Tech is cool and all, but did they really have to name it Lightning Bolt? That's one way to make an alternative technology sound like a cheap off-brand knock off.
What the hell is a thunder bolt? Thunder is the noise caused by lightning when it strikes the ground. Intel should have called their stuff Lightning Bolt. Maybe they should have kept it Light Peak.
BTW thunder is not caused when lightning hits the ground, it is the caused by the sudden expansion of air in the path of the lightning, so the sound starts before it hits the ground.
The word exists because it was before the phenomenon was truly understood. Other languages also have similar tradition, because our ancestors used to think the sound precedes the lightning (or lightning is originated by the sound), thus the word "thunderbolt". I think the word should be deprecated today, most people would say lightning bolt because it's scientifically correct.
Actually lightning doesn't "hit" the ground. Lightning streaks inside a cloud, between clouds, and from clouds to the ground. Lightning is a flow of electrons (a negative charge) that zigzags downward in a forked shaped pattern (scientists call this a step leader). As it nears the earth, a stream of positive charges moves up to the charge of electrons (negative charge). When they meet, the power flows. We can't see this because it moves too fast (first stroke). The return flow (positive charge) moves upward more slowly. This is what we see and call lightning (return stroke). If there is a flicker, the upward stroke is repeating the process.
A thunderbolt is a clever poetic construction, like being hit by a 'wall of noise'... or being struck by a 'hail of arrows'. These comparisons conjure up evocative images in our minds and are good for the spirit.
Well Intel kinda took the "light" out of Light Peak to make Thunderbolt. I wonder what Intel will call the "real" Light Peak when it's released with the optical transceivers...
I'd laugh if one of the reasons AMD named their standard Lightning Bolt was to troll the name first ;p
Either way, I always get the feeling that Intel themselves don't really plan to continue Thunderbolt in the long term, they really want the optical interconnects. They managed to get Apple excited on it enough to take the hook, then again Apple isn't bashful about abandoning prior standards they had just embraced a generation ago.
The "Ultrathin" category is with us for at least a decade (Thinkpad S30 anyone ?) with low volume special stuff 15+ years ago.
That Intel was not satisfied with a "generic" name for its latest PR campaign so come up with "UltraBook" moniker for Sandy-power ultrathins is their problem.
To anyone with a bit of common sense the intel naming is just a plain joke. Though, one must admit MS's "MID"s are really hard to beat in this department ... :D
Seems suspiciously like they've just piped USB 3.0 over a souped-up DisplayPort Aux channel... Don't get me wrong, I'm all for competition in this arena, but this just seems to be an attempt to confuse gullible consumers with a very similar looking and sounding technology.
Is that a Mini-DisplayPort cable you got there? No. Thunderbolt cable? No. It's gotta be Lightning Bolt then!
Good concept, but these competing connectivity standards are awful for consumers.
Example: Powered eSATA ports sounded like a great idea 2 yrs ago. Fast transfer speeds, a single cable for data and power, and the same port could also be used as a normal USB port so it didn't waste space on a laptop. Awesome! BUT there are almost-zero external drives that support the standard. And only 1 in 20 laptop or desktop computers have this port.
Pick ONE standard and make sure it is dirt-cheap to implement everywhere.
Powered eSATA was NEVER planned but actually created on an ad-hoc basis.
If anything, it gives one eSATA/or/USB2 in same footprint which is good.
BTW eSATA is the ONLY native storage interface so as far as external storage goes it is the best by definition.
Though, 99% people really do NOT need its features (which come at a price of unvieldy cables). But for those of us who do, eSATA/USB is jus more versatile version of eSATA ... a shame that USB3 garbage is taking over this niche. :(
Is to add Ethernet to the mix. If it also had the network connection you could use it in place of a traditional docking port. Just put the laptop on the desk and plug in a single cable and you are in business. But without the network connection...doesn't work for me.
Just pure stupidity if they proceed with this product. The reason why TB is/will be better than LB is pure and simple TB is PCIe based and so can actually help externalize hardware such as GPU which is something notebooks can't do. At least not well - usb 2.0/3.0 based solutions wouldn't live up to even an integrated gpu's performance.
