Intel Shows First Light Peak Laptop, 10Gbps Transfer Only Beginning!

[.Markus]

Intel shows off first Light Peak laptop

Intel has provided the first hands-on demonstration of a laptop running its Light Peak technology, at the company's inaugural European research showcase here in Brussels.

Light Peak is an optical interconnect that can transfer data at 10Gbits/sec in both directions. Intel hopes Light Peak will one day replace the host of other PC interconnects, including USB, DisplayPort and HDMI.

Intel has fitted Light Peak into a regular USB cable, with optical fibres running alongside the electrical cabling. Intel provided a visual demonstration of how data is passed through the cable, by shining a torch into one end of the cable, with two little dots of light visible to the naked eye at the other end.

The demonstration laptop was sending two separate HD video streams to a nearby television screen, without any visible lag. The laptop includes a 12mm square chip that converts the optical light into electrical data that the computer understands. The technology hasn't yet been integrated into the screen, which explains the ugly black box sitting between the two devices.

Intel's chief technology officer, Justin Rattner, claimed that the bandwidth afforded by the optical technology is practically unlimited. "Light Peak begins at 10Gbits/sec, simultaneously in both directions," he said. "We expect to increase that speed dramatically. You'll see multiple displays being served by a single Light Peak connection. There's almost no limit to the bandwidth - fibres can carry trillions of bits per second".

Rattner said the technology will find its way into everything from home PCs to server farms. "The potential of that headroom will lead people to rethink the design of their systems," he said. "We've very, very excited about the potential of Light Peak."

An Intel spokesman said Light Peak hardware should start to become available to manufacturers by the end of this year.

Article @ PCPro

''There's almost no limit to the bandwidth - fibres can carry trillions of bits per second".

[itzwolf]

Now I like this idea!

[Aldur82]

''There's almost no limit to the bandwidth - fibres can carry trillions of bits per second".

But there is a limit on the conversion of those light pulses into electrical signals, and then the bandwidth of the BUS on both sides of the connection.

But it's still pretty awesome!

[ahhell]

It's about damn time that everything moved in the optical direction.

HDMI was a lame stop gap. It should have been optical to begin with.

[Knife Party]

great (Y)

[PixilEyes]

If Itel does this right USB 3.0 is pretty screwed.

[hdood]

So... Is it going to be an open and reasonably priced standard?

Intel hopes Light Peak will one day replace the host of other PC interconnects, including USB, DisplayPort and HDMI.

Because you own it and are hoping to rake in the money?

[Elliott]

Because you own it and are hoping to rake in the money?

Probably, but it makes perfect sense. With this much bandwidth, there's not a single connection that couldn't benefit from this.

[hdood]

Probably, but it makes perfect sense. With this much bandwidth, there's not a single connection that couldn't benefit from this.

There are cons too though. Most hardware wouldn't benefit from it at the moment, because they simply can't (and don't need to) operate at those kinds of speeds. Your keyboard now needs more expensive hardware, including an optical transceiver. At the moment ASICS, microcontrollers, and application processors have on-chip USB. The only external electronics normally used is a couple of resistors. That makes for cheap and accessible hardware, even for hobbyists.

Of course, while this is an unfortunate side effect, it's not a real argument against new technology. The technology isn't really the problem. The problem is it being owned by a single corporation and not open. I don't know if that's true for LightPeak, but if it is, it's not good. I don't want everything to have to have Intel-made LightPeak controllers. In that scenario, I think it might be more preferable to go with something else, even if it means reducing the bandwidth in half.

[qdave]

Hopefully in 5 years of time, it wont be in research-stage.

[c3ntury]

Now I like this idea!

Same, im hoping it will work its way to server farms and datacenters very soon! :D

[Rudy]

If Itel does this right USB 3.0 is pretty screwed.

Why do you think there's no USB3 on most new motherboards? They made it a lot harder for companies like ASUS to add the technology because unlike USB2 they have to add another chip made by someone else (because Intel hasn't bothered to include USB3 support yet).

If Intel plays their cards right USB3 will never really take off and they'll push Light Peak to everyone

[PixilEyes]

Why do you think there's no USB3 on most new motherboards? They made it a lot harder for companies like ASUS to add the technology because unlike USB2 they have to add another chip made by someone else (because Intel hasn't bothered to include USB3 support yet).

