Exclusive: Apple dictated Light Peak creation to Intel, could begin migration from other standards as early as 2010
By Joshua Topolsky 
posted September 26th 2009 5:02PM
posted September 26th 2009 5:02PM
According to documents we've seen and conversations we've had, Apple had reached out to Intel as early as 2007 with plans for an interoperable standard which could handle massive amounts of data and "replace the multitudinous connector types with a single connector (FireWire, USB, Display interface)." From what we've learned, the initial conversations (and apparent disagreements) were had directly between Steve Jobs and Paul Otellini. If you were wondering about that Apple-blue motherboard we saw at IDF or the aforementioned Hackintosh demo, this should explain everything. Cupertino apparently had specific demands for the standard, including the desire for a single port solution, and an insistence that optical was the only logical choice for such a connector type. Based on the documents we had a look at, the short-term plans seem to involve a one-size-fits-all solution (somehow allowing for multiple connections but avoiding "double dongles") which would enable users to connect a variety of devices into a single Light Peak port, while slightly longer-term plans will mean Light Peak obviates the need for almost every type of connector you use today. Translation: Apple products in the near future could come equipped with only a Light Peak port (or ports) to handle your networking, display driving, and general connectivity.
What happens next, however, is where the story really gets interesting. Based on what we've learned, Apple will introduce the new standard for its systems around Fall 2010 in a line of Macs destined for back-to-school shoppers -- a follow-up to the "Spotlight turns to notebooks" event, perhaps. Following the initial launch, there are plans to roll out a low-power variation in 2011, which could lead to more widespread adoption in handhelds and cellphones. The plans from October 2007 show a roadmap that includes Light Peak being introduced to the iPhone / iPod platform to serve as a gateway for multimedia and networking outputs. While the timing doesn't line up, a low-powered Light Peak sounds like the kind of technology that would be perfect for a device with a need for broad connectivity but limited real estate for ports... like a tablet.
Apparently, there are also plans for a new, low-power Atom chip due next year meant to compete with current CPUs driving mobile devices like the iPhone, netbooks, and set-top boxes. The indication we've been given is that that product (coupled with the Light Peak standard) could provide the basis for some "big" MID news in 2010... but not from Apple. As we saw at the announcement, Sony is also a first-tier partner (though not nearly in the connected way Apple is), and it looks likely the company could find use for that pairing (especially with its entry into the netbook market this year). We can't see Apple utilizing an Atom chip in anything, especially when it's busy cranking away with P.A. Semiconductor and its rumored projects.
In the end, we can't say we know or understand Apple's complete plans for the standard, but what is notable is that the company is pushing for an all-on-one connectivity solution, and pushing hard. That means Cupertino is at least prepared to abandon the standards we now know for a singular solution, and potentially just skip over forthcoming offerings like USB 3.0 (or at least downplay their use). It's not an uncommon move for Apple, but in the past those choices have had major repercussions -- and this pairing is extremely reminiscent of the USB / iMac story (Intel also created that standard, which Apple then popularized). We're going to keep our ears to the ground on this news, but for now, it's an interesting peek behind the curtain, and a possible glimpse into the future of connectivity.
So is this the first Apple-approved hackintosh?
It was a taken apart Mac Pro, not a giant hackintosh as previously thought.
Ah, shame. Well, the spirit lives on.
Apple leads, Microsoft follows....years and years behind.
The Highlander speaks.... "There must be only one (port)."
Ballmer is going to either start laughing himself to death or have a fit. No way is the bald master of sweat going to allow Apple to get control of anything that runs on a Windows machine. It's always better to have a half-dozen cheap interfaces on a Windows computer than an expensive one. In many ways that does make sense because almost nobody wants to have to upgrade their external devices with new ones just because the connectors have changed. Anyone selling a Light Peak to USB 2 dongle yet?
I can't wait until 2011 when all the Macs come equipped with Light Peak as their only connector. Hey, wait a minute, there are no companies that make peripherals that use Light Peak, except Apple.
^God forbid Apple push for a connector that seems good. I mean look at how USB failed after Apple made that wierd computer called "iMac" which has all those USB connectors on the keyboards and on the machine itself. Oh wait...
http://news.cnet.com/FireWire-to-get-big-push-at-Comdex/2100-1001_3-247169.html
"FireWire is Apple's name for the IEEE 1394 High Speed Serial Bus. It was initiated by Apple (in 1986[2]) and developed by the IEEE P1394 Working Group, largely driven by contributions from Apple..."
http://en.wikipedia.org/wiki/Firewire
http://edition.cnn.com/2001/TECH/computing/01/24/apple.firewire.idg/index.html
http://www.macdailynews.com/index.php/weblog/comments/5062/
USB was the alternative standard. Firewire was the standard Apple was pushing. And it lost. They didn't invent USB, they adopted it.
