Many people here seem to think all it takes is mass adoption for good prices.
Just look at firewire or other more complex chip designs.
Thunderbolt needs insanely expensive active cables that cost more than the controllers. Optical solutions are apparently still not working and would be even more expensive.
The current controller is still quite huge and expensive. Only made by one company that is very fond of its margins. Those won't ever be anywhere near as cheap as DP, USB 3.
It is and will stay a really expensive tech that is only worth it for very few applications even if you fast forward 5 years into the future.
It is a tech that sound good but doesn't deliver what it was supposed to. It sucks too much energy.
The only real purpose are docking stations and some workstation external drive array uses. The former is the only thing that may be of use for many people but I don't see them coming down in price.
You need a TB controller, a PCB that supports via additional controllers HDMI, DP, DVI, VGA, USB 3, Audio and if that wasn't enough a cable that will never be much cheaper than double or tripple a still not very cheap HMDI cable.
All that in a box plus margin for the company and it is a freak-in expensive piece.
Apparently the external GPU solutions have big troubles with BSOD and other driver problems that aren't that easy to fix. TB is just not that great at plug&play.
TB was once touted as the one size fits all future wonder for everything. USB 3 is worlds cheaper, needs less energy, 2$ cables and is fast enough for anything that isn't Displays, external GPUs or really fast RAID arrays. Current USB 3.0 controller can handle close to 300 MB/s which should be enough for any single drive and even dual HDD RAID 0/1 arrays. 400 MB/s is about what future controllers will possible achieve as a maximum.
That is enough for almost anybody for very little money.
TB will have a hard time justifying its cost on the mass market and will therefore likely never really become a true mass market product.
What it needs is optical cables with unlimited length. Cheaper controllers. More stable plug & play. Cheaper cables. Willingness from Intel to subsidize the adoption in the mass market somewhat.
That would at least make it a great future standard for universities, companies, events, where people need often work with projectors and other equipment.
Somehow I think even here some wireless Display tech will show before TB really becomes a standard.
I wonder if it would be possible for a optical TB to enable wireless optical bridges. You just put the Notebook next to a sensor built into the table. The optical signal travels some 3-50cm over the air (or directly even further) to the cable and than goes to the projector, display from there. No need for adapters, plugs at all, and possibly much less energy use.
Wireless is the future. I agree with some posters that TB is more or less doa. It will find a use by a very limited number of people but won't penetrate the mass market because of the costs and lack of real need for its features.