This is one of the craftiest, wisest gimmicks of all time. Provide what appears to be what Intel is pushing, for MUCH cheaper, riding on the back of USB3.0.
I'm entirely happy with my USB 3.0 external storage and 5870. If Tbolt can even be limited or stopped, everyone besides Intel wins. Brilliant honestly.
How does everyone besides Intel win if we limit or stop the inclusion of 20Gbps I/O interfaces on laptops and all-in-ones?
Have you considered that there are uses for a 20Gbps interface besides external storage or GPUs? Just cuz you don't have need of it, doesn't mean it's not useful.
If Intel wanted to make the world better with thunderbold they should have followed USB's example. Instead they follow firewire's example. USB won because it is cheap and anybody can implement it. Thunderbolt will only be available on more expensive devices. And if Intel integrates TB into their chipset it will be an Intel only technology.
Last I checked, Intel was on the panel that created USB, they currently integrate USB into their devices, and USB is not an Intel only technology. USB was not initially cheap compared to existing standards, nor was it immediately adopted en masse. Thunderbolt is currently available on the $599 Mac Mini, the cheapest PC that Apple offers.
The only statement you made that makes sense to me is that Thunderbolt bears some resemblance to FireWire.
You said it. Apple's cheapest PC is 600$, while you can get any other brand for half as much (or just buy parts). Won't be as powerfull, but it's not for gaming anyway.
Also, Fergy made perfect sense. TB is not an Apple exclusive because they lerned their lesson with FireWire. Now they let Intel try their luck with an expensive port with slow adoption rate because of it. Intel want's to get paid, so they charge for using TB, and that means the price is higher then it could be. The reason Apple is paying anyway is because the have to be the cooltest and they overprice treir products anyhow so it doesn't show on the pricetag.
In case you haven't figured out, the whole Ultrabook concept is an attempt to drag the PC market out of the cheaply made commodity rut they've been in for the past decade. Intel wants 40% of notebooks to be Ultrabooks, and they will average around $800-1000 once the first generation has filtered through. The higher price point ought to allow use of higher end technology like Thunderbolt.
TB could be a great interface for docking stations, and could provide a standard for the entire industry. Apple's Thunderbolt Display provides a template for the rest of the industry to consider.
That's the thing, there are so few applications for Thunderbolt (over USB3.0) that the only real purpose it serves is to make Intel money. If you really think you need Thunderbolt, nothing stops you from buying a premium model with it included. Much like FireWire, it will never see adoption on consumer devices because of high costs and lack of benefits over USB.
Well, if you don't need it, then it isn't entirely useful. For me fragmenting connectivity standards is worse than any possible gain. This wasn't some brilliant idea by Intel that will also happen to make them money, it was an idea to monopolize another sector USB has been evolving, and we have other connectivity methods for remaining requirements. Fiber optics is nothing new.
I'm not sure how USB has been evolving the high-bandwidth digital display interface combined with external PCIe connectivity sector. I'm also not sure how you reckoned that Intel planned on squashing the poor USB-IF (which Intel formed in the first place) with an interface that only supports a total of 6 connected devices.
It's interesting that you find more than one connectivity standard to be the worst possible thing, yet you're afraid of Intel monopolizing the market with a new standard... The bottom line here is that lack of choice is bad. You can't want USB to have a monopoly on the one hand and then not want a different monopoly because Intel's involved on the other.
But I guess we should still probably get rid of all of the other interfaces like SATA, HDMI, Ethernet, and ExpressCard, and just focus on USB. There's no reason to have more than one connectivity standard after all.
You seem to be either very confused, or simply kneejerk in being argumentative and defensive to your precious Thunderbolt/Intel (not sure which). Did you already foolishly spend money on that crap?
I was the one who said we have interfaces for all those things, SATA, HDMI, ethernet, USB. 'we have other connectivity methods for remaining requirements.'
