Argh it hurts to see Thunderbolt (Light Peak) but no Thinkpads have it...

It might/ might not be depending on market reaction. How many storage company will follow suit or are they going to stall until a better technology came out.

 

I'm still pretty happy with eSATA. The USB 3.0 2.5" enclosure I have is also pretty nice but I'm not running out the door to buy a Mac so Thunderbolt will have to wait until the market matures.

And by mature I mean there are affordable products. I'm not really going to toss my eSATA or USB 3.0 enclosures anytime soon so unless one breaks I'll sit on the sidelines and watch the show.

If I were a pro photographer or videographer, I have no doubt the gravitational pull of the MacBook Pro 17" with Thunderbolt would be a solution I'd strongly consider.

 

I remember working in a mom-n-pop computer shop. I'm trying to remember how long it was once USB showed up on Pentium II mainboards before useful peripherals came along.

I'm thinking I didn't have a USB anything until the Pentium III, or perhaps the Athlon. But man, were those Pentium-II mainboards future-proof.

 

My family had a Pentium desktop from 1997 to 2001. I think we got a USB mouse when we got a new computer (an Athlon 1 Ghz). Yeah I remember on Windows 95 that damn thing always demanded to Scandisk and froze at 98% or somewhere around there.

 

What I don't get is why people think the MacBook Pro is "future proof" because it has Thunderbolt, whereas something like the T420s is not because it has USB 3.0 instead.

USB 3.0 is going to win. It's supported by every major manufacturer except Apple. It's compatible with USB 2.0, the most popular external I/O technology ever. And it's plenty fast for whatever people want to use it for in the immediate future.

Even Intel's next-generation chipsets (7 series) are supposed to have built-in USB 3.0.

If you want a future-proof interconnect, you want USB 3.0. Thunderbolt may make it to some PCs, but consider this:
- Thunderbolt requires DisplayPort, which most PCs don't have.
- Thunderbolt controllers appear to require their own heatsink (unlike USB 3.0), adding cost.
- Thunderbolt controllers are made only by Intel, unlike USB 3.0 where there are a variety of manufacturers.
- USB 3.0 ports can replace USB 2.0 ports, saving board space and cost.

FireWire at 400Mbps was way faster than the competition (USB 1.1 at 11Mbps) but it still failed to get market traction because it didn't have broad industry support. That's not to say that FireWire was a failure, because it certainly has done well in specific markets (like video/audio production).

In my opinion, Thunderbolt will be the same. It's faster than USB 3.0, but USB 3.0 is fast enough that most people don't care (both are way faster than any hard drive). It's again used primarily on Apple computers, although (as with FireWire) some other manufacturers might eventually follow suit. It's again not compatible with the sorts of devices people actually have (USB keyboards/mice/headsets/hard drives/printers/cameras/mobile phones/etc.), and again the competition is.

That doesn't mean that Thunderbolt won't carve out a niche as a mini-docking solution primarily for Apple users. But for everyone else, USB 3.0 is the way to go.

 

I'm not going to join this religious war USB vs ThB. But I'd like to mention that you missed an important point in your (possibly correct) assessment. ThB sockets are same as Light Peak - they are independent of the connector tech, fiber or wire/copper. Thus, it will be transparent for peripheral manufacturers when ThB gets upgraded to LP. And most people are pretty sure that it's only a matter of time before fiber optic TX/RX are cheap enough to supplant wire. Thus, it is in the business interest of peripheral vendors to support ThB. The only thing they'll be concerned with is Intel's lockin. If not for that they'd be all over ThB.

 

I read the is a rumor that the Elitebook Workstation models will include Thunderbolt, and USB 3.0. Could be good.

 

I would argue they both have degrees of "future-proof". But it's important to note that while the Macbook Pro can ostensibly have adapters that convert Thunderbolt to USB 3.0 and deliver 5 Gbit/s speeds, there's no way to get 10 Gbit/s Thunderbolt out of the T420s.

I don't really see the two as being in competition actually. USB 3 should coexist with Thunderbolt. In fact, as Thunderbolt switches over to fiber optic, I'd prefer old fashioned USB (even 2.0 and 1.1) for low speed wired peripherals. And of course there's the problem of charging devices, something that fiber optic can't do (by itself).

USB 3 ports will replace USB 2 ports, and I think similarly Thunderbolt ports can replace Displayports (probably miniDisplayport).

Again I don't think the two are in conflict. We'll probably see a bunch of high-end laptops (PCs and Macs) that offer Thunderbolt and USB 3.0. It seems Thunderbolt is really positioned by Intel as an add on to Displayport, and as a carrier for multiple protocols. I think there's really going to be a need to keep USB for the mice and keyboards of the world and charging iPhones and what-not but there's some things that USB 3.0 doesn't do well, like latency or speed. And USB 3.0 doesn't quite do video out either, so there's still going to be a need for a video out port, so why not do a Thunderbolt port as an added benefit?

