I don't care much for speeds and feeds, but I care a lot about reliability and subsystem dis-integration.
When I get a cpu / memory 'box', that requires a Thunderbolt connection to a 'video card / cards' box, and a connection to a mass storage 'box', so that I can mix and match and replace/upgrade as needed, I'll care about Thunderbolt.
Yeah I'm with you. I honestly couldn't give a rats about thunderbolt. The only devices that I would personally see benefit from is External Hard Drives, and lets face it... USB 3 isn't exactly holding back mechanical drives much. (Except for exotic raid solutions.)
And eSATA is superior to USB 3.0 and equivalent to Thunderbolt in speed. It's also EXTREMELY cheap to implement, just costing the price of one USB/eSATA receptacle. Every chipset has at least 4 SATA ports, only 2 are usually used in most laptops (HDD, ODD). All you need to do for eSATA is route one of the SATA ports to the receptacle and you're done.
For eSATA hard drives CPU->PCIe->SATA controller->HDD
For USB hard drives CPU->PCIe->USB 3.0 Controller->SATA Controller->HDD
For Thunderbolt CPU->PCIe->SATA controller->HDD
I believe the thunderbolt multiplexing chips do not affect performance
eSATA itself is nearly useless. The two sticking points are that it's a port dedicated to nothing but external HDDs, and that it has no power and so you end up needing to plug your HDD into a wall socket or USB port anyhow. From a consumer standpoint, USB3 wins that by default because consumers don't want to carry around a power brick when they could just use USB3 and plug the drive in right there.
eSATAp (a combination USB and eSATA) port helps, and provides some power, but has issues the same. The USB standards body refuses to recognize them as valid USB ports, the eSATA part of the expansion is useless for anything but disks, and it doesn't provide enough power for 3.5" mobile drives (since it only does 2.5W). It also doesn't really jive with storage arrays, which want to expose something higher-level than SATA if they're doing hardware RAID.
USB3 helps: you get enough bandwidth for disks, but also for other high-bandwidth devices, and the 5W it provides is an improvement. It still can't do displays very well (USB displays are limited), but it's an improvement.
Thunderbolt improves more. 10W of power should be enough to power a desktop drive without external power, the throughput is plenty for anything we might want to do today, it can provide high throughput to more than just HDDs, it can do displays natively, it can daisy-chain which enables single-cable dock type things (like the Apple Cinema Display which you connect one cable and it gives you GigE, USB, display, audio, etc)...
There are also more than just HDDs that need high bandwidth connectivity too. Video capture for instance. Uncompressed 1080p60 is ~3 gigabit, or if you're doing 10-bit per channel, 3.7 gigabit. Uncompressed 4K at 60FPS like the Red One records at 10-bit is something like 17 gigabit. Thunderbolt can handle that. USB3 can't. There's also an element of forward-thinking here: it's better that you have enough throughput before you need it, rather than waiting for it to catch up.
Another thing is 10GigE. It needs a good chunk more bandwidth than USB3 can provide, and copper 10GigE PCIe cards have been on the market for years. The prices are still outside of consumer market levels, but they're coming down like anything does. Right now you can find Intel 10GBase-T PCIe cards for $400, and a 100ft Cat6a cable costs only $28 on monoprice. I'm sure other people can come up with more examples of high bandwidth things other than disk drives.
eSATA is a disgrace. What moron introduced a powerless I/O port in 2004?
And we keep seeing the same dumb comments about Thunderbolt: "Duh, no drives are this fast, so why do we need this?" Of course, that completely ignores the fact that you can daisy-chain several drives and a display off Thunderbolt at the same time.
Oh, and run USB and Ethernet over it as well.
People cheerlead AGAINST better technology now. It's pathetic.
"When I get a cpu / memory 'box', that requires a Thunderbolt connection to a 'video card / cards' box, and a connection to a mass storage 'box', so that I can mix and match and replace/upgrade as needed, I'll care about Thunderbolt."
Oh don't waste everyone's time. Out in the real world, massive numbers of laptops are sold every year, and Thunderbolt targets them. Obviously a system that has 6 PCIe slots doesn't need a PCIe-over-cable solution.
The idea is to converge your gaming computer and your low power om the go ultra-portable system. We've sort of had this for years with business class laptops and docking stations which made connecting your external mouse/keyboard/monitors/power a single action, but we were still stuck with whatever crappy IGP was in the laptop itself.
Thunderbolt will combine everything except the power cable into a single plug while also giving support for an external GPU. A docking station could use thunderbolt for part of its interconnect combining everything into one item; while the reduced engineering costs needed to design it should be able to push the sticker price down.
"Not surprisingly, the RAID 0 striped results were roughly 2x faster than those either accessing a single drive at a time (JBOD and concatenated) or both drives simultaneously but in a mirrored fashion (RAID 1). "
Judging from the table data, you meant to say 1x faster or twice as fast. Great review though.
USB has over two decades of backwards compatibility, has no patents controlling it, the royalties are optional and it's fast enough for 90% of tasks it's used for today. Thunderbolt may be faster, but it costs more in every aspect, has EXTREMELY low adoption rate across the board and still can't manage the "one cable for everything" idea Apple originally helped create it for because it uses copper wiring instead of fiber optics.
The high cost of Thunderbolt devices compared to mainstream consumer gear has nothing to do with patents or royalties. It's actually insanely cheap compared to any other 20 or more Gbps, full-duplex over a single cable technology. There are plenty of people out there for whom having this type of technology is a serious aid to their workflow.
Point to another first generation I/O technology that had a higher adoption rate in its first 12 months.
Apple never espoused a "one cable for everything" idea, as is clearly evidenced by the fact that the Thunderbolt ports they added to the 2011 Macs are right next to... well, a bunch of other ports, including USB.