"USB was created by a core group of companies that consisted of Compaq, Digital, IBM, Intel, Northern Telecom, and Microsoft."
http://en.wikipedia.org/wiki/Universal_Serial_Bus
By the way, Apple didn't invent this crap either. It's an IBM tech from 15+ years ago. Optical Computing. IBM and Intel were working on optical interconnects forever, so please stop making shit up.
Looks nice. Hopefully we'll see it on a range of machines in the future - personally I don't give a toss who invents it as long as it works.
@Please forgive Me
I dont think Gamechld said Apple invented USB technology just that they push it into the mainstream with the IMac at a time when MS and PC Hardware manufacturers where dragging their feet and still fussing around with devices like the floppy drive. You can't knock Apple for that.
The iMac only constituted about 2-3% of the sales in the computer market, how in the hell can anyone with half a brain even reach the wild conclusion that a machine that only appealed first of all to Mac fans, and secondly only made up 1/50 of the market, was somehow responsible for popularizing USB???
Do some of you Mac zealots even live on the same planet as the rest of us?
Ahhh the ol Windows FTW
2-3% of sales across the PC market and 95% of sales across the USB fitted market. Nothing like a good ol zealot to make a senseless comment and omit half the facts. (My percentage is just as made up as yours just to keep with the theme)
At the time Windows boxes were all parallel and serial with the very advanced PS2 connection for keyboards and mice, they even scoffed when the iMac mentioned didn't come with a floppy drive. You could get plug in cards with USB or buy specialised ASUS motherboards and build your own box.
Anyway, optical interconnects were also used by Cray, purchased by Silicon Graphics and called Cray Link, they allowed for massive parallel computing without each processor needing to be next to each other. I'm with Mark, who ever invented it, great, let's just get it and use it so we can stop dicking around with USB, FW400, FW800, FW3200, ESATA, HDMI, DP, DVI, RJ45, VGA and FC. It's a mess of purpose built high speed standards at the moment. A single cable will be gold
..it will probably also cost like gold in the first instance as well. Sigh, time for another round of early adopter pain.
Some fun facts:
http://www.computersciencelab.com/ComputerHistory/History.htm
http://www.computerhistory.org/timeline/
http://www.phy.ornl.gov/csep/ov/node8.html
http://www.levenez.com/lang/lang.pdf
http://en.wikipedia.org/wiki/History_of_computing_hardware
Just to clarify, I am not a Mac Zealot. I don't own a mac. I'm just saying, if they want to try and bring attention to another possible standard change, that could be great for everyone. I know the iMac did not usher in the era of USB single handedly, but they did embrace it more avidly than any other system at the time. If Light Peak ends up replacing USB, SATA, Firewire, etc. all the while increasing speed overall, sounds good to me. I was mainly using the success of USB killing of parallel ports and PS/2 ports, and who knows what else, to respond to this comment:
"No way is the bald master of sweat going to allow Apple to get control of anything that runs on a Windows machine. It's always better to have a half-dozen cheap interfaces on a Windows computer than an expensive one. In many ways that does make sense because almost nobody wants to have to upgrade their external devices with new ones just because the connectors have changed. Anyone selling a Light Peak to USB 2 dongle yet?
I can't wait until 2011 when all the Macs come equipped with Light Peak as their only connector. Hey, wait a minute, there are no companies that make peripherals that use Light Peak, except Apple."
^Because clearly, Apple has never backed a new standard and had it pan out.
Additionally, in response to Firewire losing to USB:
Firewire did great against USB 1.0, and helped people get a foothold into the digital home movie scene. USB 2.0, is where Firewire lost some steam. I assume Firewire800 is a nice interface for content creation, but USB 3.0 and eSATA seem like rising stars that can kill off Firewire for good. Not that thats a bad thing, it's served its purpose well enough.
@ Cy Starkman...Not all PCs at the time were still using Serial and Parallel interfaces at that time. In fact, many PCs that predated the iMac (released August 1998) had motherboards with onboard USB. My K6-2 machine of the day (purchased in June 1998) had a Super Socket 7 Mobo with two onboard USB ports.
lol, and people were disappointed when there were no express card slots / HDMI. Imagine no USB 3.0 without an adapter.
Yeah, with USB 3.0 external HDDs, flash drives etc. this might be another bag of hurt for Apple.
