized technology, and there are only a few people who make products in that space. If these researches start to show more promise than theory, I'm sure they'll patent it, and they'll get gobbled up if the current manufacturers aren't already going down this path on their own.

 

From my experience, there are two localized problems with fiber in the home: micro bends and dirty connectors.

 

Very good post. The link in the OP references ONLY the fiber optic cable. The C and S bands have been unavailable because they haven't had the ability to modulate the laser light at those frequencies fast enough to achieve the data rates. Previously, there were no materials capable of modulating in these bands at high enough rates. Fiber optics don't care about the rate. They have always been able to transmit these frequencies. They managed to develop a technology which allows them to modulate the light in these bands at a high rate. Like Cornpop so eloquently stated above, this is a long way from having these data rates to your home.

 

I've been trying to keep the concepts and jargon minimal so everyone can participate. You're not wrong about what you've said but I was attempting to interpolate those concepts into something easily understood. I don't disagree that a wired connection is always a better choice for mission critical work. But most homes even with teenagers have minimal throughput users. I have a buddy that has 4 kids all on all the time, gaming, school and social. They use what was "the living room" as a internet room for the kids probably to isolate the noise and have everything out in the open so it's easy to "monitor" by the parents. IF I were him I would have my IP DMARC go into that room and run through a gig e wireless switch that the kids could plug directly into. I might run some cat 5 to my office or my wife's and into a gig switch, but it's just not necessary and we're pretty heavy users, not gamer's though. He's more tech than I am and works from home so he may be wired off the main router somehow; probably. His wife could care less so as long as alexa is happy.

Anyway back to the point, yeah fiber is a PITA that's why, in some future world were fiber is common to a home, I'd have my DMARC in the high use room and serve everyone else with cat 5 or wireless and expect no problems. But, the choke point is still going to be the cat 5 patch cable into you PC ... unless or until technology gets to the point you have a optic card in your PC. Even in a business situation I don't see anyone needing that kind of workstation. Sure, it will happen eventually, and if the past predicts the future, it will be driven by recreational users; either for gaming or simulators of some kind. Between AI and VR the day will come when there will be a demand and need for that kind of through put, but that's not today or any day in the near future.

But, by then there may be some new tech that we can't even imagine.

 

While I don’t disagree I feel you are trying to diminish how absolutely amazing this is. They were obviously able to capture the speed of this download time meaning that some device they had was quick enough to gather the sample at that speed. This may not be in civilians hands fast but it won’t be slow either.

 

I'm not trying to diminish this at all. I recall when people could only get dial up internet and the speeds were 1.5 megs. Then along came cable TV internet and the speeds at the time were 5 megs. Now we have fiber, sometimes to the home, with 1 gig speeds. All these advancements first started in a lab and were then rolled out to the public. But going from 1.5 to higher speeds at best only required inserting an ethernet card in a computer that might not have come with one. With the speeds we're talking about in the article, I'm trying to think through what needs to be done on the consumer end to make those speeds feasible.

 

Based on images of their setup I don't believe they captured the download speed on a client. It seems more likely they proved they could use the bands and measured the optical modulation. The maximum speeds are likely theoretical. They would need to create a switch, transceivers, network interface cards, non-bottlenecked storage, and a system to support them to measure client download speed. If you look at the image of their setup it's just fiber patch cables hooked up to what is most likely an optical amplifier or something similar to one. Carrier grade optical network switches look nothing like that.

 

I work with fiber everyday that I work. And yes, the switches do look like that.

 

Switches generally have interfaces, transceivers, fans, ventilation, status lights, and use higher gauge power cables than anything coming out the back of that on them.

 

I guess to clarify what I believe they have done in case I wasn't clear before.

Those look like fiber patch panels which are used to identify fiber ports for small fiber cables to interconnect fiber connections. Fiber switches have fiber interfaces on them on standardized interfaces. Those look like they could be MPO cables based on the horizontal shape, but MPO transceivers have pull tabs to remove them that would be clearly visible. Additionally, other transceivers would protrude from the panels slightly and you would see some metal showing. Switches and routers also have status lights and the interfaces are typically laid out in a predicable pattern. They also use high gauge power cables due to the power requirements.

Fiber patch panels have none of those things and look closer to what is shown into the image. What it seems they're doing here is using fiber patch panels, fiber patch cables, and devices to push light through the patch cables and devices to measure the reception on the other end.

Those look like 32 port MPO patch panels with the front protection panel removed for easier access like the image below.

There's nothing in the image that appears capable of pushing actual data. It seems like they are doing experiments with light and cables to verify they can transmit and receive on specific bands.

Carrier grade optical switches can push light on multiple wavelengths at the same time and some transceivers can do this to a small scale as well. By proving there are more wavelengths that can be used they have proven that we can multiply the amount of traffic that can be pushed over the same cable. Hence why I keep referring to carrier grade equipment and this technology not expected to hit the consumer home market.

If any of this doesn't make sense please let me know and I can expand. What they seem to be doing isn't actually that complicated from a high level perspective. The trick will be replicating it into actual network equipment.

 

I've set up a fiber optic network before and it turns out that optic cable is very difficult to troubleshoot connection issues with, and you have to get everything to match exactly. There are several standards of optic cables that aren't compatible with each other, and they have to match across all your equipment or else the connection speeds are even slower. That's why, when gigabit transition speed came to ethernet and wireless adapters optic cable networking fell away into obscurity. I'd love if optic cable networking was easier to use.

 

To be successful with fiber it takes knowledgeable planing, deep pockets and well defined goals. None of which are surmountable but it's a serious commitment. Setting up a fiber network reminds me of this scene from "Moneyball".



It's really not a home networking solution. That's not to say it won't be some time in the future, but there will need to be some standardization and dumbing down done to make it practical and a valid reason for it; right now there's neither. Getting the kind of speeds described in the OP to today's desktop is as unnecessary as it is impractical.

 

I rarely had problems troubleshooting fiber. Most of the problems either come from bends tighter than the radius of a baseball or dirty connectors. In the field, the first thing we did was check the light levels and clean the connectors. If that didn't fix the problem, and most of the time it did, we'd break out an OTDR and shoot the fiber and look for any events. Any anomaly will cause a reflection and the OTDR will tell you in feet how far away the problem is. You can shoot fiber for miles. The only problem here is if the reading comes back with an anomaly at 39,875 feet, where is that? That reading will get you in a general area, but after that you have to start looking for splice enclosures, signs of digging, or even a tree falling on the cable. Finding a bullet hole, something that happened too often in Pomona, CA, was hideously difficult, especially if the utilities are in the backyards. That's when you had to pick a spot in the fiber, cut it open, connectorize it and shoot it again and see if that helps you zone in on it. If worst came to worst, we would cut two spots and shoot it both ways from there. If the event or bullet hole was in the middle, it's quicker to run new fiber.

 

You are spot on! I'll take one step further, unless the infrastructure is completely fiber it will be like super highway being funneled into old two-way streets.. Granted it will be a good start, money well spent, but how long will it take to bring fiber into neighborhoods house by house?

Someday though!

Reality is, the U.S. is struggling to keep up roads and bridges..

 

Wonder if the same technology used in fiber optics can be used in lasers between Starlink satellites.

 

I hate to pour some cold water reality on this interesting post, but the speed difference has to come from the coding and terminal equipment. Fiber optics at any bandwidth has a slower EM transmission speed than any copper or over the air microwave facility.