100 Gigabit Ethernet standard ratified It's official: the IEEE 802.3ba 40Gb/s and 100Gb/s Ethernet standard has been ratified by — who else? — the IEEE P802.3ba 40Gb/s and 100Gb/s Ethernet Task Force. "Ubiquitous adoption of bandwidth-intensive technologies and applications, such as converged network services, video-on-demand, and social networking, is producing …
"...When Apple's Macintosh was released in January 1984, it was the first mass-market PC to have networking built in.."
Not sure what you mean by the 'first' - you could buy a BBC B with built-in (on the motherboard) EcoNet in 1981-2. Not many of this varient were sold in the UK (not many UK homes had a multi-system network requirement then), but all the US export ones had this in 1983....
When I try to copy gig's worth of data across my 100Mbs switch I see a huge difference to when copying it ovew direct 1Gbs link between my two computers. And that is on 4 year old laptops (One Core Duo, the other Core 2 Duo).
Not quite top of the range any more...
And 6 years ago cheap Xeon servers could also transfer huge files much quicker across a 1Gbs connection than across 100Mb connection.
434,027.8 times faster - not bad.
Moores law (calclated at 18 months) would only have it 165,140 times faster. It shouldnt be getting that fast till 2012. Mind you, that might be the time that the first consumer kit is released to he mass market with it built in.
Its pretty accurate, moores law.
Until FTTP or FTTH connections become available in your neighborhood, don't count on it. Even this announcement (mostly) assumes fiber optics for transmissions (the copper specs don't permit more than 10m for 100GBASE-CR10 connections). It's a safe assumption that all that copper used for local loops won't ever likely support much more than 6Mbps unless you're super-close to your Telco central office (less than 1km).
This post has been deleted by its author
Geezus... "Moores law" was only ever about the number of transistors in a single integrated circuit.
If I might equally inappropriately invoke "Newton's inverse square Law", it [Moores Law] has the square root of F-All to do with network throughput, speed, capacity or anything else other than that to which Moores Law actually refers.
I wish people would stop invoking Moore's Law every time some spurious aspect of computing technology/specification comes up, especially SPEED, which for some reason seems especially to attract abuse of poor old Moore.
The connection over the years between Moore's Law and any metric other than transistor count is entirely coincidental.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