I also, never claimed USB was evolving into external PCIe, I simply said it's cheap, does what it's supposed to do and more, and there plenty of other options for connectivity if you need something else.
This new standard is doomed to fail in the face of USB3 and everything else that's already well-established.
Enjoy buying all those Firewire.. I mean Thunderbolt devices.
I think you're missing the point. TB is simply extending the low-level PCIe bus from the motherboard and/or the CPU to the outside world. In other words, plugging a device into TB is analogous to plugging a PCIe adapter card into a PCIe slot in a PC. That PCIe device could be anything. When you consider that all of the interfaces (USB, SATA, etc.) are already connected to the CPU/GPU via PCIe bridges, even if that bridge is internal to the chipset, then you can see that TB is protocol agnostic.
Fiber optics is not new. However, silicon photonics is very new. It is not about fiber optics. It is about how the light pulses are generated and detected. Current high speed optical i/o devices, like fiber channel adapters, require very expensive laser driver and detector hardware. Silicon photonics promises integrating the laser, detector, and modulation on the silicon. In other words, it makes it cheap to use fiber optic cables at tremendous bandwidths.
You are so correct. TB's great innovation is to externalize the guts of a PC design by putting PCIe onto a serial cable. Everything else follows from there.
USB on the other hand is mostly interrupt driven and therefore inherently a bad match to many of the uses we want it for, external disks being one prominent example. Sure, external USB drives work, but it's still hobbled and provides variable, hiccupy throughput. TB, and Firewire before it, solves this cleanly and effectively.
My 8GB Supertalent Pico-C and 1TB WD Passport (USB 3.0) disagree with you.
My external storage averages ~100MB read speeds and ~70MB/sec write, a self powered 5400RPM drive.
Sure it's no SATA drive, but it's more than effective and convenient on technology widely available today. In fact, I've been using it for 2 years now while everyone waits and waits for Tbolt to dominate.
You misunderstand the bigger point being made here. Sure USB 3.0 is fast for most of the tasks, however, it is not fast enough to allow a decent GPU to be hooked up to the computer externally. With TB, due to its implementation, not necessarily due to the bandwidth, you can gut out almost all parts of your computer that are currently only viable as internal devices while keeping the performance experience close to normal. A simple example is being able to use an external GPU on a laptop/netbook/tablet to boost the GPU performance considerably - something that USB solutions cannot do and USB doesn't seem to be driven to tackle that front.
This isn't to say that TB is perfect but it is at least one step ahead of USB now. Companies need to throw more weight behind TB to create more products that end-users can make use of to simplify devices portability and upgradability.
I believe USB would be great as a peripherals connectivity (displays, hdd, kb/mouse, camera, printer, ...) and TB for computer hardware (videos, raid cards, accelerators, ...).
Exploiting such capabilities of TB would be a two edge sword - provide products with greater benefits (portability for re-use) to end-user, however, at the same time lose revenues from end-users as they will buy only one expensive solution for more than one computing system instead of for each system.
Thunderbolt and Lighting Bolt seem to be completely unrelated, despite initial impressions. Lightning bolt basically combines two existing external interfaces into one cable, for the sack of convenience (docking). It's purely a physical thing.
Thunderbolt on the other hand makes the bandwidth provided by PCI-e, available externally. It provides capabilities that are otherwise unavailable over any other existing interface (or combination of interfaces). If you need to send more than 5/6gbps to an external device, Thunderbolt is the only solution.
I think people associate the name Thunderbolt with comics and unscrupulous PC electronics makers, not a solid produce, a lot like Ultra-Flop notebooks.
With yet another incompatible technology vying for use of the mDP, this is bound to just lead to more customer confusion. It was bad enough that Intel and Apple settled on mDP, but at least few PC manufacturers besides Apple had used the port prior to the introduction of Thunderbolt.
It does seem to just be a merging of USB 3.0 and Display Port. I agree with others about the name. Intel started with "Lightpeak" and that made some sense because it was an optical interface that also supported PCIe, etc.
