Remember Intel's LightPeak, that optical interconnect which promised incredible throughput that Apple somehow managed to quash? Thunderbolt is interesting, though certainly expensive and offers only a part of what we were promised at what seems an exorbitant amount of money. At the upcoming IDF Intel promises to introduce an optical connector which is similar to what LightPeak was although it will be intended for server interconnects as opposed to removable devices. However at 1.6 Tbps MXC will be impressively fast and Corning's new ClearCurve LW fibre technology will prove to be rugged enough to survive through the bends and snarls which inevitably occur when two or more wires are put in close proximity. Check out the link to the abstract through ExtremeTech.
"Ahead of the Intel Developer Forum next month, Intel and Corning are teasing a new optical interconnect technology capable of 1.6 terabits per second. Dubbed MXC, the interconnect is designed to supercharge the interconnection of servers in data center environments, where current networking technologies are struggling to keep up with the massive growth of cloud computing."
Here is some more Tech News from around the web:
- Radeon samurai and Linksys bowman cross swords in battle of PC scrap art @ The Tech Report
- Xerox admits there's no fix yet for number-fudging copiers @ The Register
- Google: Cloud users have 'no legitimate expectation of privacy' @ The Register
The reason why LightPeak
The reason why LightPeak failed is predominately because they tried to enter the USB market when it wasn’t needed. In fact, Intel intentionally held back USB 3.0 parts for nearly 2 years in a massively failed effort to push Thunderbolt onto consumers. It’s sickening a company will hold back a technology in that way just to push their own product that at the time offered no real difference.
Other reasons it failed is it compromised itself from being a solid optical only interface, to one that is tied down by copper. This was done because they knew it wouldn’t have a chance at replacing USB devices that are often times powered by USB controllers. That compromise further lost them the reasoning of why people should replace something that is already blazing fast by today’s terms (USB 3.0 5Gbps) for something that is simply twice as fast. Big whoopie.
Lastly, if they did stick with a pure optical interface, sure they could boost speeds incredibly, but then you’d still need a separate wire connecting each device. Something in this day and age is just not going to be taken seriously in the consumer market.
It could have a major impact if they went the Displayport/HDMI route where bandwidth is in short supply and going to be drastically needed for future resolutions. Even Displayport 1.2 which far exceeds HDMI 1.4 specs, will only have HDMI 2.0 to compete with on the same grounds. Neither of those are going to be serious in terms of gaming at UHD resolutions at 60FPS. Not 4K and certainly not 8K. This could raise the bar and dominate the market early if they wanted to and got the right people lined up, which they could with their fat wallets.
All this of course goes
All this of course goes without saying. In the Networking world and for backplanes this could be a huge deal and it probably will be.
While I have no insider
While I have no insider knowledge, its always appeared that Intel wasn’t particularly interested in Thunderbolt. It seems that they’ve intentionally made it difficult for vendors to get certified (e.g. the ASUS add-on) and generally prevented AMD partners to get involved.
In short, I think Apple pushed them into it and Intel has been dragging their feet. Really though, USB 3.0 is fast enough for nearly all peripherals and given its backwards compatibility Thunderbolt & LightPeak will probably remain a niche market.
A bad moves intel !!!
A bad moves intel !!!