First Thunderbolt Peripherals Arrive To Market

MojoKid writes "Promise Technology recently launched the first Thunderbolt-compatible devices; the company's Pegasus RAID R4 and R6 storage solutions can now be ordered from the Apple Store. There's a catch, however. In order to use either storage array, one must first purchase a cable directly from Apple. The company has priced the two-meter cable at $50. As it turns out, Thunderbolt uses what's called an active cable. Inside the cable there's a pair of Gunnum GN2033 transceivers. The GN2033 is a tiny, low power transceiver chip designed to be placed inside the connectors at either end of a Thunderbolt cable, enabling dual bidirectional 10Gb/s concurrent links over narrow-gauge copper wires. The cable's $50 price may be justified, but it's also a further reminder of why Thunderbolt may follow FireWire's path into obsolescence. Apple is the only company currently selling Thunderbolt cables."

Comments

or maybe

[Anonymous Coward]

or maybe, once production is ramped up, prices will go down. Since that's what generally happens with new technology.

Re:or maybe

[Serenissima]

That's crazy talk. This is Slashdot. Where anything remotely related to Apple or Microsoft must be met with derision! There's no need to bring logic or common sense into the discussion!

Re:or maybe

[tlhIngan]

The summary says it's currently being used for RAID configurations. That's a sensible use. But I doubt it will make much headway with consumers.

How about some high-bandwidth situations? Like perhaps having a nice mobile device with Thunderbolt with long battery life, then plug it into your Thunderbolt dock and you suddenly have kickass gaming graphics and all that fun stuff?

Hell, perhaps we'd see stuff like GigE network dongles and stuff - if you're mobile and using WiFI all day, then plug it in at home and you have gigabit connectivity.

Right now, people use it because it's crazy fast for drives. But it's likely Intel sees it as the future of mobile devices - optimized highly for mobile use with long battery life by keeping all the power hungry stuff in a dock - high-end graphics, wired networking, etc.

It's basically a cable-ized version of PCIe.

Re:or maybe

[RoFLKOPTr]

Not without some competition they won't. And Apple's patents will ensure there's very little competition

Intel owns the rights to Thunderbolt technology and trademarks. Apple helped develop it, which is why they happen to have the first cables and devices on the market.

Re:or maybe

[scubamage]

The design behind thunderbolt is intel, not apple. They just partnered with apple to get it out into production. The technology is actually pretty nifty; unlike other interfaces where you're dealing with a separate controller, thunderbolt basically creates an external pci-express port. So, anything a normal expansion card can do can now be made modular. It's got some very sexy potential. Imagine never having to get more than a decent proc and ram because video cards now plug into your laptop's thunderbolt port. You could have thunderbolt enabled televisions which include a graphics adapter. There's some cool potential. Let's see if it actually gets off the ground though.

If you think $50 for the Apple cable is bad...

[crafty.munchkin]

Just wait for Monster Cables to bring out their gold plated $800 Thunderbolt cable!

Re:If you think $50 for the Apple cable is bad...

[bucky0]

what matters is that it is special, Apple-born and exlcusive therefore carrying high profit margin.

Intel made it and owns the rights to it, apple just helped develop it. It's hardly "apple-born"

No thanks

[Dishwasha]

I'm going to hold off on buying these because everybody knows Monster Cables are the best. Their sweet gold-plated impedance really accentuates the harmonics of my digital bits, giving my data soft warm tones and the largest acoustical threshold range that guarantees that my ones are as oney as they can be and my zeros actually stop the measurements in my voltmeter because all the electrons are at a complete standstill. I mean seriously Apple, $50? You're practically admitting that these cables are just junk.

passive was too hard.

[cheeks5965]

my team did a lot of the ground research for the light peak spec. the greatest challenge was shoving enough bits through the wire -- we couldn't find a way to do it passively. That's why it's $50.

Re:passive was too hard.

[toQDuj]

The chips are tuned *per cable* as far as I heard, and thus cannot be included on-board. They would've if they could've.
regarding the fibreoptics, the cost was much higher than for copper. Not rocket science, but not exactly consumer-price either.

Re:passive was too hard.

[itsdapead]

Because different cables can use different chips or firmware. The initial Intel release said that optical cables might be available in the future - same (electrical) sockets, but with an optical transciever built into each plug.

Also, Thunderbolt is not a USB replacement for attaching mice and cheap memory sticks - its an external PCIe bus and its killer apps will be things that you can't do with USB. Hence the first peripherals are things like kick-ass RAID arrays, fast SSDs, high end video capture/editing kit etc. One of the forthcoming peripherals is an external case to take a full-size PCIe card (try that with USB!)

So, lots of MacBook users are not going to use TB as anything other than a monitor port, so it makes sense to shift some of the component costs to the cable rather than the motherboard.

Re:passive was too hard.