Lightening Bolt doesn't seem to have anything to do with optical other than trying to get people to believe its similar to Intel's LightPeak and ThunderBolt.
I sense some patent lawsuits here... lol Intels thunderbolt was originally called lightning bolt from what i remember and i think that it was already patented? I think the plan is to call it thunderbolt while they still use coper wires and lightning bolt when it reaches full fruition of using optics.
Physics fail. Signals travel down copper at somewhere between about 65% the speed of light for coaxial cables and 97% for a free conductor. Not sure were you got a 100X difference from.
The "point of fiber" is absolutely bandwidth. The maximum theoretical bandwidth limit is far, far higher for fiber optic cable than a copper wire (again, those pesky physics).
Easily isolated fiber optic cables also do not induce nor pick up interference from the surrounding environment, while every copper wire doubles as an antenna. Noise immunity is a giant hurdle in any copper communication link. Fiber links simply don't have to deal with it.
You can make cheap copper cable but you need complex demuxes on the ends (which NEED time to do its work) plus you have to run at low freq.
TB would have been wonderfull had the actualy been a primarily fiber interface. You take optics from TB and you get an inflexible useless eSATA/ePCI/eSCSI ... you name it.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
54 Comments
Back to Article
shadowofthesun - Thursday, January 12, 2012 - link
Tech is cool and all, but did they really have to name it Lightning Bolt? That's one way to make an alternative technology sound like a cheap off-brand knock off.chizow - Friday, January 13, 2012 - link
Pretty much, I had the same reaction a few weeks ago when AMD decided to use "K" of all available letters to differentiate their unlocked APUs....These classic clips from Coming to America come to mind immediately when I see AMD (or any company) do this kind of thing:
Explaining Lightning Bolt name: http://www.youtube.com/watch?v=CGRiTpCji0I
AMD marketing department hard at work: http://www.youtube.com/watch?v=XrwZORNA-k0&fea...
sigmatau - Friday, January 13, 2012 - link
What the hell is a thunder bolt? Thunder is the noise caused by lightning when it strikes the ground. Intel should have called their stuff Lightning Bolt. Maybe they should have kept it Light Peak.Camikazi - Friday, January 13, 2012 - link
http://www.merriam-webster.com/dictionary/thunderb...Webster dictionary seems to say that Thunderbolt is a word.
http://oxforddictionaries.com/definition/thunderbo...
As does Oxford.
BTW thunder is not caused when lightning hits the ground, it is the caused by the sudden expansion of air in the path of the lightning, so the sound starts before it hits the ground.
tim851 - Saturday, January 14, 2012 - link
Thunderbolt and lightning,Very very frightening me.
Galileo...
[deXter] - Sunday, February 5, 2012 - link
GALILEO! Galileo, GALILEO! Galileo, Figaro, Magnificoooo....(Glad I'm not the only one who thought of this song when reading this article ;) )
Th-z - Wednesday, January 18, 2012 - link
The word exists because it was before the phenomenon was truly understood. Other languages also have similar tradition, because our ancestors used to think the sound precedes the lightning (or lightning is originated by the sound), thus the word "thunderbolt". I think the word should be deprecated today, most people would say lightning bolt because it's scientifically correct.tekkitan - Wednesday, January 18, 2012 - link
Actually lightning doesn't "hit" the ground. Lightning streaks inside a cloud, between clouds, and from clouds to the ground. Lightning is a flow of electrons (a negative charge) that zigzags downward in a forked shaped pattern (scientists call this a step leader). As it nears the earth, a stream of positive charges moves up to the charge of electrons (negative charge). When they meet, the power flows. We can't see this because it moves too fast (first stroke). The return flow (positive charge) moves upward more slowly. This is what we see and call lightning (return stroke). If there is a flicker, the upward stroke is repeating the process.kingius - Friday, January 20, 2012 - link
A thunderbolt is a clever poetic construction, like being hit by a 'wall of noise'... or being struck by a 'hail of arrows'. These comparisons conjure up evocative images in our minds and are good for the spirit.xrror - Friday, January 27, 2012 - link
Well Intel kinda took the "light" out of Light Peak to make Thunderbolt. I wonder what Intel will call the "real" Light Peak when it's released with the optical transceivers...I'd laugh if one of the reasons AMD named their standard Lightning Bolt was to troll the name first ;p
Either way, I always get the feeling that Intel themselves don't really plan to continue Thunderbolt in the long term, they really want the optical interconnects. They managed to get Apple excited on it enough to take the hook, then again Apple isn't bashful about abandoning prior standards they had just embraced a generation ago.
