Thunderbolt Adoption Reportedly Slowed by Intel's Licensing and Certification, But Improvements Coming

bungiefan89 said:

Are you saying it's possible for me to use my 2011 27" iMac as a screen for my Xbox 360 or my 2009 Macbook?

Click to expand...

No. There seems to be a misconception of what Thunderbolt is. Thunderbolt is an extension of the PCIe bus, allowing external peripherals to operate as though they are connected to a computer internally. Thunderbolt is not natively a video protocol, but it allows Displayport data to be multiplexed into the Thunderbolt data stream. At present Thunderbolt only carries a 4x PCIe signal, but if it reaches the full 16x, it would be possible to have an external card cage with multiple nVidia GTX 680 desktop video cards in SLI mode, connected to an iMac for gaming, as iMacs have desktop CPUs.

The Thunderbolt display and iMac have minidisplayport output, but not input. That's an Apple decision. If they accepted minidisplayport input, then any device with minidisplayport output could be connected using a minidisplayport cable.

I'd just like to see a cheap Thunderbolt dock with a few USB 3 ports, a few FW800 ports, and a thunderbolt passthrough port for daisy chaining. Maybe add GigE since the few docks out there also have it and it's probably cheap to include.
Oh, and bacon.

Its more than that... its what Apple advertised thunderbolt as, which has clearly failed.

The reality is, TB will remain a niche product, more so than FW. Neither Consumers, or the vast amount of other PC manufacturers care about TB. Only a small amount of pro [-consumers], and a handful of PC and peripheral manufacturers.

USB3 and HDMI connectors provide enough functionality and at a great price point for the consumer market ( and pro-consumer too, for that matter ) , which is far larger than the target market for TB.

TB may be technically better than other solutions, but in the market place, this has little relevance if no one cares about it.

deconstruct60 said:

Like Steve Jobs never sold kool-aid; reality distortion field.

Apple was spinning doing Thunderbolt first and while USB 3.0 late. It always was going to be two over the short-medium term future. The great distance future nobody, including Apple, knew about.

Apple, Mac's specifically, has a much deeper need Thunderbolt sooner rather than later because they were largely limited to no PCI-e expansion. They needed some ability in that area to be competitive in the general market. So it was made an extremely high priority.

Click to expand...

bungiefan89 said:

All Thunderbolt needs is an adapter so it can accept a signal from HDMI, VGA, and/or DVI.

Click to expand...

and USB3!!!

CoolSpot said:

TB is dead and USB3 killed it. If you are an OEM, are you going to waste time and money chasing a connector that nobody uses, or are you going to load up with USB3.

Click to expand...

It is not a time wasting issue. It is a "time priority" issue. For Apple there are very significant and well grounded reasons to do Thunderbolt first and USB 3.0 second. For general PC vendors (with many system with several ports and deep fixation on classic PCI-e connectivity through slots and ExpressCard ) there is quite another order that presents much lower risk and short term need match.

That hardly means Thunderbolt won't grow over time. The primarily remaining high growth form factor in the general PC market is the laptop. The PCI-e connectivity solution for laptops is largely ExpressCard. ExpressCard versus Thunderbolt is relatively a revolutionary increase in speed.

In general the average size of the enclosure for a Personal computer is going to get smaller. That is going to squeeze out the classic PCI-e card as being the most widespread solution. Thunderbolt is well position to partially alievate that issue in a wide spectrum of contexts. USB 3.0 (or even USB 3.0++ ) is not.

Dragging this into a Thunderbolt as are replacement for USB is just silly. It isn't. They both have a primary area of utility that are different. There is some areas where they can overlap but to focus on that is to miss the forest for the trees.

hamean said:

Thunderbolt has been an abject failure to this point. Bytes from this article indicate that it will continue to be available exclusively at the professional price point for the foreseeable future.

Click to expand...

Yup. Personally, I don't blame Apple. The fault lies squarely with Intel on this one.

Intel engendered a magnificent technology and then almost killed it by adding so many pointless stipulations:
- Apple exclusive for 1 year (90% of the market frozen out )
- Not likely to appear on AMD platforms
- Intel needs to personally validate each product to make sure it is fully ThunderBolt compatible which is time consuming and expensive
- Not integrated into any chipsets or upcoming chipsets
- None?/Few PCIe Thunderbolt cards therefore one is not able to retroactively add Thunderbolt to computers
- General public disinterest after being frozen out for a year
- Price premium which remains artificially high due to the lack of options for Thunderbolt due to the aforesaid restrictions.

