Thunderbolt – previously known as Light Peak - anyone excited?


second exodous

Advanced Member
Joined
Sep 27, 2005
Messages
2,974
Location
Utah, USA
When I read about Light Peak a few years ago I didn't think it was going to happen or that it was so far off I didn't really care. Well, Apple using only Intel processors gave them an in and now they have it and even renamed it to Thunderbolt.


I can't wait to see what it can do, having one port for EVERYTHING sounds great but we'll see. I hope this doesn't mean more incompatibility and need funky drivers that USB used to have. One thing that bums me out is it isn't optical anymore, when it was being made a few years ago they had full speed data transfer over a spooled cable that was over a mile long or something but now it is just copper so it doesn't have the limits of the speed of light like it used to.


As only Apple has it right now it might not go anywhere outside of Apple making any device that uses it more expensive than any other new bus, i.e. USB3.0, and it will stay overpriced also.
 
Ah, another potentially Apple-only interface. Fan-*******-tastic.


Anything to stand out, I guess.
 
Well, for now, I think Apple is first because Intel is pissed at Microsoft, well, them and everyone else for one reason or another. We'll see though, it might make it over to PCs soon.
 
If it does become somewhat universal, that would be great. Personally I can't stand USB in the least: it's agonizingly slow. I suppose the biggest worry to me then is whether or not it could be a competitor to USB in terms of, well, universalness (totally a word). I think Intel have been pretty decent with at least allowing their hardware to be useable under alternative systems, but Apple is the wildcard here. They're not really friendly towards others. The fact that it's allegedly backwards-compatible with Display Port worries me. Display Port is one of Apple's proprietary doohickeys, isn't it?


Don't get me wrong, it sounds great. I just hate hardware diversification... that sounds bad. What I mean is, whenever it comes to these sorts of things - extension ports, I/O ports, things like those - I really love when things are standardized. In my opinion, those things should be totally universal. That's why this, while seeming good, feels pretty ominous to me. But hey, changes have to start somewhere, right?


I really wish that USB could be brought up to speed. Hell, USB 3.0 uses fiber optics, right? And it's still slow I've heard, though I don't have any experience with it myself.
 
They're not really friendly towards others. The fact that it's allegedly backwards-compatible with Display Port worries me. Display Port is one of Apple's proprietary doohickeys, isn't it?
No. You're thinking of the miniature one, which used to be, but ended up becoming part of the official specification years ago. :p

How dare you assume that I do basic research before I speak!


Heh... thanks, especially for the sources. :p


I guess I have less to worry about now.
 
Given its features, I hope it doesn't wind up a dead horse like FireWire (btw why USB3 when there's been FW800 for years... sometimes I gotta scratch my head at 'newer technologies' cause' we're sometimes given all but gimmicks and slurp it up like it's something new).


But FW took them what ~9 years to get out the door to begin with (from development stage). LP took ~2.
 
Bye-bye USB - no more slave device discrimination FTW!</sarcasm>


And bye-bye firewire, devices can now be compromised with a much more shinier connection ( http://erratasec.blogspot.com/2011/02/thunderbolt-introducing-new-way-to-hack.html )! And hey, it's even less ovious since it may be used to connect SO many different kinds of devices.


Yeah, I know, we've got nothing to hide.

I wonder if there's a further way around this than a microprocessor - say, blocking/locking said port as well? Didn't know that about FW, good to know (so I can see the argument USB>FW). I don't think security was a concern in the 80's when they first started working on the tech, just sayin'


I don't quite understand how they can theorize an exploit without actually doing anything with said hardware first though, that's kinda biased (and lazy writing to stir up a debate/conversation obviously) isn't it? Need more sources/data to back that claim up.


Edit: One person said: "When you enable a firmware password on a Mac this "feature" of Firewire is disabled and it goes into lockdown mode ... " Don't know anything about that though, this is also one thing people should test too (thus removing any bias).


Also this is probably one of those things you want to wait for it to be out in the market a while, before spending on it.
 
Last edited by a moderator:
I wonder if there's a further way around this than a microprocessor - say, blocking/locking said port as well?
Like it's stated in the blog post that I linked to above, a specific kind of hardware virtualization could be used to make Lightbolt a bit more secure. However, this specific line of Macbooks doesn't seem to support this kind of virtualization, but as you guessed correctly, this doesn't necessarily mean that Apple might not secure the Thunderbolt port in another way. From what I know about Apple this seems very unlikely to me though. They have just begun to pay attention to security about a year ago and have a history in terms of security that would be amusing if it wasn't so sad.


In any case, Thunderbolt itself is insecure by design because it trades security for speed.

Edit: One person said: "When you enable a firmware password on a Mac this "feature" of Firewire is disabled and it goes into lockdown mode ... " Don't know anything about that though.
I don't know either, but it sounds logical that the firewire port could even be automatically disabled once you use the (insecure) BIOS password protection feature. On some computers (typically those aimed at business use), hardware ports can be manually disabled separately through the BIOS.


For various reasons, that alone won't give you automagic security either though. In particular, security isn't a solution - it's a concept. You always need a security concept because addressing a single security problem won't make any system secure.


------------------------------------------------------------------------

I don't think security was a concern in the 80's when they first started working on the tech, just sayin'
Other famous victims of "currently no need to address that" (pun seriously not intended) include DNS, IPv4, and Unix time; to name just a few...
 
10Gb/s, 40Gb/s, and 100Gb/s IEEE 802.3 ethernet is a reality, as is Cat7 (10Gb/s) and Cat7a (40-100Gb/s), and HDMI/Display Port has already led to cabling that's rated to greatly exceed the 10.2Gb/s spec if you don't have to run it long distance. So hitting those speeds isn't novel, even with copper. Given one isn't exactly going to be running a RAID array of RAM SSDs off one of these ports, there's not exactly that much point to that kind of bandwidth outside of the stuff that already has or exceeds that bandwidth.


USB 2.0 practical max was 320Mb/s or 40MB/s with USB 3.0's theoretical max at 4.8Gb/s or 600MB/s. Even if it had 50% overhead you'd need a fast SSD, that you should be running off a eSATA anyways, in order to saturate that. The upcoming PCIe 3.0 lanes overhaul is much more meaningful to the realm of PCs then Apple buying an Intel gimmick of no real practical value outside of being a proprietary port to force Apple customers to buy derivative products from Apple.
 
Last edited by a moderator:
Back
Top