sigmatau - Friday, January 13, 2012 - link
BTW, I do agree it sounds missleading at the least. Imagine if Microsoft came out with the mePhone after Apple introduced the iPhone.tim851 - Saturday, January 14, 2012 - link
Agreed."Ultrathins" is another example of AMD trying to knock off Intel.
mino - Tuesday, January 17, 2012 - link
HUH?The "Ultrathin" category is with us for at least a decade (Thinkpad S30 anyone ?) with low volume special stuff 15+ years ago.
That Intel was not satisfied with a "generic" name for its latest PR campaign so come up with "UltraBook" moniker for Sandy-power ultrathins is their problem.
To anyone with a bit of common sense the intel naming is just a plain joke.
Though, one must admit MS's "MID"s are really hard to beat in this department ... :D
MobiusStrip - Friday, January 20, 2012 - link
Not cool at all. They took three existing things and crammed them into one connector, degrading their performance as a result. This sucks.USB 3 is NOT a substitute for Thunderbolt. Any reader of this site should know the multiple reasons by now.
repoman27 - Thursday, January 12, 2012 - link
Seems suspiciously like they've just piped USB 3.0 over a souped-up DisplayPort Aux channel... Don't get me wrong, I'm all for competition in this arena, but this just seems to be an attempt to confuse gullible consumers with a very similar looking and sounding technology.Is that a Mini-DisplayPort cable you got there? No. Thunderbolt cable? No. It's gotta be Lightning Bolt then!
TrackSmart - Friday, January 13, 2012 - link
Good concept, but these competing connectivity standards are awful for consumers.Example: Powered eSATA ports sounded like a great idea 2 yrs ago. Fast transfer speeds, a single cable for data and power, and the same port could also be used as a normal USB port so it didn't waste space on a laptop. Awesome! BUT there are almost-zero external drives that support the standard. And only 1 in 20 laptop or desktop computers have this port.
Pick ONE standard and make sure it is dirt-cheap to implement everywhere.
mino - Tuesday, January 17, 2012 - link
Powered eSATA was NEVER planned but actually created on an ad-hoc basis.If anything, it gives one eSATA/or/USB2 in same footprint which is good.
BTW eSATA is the ONLY native storage interface so as far as external storage goes it is the best by definition.
Though, 99% people really do NOT need its features (which come at a price of unvieldy cables).
But for those of us who do, eSATA/USB is jus more versatile version of eSATA ... a shame that USB3 garbage is taking over this niche. :(
ricardodawkins - Thursday, January 12, 2012 - link
AMD trying to out-niche another niche.taltamir - Thursday, January 12, 2012 - link
that's remarkably astute... How many people even need such a thing?Ratman6161 - Tuesday, January 17, 2012 - link
Is to add Ethernet to the mix. If it also had the network connection you could use it in place of a traditional docking port. Just put the laptop on the desk and plug in a single cable and you are in business. But without the network connection...doesn't work for me.mino - Tuesday, January 17, 2012 - link
There is USB3 in there which is more than enough (even at reduced speed) for ethernet or whatever IO you might need ...teng029 - Thursday, January 12, 2012 - link
And here I thought Thunderbolt was a silly name...SilthDraeth - Friday, January 13, 2012 - link
They are both silly names, but Thunder doesn't come in bolts, and lightning does.Just saying...