 

These are all very good points. I would add that TB looks like yet another port for which Apple will provide a plethora of expensive adapters. Technically it offers little more than USB 3.0, but asks a lot more in return.

 

Business oriented notebooks won't add a connector unless it shows it have proven returns. Not everyone is jumping on the Thunderbolt and soon Light Peak boat just yet.

Again as it's been stated before, once USB 3.0 gets mainstreamed, I can see LP market share plummet. In order for it to expand it needs customers. If people don't see a need, the technology will die.

 

Businesses will be all over this connector as will allow them finally to replace all desktops with notebooks at little loss to performance.

This is a general purpose interconnect that can replace all other interconnects allowing for a significant reduction in ports compared to current machines. [This reduces space and manufacturing costs, a win win!]

You might only need a single mini DP port to run all your current peripherals including monitors, USB 3.0 devices and add many more at speeds unimaginable just a few days ago.

Listen to this presentation for a few more insights.

YouTube - Intel Thunderbolt Media Briefing at Intel's Silicon Valley HQ

 

But wasn't USB 3.0 an unproven technology as well? Lenovo was among the first laptop makers to offer USB 3.0. And Displayport was unproven when Lenovo began offering it. They even adopted Displayport before Apple did!

 

While my previous post panned the "future-proof" thing, I did so because I think by the time the technology matures, Ivy Bridge will have everyone interested, making the point moot. Maybe by then the optical implementation of Light Peak will be nearer.

I think the potential advantage of Light Peak is lower CPU usage than USB has traditionally had. Like the adoption of USB however (which took forever, and didn't truly happen until Apple put the Bondi iMac in play in 1998), peripheral adoption will take time. I think it has a future; but like USB, I see that future being built over the next couple generations of processor technology. Perhaps it will be useful by the time the ThinkPad T430 comes to market.

 

USB 3.0 is backwards compatible and older USB will work in a 3.0 port. I was speaking on terms of Light Peak. And I never mentioned anything about being unproven technology, I was talking about $$$. If Lenovo invests 100 million dollars to put a LP connector and nobody uses it, it's a waste of money.

Displayport again is a connector for outputting video just like VGA, not a technology that is replacing the entire serial bus..

 

Usb 3.0 is an open standard and it is backward compatible with usb 2.0 thus a natural successor.

Once again Displayport is also an open standard + major display graphic card is already supporting the standard before Lenovo jump into the bandwagon (if my memory serve me correctly)

Thunderbolt however is a proprietary and the only 1 company has announce product to take advantage of the technology (lacie). Require port for usb compatibility (which hasn't been produced yet). Also it more expensive to implement into the system than any of the two technology you mention.

 

Just so I'm clear, define "proprietary". I ask because here's what I've read so far (taken from Anandtech):

Apple learned its lesson after FireWire licensing slowed adoption - the Thunderbolt port and controller specification are entirely Intel’s. Similarly, there’s no per-port licensing fee or royalty for peripheral manufacturers to use the port or the Thunderbolt controller.

 

You weren't speaking about Thunderbolt, you were speaking about Light Peak? Is that what you were saying?

Well that's what a proven technology is, and an unproven technology is one not shown to have returns. So I was saying what you were saying.

And there's a difference? Displayport was completely new, and the market was really slow to adopt it. it came out in 2006 and products only came out in 2008. And anyways, Thunderbolt is an add-on, so it's not like you need to make room like if there was a whole new connector (like you see with Displayport).

 

Cost isn't known yet. My estimation is that if it was so expensive, Apple would've increased their prices by a slight bit, but they haven't. And if implementing the technology ate into their profit margins by a great deal, I don't think they would have adopted it at this point.

And USB is just as proprietary, in fact, a while ago Intel was being a real pain in the about USB 3.0 specifications, they were refusing to release them to AMD and other companies. So much for open standards.

 

Intel owned everything about thunderbolt. Secondly you still have to buy controller which is made only by Intel at this point. Thirdly Intel could pull a time exclusion deal with apple for tb 1.1 if they wanted.

While usb 3.0 is an open standard, which is rule by USB Implementers Forum (Notable members include Apple Computer, Hewlett-Packard, NEC, Microsoft, Intel, and Agere Systems).

You could get couple thousand dollar licensing and create as many devices as you like as long as you follow their specs.

 

"Obviously within the life expectancy of the device. But that is usually understood in tech circles. We aren't talking about Eniacs here."

No, I'd say real tech circles know that nothing is "future proof" - end of discussion.

At this point any laptop not supporting Thunderbolt should be considered obsolete already. Intel, not by accident, calls this revolutionary technology for laptops.

Although I intend to buy high-performance Sandy bridges, I have a t61p and an extreme that ARE NOT obsolete.

IMO this development is so disruptive that it will soon affect higher end sales of Apple's competitors."

I'm not disrupted, you must be.

Renee

 

Two messages which were unnecessarily argumentative were removed. Keep it chilly, guys.

 

I don’t want to divert the thread but I simply couldn’t let the statement below stand.