Asus, the number one motherboard manufacture, state that once one in four people require a certain technology, it becomes feasible to add it to a motherboard. ThunderBolt doesn't look likely to get above the 25% which will severely limit its ubiquitousness. Niche products beget niche prices so I doubt the price will fall a lot.

If they want Thunderbolt to catch on, they need to slap it on the Haswell Z77-equivalent chipset and swallow some of the cost. Next to no-one is going to pay $100 more for a Thunderbolt motherboard or a $100 more for a computer with Thunderbolt on it.

The absolute antithesis of ThunderBolt is DisplayPort. Like TB, it is a superior technology but it is royalty-free which means that manufacturers incur minimal costs to add it to motherboards. Despite becoming mainstream after HDMI, it appers as though it will usurp or at least complement HDMI in the display space in the coming years.

Stella said:

Its more than that... its what Apple advertised thunderbolt as, which has clearly failed.

Click to expand...

There is nothing on this page

http://www.apple.com/thunderbolt/

that says "one port to rule them all". Apple says it is small, fast, 'giant leap in possibilities' , 'revolutionary'. All standard Apple hyperbole.

This notion of this will be the one port you only need is largely made up. Not even the vendors are saying that. It certainly sells page views on websites ( tech news and places like Macrumors) but it is primarily imagined "filler" that folks are constructing.

Messages like this (which is on the page ) are misinterpreted

"... You also have more than enough bandwidth to daisy-chain multiple high-speed devices without using a hub or switch. For example, you can connect several high-performance external disks, a video capture device, and even a display to a single Thunderbolt chain while maintaining maximum throughput. ..."

This is highlighting the diversity of protocols/devices you already use being consolidated onto a single cable daisy chain. It does not say that ports are being consolidated. Just the traffic onto one cable. There is a distinct difference between those two that is lost on those primary interested in conducting a jihad against some other port as opposed to looking at what the intended uses are.

The reality is, TB will remain a niche product, more so than FW.

Click to expand...

FW wasn't a niche. It was very pervasive on video cameras and FW400 became commonplace on many PCs (that were designed to be a cheap as possible).

It wasn't on the majority of PCs but neither was it a relatively narrow percentage either.

Thunderbolt has similar possibilities. Probably never dominate ( 80+% ) the market but quite possibly more than a narrow range of PCs (e.g. 30-45% ).

Thunderbolt is likely to remain a discrete add on controller for many years so it won't be a default addition, but it will be a differentiator.

Make it affordable then it will take off. Very simple.

drambuie said:

No. There seems to be a misconception of what Thunderbolt is. Thunderbolt is an extension of the PCIe bus, allowing external peripherals to operate as though they are connected to a computer internally.

Click to expand...

Thunderbolt converts and transports two protocols. PCI-e and DisplayPort. Not just one. The misconceptions largely flow from other misconceptions being propagated. Thunderbolt is not PCI-e data... it is in a new format that is Thunderbolt. That new native format is only used to get the data from one controller to another. The data reappears on the other side in one of those two native formats.


If a DisplayPort only cable is hooked to a Thunderbolt port the controller can to some extent just pass through the DisplayPort formatted video data.
They confusion in this context arises largely from this backward compatible mode and inattention to the direction of video traffic (or belief that direction doesn't matter... it does. ).

Display Target Mode (whether tracked through Thunderbolt or a regular Display Port cable ) is not a standard. It is a proprietary feature that Apple implements on some iMacs. So no it doesn't "have to work" because it technically isn't even part of the Thunderbolt standard.

ArchAndroid said:

Yup. Personally, I don't blame Apple. The fault lies squarely with Intel on this one.

Intel engendered a magnificent technology and then almost killed it by adding so many pointless stipulations:
- Apple exclusive for 1 year (90% of the market frozen out )
- Not likely to appear on AMD platforms
- Intel needs to personally validate each product to make sure it is fully ThunderBolt compatible which is time consuming and expensive
- Not integrated into any chipsets or upcoming chipsets
- None?/Few PCIe Thunderbolt cards therefore one is not able to retroactively add Thunderbolt to computers
- General public disinterest after being frozen out for a year
- Price premium which remains artificially high due to the lack of options for Thunderbolt due to the aforesaid restrictions.

