News to make file sharing opponents poop their diapers

[sonaboy]

There is a new type of Hyper Text Transfer Protocol on the horizon that can travel and be decoded hundreds of times faster than traditional HTTP.

Top speed reached in closed tests at this point say the max speed attained was something like 8,000+ Mbps.

Entire DVDs of info can be downloaded in under 6 seconds.
wow....*rubs hands* bring it on!!
http://news.nationalgeographic.com/news/2003/03/0318_030318_internet.html

[LiquidX]

ahhh that would be great...get my porn in seconds instead of minutes. I doubt the public will get it in there homes anytime soon though.

[sytaylor]

i remember in 1997 seein on tv future technologies that would be "drastically faster than 56k modems".. called DSL...

So if thats anything to go off we should have it by 2009..

[sonaboy]

Originally posted by sytaylor
i remember in 1997 seein on tv future technologies that would be "drastically faster than 56k modems".. called DSL...

So if thats anything to go off we should have it by 2009..

?? i dont' know about YOU, but i got my cable setup in about 1997.
besides...that's not the point of this article..this is a whole new LANGUAGE, not hardware.
DSL took so long to implement because the burden was on the telecommunication companies to upgrade the phone lines to handle the voltage etc..
this kind of leap in tech requires only a new set of instructions for windows, linux and macs to recognize.

[LiquidX]

Originally posted by sonaboy
?? i dont' know about YOU, but i got my cable setup in about 1997.


1997......wow thats news to me, I must have been in my cave. I didnt even there was anything faster than 56k back then. But then again; I am late, I have had dsl for less that a year now.

How much did cable cost back then?

[sonaboy]

Pretty much the same. 45 a month.
It didn't perform as well back then, because most of the nodes were bottlenecked from the user expansion...they figured that out pretty quick though...i can't think of how many times i sent in a complaint..

I also live in Sprint's hometown, Kansas City, and we had broadband DSL (Ion, i think) before most other people did.

also a cool thing back then, was that the signal from the cable also had a standard tier piggy backing on it..
so if you bought the right kind of splitter, you could get "free" cable from the company without them knowing.
nowdays, they separate the package - especially since fiberoptic/digital cable came into play.

[Smokey]

Originally posted by sytaylor
i remember in 1997 seein on tv future technologies that would be "drastically faster than 56k modems".. called DSL...

So if thats anything to go off we should have it by 2009..

Im sure I pre ordered my ADSL with BT in 2000, then I left and moved to France, but Im sure the UK got ADSL in 2000 :confused: well at least in London!

Still 2-3 years wouldnt be bad :p

[Son Goku]

Originally posted by sonaboy
There is a new type of Hyper Text Transfer Protocol on the horizon that can travel and be decoded hundreds of times faster than traditional HTTP.

Umm, new type of HTTP? Not to nit pick too much, but where did you read this from, the article mentions a new type of TCP. At first reading this thread, I was thinking that smoke was being blown up our butts from somewhere, in that, well though one could potentially decode faster, HTTP won't result in data travelling down the physical pipe faster. I nonetheless took a look at the article, and it doesn't mention http

The data was transferred over shared research networks in the presence of background traffic, suggesting that FAST can be backward compatible with the current protocol, Caltech said. The FAST team has started to work with various groups around the world to explore testing and deploying FAST TCP in communities that need multi-Gbps networking urgently.

BTW, just for those who might not be most familiar with the OSI model, TCP or transport control protocol is a protocol that operates at layer 4, or the transport layer. HTTP, operates at layer 7 (or the application layer) of the OSI model. These aren't the same thing.

Whereas HTTP is more involved with the displaying of web pages, TCP is more involved with flow control, establishing a connection or "virtual circuit" between peer systems/applications, and provides acknowledgements for received segments, so the one's not recieved can be re-transmitted, as well as re-assembling these segments at the receiving end. Note, this differs in some respects from UDP which also operates at the transport layer.

Looking over the article, I would assume, but couldn't say without looking into their research further, that there is also support for what they're doing among the lower layers (network, datalink, and physical layers) or other improvements they're working on in their research project also. One shouldn't expect a improved TCP protocol to give one giga-bit network transfers on a 28.8 modem, etc... I haven't looked beyond the article to confirm what they're doing beyond what was stated however...

[LORD-eX-Bu]

wow, I actually understood alot of what you posted. Guess these network administrator certification courses I took a while back are paying off:D

[sytaylor]

Originally posted by Smokey
Im sure I pre ordered my ADSL with BT in 2000, then I left and moved to France, but Im sure the UK got ADSL in 2000 :confused: well at least in London!

