Intel's grip on Thunderbolt keeps accessories off the market
While Apple has integrated Intel's Thunderbolt connection standard in every model of Mac the company has introduced since 2011, the technology has yet to truly take off in the computer industry, and one report says it is languishing due to Intel's tight controls over the standard.
Update: Intel contacted AppleInsider to express the chipmaker's reservations with the information presented in DigiTimes' report. The company's response to the report is detailed in this feature piece.
Apple brought Thunderbolt to the Mac shortly after Intel officially introduced the technology, which the two companies developed in collaboration. Thunderbolt allowed for speeds 12 times faster than FireWire 800 and twice as fast as USB 3.0, as well as the ability to daisy-chain multiple devices without using a hub.
Apple secured an exclusive license on the technology shortly following its unveiling. Even when that license expired, though, other PC manufacturers didn't begin to adopt the standard en masse.
Additionally, the number of peripherals supporting the standard has been well below what some expected given its specifications. A research note from DigiTimes points the finger at Intel's unwillingness so far to license the technology behind Thunderbolt to other parties, a factor that keeps the cost of Thunderbolt high and wider adoption low.
Intel, according to the report, uses independent packaging for Thunderbolt chips, and the chipmaker maintains a tight grip on the technology. The specialized chips Intel places near Thunderbolt configurators are proprietary designs, meaning that Intel can maintain control over the pricing on Thunderbolt cables.
A four-channel Thunderbolt chip component has a wholesale price of $35, while a two-meter Thunderbolt cable has a recommended retail price of $39. These price points are keeping some smaller manufacturers from entering the Thunderbolt device segment, leaving it largely to established players.
Other chipmakers, including ASMedia, are trying to convince Intel to license the silicon intellectual property rights related to Thunderbolt. Licensing the technology, they argue, would allow other companies to make smaller-sized and cheaper peripherals. Intel has yet to respond publicly to such requests.
Update: Intel contacted AppleInsider to express the chipmaker's reservations with the information presented in DigiTimes' report. The company's response to the report is detailed in this feature piece.
Apple brought Thunderbolt to the Mac shortly after Intel officially introduced the technology, which the two companies developed in collaboration. Thunderbolt allowed for speeds 12 times faster than FireWire 800 and twice as fast as USB 3.0, as well as the ability to daisy-chain multiple devices without using a hub.
Apple secured an exclusive license on the technology shortly following its unveiling. Even when that license expired, though, other PC manufacturers didn't begin to adopt the standard en masse.
Additionally, the number of peripherals supporting the standard has been well below what some expected given its specifications. A research note from DigiTimes points the finger at Intel's unwillingness so far to license the technology behind Thunderbolt to other parties, a factor that keeps the cost of Thunderbolt high and wider adoption low.
Intel, according to the report, uses independent packaging for Thunderbolt chips, and the chipmaker maintains a tight grip on the technology. The specialized chips Intel places near Thunderbolt configurators are proprietary designs, meaning that Intel can maintain control over the pricing on Thunderbolt cables.
A four-channel Thunderbolt chip component has a wholesale price of $35, while a two-meter Thunderbolt cable has a recommended retail price of $39. These price points are keeping some smaller manufacturers from entering the Thunderbolt device segment, leaving it largely to established players.
Other chipmakers, including ASMedia, are trying to convince Intel to license the silicon intellectual property rights related to Thunderbolt. Licensing the technology, they argue, would allow other companies to make smaller-sized and cheaper peripherals. Intel has yet to respond publicly to such requests.
Comments
This is, indeed, what held FireWire back. It might be too late to fix it by now, too; Apple eventually let go of some of its grip on FireWire and made it more attractive to peripheral makers, but by that time USB 2.0 had come out and was "good enough." Now, USB 3.0 is out and is "good enough," and Thunderbolt doesn't even have the depth of support that FireWire had in its early days. Thunderbolt is, sadly, probably DOA at this point.
It's so frustrating to see people repeat the mistakes of the past.
Since intel is doing poorly lately, I would suspect that what's happening here is that they are pricing the chips so as to recoup what they feel is their "investment" in developing the technology. Basically keeping the price high to make their books look good and to pay for research that probably isn't even related at all.
R&D costs however are the biggest boondoggle that ever existed and can easily be quoted to be anything from nothing at all to 90% of the cost of the chip. Most modern thinking on this says that R&D costs should actually *never* be added into product cost but rather absorbed by the company as basic operating costs.
There are PCI expansion boxes that are kind of expensive but geared towards the professionals.
The external hard drives are pretty much RAID systems which are geared for professionals.
There are boxes like Universal Audio which is geared towards the Professional market.
Then there are these new Belkin docking boxes, which are more expensive than a typical USB hub, but much more functional.
Apple was smart in integrating this technology and so far it gives Apple a differentiator that makes a difference for the professional markets, Apple just needs to squeeze out a MacPro that's Thunderbolt equipped as fast as possible to keep their high end tower workstation market.
Quote:
Originally Posted by drblank
It's meant more for the Professional market, at least that's what the majority of the devices on the market are geared towards.