BSMonitor - Monday, January 16, 2012 - link
Enough with the "Just Saying"Just Saying, just saying
XZerg - Thursday, January 12, 2012 - link
Just pure stupidity if they proceed with this product. The reason why TB is/will be better than LB is pure and simple TB is PCIe based and so can actually help externalize hardware such as GPU which is something notebooks can't do. At least not well - usb 2.0/3.0 based solutions wouldn't live up to even an integrated gpu's performance.redisnidma - Thursday, January 12, 2012 - link
Yeah, but remember that it's intel who's in dire need of external GPUs because of their IGPs being crap. That's why they came out with this gimmick.Obsoleet - Thursday, January 12, 2012 - link
This is one of the craftiest, wisest gimmicks of all time. Provide what appears to be what Intel is pushing, for MUCH cheaper, riding on the back of USB3.0.I'm entirely happy with my USB 3.0 external storage and 5870.
If Tbolt can even be limited or stopped, everyone besides Intel wins. Brilliant honestly.
repoman27 - Thursday, January 12, 2012 - link
How does everyone besides Intel win if we limit or stop the inclusion of 20Gbps I/O interfaces on laptops and all-in-ones?Have you considered that there are uses for a 20Gbps interface besides external storage or GPUs? Just cuz you don't have need of it, doesn't mean it's not useful.
Fergy - Friday, January 13, 2012 - link
If Intel wanted to make the world better with thunderbold they should have followed USB's example. Instead they follow firewire's example. USB won because it is cheap and anybody can implement it. Thunderbolt will only be available on more expensive devices. And if Intel integrates TB into their chipset it will be an Intel only technology.repoman27 - Friday, January 13, 2012 - link
Last I checked, Intel was on the panel that created USB, they currently integrate USB into their devices, and USB is not an Intel only technology. USB was not initially cheap compared to existing standards, nor was it immediately adopted en masse. Thunderbolt is currently available on the $599 Mac Mini, the cheapest PC that Apple offers.The only statement you made that makes sense to me is that Thunderbolt bears some resemblance to FireWire.
SleepyFE - Friday, January 13, 2012 - link
You said it. Apple's cheapest PC is 600$, while you can get any other brand for half as much (or just buy parts). Won't be as powerfull, but it's not for gaming anyway.Also, Fergy made perfect sense. TB is not an Apple exclusive because they lerned their lesson with FireWire. Now they let Intel try their luck with an expensive port with slow adoption rate because of it. Intel want's to get paid, so they charge for using TB, and that means the price is higher then it could be. The reason Apple is paying anyway is because the have to be the cooltest and they overprice treir products anyhow so it doesn't show on the pricetag.
KPOM - Friday, January 13, 2012 - link
In case you haven't figured out, the whole Ultrabook concept is an attempt to drag the PC market out of the cheaply made commodity rut they've been in for the past decade. Intel wants 40% of notebooks to be Ultrabooks, and they will average around $800-1000 once the first generation has filtered through. The higher price point ought to allow use of higher end technology like Thunderbolt.TB could be a great interface for docking stations, and could provide a standard for the entire industry. Apple's Thunderbolt Display provides a template for the rest of the industry to consider.
FaaR - Friday, January 13, 2012 - link
Neither you, nor that other guy make any sense at all. TB is royalty-free, any comparisons to firewire is therefore immediately dead in the water.You guys simply couldn't be more wrong. TB's gonna go places and you'll come around liking it eventually once you get over your blind apple hate.
LtGoonRush - Friday, January 13, 2012 - link
That's the thing, there are so few applications for Thunderbolt (over USB3.0) that the only real purpose it serves is to make Intel money. If you really think you need Thunderbolt, nothing stops you from buying a premium model with it included. Much like FireWire, it will never see adoption on consumer devices because of high costs and lack of benefits over USB.Obsoleet - Friday, January 13, 2012 - link
Well, if you don't need it, then it isn't entirely useful.For me fragmenting connectivity standards is worse than any possible gain. This wasn't some brilliant idea by Intel that will also happen to make them money, it was an idea to monopolize another sector USB has been evolving, and we have other connectivity methods for remaining requirements. Fiber optics is nothing new.
repoman27 - Friday, January 13, 2012 - link
I'm not sure how USB has been evolving the high-bandwidth digital display interface combined with external PCIe connectivity sector. I'm also not sure how you reckoned that Intel planned on squashing the poor USB-IF (which Intel formed in the first place) with an interface that only supports a total of 6 connected devices.It's interesting that you find more than one connectivity standard to be the worst possible thing, yet you're afraid of Intel monopolizing the market with a new standard... The bottom line here is that lack of choice is bad. You can't want USB to have a monopoly on the one hand and then not want a different monopoly because Intel's involved on the other.