How exactly would Thunderbolt be better if it used fiber optics? For the short cable runs most users typically require, copper is still king. You can use a fiber optic cable with current Thunderbolt controllers if you like, it'll just cost more.
If the performance gain you get from a Thunderbolt enabled workflow isn't worth the couple hundred dollars extra that it would cost you, then don't shell out for it. But at least recognize that for some people, it's well worth it.
Just because you don't use something yourself, does not make it a failure. I've never personally needed to drive a semi-trailer truck, but I'm sure glad others do every day, and I would hardly categorize the entire class of vehicles as a failure. You may not realize how ubiquitous 1394 is, but rest assured, it continues to be quite useful to many folks out there.
1) seagate thunderbolt 2.5 inch adapter apple cable and a 256gb ssd if sata 3 any larger has stability problems. I use a mushkin 240gb and a crucial m4 256gb fast stable use as os drive with mac minis. cost if you got the deal with the crucials at 199 the cost is 350 for a stable external boot drive.
2) next a lacie little big disk pull the 500gb drives and put in a pair of samsung series 810 ssds 256gb. in fact I am typing on that setup right now. cost 400 for the little big disk 50 cable and 600 for the ssds. 1050 and I sold the spare 500gb hdds. for 150 net is 900. pull the insane fan sounds like a effing mosquito on steroids. this was 900.
now you could build this for less if you got the 256gb crucials at 200. this would be 850. this is an outstanding os drive. rock steady in raid0
3) last buy the pegasus r6 use the software to run 2 raid0 setups. one with a raid0 for lion one with a raid0 for boot camp the other 2 drives are backup use it with a mac mini server or an imac cost is 1500 plus the ssds 4 of the crucials would have been 2300.
these units tested today are pretty much not getting what t-bolt is good for external ssd boot drive along with a raid0 for storage. they all run hot and they may not be stable. (time will tell) any of the setups I mention work and work well. they have ended the iMac my internal drive is trapped problem. Now usb3 would be almost as good for the 3 above it would only lag a lot to the pegasus.
Hi StormyParis, I mention CPU load to some degree on the 'Software RAID' page, related to the Thunderbolt Duo. Suffice it to say that in RAID 0 mode, the incremental CPU load specifically due to software RAID was miniscule (a few percentage points of one of the four 'cores'...two physical cores, further doubled to 'four' courtesy of HyperThreading support in the CPU), and in non-RAID modes for both the Seagate and WD drives, it was imperceptible for Thunderbolt alone.
Thanks for the feedback, isulzer, but the MacBook Pro SSD (which I was able to successfully run AJA on) is also that particular system's boot drive. Is the particular O/S part of the reason for the issue? The MBP runs Snow Leopard, the Mac mini runs Lion
In lion and mountain lion, permissions on the boot drive are slightly different. Only System has write access. Everyone else has read only. So if you try to write a file to \ in Lion, you will have to authenticate.
One thing that would be incredibly useful is direct PC to PC connection with USB 3.0 networking. Anybody else down for 4.8 gbps transfers with a cheap $15 cable? Apparently, all you need a USB 3.0 crossover cable and each computer motherboard with capable drivers. Anandtech, can you guys take a look at this? Some things in this industry just boggle me, such as a lack of an obvious USB 3.0 networking solution... Thanks!
WHY does Thunderbolt need to carry DisplayPort AND PCIe. Wouldn't it have been easier to add pins to a displayport receptacle or create a new port for the PCIe data? Why do we need these multiplexing chips in each device when we could have just used straight PCIe? Even better, integrate the PCIe lanes right into a USB receptacle. It's a data bus, should it not be in a data port?
If you want one cable to a docking station/display, they could have easily mandated that one Thunderbolt port must be right next to a displayport port.
Apple was already shipping Macs with mini-DP connectors that were capable of pushing 17.28 Gbps worth of packetized data over a single cable. It's only common sense to look at that and say, "Wait a minute, why just use this for display data? Why not make it full-duplex and use it for transporting PCIe packets as well?"
Digital display interfaces are the only ports commonly found on PC's that are even in the ballpark of Thunderbolt from a bandwidth perspective, so the marriage of the two is actually fairly logical.
Mini DisplayPort connectors already pack 20 pins in a 33 mm^2 cross section. That's pretty dense, there's not really room for adding more pins. Plus, by keeping the same physical connector as mini-DP, it's easier to maintain backwards compatibility with existing DisplayPort gear.
Besides, adding more pins to the connector would mean adding more conductors to the cable, and Thunderbolt cables are already complex and expensive enough as is.
So you'd rather have two cables hanging out of your computer than one? That's a silly argument. Why not have three or five then?
I don't get all the hate for thunderbolt. When intel first demoed it everyone was raving about how great it would be, but once it came on macs first everyone said how dumb and stupid it was. At least pcs are going to get it now so we can get more devices for it and then it can suddenly become "cool" again.
Actually, if I remember correctly, Intel tried to get LightPeak (as it was called while under development) integrated into a next generation USB port, but the consortium told them to take a hike - patents/royalties and customer confusion being named as primary issues with that implementation. That's when Apple stepped in and offered their mDP design - Apple was one of the originators, but it's a royalty free design as far as I've heard - part of the reason why AMD uses it for Eyefinity. Intel develops the tech, Apple offers the plug design, and in turn, gets first dibs on it.
Sony *did* implement a variant of TB/LightPeak on its Vaio Z (the one that has the external 6630m) that utilizes a USB port, but that's not standard.
By "the consortium" I reckon you're referring to the USB-IF, which was actually formed by none other than Intel. I know that the USB-IF and Intel both frowned on Sony's non-standard implementation, despite the fact that it closely resembles a lot of the early demonstration hardware, but did Intel ever actually try to convince the group to sanction the adoption of a USB style connector for Light Peak?