[toQDuj]

Well, as they say in the linux help forums: If you can think of a better way to do it, feel free to make and implement your own.. I am of the opinion that thunderbolt has potential and will wait a while before making opinions. I also think that there are a slew of engineers who have worked on this, and it seems to me a bit insulting to them to hear people here go "why did you not just do A or B". Do you really think they did not think of this thing you just thought up in 5 minutes? The hubris of the unwashed masses..

Actually it fits quite nicely w/ Apple's strategy

[antifoidulus]

well, at least part of it anyway. With the departure of the XServe from Apple's lineup and their promotion of the mac mini server, it's obvious Apple is really trying to go for the small-medium business market with their server offerings. As part of that, Apple has been trying to convince owners/IT people who work at said businesses that you can essentially create the same "infrastructure"(hardware/software/workflows etc) as the big enterprises do without having to spring for enterprise level hardware. Even with the cable, this RAID is still cheaper than a fiber channel card, and of course actually allows people to connect real storage to the mini-server(provided they throw a thunderbolt port in the next mini, which they would have be insane not too).

While I certainly don't see anything that requires a $50 cable to totally usurp USB anytime soon, that doesn't mean it won't be successful or fit in well with the type of product lineup Apple is trying to build.

Too early yet to bury Thunderbolt

[tftp]

The cable's $50 price may be justified, but it's also a further reminder of why Thunderbolt may follow FireWire's path into obsolescence.

Firewire went to silicon heaven because USB was cheaper, smaller (connector-wise and cable-diameter-wise) and fully embraced by Intel. Will you make a FireWire mouse? Probably not; you can hoist a cow on a standard FireWire cable. But once you have a USB mouse, why to get Firewire? Note that speedy peripherals were uncommon back then, except video cameras. And USB 3.x attacked that market; I have one USB 3.0 device here, an HDD, and it is backward compatible to USB 2.x.

However 2 x 10 Gbps is some good increase in speed. You don't need it for 99% of peripherals on the market; but when you need it you need it - like that RAID thingy which can generate and consume that much data. Your choices there are simple - either this Thunderbolt, which is more or less fixed, or a variety of 10 Gbps connections, copper or fiber, SFP+ or XFP or whatever. They all are very much different, locking you into some specific hardware, and they all run hot - bad news in a notebook.

10GBASE-T is one of competitors; it runs on slower clock and requires more pairs. But as long as it works, who cares? The twisted pair cable, even category 6A, is cheap, and the distance up to 100m is what you want in any reasonable setup that includes more than two boxes on top of each other. 10G Ethernet is also switchable and routable. Considering that Thunderbolt is a point to point transport for DisplayPort and PciE, it's use is probably limited to expansion ports; but it's probably pretty good in that role - even if majority of computers can't even handle the bandwidth, let alone have a need for such a thing.

Re:Too early yet to bury Thunderbolt

[tftp]

But do you really need it?

It beats SATA because it is not locked into ATA command set. Thunderbolt routes PCIe I/O, which means you can build any PCI peripheral and it will work as if you plugged it into the main board. You can have access to the RAM, use interrupts, DMA and whatever. There are many I/O devices out there that generate lots of data, and they are not disks. Medical sensors, scientific equipment, software-defined radios, high resolution / high frame rate cameras (for security and for machine vision,) external video cards and GPU... I can think of many examples.

Another item of interest is the DisplayPort channel. SATA doesn't support it, Thunderbolt does. Sure, you can always have a second cable... but why to use two when one works fine? The need for remote display devices is quite obvious, and one Thunderbolt jack can replace DP and SATA ports - something that a small device will appreciate.

Re:Reason? It's Smart

[nonsequitor]

Putting the transceivers in the cable itself could mean that upgrading the bandwidth is as simple as getting a better cable and upgrading the thunderbolt driver.

like the fella said

[callmebill]

Standards are like toothbrushes: everyone agrees you should have one, but no one wants to use yours.

Still waiting for the magic box...

[hile]

What I would like to have with thunderbolt is fancy magic breaker box, which would for example include:
- 4 firewire 800 ports
- 8 USB2 / USB3 ports
- 2 ESATA ports for disks
- maybe connector for external display as well

Connecting such box to your laptop might sound silly for most users, but my use would be to hook this to my music hardware rack, having all of the audio hardware connected to your gig laptop with one cable. Like, all various MIDI controllers (usually USB), audio recording interfaces (usually firewire), instruments (my line6 guitar amp has USB) and external disks for recording.

Usually you only use one or two of these devices at a time, but the cables can be really a PITA: having one magic box bolted to your audio rack, connecting everything there permanently makes things so much simpler. Of course, I would like the magic box to come in 1U form factor, or with rack mounting kit.

If such box is made available, I seriously might be tempted to get a new MBP, just to be able to use it.

This is not going to make thunderbolt a must for all users, but it's wonderful technology to replace firewire (which is certainly not dead yet in pro audio market!). Everything doesn't have to be The Big Thing for everyone. I'm not sure about USB3, but I though it still has latency issues like USB2 for multichannel audio (like 32 channels, not your average gaming rig...), which are not solved by higher transfer rates. Might be wrong of course regarding USB3...