This will definitely be another bag of hurt for Apple. Yet another connection type we don't need, don't want, and Apple will try and force upon us. I imagine beginning with ipods and iphone....
Stupid company, standards are what people want, standards are what people need. Stop trying to create your own formats like Sony does and just give us what we want.
"for an interoperable standard which could handle massive amounts of data and "replace the multitudinous connector types with a single connector "
Ever hear of USB, Apple? Maybe, you know, USB 3.0. A standard connection that is compatible with past devices with the same type of plug. Not so difficult is it.
@ Jordan
If more people thought like you then the world would still be stuck on serial/parallel cables and PS/2 connectors. It's also funny that people like you love to bash Apple and say not a peep about Intel, you know, the company who will actually inject Light Peak into computers in the coming years. I bet you probably don't know that, much like USB, Light Peak will be a standard available for all manufacturers to use. So quit living in the past or the future will just pass you by.
I'm not gonna weigh in on the LightPeak debate until I know how much it'll cost early on. I've always been about the bang/buck ratio (yeah yeah that's what she said etc).
So, if I get new "LightPeak" only machines to replace what I currently have, I can look forward to buying multiple:
LightPeak to USB adapters
LightPeak to HDMI adapters
LightPeak to ethernet adapters
LightPeak to SPIDF adapters
OR invest in brand new HDTVs and AV receivers that have LightPeak built into it...
Yay for the aftermarket adapter business, I guess... :/
Think I'll wait out the revolution until it's proven to 'stick'...
Sounds like a lot of money is being spent on R&D for something with very little use. Yeah, it would be nice to have the equivalent of 5 USB ports on my machine when I'm using WiFi, not hooked into an external monitor, speakers and muted etc. but it's not going to change the way I use computers.
"...very little use..." huh? I bet you didn't know that, when you're using your laptop on WiFi, even with no external peripherals connected, you're still likely using USB. On an Intel Mac, for example, the keyboard, trackpad, iSight camera, IR receiver, and Bluetooth controller are all connected via USB internally. Imagine what form factors industrial designers and engineers will be able to come up with using an ultra-high bandwidth optically-based muliplexing connection standard like Light Peak. With something like this, you could literally have the processor in the basement, the motherboard in the attic, and the rest of the components scattered around your house with no ill effects. I don't think it's hyperbole to say it will "change the way you use computers".
@krbanton, I know you were trying to support your point, but you in essense just did the opposite. wickedpheonix, said "R&D for something with very little use", that is exactly what is happening. This could be nice for some larger computer system, but for the end consumer it has "very little use".
This is like when they bridged the components on the motherboard with light, but this is like bridging completely separate devices. They appear to be using their strength of controlling their hardware to their advantage.
@krbanton
1) But it doesn't matter if something is connected via USB internally, because I never have to connect or disconnect it. Keep in mind, even USB 2.0 can handle 480 Mbps - well over even 802.11n standards. Making a huge deal out of this for WiFi purposes is like making a big deal out of 6 Gbps transfer rates for hard drives - just way too much bandwidth for what is even theoretically required.
2) You both couldn't and wouldn't want to wirelessly connect the parts of your computer. Part of the problem that WiFi and indeed, IP has to overcome is the issue of dropped packets. If the very basic lowest-level processing had a lost packet, I'd be more worried about system stability. And even if there wasn't any such worries - why would you do such a thing? Why would I ever want to spread out the components of my computer - it just makes it harder for me to move my computer when I need to do so. The only reason I could think of is if you wanted to communalize computing resources within your own home - but the barriers for that are more software oriented than hardware oriented, especially when we have Gigabit wired technology readily available for residential purposes.
actually they use mini pci-e slots designed specifically for them. If you open it up you can upgrade your wireless module on laptops... i believe its something to do with bandwidth/power consumption compard to USB.
@wickedpheonix: Saying SATA, even at 6Gbit speeds is "just way too much bandwidth for what is even theoretically required" is very shortsighted. Sure, no single platter based storage medium is likely to tax the bandwidth at a sustained rate, but the burst rate into the cache is very beneficial. Add to that multiple drives on a single link (external enclosure), or these new fangled SSD things, and soon your 6GBps pipe looks to be barley enough to contain what is needed in the very near term.
Faster interconnects like USB 3.0, and then down the road with Lightpeak are important. The use to you may not be obvious in your home life, but somewhere out there, your using service run on a server that can take advantage of this bandwidth today. And in a few years time, those benefits will roll down into the consumer space.