No, I'd say real tech circles know that nothing is "future proof" - end of discussion.

Let me elaborate why I consider this statement completely out of touch. The highest priority for an architect/designer/developer is to future proof his creation as much as possible. This is of even more importance in the case of infrastructure products like bus protocols, interconnects and operating systems.

During the feasibility and design phase, future scenarios need to be widely debated in order to future proof the product, i.e. provide the product with a life expectancy that will provide a positive return for as long a period as possible. This is the time where the term “future proofing” is a constant companion in competent tech organizations.

It’s just as important to discuss the time window (generation) for which future proofing should be attempted as future proofing can have significant costs associated with it. This is especially true in the software industry.

I architected and developed software products that lasted more than 20 years in the marketplace and one product is still competitive against major competition (Oracle, etc.) after 30 years. They key was not to be hardware platform and operating system dependent as well as making the correct bets on future technology trends so that processes that were likely to see major evolutions could be isolated and therefore easily adapted to generational change.

To deny that future proofing is essential to product developers as well as consumers is silly. Change happens and there are no guarantees but future proofing an investment is the best way to receive a solid ROI.

It should be obvious that what applies to product developers applies just as much for product consumers. If you pick the right product it might even survive generational change. Future proofing your purchasing decision is simply good management.

 

if this thread can't continue as a discussion rather than an argument then it's getting locked.

for your own sakes, please stop arguing.

if someone takes personal issue with something said then please take it to PM.

thank you.

 

Future proofing may work if the consumers:

1. Know enough of what you are purchasing. Lot of people buy blindly into what marketers tell them to buy, and not really understand what their own needs are.

2. Whether they have enough knowledge of the technology behind the stuffs they are purchasing. Or research into future trends for software or hardware development. But whether or not a person would want to invest that much time and attention to such endeavour would depend on their habit, interests and their opportunity cost.

3. Some people don't need to future proof their technology since they are incentivized to replace their technology every financial year.

4. Future proofing from a perspective of consumer and developers are different so they are not directly compatible.

 

"4. Future proofing from a perspective of consumer and developers are different so they are not directly compatible."

Exactly Lead_org! That was my point. I've always been technical and a feminist which I will address any sexism.

"I architected and developed software products that lasted more than 20 years in the marketplace and one product is still competitive against major competition (Oracle, etc.) after 30 years. They key was not to be hardware platform and operating system dependent as well as making the correct bets on future technology trends so that processes that ..."

Reglli, since you are "technical" you know that hardware evolves faster than software. My statement was not out of touch. We have been speaking of hardware.

"To deny that future proofing is essential to product developers as well as consumers is silly. Change happens and there are no guarantees but future proofing an investment is the best way to receive a solid ROI."

I never said that. You must be talking to someone else. I did say that 'future proofing' is not possible.

Renee

 

So as I'd suspected, SSDs are getting super fast.
YouTube - Macbook Pro 15" 2011 OCZ Vertex 3 benchmark - fastest SSD ever (watch in HD)

about 490 megabytes/second for the Vertex 3, or almost 4 gigabits/second for just one drive. That's practically double what a previous generation drive would get. So maybe in another 1.5 years, we'll see a 1 gigabyte/second drive, which will need 8 gigabits/second, and that would probably max out current Thunderbolt/Light Peak. At which point, I would hope 100 gigabits/second Thunderbolt comes out later on. I see no other peripheral bus standard that would be able to handle such high data rates. USB 3 can do typically 3 gigabits/second, maybe, eSATA is still only based around SATA II, Firewire is going nowhere, 10 gigabit ethernet is going nowhere, etc. So really, I am now seeing Thunderbolt/Light Peak as our only real "future-proof" peripheral bus standard.

Now look, I can see the arguments against, but really, I just want the best technology available right now (by the way, would it be a problem for current Thinkpad users because it uses miniDisplayport and not the full Displayport connector?)

I don't know, I guess in total what you can say is that PC manufacturers were probably left in the dark about Light Peak (pun intended). I read an article in an electronic engineering magazine about Thunderbolt. The critics (speaking anonymously) think Intel should forget about it and instead focus on USB 3.0. The case against Thunderbolt

 

How do I get that gig?

 

just to point out with the sony rumblings are that with light peak would be great with a EXTERNAL GPU for laptops becuase it has the bandwidth which would be cool if it could come as a unified standard with laptops buy 700 $ gpu only laptops now and then buy a 200$ gpu latter

also DisplayPort v1.2 has speed rating up to 17 Gbit/s

but thunderbolt has 10w of power threw it
Dp=.5w
USB=.5-.9 w

 

USB 2.0 is 2.5W (500mA @ 5V). USB 3.0 is 4.5W (900mA @ 5V).

I agree that external GPUs are one of the more interesting possibilities with Thunderbolt. Especially on an ultra-light laptop.

But you still have the problem of needing to lug an external PSU and GPU around. And you need proper driver support, too, or you're left with hacked-up unstable solutions.