Light Peak has potential data speeds capacities, however higher cost of parts for optical connections will keep it back. USB offers powering of devices as well as connection, I don't see anything on this matter concerning Light Peak. I reckon USB still be around another good 10 years or at least till light speed come into mainstream (due to reduced pricing/increase popularity) maybe 2016, long time. Or lack of adaptation could see it never really taking off.
Thanks for the update! I think USB 3.0 will certainly have an edge considering the immense popularity of USB. Though Light Peak definitely has its advantage in terms of proposed trasfer speed and future potentials, but initially it might take some time before it could overshadow USB 3.0.
Completely agree with Sanjib. With the wide popularity of USB than any other external I/O interface on PC, the technology is going to stay here for long time. Unless if there is a necessity of huge data transfer speed not all the OEM's will bring this light peak technology into the products. Thinking from the cost point of view usage of optical cables will also increase the price of the products. which is a disadvantage ...
I see this as a crossover point between these two technologies. USB 3.0 is pushing the limits of electronic signaling, while Light Peak is on the early end (relatively speaking) of photonic signaling. This generation of each will be competitive, but the latter should have a lot more headroom to grow.
This assumes, of course, that later USB stacks are still based on electrical signaling. There is probably no good reason to make that assumption. In fact, what might make the most sense technically is to build future USB versions on top of a Light Peak physical layer. Unfortunately, business concerns would make that difficult.
The primary one will be licensing revenue. If Light Peak is strictly Intel, then all (or at least most) of the licensing revenue goes to them. If it is a portion of the USB stack then Intel would just be one of the licensing pool members.
intel has been known to introduce some exclusive thing to help dominate the market. Like when AMD and INTEL partnership for MMX.
MMX was a nice marketing plow to help put those two CPU manufactures ahead of everyone else. In reality there was only a dozen software titles actually used that MMX extensions, it was short lived for like 6 months or something but it was a awesome strategy. Who cares about the mmx extensions now? LOL!
Over 2 billion USB devices are sold every year, and USB 3.0 will be compatible with all of those legacy devices. So it's "only" 5G/sec instead of 10G/sec -- the speed deficit is not nearly as significant as the entrenched base of compatible devices.
Yes, Light Peak and USB 3.0 can coexist in the same system, but why would they?
Let’s see, I see a possible Intel MOBO with a Optical link to my Sandy Bridge, HD, HDTV, and a link to any peripheral at up to 100G/s vs. USB 3.0 so I can use my old mem stick, don’t rush me I’m thinking.
Does anybody have any specs on the required fiber to be used with Intels Light Peak?
I can see that as being an issue with uptake if high coherency fiber is required rather than the less expensive fiber.
I also agree with the licensing. If Intel makes it available as an open license then it may pick up for remote disk arrays.
I think the comparison here in the article is not done in the proper way as USB 3.0 is looked as high speed solution for desktops and portable devices. Where as LightPeak will move towards network devices finding it application.
This will surely reduce the cost of optical interconnects compared to present scenario.
What does Intel want? Lightpeak is anyday better (more $$ in profits) for them, so why would they push USB 3.0? However, if Intel doesn't support USB 3.0, it will be a golden opportunity for AMD (& others perhaps) to gain market share in desktop & laptop systems. If Intel pushes Lightpeak more, they will have a strategy to make it popular. Perhaps, incubate/support Lightpeak-to-USB converter makers while trying to convince the device makers to shift to Lightpeak sooner.
There's already USB3.0 solutions available so I'm not too concerned about Intel right now. I agree that AMD has an opportunity here if they choose to take it. I, for one, will likely skip USB3.0 and wait for LP -- unless I get a USB3.0 equipped board for other reasons. A great thing about LP is that it can be used to support all protocols including USB3.0 LP is an optical pipeline, not a protocol by itself. It leverages existing PCIe & USB protocols but can collapse all the separate wires for these into a single optical cable.
Intel is showing lots of power to continue to dominant. They have been moving strategically and developing products which make sense. It is good to see them drive the market . Light Peak with become a game changer.
Interesting, I got the Camera Connection Kit for my iPad 2 and when attaching 2.5 inch external hdd on it, the iPad complains the attached device is requesting more power then it's prepared to give. It occurred to me that port doesn't meet USB 2.0 standards, maybe only meeting USB 1.1. Did refusing to go USB 2.0 help reduced it's battery consumption rate for the iPad?
Has anybody asked what sot of bus is going to feed 100Gb/s, or 1Tb/s (the photonics link). Intel has aimed for the 100Gb/s on system bus years ago, but what new (optical) bus might they have up their sleeves now? All I can tell you is AMD should worried and worried about being trumped. Steve Jobs went to see the cell even though he did not want to, he had already abandoned Power PC, because he had already seen what Intel was promising (not that it all came through as expected). If he seen all this, Terascale and all, no wonder.
A Book For All Reasons Bernard Cole3 comments Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...