Perfect example, I bet in 2001 you would have said 64 bit is pointless. However, even at that time plenty of servers were needing more then 4GB of ram per application. That groundwork for 64 bit (going back even into the 90s) is slowly working its way into the home.
Don't dismiss the need of future technology so fast, otherwise your quotes will sound just as dated as the often quoted "Noone will ever need more then 640k of memory" comment.
i agree with wicked pheonix usb 3.0 is more than enough and versatile to handle the need of devices today and even in the future. this light peak my be superior but that doesn't mean its going to be a new standard. anyone remember fire wire and how great that was suppose to be? a few years later and no one uses it and its nothing more than a huge failure.
Well, this light peak technology looks very promising, still, it's not as easy as krbanto believes.
Optical is something different and more difficult and more expensive to build. And with optical alone you aren't able to power a device.
Also think about all the small developers who use microcontrollers to create new useful devices. It's much easier to do this with USB than with any optical solution, especially because they don't need this huge data transfer. So I don't expect that it will replace USB, rather add another connection which maybe replaces eSATA (and maybe SATA) and Display Ports and adds a high bandwidth interface for external GPUs.
You also shouldn't forget that the used network cables are still CAT cables, so if you have a notebook with such a light peak connection only, you'll always need an additional external adapter, cumbersome.
And finally, back to all those internal peripherals like mouse/keyboard/buttons and other things, they don't need the high bandwidth, and to keep the costs as low as possible they will continue using USB, or the LPC interface they currently use.
Optical transmission in the consumer sector is finally possible and a good technology, and I hope it will replace all the useless other high speed ports (Firewire, USB3, eSATA, SATA, DVI, DisplayPort, HDMI), but the low speed ports (USB2) will stay.
i think this is actually really great, because people are not realizing the real applications of this technology. right now you're thinking spreading out your computer parts, or the internals of your current laptop and things like those, but as always with new technologies you have to think beyond.
imagine having one central processor in the basement which could be easily upgraded, which also handles ram and graphics, and the actual computers could just be input output peripherals. think how much this would advance home automation systems, how much it would take that step forward of harmonizing your devices. imagine coming home and dropping your Zune onto your surface and suddenly having your whole wall turn into your jukebox, and all this without any of them needed processors, or other wired components that today's technology needs.
i think the biggest obstacle is the fact that there is no good wireless power solution, and the bandwidth that this allows for, whereas the advancement of usb may solve all of this. with home automation you could just wire usb into the house walls, and with portable devices you'd want them to have their own processors regardless. i suppose they should work with the usb guys and go for a hybrid wireless wired system, and make that the standard for all types of data and power transfers for any electronic. i mean, intel does employ the co inventor of usb does it not?
@jay jay Firewire is by no means a failure. It's still being shipped as part of many computer systems, video cameras, and other devices. And became widespread enough to be mandated in cable boxes for a while.
No firewire was pretty much a failure. Besides camcorders, what devices use firewire today?
The only other devices I can think of are external firewire hard drives for those mac users who refuse to use usb
In the beginning Firewire was supposed to be this grand connection....and no, it definitely wasn't.
Jordan: FireWire is used all over in the pro audio and pro video worlds. USB just doen't cut it here, and no other port is really suitable for this (besides the "PCI card + proprietary connector" approach some companies take, but I, for one, hate that approach).
This could change with USB 3, but if it does, it will be very slowly, because gear compatibility is a serious issue in the pro world.
On topic, I'll just say I'm a strong supporter of optical connections, and I wish they'd have been widely adopted earlier.
So then what consumer products, aside from camcorders, do you know that use firewire?
The answer: None
When firewire first came out it was supposed to be the holy grail and in all of our modern products Even the first few iPods had it until Apple dropped the support for it because USB far surpassed it. By standards of what it was supposed to be, firewire was a failure.
The same will happen with this. USB is way too common and way too universal. Here we have yet another connection type that's supposed to replace a bunch of other connections. Fact is, it won't happen. USB 3.0 is obviously going to be around a while before this gets released. By that time USB 4 will probably be on the horizon, probably able to do all light peak is proposing, and again, beat out the competition. This is just yet another connection in the world of connections that is not going to deliver.
@ DarkLight
Saying that Firewire was not a failure because it's still used in the pro video market is like saying OS/2 was not a failure in trying to be the dominant desktop OS back in the day because it's still used in many ATM's.