Asus, the number one motherboard manufacture, state that once one in four people require a certain technology, it becomes feasible to add it to a motherboard. ThunderBolt doesn't look likely to get above the 25% which will severely limit its ubiquitousness. Niche products beget niche prices so I doubt the price will fall a lot.

If they want Thunderbolt to catch on, they need to slap it on the Haswell Z77-equivalent chipset and swallow some of the cost. Next to no-one is going to pay $100 more for a Thunderbolt motherboard or a $100 more for a computer with Thunderbolt on it.

The absolute antithesis of ThunderBolt is DisplayPort. Like TB, it is a superior technology but it is royalty-free which means that manufacturers incur minimal costs to add it to motherboards. Despite becoming mainstream after HDMI, it appers as though it will usurp or at least complement HDMI in the display space in the coming years.

Click to expand...

Oh really? I am pretty sure that one year exclusive was Apple's idea. Besides, the only problem with TB and Macs is that Apple was stupid/lazy/greedy to add USB3 ports to Macs when all other PC manufacturers did it and not to stick with outdated USB2.

Radio said:

Well that's unfair.....Nothing is compelling about tb over USB 3 for the average consumer

Click to expand...

Uhmm.....
Peer to peer, no tying up CPU resources
Speed, two bi-directional data lanes at up to 10 Gbps, for now
With electrical connections, 10W of power
Daisy-chainable
Simplicity, one connector for all peripherals and accessories
With optical connections, future speeds of up to 100 Gbps and beyond.

True, the average consumer probably doesn't need this right now, but the way we use technology evolves constantly, and Thunderbolt is tech that will make us better prepared for a future of interconnected devices all communicating back and forth at blazing fast speeds to satisfy our ever-increasing need for information.

This could have left USB3 in the dust.

Intel can read the writing on the wall. USB will eventually do in Thunderbolt. They are pricing it out of existence like SCSI and firewire before.

rmwebs said:

Whilst I think of it - anyone who's in the market for external drives for home/work - dont bother with Thunderbolt. USB 3.0 is perfectly fine. I transferred 15 gigs of apps I rarely use to a USB 3.0 drive (Western Digital Backup Plus) and it took about a minute. That's perfectly acceptable.

Click to expand...

What hard drive did you have that wrote at ~250MB/s?

Let this be the thread that decides if Thunderbolt dies or lives.

I'm sure it's been brought up a thousand times, but for those of you who haven't heard me (or someone else) say it already.

Thunderbolt is already a crucial can't miss feature for my clients.

It, perhaps more than any other single thing, keeps Apple in our long term purchasing plans.

With it we have Docking stations, SAN (FC) adapters, DAS-Raid options, PCIe expansion options, etc (all of those are already shipping and currently supported).

In almost all cases the thunderbolt accessory and computer (MBP/iMac/Mini) would be cheaper than just the computer portion of our old solution(s) (using MP or xServe + addon cards)

Yes, there are times where the job requires a tower or other heavy weight solution, but many times we don't need that anymore.

In fact I just recommend we buy halve a dozen or so iMacs and Mini's for a particular need because they have thunderbolt ports and they offer the right mixture of size/power/capabilities for the given task.

Thunderbolt, to us, represents a new era of being able to live with "appliance" machines but still get our work done. There is even quite a bit of upside in lower physical space, cooling, and power requirements in many cases.

Thunderbolt isn't for every scenario. I very much doubt I will ever see a thunderbolt keyboard, printer, or thumb drive. I do however expect to see it in an ever broadening array of devices which need high speed, low latency, low overhead interconnect.

Thunderbolt and USB3 don't compete in my mind. USB3 is the natural progression of a widely used lowest common denominator peripheral standard. Thunderbolt is the solution to allow those of us who need the PCIe format and the advantages it brings to the table in an ever appliance-fied world.

Just some insight from someone who uses TB plugs everyday.

Karl P

Dream on. This is never going to take hold. The ship has sailed.

rmwebs said:

It was per unit. As I recall a few figures were quoted - all of which were around $75 or above.

Click to expand...

per unit at what volumes? That is unlikely the price at the 100,000 per order pricing level.

