Is Intel’s Thunderbolt Dead in the Water?

When Intel unveiled its ‘Thunderbolt’ connector this past February, it seemed there were reasons to be both excited and a little disappointed. The latter was due to the fact that Thunderbolt wasn’t quite the ‘Light Peak’ we had come to know. While offering similar functionality, it totally lacks a fiber-optic connection, which was the reason for the ‘Light’ in its codename. Instead, we received a much slower copper-based solution, though a still rather impressive one.

The big reason Thunderbolt sounded so impressive out of the gate is that it has huge bandwidth – 10Gbit/s bi-directional. Theoretically, that’s 4x the total bandwidth of USB 3.0, minus the overhead each of these standards bring. For those who love the idea of having such massive bandwidth available through a simple connector, Thunderbolt gave them a good reason to be excited.

During the months after its launch, however, I began to wonder if Thunderbolt was indeed that important, or interesting. It does offer much more bandwidth than USB 3.0, and reduces latencies, but are either of those things important? On the MacBook Pro, for example, while Thunderbolt is available, USB 3.0 is not. One is able to offer people super-fast transfer speeds today, while the other is experiencing slow adoption. For media-hungry consumers, having Thunderbolt today seems to be more of a burden than a benefit.

It seems I wasn’t alone in this thinking, as ExtremeTech has gone as far to call Thunderbolt ‘Dead in the water’. The reasons are similar to what I mentioned above, but also the fact that Thunderbolt today is far more expensive than USB 3.0. While a typical USB 3.0 controller runs around $3 USD, Thunderbolt hardware hovers around $90. In fact, Matrox’s latest Thunderbolt-enabled products cost about $200~$300 more than similar eSATA or USB equivalents.

There is such thing as an “early adopter’s tax”, but even so, in today’s landscape Thunderbolt’s true use is limited, and those who can take advantage of it are likely to be limited to film producers or media developers in general who demand the massive bandwidth it offers. The question still has to be begged – what’s wrong with USB 3.0’s ability to transfer at speeds higher than 500MB/s? With storage, latency isn’t a major issue, and at the end of the day, hard drives are the biggest bottleneck there, not the interconnect.

As a consumer, I still like the idea of Thunderbolt because choice is good. But I can’t help but feel like it’s a tech that right now isn’t important, though the potential in the future could be huge. How do you guys feel about Thunderbolt? A possible “must-have” product in the near-future, or one that fills a non-existent void?

Light Peak could even piggyback on top of USB cables, providing socket backwards compatibility — and to top it off, Intel said that 10Gbps was just the beginning: 100Gbps would be possible in the next decade! Light Peak, in short, delivered a delicious hint of what a fast, flexible, and future-proof interconnect could do. Then Apple came along and ruined everything.