Also, DisplayPort is a royalty free specification maintained by VESA. Apple developed the mDP connector and agreed to license it for free, and VESA subsequently included it in the DP 1.2 specification. This all happened over a year prior to Thunderbolt arriving on the scene.
I'm pretty sure Apple got first dibs on Thunderbolt because they agreed to buy 18 million additional chips from Intel in 2011 if they got an exclusive on it.
Nice review and flyby of the current Thunderbolt market... The overuse of parenthesis on the first couple of pages was odd tho (almost like some were added by an editor?).
Anandtech seems pretty bullish on the virtues of Thunderbolt in general but outside of laptop/docking station scenarios I still see very little upside to it, and that prospect alone (plus the super high adoption cost) won't help propel the interface to wide adoption anytime soon.
It's a shame too because some of the long term applications that have yet to materialize are very appealing, things like external GPUs, etc. If costs don't drop fast enough it's gonna be Firewire all over again, hopefully Intel stands behind it long enough for that not to happen.
Just like with WiFi, please for the love of god explain this stuff correctly. The issue is not that this is a "theoretical" or a "peak" bandwidth, it is that this is the PHY bandwidth, that is, it is the speed at which bits (all the bits) are dumped onto the wire (or into the air). So why doesn't that match the actual goodput (ie the throughput of my file bits)?
The minor factor is that there is a variety of overhead that is required to correctly frame and describe each packet. There are bits that describe the purpose of the packet, its length, its source and destination, etc. In the case of WiFi (I don't know about USB2, but I expect these are necessary at the speeds of USB3) there are bits that allow the receiver to track how the signal is degraded as it travels and thereby compensate, and so on and so on.
The major factor is that these are not point to point links, they are shared, which means that some sort of protocol (a MAC) is required to decide who gets to talk at any given time, and how to recover when a mistake is made and two parties talk at the same time. Unfortunately the general rule for MACs is usually that people seem far less interested in working hard to make them efficient than they are in working hard to make the PHY efficient. So what one sees is that an awful lot of time is wasted in MAC overhead.
The historical pattern for WiFi has been that each release includes various new smarts that can allow the the MAC overhead to drop to about 25% (ie actual goodput under optimal conditions is around 75% of the PHY rate), but that the chip vendors actually implement only enough of these smarts (which are always labelled optional in the spec) to get this goodput to a little over 50%.
The historical pattern for USB, which has stood still for so long, is that at first we had the same sort of situation --- crappy MAC implementations that gave us a real world throughput of about 50% --- but, with nothing else to do, the chip vendors ramped up their chips to use every smart possible in the MAC, given us the current goodput under optimal conditions of about 65%. Point is --- the correct way to describe this is that USB2 has a PHY rate of 480Mbps, and an maximum goodput (taking into account both overhead bits and the MAC) of around 320Mbps.
The same is going to be just as true of something like Thunderbolt, when we get to the point of stressing it hard. There is a PHY rate, there are overhead bits for addressing, framing, packet description, training, etc; and there is some sort of MAC. I know nothing about the Thunderbolt MAC, and so have no opinion about whether it's an inspired or a crappy design (or the extent to which chips can make more or less efficient use of it). These would all be good topics for a future article.
<blockquote>Granted, Thunderbolt's 8b/10b encoding scheme reduced the effective peak bandwidth to 8 Gbps...</blockquote>
Thunderbolt provides a full 10 Gbps per channel to the upper layers—there's no 8b/10b in that figure. Anand achieved just over 8 Gbps of throughput with the Pegasus R6. Taking in to account other sources of PCIe overhead, such as packet framing and link maintenance, that number is about spot-on for a 10 Gbps link. Also, the PCIe data may be transported on more than one channel of a Thunderbolt cable, however the current controllers are limited to 10 Gbps of PCIe throughput because they only contain a single PCIe to Thunderbolt protocol adapter.
<blockquote>...perhaps, alternatively, the HDD is getting a portion of its power allocation supplied directly over the Thunderbolt link from the connected computer system.</blockquote>
I doubt either of these devices utilize bus power to a significant degree, because according to Apple, only the first device in a Thunderbolt chain can be bus powered.
<blockquote>Such a feature allows the system-side Thunderbolt controller to "see" each Thunderbolt Duo as a single bus peripheral, thereby enabling WD to accurately claim that you can daisy-chain numerous drives to each other and the system before you violate Apple's six-max specifications.</blockquote>
The 6 device limit only pertains to the number of Thunderbolt controllers that can be in a chain. You can hang as many PCIe attached devices as you like off of a single Thunderbolt controller. A good example of this is the Apple Thunderbolt Display—in System Information it shows up as many separate devices, but it still only counts as one Thunderbolt device.
As far as Thunderbolt product pricing goes, I think we're just looking at a scary BOM cost, which will hopefully be alleviated some with the arrival of Cactus and Port Ridge controllers.
And you can use the GoFlex desk adapter with any old drive, so I'm surprised you didn't give that a whirl.
"Given that the PCI Express channel of the Thunderbolt interface delivers 10 Gbps of peak bandwidth in either direction, I next decided to see what would happen if I tethered the Seagate and WD drives together."
You can do even better than this. You should have been able, in Disk Utility, to create a striped drive consisting of the two WD drives and the Seagate drive. This would be gated by the speed of the slowest drive, so should give you around 360MB/s read and write speeds.
Disk Utility is not perfect, and god knows I've sent Apple long lists of ways it can be improved; but it is pretty awesome for basic things like creating either striped arrays or large concatenated arrays.
Dear name99, that's a great idea. I'm traveling at the moment but will give it a shot when I get back in front of the Mac mini. I agree with you that it conceptually should work, and will be performance-gated by the drives in the Thunderbolt Duo.
You talked a bit about the impact of running a soft-RAID device being negligible, it would be nice to know the impact on the OS when pushing those mass storage devices to the limit.