Still 2-3 years wouldnt be bad :p

For it to filter down to a price most can afford its taken until now. Try asking BT when theyre gonna roll out optical fibre the luddites break out their bats :rolleyes:

[StealthHawk]

i wouldn't be too concerned about this in either case. hard drives are just so terribly slow in comparison that consumers will never get close to the maximum transfer speed. bandwidth will still cost money.

i'm willing to bet that the average person's hard drive write speed is under 30MB/sec. in other words, even if this technology became widespread in 5 years, it still wouldn't have quite the impact that you suggest ;)

[sonaboy]

Originally posted by StealthHawk
i wouldn't be too concerned about this in either case. hard drives are just so terribly slow in comparison that consumers will never get close to the maximum transfer speed. bandwidth will still cost money.

i'm willing to bet that the average person's hard drive write speed is under 30MB/sec. in other words, even if this technology became widespread in 5 years, it still wouldn't have quite the impact that you suggest ;)

All I know is, with cable - the fastest DL speed i've hit thus far, is around 440 Mb per second - and if i could sustain that for ALL downloads regardless of size, this would be a welcome feature in all my net surfing and file transfers.

As for the diff between HTTP and TCP - whichever, whatever.
The whole point of this post is that they're working on an instruction set that transfers data hundreds of times faster than what the norm is now, and that it's worked well in closed tests.
If you're curious about "where i got the article from" well...that's why i POSTED THE LINK with my original post. Read! :spank: :POKE:

[Son Goku]

Yes, I know what is in the article. They've mentioned a technology they're working on, and a TCP protocol specifically. It makes sence someone isworking on something, because technologies such as giga-bit ethernet, OC-92, etc already exist. These aren't technologies that the consumer will be seeing anytime soon. An OC-3 line for instance (which one of the universities I'm going to has) is prohibitively expensive, though the BW available there is nice. Back when I was taking network topologies, someone came in one of our classes, and mentioned that Juniper was working on a tera-bit network technology, though I don't know what has become of it... These aren't the sorts of things home consumers would tend to buy, unless they're Bill Gates, or have near his wealth at their disposal.

BTW, in the case of giga-bit ethernet (the cards aren't so prohibitively expensive as OC-x and they can use CAT 5e cable), though ethernet is only a LAN technology and not what one would connect to an ISP with. There is another issue, and that is the PCI bus. Gigabit ethernet can transfer 128 MB/sec ((1024 MB/1 GB) / (8 bits/byte)). This is rather close to the 133 MB/sec one is talking about with a 33 MHz 32-bit PCI bus...and though it is slightly below, the bus bandwidth is shared among other devices... Yes, we could use PCI-X or NGIO comming on our motherboards as time moves on...

A WAN technology would still be needed to see these transfer rates, and unless prices drop rapidly or a cheaper form of high speed WAN technology is released, this is what could break the bank. Currently things like OC-92 are posessed by the telecommunications companies at their own back bone, not home users...

[The Baron]

Hmmmm... it's not like anyone (except business consumers) really NEED gigabit ether. Unless you're goofy/have way too much pr0n/stuff like that. ;)

Of course, nobody needs more than 640k either.

[dohcmark8]

I need gigbit internet, all mt pron 24/7. Gigys of it every single day. Then any harddrive smaller than 400GB would become obsolete.

[Ryoko]

Originally posted by The Baron
Hmmmm... it's not like anyone (except business consumers) really NEED gigabit ether. Unless you're goofy/have way too much pr0n/stuff like that. ;)

well, it would be nice to run a backup in under an hour...

[Son Goku]

Originally posted by The Baron
Hmmmm... it's not like anyone (except business consumers) really NEED gigabit ether. Unless you're goofy/have way too much pr0n/stuff like that. ;)

Of course, nobody needs more than 640k either.

Hehe, try telling me that when I downloaded the ISO images for Linux (several CDs worth), or the OS betas and RC's from MSDN subscriber downloads. Even at 200 KB/sec+ when I lived in the college dorms, the thing took rather long to download. I wouldn't have minded the entire OS in less then an hour :p Need, perhaps not, but would like or want... Since when do people in western culture settle for absolute necessities anyway? hehe People don't technically need technologies our ancestors didn't have thousands of years ago for a human being to survive either. But then again, without quality to life, how many people would want too? :lol: