World's Fastest Internet Arrives in Tokyo

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.


40Gbps lol... I think as of now, anything beyond 1Gbps is practically useless. Atleast for a standard user. Most of the computers today only have 1 HDD, which most modern HDDs nowdays can probably write at about 100 - 120 MB/s. 1Gbps equals about 125MB/s, which equals about the same speed as a HDD can write. If you are downloading something at that speed, your computer will lag, because the HDD is being written so much data, and plus the operating system is writing data also. Until SSDs becomes the standard for storage, 40Gbps will be still too fast to write on a SSD, which most SSDs writes at 300 - 550 MB/s depending on the SSD. So, internet too slow may be a problem, but too fast can be another problem as well.
 


My point was, that for $69.95/m, on a 2 year contract in AUS, I get 200Gb @ 10Mb/s.
Hardly groundbreaking. I can get faster speeds on 3G during off peak hours.

Assuming a decent quota, I would certainly pay $3/m more for this.
 
ikyung, There is sushi that costs double then that in Japan and we don't complain about it, besides if there was an internet that was that powerful in the USA I'd pay triple that amount in one week then one month.
 
ikyung, There is sushi that costs double then that in Japan and we don't complain about it, besides if there was an internet that was that powerful in the USA I'd pay triple that amount in one week then one month.
 
No one in my neighborhood could afford to even get fiber lines out to my house -.- My family would be willing to pay up to probably 200$ for a 1Gbps connection.
 
Recently traveled to China, and in some areas, they even have 100Mbit/s fiber optics internet for a reasonable price around $35 U.S. dollars per month. Why can't the US have those prices? I think lack of competition is a huge problem. ISPs even sign agreements to controll each area they serve. And FCC isn't even doing anything about it.
 

Gotta love duplicated effort.

Most of the internet runs over Ethernet-based routers and switches. Ethernet already has specs and production equipment for 100Gbps with specs for 400Gbps in the works.

Seems a bit late to invest in 100Gbps research when 100Gbps has been a production standard for over a year; just not too popular yet due to cost and low line card port density partly because the current QFP optics are huge and partly due to limited backplane bandwidth - most of the newer modular routers cannot handle more than 2-3 non-blocking 100GbE ports per slot and not-so-old ones can only accommodate one due to their 100-160Gbps backplane interface.
 


Ehm, I don't think that EU and Japan don't know what they are doing...
The record for multicore fiber is atm at 1,05Pbps. But that doesn't mean its cheap. What EU and Japan want to do probably is to bring such high speeds from theory to practical and cheap solutions. As the article states the average user has 19Mbps. So trying to provide cheap 100Gbps to every residence isn't something that its already done....
 

Except the (EU/Japan 100Gbps) article is about upgrading ISP, carrier, transit, etc. networks' capacity. Not bringing 100Gbps to everyone's door.

I was calling it an unnecessary duplicate effort because every network operator who handles enough traffic to use or want to use 100GbE anywhere in their network wants it to get cheaper too. So you have all the major ISPs, carriers, transit providers, etc. all over the world interested in cheaper 100GbE.

There is no point in thinking about 100Gbps for the masses when it is still far too expensive for carriers to justify using it beyond their most heavily loaded trunks. It has to become much cheaper at the (inter-)carrier level before it has a chance of getting anywhere close to the general public.

That's why data rates in telecoms are usually ~15 years ahead of data rates on the desktop: 10GbE became an official standard in 2001 but 1GbE only became standard on PCs in 2005... 1/10th the speed four years later. If this repeats after 100GbE becoming an official standard last year, I would expect 10GbE to start popping up on desktops in about four years.

With 8-ports 10Gbps switches still costing over $800 despite being decade-old telecom technology, I'm not expecting people to rush into upgrading their home/office LANs to 10GbE.
 


That isn't too terrible especially in a developing country like Pakistan. What about the upload speeds?
 
Status
Not open for further replies.