I've bought an USB3 extension card for my mac and while transfer speed are great, OS responsiveness takes a hit every time a big transfer occurs (TM backups are a pain).
A formal test of the impact of USB 3, ThunderBolt would be nice.
Have you watched Activity Monitor during a TM backup to see what's being stressed? I usually associate the performance drop during TM backups with the crazy amount of small random reads that need to be performed on the system volume while you're still trying to use it.
So basically, if you're really concerned about performance, an SSD using an old 3 Gbps connection will read and write twice as fast as a HDD using a Thunderbolt connection?
I would be interested in knowing what the results would be with a HDD using an eSATA connection. And of course USB 3.0.
Each connector of the thunderbolt cable has a small chip, which is why the cable costs so much. It's not just a $50 cable. iFixit has done a tear down of the cable if you want to check it out.
There are quite a few factors which make Thunderbolt cables expensive, being an active design is just one of them. If you ignore the bit about Thunderbolt being intended for iOS devices, the following blog post reveals quite a few details:
I think the snake oil salesmen that prey on the audiophiles will avoid Thunderbolt for some time, specifically because it is an expensive to manufacture, high bandwidth cable that can't be made for $3. A 2.0 m copper pipe that can carry 20 Gbps, full-duplex, plus bus power and out-of-band signaling just doesn't exist yet in the sub $20 realm.
Just like on the MAc centric forums, people can not look past their narrow usage scenario to accommodate other perspectives. I work at a TV station, but have a background in audio production. In most prosumer studios, and quite a few mid level studios, firewire is the connection of choice for audio interface (soundcard). The high end audio interfaces have until recently all been either a PCIe card or Firewire. Add to the pile firewire based DSP such as the UAD-2 and firewire disk arrays used to insure fault tolerant and fast nearline storage. In video production we really don't use 1394 anymore. But there are still some old boxes in use that have the mini connector. Often they get pressed into use as a transcoder or layback machine for formats we dont use natively anymore. We are all on Sony XDcam disk, which I would love to see as a consumer format! I personally can't wait to see wider adoption of thunderbolt. I plan on upgrading to a T-bolt equipped mac after the next refresh of the MBP line.
This sentence is a bit misleading: "a specific RPM at any particular time is dependent on both access requirements and sensed operating temperature." - This would suggest that this HDD runs with varying RPMs. This is not true. One HDD runs with one RPM. However, between the various HDDs among one charge there can be different RPM settings. So one Caviar Green "IntelliPower" runs with 5400 while the next runs with 5900. You will not see one 5400 unit suddenly run at 5900 because the temperatures are particularly low.
Dear Death666Angel, I don't think you're correct. WD has told me on many occasions that the IntelliPower algorithm varies a drive's rotational speed over time as a function of performance demand and operating temperature. And past AnandTech writeups make similar statements:
The real advantage of Thunderbolt should be that it carries power, but special this feature is very limited, as the power pin on the connector is very small and shall carry 10Watts, Sata connectors, USB and other have bigger contact surfaces and/or multiple pins to carry power. The display port adapter was designed for a lot less power then 10watts, that's the current issue with Thunderbolt today. Some details can be found here: http://technology.coolodman.com When a self powered thunderbolt device is used it works fine and uses only about 22% overhead to encapsulate the PCIe bus, so the speed with fast SSD's like Samsung 830 series is really not bad.
Two months after this article was written, I find Thunderbolt options at the same phase, quite limited. So I bought the Seagate GoFlex Thunderbolt adapter for $$$ from B&H, plus the cable. Although expensive, it seems for the time being the least expensive approach, albeit a DIY approach, to use raw hard drives, of which I have an expanding collection for my video and audio projects.
If my research is correct, there are no official generic Tbolt docks, unlike the common USB3 (and earlier interfaces) docks on Tiger Direct. Although LaCie 2big bays appear to physically allow users to swap raw hard drives, LaCie voids the warranty in doing so. Obviously LaCie discourages a DIY approach and of course favors consumers to procure LaCie's aftermarket harddrive-installed bays. This is less of a diehard obstacle as opposed to WD's Caviar requirement. Conversely, I'm not into opening up enclosures to swap drives. It requires tools, is time-consuming, and is semi-permanent. My preference is to swap raw drives on-the-fly.
At least the GoFlex DIY hacker method works, and I secure the drive on the dock via rubber band technology. I report 'so far so good' with raw drives. I don't know precise speed details as long as I'm able to put the Tbolt bus to use, and it appears faster than any other bus I've used with FCPX. Too many companies try to hijack Tbolt by instilling proprietary constraints. Nevertheless, I find the GoFlex design with its generic SATA interface to be an extremely niche approach because most consumers won't want to seat a raw hard drive somewhat precariously in a dock not designed for such use.
Why are people compairing the two. Yes USB is great if you are conneting things like HHD, Mice, Tablets, and the like, but remember you only have 5Gb in a single channel, one way maximum to play with, while thunderbolt has 20Gb when using both channels. True neither USB 3 nor Thunderbolt would be maxed out by a single HDD, but when connecting several data hungry devices at once, (which with thunderbolt can include monitors, and possible legacy port hubs, if they ever arrive) then USB will show the strain a lot quicker than thunderbolt.
Hence why you need a thunderbolt port? Well if all you are doing is plugging in one HHD over it it is probably overkill, BUT, if you like to have one of those new shiny Ultra books to use on the run but then want a fully fledged PC when you get home, Thunderbolt gives you this potential, but only having to connect 1 cable, rather than 3 or 4. Plus with it being compatible with PCi then you can have expansion cards to further enhance performance with out the need for a completely separate PC. Now before you complain ultra books have low spec CPUs, go and play with one that uses a SSD and see how much speed a traditional HDD sucks up while the system waits for data. As an example, my 2011 Mac book air has a 1.7 GHz dual core i5 with integrated graphic and only 4GB of memory, and on day to day tasks, is faster than my 2011 mac min server with a 2.0Ghz Quad core i7 with the same integrated graphics but 16 GB of RAM. In fact my macbook only struggles when the RAM is all used up, which is not very often .