But I guess we should still probably get rid of all of the other interfaces like SATA, HDMI, Ethernet, and ExpressCard, and just focus on USB. There's no reason to have more than one connectivity standard after all.
FaaR - Friday, January 13, 2012 - link
I demand you stop making sense!Obsoleet - Friday, January 13, 2012 - link
You seem to be either very confused, or simply kneejerk in being argumentative and defensive to your precious Thunderbolt/Intel (not sure which). Did you already foolishly spend money on that crap?I was the one who said we have interfaces for all those things, SATA, HDMI, ethernet, USB.
'we have other connectivity methods for remaining requirements.'
I also, never claimed USB was evolving into external PCIe, I simply said it's cheap, does what it's supposed to do and more, and there plenty of other options for connectivity if you need something else.
This new standard is doomed to fail in the face of USB3 and everything else that's already well-established.
Enjoy buying all those Firewire.. I mean Thunderbolt devices.
Jaybus - Friday, January 13, 2012 - link
I think you're missing the point. TB is simply extending the low-level PCIe bus from the motherboard and/or the CPU to the outside world. In other words, plugging a device into TB is analogous to plugging a PCIe adapter card into a PCIe slot in a PC. That PCIe device could be anything. When you consider that all of the interfaces (USB, SATA, etc.) are already connected to the CPU/GPU via PCIe bridges, even if that bridge is internal to the chipset, then you can see that TB is protocol agnostic.Fiber optics is not new. However, silicon photonics is very new. It is not about fiber optics. It is about how the light pulses are generated and detected. Current high speed optical i/o devices, like fiber channel adapters, require very expensive laser driver and detector hardware. Silicon photonics promises integrating the laser, detector, and modulation on the silicon. In other words, it makes it cheap to use fiber optic cables at tremendous bandwidths.
ChristophWeber - Friday, January 13, 2012 - link
You are so correct. TB's great innovation is to externalize the guts of a PC design by putting PCIe onto a serial cable. Everything else follows from there.USB on the other hand is mostly interrupt driven and therefore inherently a bad match to many of the uses we want it for, external disks being one prominent example. Sure, external USB drives work, but it's still hobbled and provides variable, hiccupy throughput. TB, and Firewire before it, solves this cleanly and effectively.
Obsoleet - Friday, January 13, 2012 - link
My 8GB Supertalent Pico-C and 1TB WD Passport (USB 3.0) disagree with you.My external storage averages ~100MB read speeds and ~70MB/sec write, a self powered 5400RPM drive.
Sure it's no SATA drive, but it's more than effective and convenient on technology widely available today. In fact, I've been using it for 2 years now while everyone waits and waits for Tbolt to dominate.
XZerg - Friday, January 13, 2012 - link
You misunderstand the bigger point being made here. Sure USB 3.0 is fast for most of the tasks, however, it is not fast enough to allow a decent GPU to be hooked up to the computer externally. With TB, due to its implementation, not necessarily due to the bandwidth, you can gut out almost all parts of your computer that are currently only viable as internal devices while keeping the performance experience close to normal. A simple example is being able to use an external GPU on a laptop/netbook/tablet to boost the GPU performance considerably - something that USB solutions cannot do and USB doesn't seem to be driven to tackle that front.This isn't to say that TB is perfect but it is at least one step ahead of USB now. Companies need to throw more weight behind TB to create more products that end-users can make use of to simplify devices portability and upgradability.