Furthermore, a lot of stuff just doesn't need that kind of throughput, even theoretically. Mice, numeric keypads - how are these supposed to even max out the throughput of USB 2.0? Some stuff just plain does not and cannot benefit from godly amounts of throughput. My response about 6 Gbps was taken out of context - I said for HARD drives, not SSD's. Even Raptor drives barely push the throughput of SATA 1.5. If you really want to get more throughput, then put integrated RAID on the drive - make each platter a separate drive and have RAID 0 between them, so that you can double your performance and still have the drive look like a single HDD to the OS and BIOS. Even if you had to make it a 5.25" HDD (with 3.5" platters) so that you could have room to put in the integrated RAID controller, it would still be worth it, especially until miniaturization brings the space requirements back down to 3.5" and eventually 2.5".
And because Engadget through one of my comments away, I'm just going to say again - I'm not against future technology. There will be a load of stuff that WILL need more throughput, and can benefit from optical tech. But that's not what Apple is trying to do here - Apple is currently trying to use optical tech to eliminate the need for different connectors on your computer. In Apple's future, your Ethernet, mouse, and external display all plug into identical connectors, which means that if you have 4 of these connectors you'll have the flexibility to connect, say, either 4 displays or 4 gamepads, for more flexibility. That's all well and good, but it costs a ridiculous amount of money on R&D for this for what is a mere convenience most of the time. If it was being developed as a Firewire sequel, then great - but that's not the foundational purpose here apparently.
@Jordan: If we kept thinking that what we can see today is enough for tomorrow, we'd still be using RS232 interfaces. FireWire, after 14 years, remains superior to USB in many applications due to its computer to computer interfacing and dedicated asynchronous transfers. USB is more popular due to the fact that the advantages of FireWire are not needed by many types of interconnect and people don't want more cables than necessary. Moreover, some of the advantages of FireWire are now supported by USB (booting from a hard drive). The key to USB's success is the ubiquity of devices that can connect over this interface. However, who wants to have four different cables when one could do? If Light Peak can replace USB, HDMI, SATA, DisplayPort, FireWire, etc., it makes device connectivity friendlier for consumers. That is what USB did and that's why this will replace USB eventually (there will be plenty of overlap of the standards). It will drive innovation. Who cares who invented it.
I'm not saying USB the way it is will be good enough for the future, but what I am saying, and people are not understanding, is that USB 3.0 is coming very soon. USB is universally adopted. USB 3 will provide faster speeds and by the time light peak comes to fruition we'll probably have USB 4 on the horizon. If what Apple is pushing for is one way to connect, USB can be that way and is already here and widely adopted. All new USB ports are backwards compatible with past USB devices.
Though light peak looks good now, it will never become what it is proposing to be. USB 3 will overshadow it in the market and whatever USB evolves to next will further that.
I'm also saying, yes firewire is used a lot in the pro world, namely camcorders and mac users who use external drives, but as far as consumers (the majority market) are concerned, firewire what?
@Tom
"Noone will ever need more than 18 exabytes of memory."
Yippee, Apple. Now please support USB 3.0 ASAP.
Whoa!!
Hold on there captain enthusiasm...
Apple just got around to supporting MMS on a cell phone..
its gonna be a many millenia before you see USB 3 on anything Apple...
But it is nice to dream
@BigAssTroll (i ain't got time for underscores).
I think Apple may skip USB 3.0 entirely for Light Peak, from my understanding of this article. If they do go with USB 3.0 it'd be sooner or never at all.
@bigasstroll
Right because before that they had EMAIL. Some might argue are more important and useful form of communication.
And Steve said, "What? MMS? People actually use that...?"
Apple is so low-tech they brought out a laptop that's half an inch thick. And you guys are fretting about USB plugs? Lame.
@ nub above...
Yes because Apple did never release a 3.5mm Heaphone jack...........THAT WAS RECESSED and needed an iAdapter
OH yes because Apple is so high tech that they release a laptop that fir into a manilla envelope..... THAT NEEDED A USB iAdapter to access it...
Hmmm you see a pettern here
nub?
Yet another reason to wait for the new macbook pros. Come on already! Core i7, USB3.0, and now this!
Expect none of those things on the next macbook pro.
The new Macbook Pro might do MMS
Keep your fingers crossed, I hear this "push notification" thing is going to show up any day now.......
Wow, is all I can say.
Not only was this not Intel's idea, it wasn't Apple's either.
I knew I saw you post this idea a few years back as a comment on another similarly related Engadget article. Phew, case solved.
@Alex Chopra: Did we? It's possible, not sure. We may reveal more in the future, perhaps through an exclusive to the press. We'll see. Our invention was disclosed as a subset of our UDiD Media proposal, created in 1988-1991 and made available to Apple and others. Either way, it's something that needs to happen as there are way too many I/O 'standards'.