I guess this is exactly what Dan Neely meant. Good on you Dan, I guess the USB people have not quite grasped the fundamental difference yet.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
46 Comments
Back to Article
rruscio - Sunday, May 13, 2012 - link
I don't care much for speeds and feeds, but I care a lot about reliability and subsystem dis-integration.When I get a cpu / memory 'box', that requires a Thunderbolt connection to a 'video card / cards' box, and a connection to a mass storage 'box', so that I can mix and match and replace/upgrade as needed, I'll care about Thunderbolt.
Till then, thanks, but really ...
rr
StevoLincolnite - Sunday, May 13, 2012 - link
Yeah I'm with you. I honestly couldn't give a rats about thunderbolt.The only devices that I would personally see benefit from is External Hard Drives, and lets face it... USB 3 isn't exactly holding back mechanical drives much. (Except for exotic raid solutions.)
peterfares - Monday, May 14, 2012 - link
And eSATA is superior to USB 3.0 and equivalent to Thunderbolt in speed. It's also EXTREMELY cheap to implement, just costing the price of one USB/eSATA receptacle. Every chipset has at least 4 SATA ports, only 2 are usually used in most laptops (HDD, ODD). All you need to do for eSATA is route one of the SATA ports to the receptacle and you're done.For eSATA hard drives
CPU->PCIe->SATA controller->HDD
For USB hard drives
CPU->PCIe->USB 3.0 Controller->SATA Controller->HDD
For Thunderbolt
CPU->PCIe->SATA controller->HDD
I believe the thunderbolt multiplexing chips do not affect performance
Guspaz - Monday, May 14, 2012 - link
eSATA itself is nearly useless. The two sticking points are that it's a port dedicated to nothing but external HDDs, and that it has no power and so you end up needing to plug your HDD into a wall socket or USB port anyhow. From a consumer standpoint, USB3 wins that by default because consumers don't want to carry around a power brick when they could just use USB3 and plug the drive in right there.eSATAp (a combination USB and eSATA) port helps, and provides some power, but has issues the same. The USB standards body refuses to recognize them as valid USB ports, the eSATA part of the expansion is useless for anything but disks, and it doesn't provide enough power for 3.5" mobile drives (since it only does 2.5W). It also doesn't really jive with storage arrays, which want to expose something higher-level than SATA if they're doing hardware RAID.
USB3 helps: you get enough bandwidth for disks, but also for other high-bandwidth devices, and the 5W it provides is an improvement. It still can't do displays very well (USB displays are limited), but it's an improvement.
Thunderbolt improves more. 10W of power should be enough to power a desktop drive without external power, the throughput is plenty for anything we might want to do today, it can provide high throughput to more than just HDDs, it can do displays natively, it can daisy-chain which enables single-cable dock type things (like the Apple Cinema Display which you connect one cable and it gives you GigE, USB, display, audio, etc)...
There are also more than just HDDs that need high bandwidth connectivity too. Video capture for instance. Uncompressed 1080p60 is ~3 gigabit, or if you're doing 10-bit per channel, 3.7 gigabit. Uncompressed 4K at 60FPS like the Red One records at 10-bit is something like 17 gigabit. Thunderbolt can handle that. USB3 can't. There's also an element of forward-thinking here: it's better that you have enough throughput before you need it, rather than waiting for it to catch up.
Another thing is 10GigE. It needs a good chunk more bandwidth than USB3 can provide, and copper 10GigE PCIe cards have been on the market for years. The prices are still outside of consumer market levels, but they're coming down like anything does. Right now you can find Intel 10GBase-T PCIe cards for $400, and a 100ft Cat6a cable costs only $28 on monoprice. I'm sure other people can come up with more examples of high bandwidth things other than disk drives.
MobiusStrip - Wednesday, May 16, 2012 - link
eSATA is a disgrace. What moron introduced a powerless I/O port in 2004?And we keep seeing the same dumb comments about Thunderbolt: "Duh, no drives are this fast, so why do we need this?" Of course, that completely ignores the fact that you can daisy-chain several drives and a display off Thunderbolt at the same time.
Oh, and run USB and Ethernet over it as well.
People cheerlead AGAINST better technology now. It's pathetic.
peterfares - Monday, May 14, 2012 - link
"When I get a cpu / memory 'box', that requires a Thunderbolt connection to a 'video card / cards' box, and a connection to a mass storage 'box', so that I can mix and match and replace/upgrade as needed, I'll care about Thunderbolt."So kind of like a desktop PC?
Kiste - Monday, May 14, 2012 - link
Eh, you should try to get into the Mac user mindset here... it may not be rational, but it is what it is.name99 - Monday, May 14, 2012 - link
Oh don't waste everyone's time.Out in the real world, massive numbers of laptops are sold every year, and Thunderbolt targets them. Obviously a system that has 6 PCIe slots doesn't need a PCIe-over-cable solution.
Kiste - Monday, May 14, 2012 - link
So it would make more sense to haul around a CPU-box, a GPU-box and a storage-box that you can mix and match with $50 cables?That's mobile computing?
DanNeely - Monday, May 14, 2012 - link
Are you being deliberately obtuse?The idea is to converge your gaming computer and your low power om the go ultra-portable system. We've sort of had this for years with business class laptops and docking stations which made connecting your external mouse/keyboard/monitors/power a single action, but we were still stuck with whatever crappy IGP was in the laptop itself.