Indirect comments I've seen (nothing like an Intel listing) is that the volume licensing thresholds are much higher than other Intel products. Where buying in lots of 1,000 will get you a discount on CPUs that does nothing for Thunderbolt.

Again that kind of thresholding is far more about delaying the race to the bottom than a long term impediment.

However that does shift the risk of getting over that threshold in sales onto the peripheral maker. If they don't sell in volume then they'll have to eat the cost. Few can make that sort of bet or have a product that is value added enough to offset the costs.

At this early stage Thunderbolt does not need the maximum number of devices that overlap with with USB 3.0. That is the wrong strategic focus area for both Intel and the peripheral vendors with any sense.

In comparison, USB 3.0 is something like $5 (on average) and USB 2.0 is next to non existant (a lot of places dont bother licensing it).

Click to expand...

If there is no licensing and certification then it is highly likely that will see substandard products. Quirky products that don't work or cause problems isn't going to help Thunderbolt in its initial stages.

There are some minor problems now even with somewhat stringent certification. If folks were just slapping Thunderbolt logos on stuff willy-nilly then there was be a significantly more junk out on the market.

At last, the truth ....

At least we now know it is not Apple who has been holding back TB.

Xero910 said:

I'm surprised by the number of Thunderbolt-haters in this thread. Comparing it to USB3 as if there is a direct comparison. It's apples and oranges. Thunderbolt is fundamentally a very different connector. You can hate it all you want, but it's here to stay and will be the preferred connector in the years to come.

Click to expand...

I hope you are right, but it would be better if TB found a consumer-level application sooner than later. As tech savvy as I am (compared to most consumers), I have been in part watching to see what will happen with TB, but I still have no real need for it. Rather, I need a FW800 port and I appreciate Apple's inclusion of HDMI on newer Macs. Two TB ports on current Macs to the exclusion of FW seems wasteful of the available space and it only forces me to have to buy adapters.

Kaibelf said:

It's been like... one year. I remember when USB came out, it didn't magically appear in every machine or on all peripherals for about 4 years, so to call it an "abject failure" before it's even gotten a chance to go either way is pretty dumb.

Click to expand...

Not to quibble too much, but wasn't TB released on Macs in 2011? That makes TB almost two years old. Your reference to the time it took USB to become ubiquitous might be accurate, but I can also understand people who are upset at the limited use for TB given the time already elapsed. After two to three years, many people are looking to upgrade and many have been unable as yet to find a use for TB.

coolspot18 said:

The one and only application keeps getting repeated over and over. Are there any other (interesting) applications for Thunderbolt?

Where are the external PCI-e enclosures, or docking devices that actually have a decent set of ports?

Click to expand...

Ditto.

coolspot18 said:

Where are the external PCI-e enclosures, or docking devices that actually have a decent set of ports?

Click to expand...

http://www.sonnettech.com/product/echoexpresschassis.html (but if you need to ask the price you can't afford it).

...but yes, I don't understand what's preventing someone doing a decent dock.

unplugme71 said:

Apple killed Thunderbolt when they included USB 3.0 in their devices. Now with USB 3.0 r2 coming in 2014 that offers same speed as Thunderbolt, it will officially die.

Click to expand...

...but if you've got Thunderbolt then there's a fighting chance that you'll be able to add a USB3r2 controller to your 2012/2013 laptop.

USB3 is the best bet for adding individual peripherals. Thunderbolt is ideal for adding additional ports. They're not really in competition.

With a commitment like that from Intel, it sounds like Thunderbolt will remain a niche standard (at best).

USB3's lead is just about insurmountable now. The race is already over. Even if things change drastically for the better for TB in a year or so (unlikely), I don't think that will make a difference.

Not surprising... but quite disappointing.

John.B said:

Thunderbolt is a prosumer-level technology, it's not designed to be a USB replacement. It's effectively a PCI-e bus extension. If you don't need what that offers then a USB 3.0 external drive might be all you ever need.

Click to expand...

^^^ THIS.

For what it does, Thunderbolt, was exactly what was promised. It's real-world applications will be cost-prohibitive and arguably even unnecessary for consumers.

theluggage said:

http://www.sonnettech.com/product/echoexpresschassis.html (but if you need to ask the price you can't afford it).

...but yes, I don't understand what's preventing someone doing a decent dock.




...but if you've got Thunderbolt then there's a fighting chance that you'll be able to add a USB3r2 controller to your 2012/2013 laptop.

