LinkBack URL
About LinkBacksRead more.Intel's Thunderbolt claims to be one of the fastest, most versatile PC interfaces ever produced. But how does it work and what does it mean to you?
Read more.Intel's Thunderbolt claims to be one of the fastest, most versatile PC interfaces ever produced. But how does it work and what does it mean to you?
Is this going to be an industry standard or not?? If it is only available for ONLY Intel motherboards then MEH!
Looks like a sensible step forward. I hope it takes off.
I only hope it takes of if it's an open standard, as CAT says. If it is then it does look pretty impressive, but I don't see the need to have video pointlessly transmitted to your flash drive, I assume it only uses data/video/both when needed. I like how they market it as being able to back up a Blu-Ray film in under 30 seconds, what storage device can match that? It's no worse than USB2.0's quoted 480Mbps bandwidth though...
I was very excited about light peak aka 'Thunderbolt' until this launch which now makes it look gimmicky and niche. Another 1up for apple marketing.
I hope the AMD 990FX chipset has native USB3 support. Light peak is technically very promising, but then, so was firewire.
The article says it all really:Originally Posted by watercooled
So Light Peak Thunderbolt becomes OS and hardware agnostic. There won't be any complaints about Linux driver compatibility, Windows having a buggy software stack or Mac OS having a non-complete implementation. It just works.And, most importantly, by using these existing protocols, devices connected via Thunderbolt will be recognised by existing operating system drivers.
It also makes sense to use an existing connector and set of cables if it can. You already get DisplayPort to DVI/HDMI/VGA adapters and these will continue to work in the Thunderbolt port, more adapters for USB, Firewire, eSATA, Ethernet, etc. will just be a logical extension to this.
Forget about that, Intel are encouraging you to take part in illegal activities!!!111!!! The MPAA should sue them!!
If it is an open protocol, I wonder if we will see AMD graphics cards with six Thunderbolt ports instead of plain mini DisplayPorts in the future?
Last edited by Funkstar; 24-02-2011 at 10:05 PM.
watercooled (24-02-2011)
Ahhhh that's a good point - so they're using existing (and standard) software-layer protocols, just over a different medium. Does the PCI-E protocol recognise things such as mass-storage devices though? I'm currently thinking of those SSD cards so I might have answered my own question - but what about hot-swap-ability?
(btw, i added more to the post in an edit
I believe PCIe does support hot swapping, this is mainly for servers, but it is there.
I don't think it supports "mass storage" technically, but then, just put an already existing PCIe -> USB/SATA chip behind the Thunderbold chip and you have something capable of mass storage.
You need to think about this a little differently to USB or Firewire. It is only a new transmission medium (with a meta-protocol layer in there as well) not a new connection method like USB was when it was introduced. When you plug a mini-DP cable into a new Mac Book Pro, and hook a Thunderbolt chip to the other end, you have just created a 3m extension for the laptops PCIe bus and DisplayPort output nothing more. It's this with knobs on: http://uk.startech.com/product/PEX2P...-Expansion-Bay (and much more elegant)
Maybe this is why Intel haven't added USB 3.0 support to their chipsets. In theory, this should be even better, minus the backwards-compatibility of course, but our motherboards still have IDE ports so that's not really something to worry about...
I remember when USB was first publicised the argument was that there were too many connection types (Serial, Parallel, IDE, E-IDE) and USB would replace them all... then came Firewire... then ESATA... then HDMI and display port... etc...
So why should this be any different?! I expect within a few months there will be competitors touting their own 'standard' super-connector!
It never ends, companies/people are too greedy to all adopt one standard and stick with it...
eSATA only came along because USB basically stopped progressing after about 2000. Eight years later they announced USB 3, and it's taken another couple of years for USB3 ports to even start appearing on motherboards. If USB3 had been developed and released four years earlier, when it was needed, eSATA would have been thoroughly pointless - having less speed, less features (i.e. no powered bus), and less compatibility. USB should have replaced all other external data cables.
Video cables is another matter entirely.
eSATA only really came about because companies started releasing external cases with regular SATA connections which were never designed to be external. Adding a latch helped, but that's a pain sometimes. So they came up with an eSATA connector, but it is just a connector, it is identical an every other way to regular SATA. There will soon be a powered version, but it's always going to be a minority interface.
Every other external interface until Thunderbolt has required software and OS support, and none of them has ever been as comprehensive as this. USB was intended as a serial and parallel replacement and introduced plug and play. Firewire was fantastic, but its high royalties meant USB became mainstream much quicker and grew much faster. Which is a shame, Firewire was an awesome interface.
I hope Intel don't go OTT when it comes to royalties. I would like to see this succeed.
I don't understand what was wrong with SATA as an external interfaceGreat, they changed the connector. What does this achieve except annoy everyone who now has to go and buy more cables?
eSATA annoys me mostly for this reason, I'd be interested to know if there was actually some rationale behind doing it.
I've spent some time with a knife trying to persuade a SATA cable to fit in an eSATA port in the knowledge that they are basically the same thing. I failed.
SATA wasn't really designed as an external interface. When you attach a drive, your system sees it as an internal drive so it enables write-caching by default on Windows which means you risk losing data if you yank the disk. And unless you run the SATA controller in AHCI mode, there will be no safely remove button. On Linux, you could just unmount the drive like any other but there's no immediately obvious way to do it on Windows.
Surely it would have been far easier to issue a patch that allowed safe removal of 'internal' drives? System drive excepted, of course. Plug and play, well, meh TBH. Presumably has to be fixed (i.e. AHCI) for eSATA same as it would for SATA. Granted, write-caching.
you have to be running it in AHCI mode for it to be made removable at all.
don't see what the issue here is....
throw new ArgumentException (String, String, Exception)
There are currently 1 users browsing this thread. (0 members and 1 guests)
Posting Permissions
Reply With Quote