Thunderbolt will combine everything except the power cable into a single plug while also giving support for an external GPU. A docking station could use thunderbolt for part of its interconnect combining everything into one item; while the reduced engineering costs needed to design it should be able to push the sticker price down.
Grizzlebee - Sunday, May 13, 2012 - link
"Not surprisingly, the RAID 0 striped results were roughly 2x faster than those either accessing a single drive at a time (JBOD and concatenated) or both drives simultaneously but in a mirrored fashion (RAID 1). "Judging from the table data, you meant to say 1x faster or twice as fast. Great review though.
bdipert - Sunday, May 13, 2012 - link
Good point, Grizzlebee, I'll tweak the text. Thanks for the kudos!Articuno - Sunday, May 13, 2012 - link
USB has over two decades of backwards compatibility, has no patents controlling it, the royalties are optional and it's fast enough for 90% of tasks it's used for today. Thunderbolt may be faster, but it costs more in every aspect, has EXTREMELY low adoption rate across the board and still can't manage the "one cable for everything" idea Apple originally helped create it for because it uses copper wiring instead of fiber optics.repoman27 - Monday, May 14, 2012 - link
The high cost of Thunderbolt devices compared to mainstream consumer gear has nothing to do with patents or royalties. It's actually insanely cheap compared to any other 20 or more Gbps, full-duplex over a single cable technology. There are plenty of people out there for whom having this type of technology is a serious aid to their workflow.Point to another first generation I/O technology that had a higher adoption rate in its first 12 months.
Apple never espoused a "one cable for everything" idea, as is clearly evidenced by the fact that the Thunderbolt ports they added to the 2011 Macs are right next to... well, a bunch of other ports, including USB.
How exactly would Thunderbolt be better if it used fiber optics? For the short cable runs most users typically require, copper is still king. You can use a fiber optic cable with current Thunderbolt controllers if you like, it'll just cost more.
If the performance gain you get from a Thunderbolt enabled workflow isn't worth the couple hundred dollars extra that it would cost you, then don't shell out for it. But at least recognize that for some people, it's well worth it.
Just because you don't use something yourself, does not make it a failure. I've never personally needed to drive a semi-trailer truck, but I'm sure glad others do every day, and I would hardly categorize the entire class of vehicles as a failure. You may not realize how ubiquitous 1394 is, but rest assured, it continues to be quite useful to many folks out there.
philipma1957 - Sunday, May 13, 2012 - link
If you own a mac you have 3 options.1) seagate thunderbolt 2.5 inch adapter apple cable and a 256gb ssd if sata 3 any larger has stability problems. I use a mushkin 240gb and a crucial m4 256gb fast stable use as os drive with mac minis. cost if you got the deal with the crucials at 199 the cost is 350 for a stable external boot drive.
2) next a lacie little big disk pull the 500gb drives and put in a pair of samsung series 810 ssds 256gb. in fact I am typing on that setup right now. cost 400 for the little big disk 50 cable and 600 for the ssds. 1050 and I sold the spare 500gb hdds. for 150 net is 900. pull the insane fan sounds like a effing mosquito on steroids. this was 900.
now you could build this for less if you got the 256gb crucials at 200. this would be 850. this is an outstanding os drive. rock steady in raid0
3) last buy the pegasus r6 use the software to run 2 raid0 setups. one with a raid0 for lion one with a raid0 for boot camp the other 2 drives are backup use it with a mac mini server or an imac cost is 1500 plus the ssds 4 of the crucials would have been 2300.
these units tested today are pretty much not getting what t-bolt is good for external ssd boot drive along with a raid0 for storage. they all run hot and they may not be stable. (time will tell) any of the setups I mention work and work well. they have ended the iMac my internal drive is trapped problem. Now usb3 would be almost as good for the 3 above it would only lag a lot to the pegasus.
StormyParis - Sunday, May 13, 2012 - link
Can we please have the CPU load info? I remember my PC becoming so jerky as to be unusable during heavy USB2 I/O in the early days.bdipert - Sunday, May 13, 2012 - link
Hi StormyParis, I mention CPU load to some degree on the 'Software RAID' page, related to the Thunderbolt Duo. Suffice it to say that in RAID 0 mode, the incremental CPU load specifically due to software RAID was miniscule (a few percentage points of one of the four 'cores'...two physical cores, further doubled to 'four' courtesy of HyperThreading support in the CPU), and in non-RAID modes for both the Seagate and WD drives, it was imperceptible for Thunderbolt alone.isulzer - Monday, May 14, 2012 - link
Its a permissions issue.The app cannot write to the root directory, so it fails.
Easiest way to get around this (if insecure) is to grant all users write access to the drive.
bdipert - Monday, May 14, 2012 - link
Thanks for the feedback, isulzer, but the MacBook Pro SSD (which I was able to successfully run AJA on) is also that particular system's boot drive. Is the particular O/S part of the reason for the issue? The MBP runs Snow Leopard, the Mac mini runs Lionisulzer - Monday, May 14, 2012 - link
In lion and mountain lion, permissions on the boot drive are slightly different. Only System has write access. Everyone else has read only. So if you try to write a file to \ in Lion, you will have to authenticate.bdipert - Monday, May 14, 2012 - link
Thanks for the clarification, isulzer, very interesting!ssddaydream - Monday, May 14, 2012 - link
One thing that would be incredibly useful is direct PC to PC connection with USB 3.0 networking.Anybody else down for 4.8 gbps transfers with a cheap $15 cable?
Apparently, all you need a USB 3.0 crossover cable and each computer motherboard with capable drivers.
Anandtech, can you guys take a look at this? Some things in this industry just boggle me, such as a lack of an obvious USB 3.0 networking solution...