If Intel plays their cards right USB3 will never really take off and they'll push Light Peak to everyone

I agree.

[Ci7]

Why do you think there's no USB3 on most new motherboards? They made it a lot harder for companies like ASUS to add the technology because unlike USB2 they have to add another chip made by someone else (because Intel hasn't bothered to include USB3 support yet).

If Intel plays their cards right USB3 will never really take off and they'll push Light Peak to everyone

look like LP will takeover the system wiring and will use other interfaces like USB3 first

they are saying product featuring LP will start to pop out by the end of the year

[ew2x4]

I won't be impressed, until I get some sort of storage that will operate at 10 GB/second.

[Growled Member]

Light Peak is an optical interconnect that can transfer data at 10Gbits/sec in both directions. Intel hopes Light Peak will one day replace the host of other PC interconnects, including USB, DisplayPort and HDMI.

That doesn't mean it will replace these other technologies. Even if it does it's going to take years to take place.

[Ci7]

I won't be impressed, until I get some sort of storage that will operate at 10 GB/second.

there is a huge difference between Gb/s and GB/s

[Roger H. Veteran]

I won't be impressed, until I get some sort of storage that will operate at 10 GB/second.

That's Gbps people, so in reality you might get 1024MB/s (or 1GB/s) It's not too far off but then you gotta realize LP is not just for storage, your monitor and all this other stuff can be connected via LP so then the bandwidth available wont be just for storage, it will have to share. Some SSDs are pushing 300MB/s and they will only get faster by 12/2011 :)

[Ci7]

That's Gbps people, so in reality you might get 1024MB/s (or 1GB/s) It's not too far off but then you gotta realize LP is not just for storage, your monitor and all this other stuff can be connected via LP so then the bandwidth available wont be just for storage, it will have to share. Some SSDs are pushing 300MB/s and they will only get faster by 12/2011 :)

actually some SSD now push over 600MB/s if i am not wrong but that is using PCIe interface

i can think of OCZ Z drive of top of my head

[Vice]

There are cons too though. Most hardware wouldn't benefit from it at the moment, because they simply can't (and don't need to) operate at those kinds of speeds. Your keyboard now needs more expensive hardware, including an optical transceiver. At the moment ASICS, microcontrollers, and application processors have on-chip USB. The only external electronics normally used is a couple of resistors. That makes for cheap and accessible hardware, even for hobbyists.

Of course, while this is an unfortunate side effect, it's not a real argument against new technology. The technology isn't really the problem. The problem is it being owned by a single corporation and not open. I don't know if that's true for LightPeak, but if it is, it's not good. I don't want everything to have to have Intel-made LightPeak controllers. In that scenario, I think it might be more preferable to go with something else, even if it means reducing the bandwidth in half.

If Intel decide to use the USB connection like they have so far then it won't be an issue as cheap components like Mice and Keyboards can continue to use the electrical properties of the USB connector and ignore the lightpeak optical connectors in the cable. That is the great thing about this technology it only requires very little space in the connector and so they can easily insert it in to the USB connector without breaking the USB connectors compatability with pre-exsisting usb ports.

I'm really excited about this technology. I need fast networking. My options today are limited to teaming connections (low cost, lots of cables) or using Fibrechannel (high cost). Light Peak would end all that.

[ew2x4]

there is a huge difference between Gb/s and GB/s

I knew what was said, but I think I still subliminally changed it. Either way, it's obvious for where the limitations for data transfer is. And $1000 for 256 pseudo GB with transfer speeds that get up to 640 MBps is not the way to get there.

[ilev]

Everyone forgets that LP is Apple's technology , brought by Intel.

[Elliott]

Everyone forgets that LP is Apple's technology , brought by Intel.

There are rumors that Apple helped Intel out in Light Peak's creation, but no, it's very much Intel's technology.

[ilev]

There are rumors that Apple helped Intel out in Light Peak's creation, but no, it's very much Intel's technology.

It is vice versa. Intel helped Apple with LP creation.

[iwod]

10Gbps isn't enough to drive a High Resolution Monitor, so i hope they can get 20+ to start with, to match the speed of DisplayPort 2.0

Hopefully within 5 years time it will replace every external cable. Afterall using Copper to archive these kind of speed is rather expensive compare to LP which has yet to scale up in production and drive cost down.