FireWire ended up with a following in the professional market, too, and a much more widespread one than Thunderbolt now has, but I think we all agree that it could have done far better.
Thunderbolt is a failure just like Firewire. The accessories are too expensive. USB3 is all I care about now.
So you'd rather have an Apple display that uses USB3.0 to get all the display data plus all the other data instead of Thunderbolt? Protocol independence with a 40Gb/s aggregate in 2014's Falcon Ridge opens the door to a lot of things USB 3.0 can't do.
Funny, this guy thinks he knows what's going on.
Thunderbolt on my MBA alongside my LED monitor is an awesome combo. Target-mode via Thunderbolt can't be matched, and better things to come.
Maybe you should try using it before trashing it. When USB3 is outdated, TB will still be there, using the same port and be more future-proof than USB. USB in itself is necessary, but TB will always be one-level above it. It's not CPU intensive like USB is. Of course, you already knew that right?
Originally Posted by bdkennedy1
Thunderbolt is a failure just like Firewire.
So: not in the slightest. Got it.
Speaking of the same port, it uses the same mDP so it doesn't require an additional port to be included, unlike FireWire, which was obvious different from what you use to attach a display. It's a win-win.
...and we'll be able to use that speed and extra functionality so well with the Macs we bought in the three years prior to the new controller's release.
Quote:
Originally Posted by SolipsismX
So you'd rather have an Apple display that uses USB3.0 to get all the display data plus all the other data instead of Thunderbolt? Protocol independence with a 40Gb/s aggregate in 2014's Falcon Ridge opens the door to a lot of things USB 3.0 can't do.
Are any of these particularly significant for the mass market though?
Genuine question, no prejudice, I don't know the answer.
I specifically mentioned Falcon Ridge so why the hell would you complain about a Mac in 2011 not having functionality that won't be arriving until 2014?
Quote:
Originally Posted by runbuh
...and we'll be able to use that speed and extra functionality so well with the Macs we bought in the three years prior to the new controller's release.
Great... this guy's back..
Do have any clue what the current TB setup does now?? There's barely anything out there (affordability-wise) than can saturate it already, let alone what will be available in three years. Are you even remotely serious with your comment?
Where do I find examples of this alleged modern thinking? If those costs aren't reflected in the product price, exactly how are those expenses recovered? If those expenses aren't recovered, then what is the incentive to advance technology?
Mind you, I'm not happy with the progress of TB devices, I don't think costing is the problem as pretty much any other product is expensed this way.
Quote:
Originally Posted by Crowley
Are any of these particularly significant for the mass market though?
Genuine question, no prejudice, I don't know the answer.
I agree that Intel needs to get its head out of its backside and start pushing TB. I also think that Intel/Apple is setting up TB for the long haul.
Now, as far as specs go, they are great for those that focus merely on specs and nothing else... For the mass market though, I look at TB more as a general-purpose high-speed dataport. I think we still need more time for the market to figure out how to best utilize it. Apple is doing a great job with it. The PC folks thought with their razor-thin operating margins though is something else.
Outside of my LED monitor, my only TB accessories to-date is a TB->Ethernet and a male-to-male TB cable for doing target mode. It's a great way to transfer large files between to machines. I like that my ethernet adapter can do full gigabit speeds compared to my old USB->Ethernet adapter. I'm sold on the tech, it just needs to be adopted by more people.
Because the effect of what you say it to tell everyone to hold off purchasing computers until next year. So what incentive do hardware vendors have for investing in Thunderbolt today? If you get into a perpetual cycle of "wait till next year", you don't get buyers now, and hardware vendors don't invest in making Thunderbolt peripherals, so you don't get buyers, so vendors don't invest, etc.
The article is talking about the fact that Thunderbolt "has yet to truly take off in the computer industry". We're talking mass adoption of Thunderbolt, not niche markets. 40Gps peripherals are not mass market right now, and won't be in 12 months. Please suggest/offer something realistic/tactical for driving Thunderbolt adoption today. We already know it know it is technically superior. So was Betamax.
Quote:
Originally Posted by JeffDM
Where do I find examples of this alleged modern thinking? If those costs aren't reflected in the product price, exactly how are those expenses recovered? If those expenses aren't recovered, then what is the incentive to advance technology?
Mind you, I'm not happy with the progress of TB devices, I don't think costing is the problem as pretty much any other product is expensed this way.
I don't have any sources for you but the idea is that since you can't really relate generalised R&D costs to a specific product, that the companies are basically just spit-balling a large amount of money that they hope to claw back. For instance, they might have R&D costs for years for many projects that lead to no specific products, and then the one time they have a product to sell, they try and recoup all those losses by inflating the cost of the one product that's selling. There are no controls, nobody watching, and no set rules on how to recoup such R&D "losses" therefore, capitalism being what it is, there is nothing to stop some company from claiming that there is a 40% R&D recoup on some product basically forever.
Did you even read Soli's comment about next year's Falcon Ridge chipset? Like you said, there's barely anything out there and he wants us to wait for next year's chipset. I was talking about the Macs we buy today not being able to use the new speed and functionality of the new chipset.