Thanks!
peterfares - Monday, May 14, 2012 - link
WHY does Thunderbolt need to carry DisplayPort AND PCIe. Wouldn't it have been easier to add pins to a displayport receptacle or create a new port for the PCIe data? Why do we need these multiplexing chips in each device when we could have just used straight PCIe? Even better, integrate the PCIe lanes right into a USB receptacle. It's a data bus, should it not be in a data port?If you want one cable to a docking station/display, they could have easily mandated that one Thunderbolt port must be right next to a displayport port.
repoman27 - Monday, May 14, 2012 - link
Apple was already shipping Macs with mini-DP connectors that were capable of pushing 17.28 Gbps worth of packetized data over a single cable. It's only common sense to look at that and say, "Wait a minute, why just use this for display data? Why not make it full-duplex and use it for transporting PCIe packets as well?"Digital display interfaces are the only ports commonly found on PC's that are even in the ballpark of Thunderbolt from a bandwidth perspective, so the marriage of the two is actually fairly logical.
Mini DisplayPort connectors already pack 20 pins in a 33 mm^2 cross section. That's pretty dense, there's not really room for adding more pins. Plus, by keeping the same physical connector as mini-DP, it's easier to maintain backwards compatibility with existing DisplayPort gear.
Besides, adding more pins to the connector would mean adding more conductors to the cable, and Thunderbolt cables are already complex and expensive enough as is.
Glindon - Monday, May 14, 2012 - link
So you'd rather have two cables hanging out of your computer than one? That's a silly argument. Why not have three or five then?I don't get all the hate for thunderbolt. When intel first demoed it everyone was raving about how great it would be, but once it came on macs first everyone said how dumb and stupid it was. At least pcs are going to get it now so we can get more devices for it and then it can suddenly become "cool" again.
aliasfox - Monday, May 14, 2012 - link
Actually, if I remember correctly, Intel tried to get LightPeak (as it was called while under development) integrated into a next generation USB port, but the consortium told them to take a hike - patents/royalties and customer confusion being named as primary issues with that implementation. That's when Apple stepped in and offered their mDP design - Apple was one of the originators, but it's a royalty free design as far as I've heard - part of the reason why AMD uses it for Eyefinity. Intel develops the tech, Apple offers the plug design, and in turn, gets first dibs on it.Sony *did* implement a variant of TB/LightPeak on its Vaio Z (the one that has the external 6630m) that utilizes a USB port, but that's not standard.
repoman27 - Monday, May 14, 2012 - link
By "the consortium" I reckon you're referring to the USB-IF, which was actually formed by none other than Intel. I know that the USB-IF and Intel both frowned on Sony's non-standard implementation, despite the fact that it closely resembles a lot of the early demonstration hardware, but did Intel ever actually try to convince the group to sanction the adoption of a USB style connector for Light Peak?Also, DisplayPort is a royalty free specification maintained by VESA. Apple developed the mDP connector and agreed to license it for free, and VESA subsequently included it in the DP 1.2 specification. This all happened over a year prior to Thunderbolt arriving on the scene.
I'm pretty sure Apple got first dibs on Thunderbolt because they agreed to buy 18 million additional chips from Intel in 2011 if they got an exclusive on it.
Impulses - Monday, May 14, 2012 - link
Nice review and flyby of the current Thunderbolt market... The overuse of parenthesis on the first couple of pages was odd tho (almost like some were added by an editor?).Anandtech seems pretty bullish on the virtues of Thunderbolt in general but outside of laptop/docking station scenarios I still see very little upside to it, and that prospect alone (plus the super high adoption cost) won't help propel the interface to wide adoption anytime soon.
It's a shame too because some of the long term applications that have yet to materialize are very appealing, things like external GPUs, etc. If costs don't drop fast enough it's gonna be Firewire all over again, hopefully Intel stands behind it long enough for that not to happen.
name99 - Monday, May 14, 2012 - link
"Its 480 Mbps theoretical peak bandwidth,"Just like with WiFi, please for the love of god explain this stuff correctly. The issue is not that this is a "theoretical" or a "peak" bandwidth, it is that this is the PHY bandwidth, that is, it is the speed at which bits (all the bits) are dumped onto the wire (or into the air).
So why doesn't that match the actual goodput (ie the throughput of my file bits)?
The minor factor is that there is a variety of overhead that is required to correctly frame and describe each packet. There are bits that describe the purpose of the packet, its length, its source and destination, etc. In the case of WiFi (I don't know about USB2, but I expect these are necessary at the speeds of USB3) there are bits that allow the receiver to track how the signal is degraded as it travels and thereby compensate, and so on and so on.
The major factor is that these are not point to point links, they are shared, which means that some sort of protocol (a MAC) is required to decide who gets to talk at any given time, and how to recover when a mistake is made and two parties talk at the same time. Unfortunately the general rule for MACs is usually that people seem far less interested in working hard to make them efficient than they are in working hard to make the PHY efficient. So what one sees is that an awful lot of time is wasted in MAC overhead.
The historical pattern for WiFi has been that each release includes various new smarts that can allow the the MAC overhead to drop to about 25% (ie actual goodput under optimal conditions is around 75% of the PHY rate), but that the chip vendors actually implement only enough of these smarts (which are always labelled optional in the spec) to get this goodput to a little over 50%.
The historical pattern for USB, which has stood still for so long, is that at first we had the same sort of situation --- crappy MAC implementations that gave us a real world throughput of about 50% --- but, with nothing else to do, the chip vendors ramped up their chips to use every smart possible in the MAC, given us the current goodput under optimal conditions of about 65%. Point is --- the correct way to describe this is that USB2 has a PHY rate of 480Mbps, and an maximum goodput (taking into account both overhead bits and the MAC) of around 320Mbps.