USB3 is the best bet for adding individual peripherals. Thunderbolt is ideal for adding additional ports. They're not really in competition.

Click to expand...

Thank you for the clarification (which I bolded); however, the first part of your post states the problem for the typical consumer--price. If the price is too high (e.g., $300 for a port replicator), then people will simply wait and purchase a new computer with the new ports ($300 will almost buy a cheap Win PC). Intel (and Apple) needs the larger PC community to adopt this technology for it to succeed; Mac users alone don't constitute a large enough market to make this enterprise succeed. This is especially true if third-party developers are the ones assuming all of the cost and risk. They need some kind of forecastable ROI (return-on-investment) to take the plunge.

Stella said:

Its more than that... its what Apple advertised thunderbolt as, which has clearly failed.

Click to expand...

That's Apple's failure. In 2011, when Apple shipped the first laptops with Thunderbolt, Jason Ziller (same guy who's been heading TB for all these years) stated perfectly well what Intel intended for the port and technology, and the consumer market wasn't it :

http://www.pcmag.com/article2/0,2817,2380890,00.asp

[Jason Ziller, director of Thunderbolt planning and marketing], who was once chairman of the USB Implementor's Forum, said that USB 3.0 and Thunderbolt were "complementary." "We don't see this replacing USB," he said. 'We see it as complementary to USB... Intel will fully support and work with that technology."

Click to expand...

So if Apple was position Thunderbolt as a replacement for USB 3, they were doing this against what Intel was doing. Again, this interview is from 2011, when Apple launched the Thunderbolt enabled MBPs. It's not like Intel changed their minds here.

TB is pretty great.

i'm just not buying any of it until it's reasonably priced.

macs4nw said:

Uhmm.....
Peer to peer, no tying up CPU resources

Click to expand...

PCI-e devices are not driver-less. There is software running on the CPU.

Speed, two bi-directional data lanes at up to 10 Gbps, for now

Click to expand...

The indepenent SuperSpeed ( e.g. what is unique to USB 3.0 over 2.0 ) link is bi-directional. (separate set of wires than the 2.0 data transfer on in the same cable) It isn't 10Gbps but if dealing with 1Gbps devices that isn't an issue.

With electrical connections, 10W of power

Click to expand...

Again if devices require 2W of power a non issue.

Daisy-chainable

Click to expand...

Limited to 7 devices including host and may have to break chain to insert/remove device. This one cuts both ways. USB ports are typically more prevelent so don't need to chain ( 4 USB ports versus 1-2 TB ports).

Simplicity, one connector for all peripherals and accessories

Click to expand...

FUD. A TB Display has all the legacy ports on the back. It doesn't provide a reduction in total system ports. Thunderbolt merely allows those ports to be moved; not reduced.

Except for some limited corner cases, (e.g, where SATA controller is bunlded into drive enclosure), thunderbolt doesn't result n a net reduction in ports.

With optical connections, future speeds of up to 100 Gbps and beyond.

Click to expand...

The same core technological improvements that allow Thunderbolt to move forward on speed at more affordable cost are basically also available to Ethernet, infiniband, SATA , and USB. There is nothing unique there.

Thunderbolt's previous live as Lightpeak sinificantly leveraged optical USB 3.0 work that Intel did. Again there is a present distinction but nothing necessarily unique going forward.

Thunderbolt will probably go to 20 Gbps around the same time USB 3.0++ goes to 10 Gbps but that will be several years after Ethernet and Infinband went to 40-100 Gbps. It is a reasonable middle ground between top end and commodization zone.

True, the average consumer probably doesn't need this right now, but the way we use technology evolves constantly,

Click to expand...

Not only wired but wireless. There are short range 7Gbps demos last week at CES 2013. No sockets is going to beat "one socket" for some folks and some contexts.

This could have left USB3 in the dust.

Click to expand...

Perhaps in an alternative universe where USB didn't already have billions of deployed devices, but not in this one.

Thunderbolt is not a shared (committee) based standard. It doesn't have an installed base of billions. Drives up total system costs by pushing the transceivers into the cables.

Thunderbolt will be good competition for USB 3.0. They overlap enough so that USB can't go to sleep for another 8-10 years (like the gap between 2.0 and 3.0 ). However, they are significantly differently targeted that each can have a productive base to work from.