I believe USB would be great as a peripherals connectivity (displays, hdd, kb/mouse, camera, printer, ...) and TB for computer hardware (videos, raid cards, accelerators, ...).
Exploiting such capabilities of TB would be a two edge sword - provide products with greater benefits (portability for re-use) to end-user, however, at the same time lose revenues from end-users as they will buy only one expensive solution for more than one computing system instead of for each system.
Wardrop - Friday, January 13, 2012 - link
Thunderbolt and Lighting Bolt seem to be completely unrelated, despite initial impressions. Lightning bolt basically combines two existing external interfaces into one cable, for the sack of convenience (docking). It's purely a physical thing.Thunderbolt on the other hand makes the bandwidth provided by PCI-e, available externally. It provides capabilities that are otherwise unavailable over any other existing interface (or combination of interfaces). If you need to send more than 5/6gbps to an external device, Thunderbolt is the only solution.
r3loaded - Friday, January 13, 2012 - link
This is honestly the first thing that came to mind when I read the title of this article: http://www.youtube.com/watch?v=j_ekugPKqFwumbrel - Friday, January 13, 2012 - link
Return to me the dignity I lost when the people around me saw me fall to floor rolling in laughter!!!Beenthere - Friday, January 13, 2012 - link
I think people associate the name Thunderbolt with comics and unscrupulous PC electronics makers, not a solid produce, a lot like Ultra-Flop notebooks.KPOM - Friday, January 13, 2012 - link
With yet another incompatible technology vying for use of the mDP, this is bound to just lead to more customer confusion. It was bad enough that Intel and Apple settled on mDP, but at least few PC manufacturers besides Apple had used the port prior to the introduction of Thunderbolt.Hector2 - Friday, January 13, 2012 - link
It does seem to just be a merging of USB 3.0 and Display Port. I agree with others about the name. Intel started with "Lightpeak" and that made some sense because it was an optical interface that also supported PCIe, etc.Lightening Bolt doesn't seem to have anything to do with optical other than trying to get people to believe its similar to Intel's LightPeak and ThunderBolt.
In the end its just another name.
CUEngineer - Friday, January 13, 2012 - link
I sense some patent lawsuits here... lolIntels thunderbolt was originally called lightning bolt from what i remember and i think that it was already patented? I think the plan is to call it thunderbolt while they still use coper wires and lightning bolt when it reaches full fruition of using optics.
mino - Tuesday, January 17, 2012 - link
I sense AMD trolling Intel's naming here.Thre is no new tech by AMD here - they just serve the role of coordinator to prevent ODM's doing each its own ...
IMO they key would be a capability to charge the notebook via it. It they can do it , they win big.
If they cannot, they might succeed.
cdillon - Tuesday, January 17, 2012 - link
Physics fail. Signals travel down copper at somewhere between about 65% the speed of light for coaxial cables and 97% for a free conductor. Not sure were you got a 100X difference from.The "point of fiber" is absolutely bandwidth. The maximum theoretical bandwidth limit is far, far higher for fiber optic cable than a copper wire (again, those pesky physics).
Easily isolated fiber optic cables also do not induce nor pick up interference from the surrounding environment, while every copper wire doubles as an antenna. Noise immunity is a giant hurdle in any copper communication link. Fiber links simply don't have to deal with it.
mino - Tuesday, January 17, 2012 - link
Wrong, problem is frequency.You can make cheap copper cable but you need complex demuxes on the ends (which NEED time to do its work) plus you have to run at low freq.
TB would have been wonderfull had the actualy been a primarily fiber interface.
You take optics from TB and you get an inflexible useless eSATA/ePCI/eSCSI ... you name it.
A shame.
mino - Tuesday, January 17, 2012 - link
Then, he is actually advocating FOR LightningBolt(whatever they anme it) as Thunderbolt is an overkill for 99% use cases./me would love commodity TB but, well with intel holding the patents none is gonna mass produce it (on the device side), zero royalty ODM's a**.
mino - Tuesday, January 17, 2012 - link
USB3 has as much in common with USB1/2 as PCI has with PCI express ...