The same is going to be just as true of something like Thunderbolt, when we get to the point of stressing it hard. There is a PHY rate, there are overhead bits for addressing, framing, packet description, training, etc; and there is some sort of MAC. I know nothing about the Thunderbolt MAC, and so have no opinion about whether it's an inspired or a crappy design (or the extent to which chips can make more or less efficient use of it). These would all be good topics for a future article.
repoman27 - Monday, May 14, 2012 - link
<blockquote>Granted, Thunderbolt's 8b/10b encoding scheme reduced the effective peak bandwidth to 8 Gbps...</blockquote>Thunderbolt provides a full 10 Gbps per channel to the upper layers—there's no 8b/10b in that figure. Anand achieved just over 8 Gbps of throughput with the Pegasus R6. Taking in to account other sources of PCIe overhead, such as packet framing and link maintenance, that number is about spot-on for a 10 Gbps link. Also, the PCIe data may be transported on more than one channel of a Thunderbolt cable, however the current controllers are limited to 10 Gbps of PCIe throughput because they only contain a single PCIe to Thunderbolt protocol adapter.
<blockquote>...perhaps, alternatively, the HDD is getting a portion of its power allocation supplied directly over the Thunderbolt link from the connected computer system.</blockquote>
I doubt either of these devices utilize bus power to a significant degree, because according to Apple, only the first device in a Thunderbolt chain can be bus powered.
<blockquote>Such a feature allows the system-side Thunderbolt controller to "see" each Thunderbolt Duo as a single bus peripheral, thereby enabling WD to accurately claim that you can daisy-chain numerous drives to each other and the system before you violate Apple's six-max specifications.</blockquote>
The 6 device limit only pertains to the number of Thunderbolt controllers that can be in a chain. You can hang as many PCIe attached devices as you like off of a single Thunderbolt controller. A good example of this is the Apple Thunderbolt Display—in System Information it shows up as many separate devices, but it still only counts as one Thunderbolt device.
As far as Thunderbolt product pricing goes, I think we're just looking at a scary BOM cost, which will hopefully be alleviated some with the arrival of Cactus and Port Ridge controllers.
And you can use the GoFlex desk adapter with any old drive, so I'm surprised you didn't give that a whirl.
name99 - Monday, May 14, 2012 - link
"Given that the PCI Express channel of the Thunderbolt interface delivers 10 Gbps of peak bandwidth in either direction, I next decided to see what would happen if I tethered the Seagate and WD drives together."You can do even better than this. You should have been able, in Disk Utility, to create a striped drive consisting of the two WD drives and the Seagate drive. This would be gated by the speed of the slowest drive, so should give you around 360MB/s read and write speeds.
Disk Utility is not perfect, and god knows I've sent Apple long lists of ways it can be improved; but it is pretty awesome for basic things like creating either striped arrays or large concatenated arrays.
bdipert - Monday, May 14, 2012 - link
Dear name99, that's a great idea. I'm traveling at the moment but will give it a shot when I get back in front of the Mac mini. I agree with you that it conceptually should work, and will be performance-gated by the drives in the Thunderbolt Duo.Torrijos - Monday, May 14, 2012 - link
You talked a bit about the impact of running a soft-RAID device being negligible, it would be nice to know the impact on the OS when pushing those mass storage devices to the limit.I've bought an USB3 extension card for my mac and while transfer speed are great, OS responsiveness takes a hit every time a big transfer occurs (TM backups are a pain).
A formal test of the impact of USB 3, ThunderBolt would be nice.
repoman27 - Monday, May 14, 2012 - link
Have you watched Activity Monitor during a TM backup to see what's being stressed? I usually associate the performance drop during TM backups with the crazy amount of small random reads that need to be performed on the system volume while you're still trying to use it.Draconian - Monday, May 14, 2012 - link
So basically, if you're really concerned about performance, an SSD using an old 3 Gbps connection will read and write twice as fast as a HDD using a Thunderbolt connection?I would be interested in knowing what the results would be with a HDD using an eSATA connection. And of course USB 3.0.
jabber - Monday, May 14, 2012 - link
......companies selling 7N Thunderbolt cables that cost them $3 to make but selling them to audiophiles for $3000 a go.Glindon - Monday, May 14, 2012 - link
Each connector of the thunderbolt cable has a small chip, which is why the cable costs so much. It's not just a $50 cable. iFixit has done a tear down of the cable if you want to check it out.repoman27 - Monday, May 14, 2012 - link
There are quite a few factors which make Thunderbolt cables expensive, being an active design is just one of them. If you ignore the bit about Thunderbolt being intended for iOS devices, the following blog post reveals quite a few details:http://www.patentlyapple.com/patently-apple/2012/0...
I think the snake oil salesmen that prey on the audiophiles will avoid Thunderbolt for some time, specifically because it is an expensive to manufacture, high bandwidth cable that can't be made for $3. A 2.0 m copper pipe that can carry 20 Gbps, full-duplex, plus bus power and out-of-band signaling just doesn't exist yet in the sub $20 realm.
quiksilvr - Monday, May 14, 2012 - link
You pay an arm and a fucking leg for these things and they don't even have the god damn decency to give you the wire?CalaverasGrande - Monday, May 14, 2012 - link
Just like on the MAc centric forums, people can not look past their narrow usage scenario to accommodate other perspectives.I work at a TV station, but have a background in audio production. In most prosumer studios, and quite a few mid level studios, firewire is the connection of choice for audio interface (soundcard). The high end audio interfaces have until recently all been either a PCIe card or Firewire. Add to the pile firewire based DSP such as the UAD-2 and firewire disk arrays used to insure fault tolerant and fast nearline storage.
In video production we really don't use 1394 anymore. But there are still some old boxes in use that have the mini connector. Often they get pressed into use as a transcoder or layback machine for formats we dont use natively anymore. We are all on Sony XDcam disk, which I would love to see as a consumer format!
I personally can't wait to see wider adoption of thunderbolt. I plan on upgrading to a T-bolt equipped mac after the next refresh of the MBP line.
Death666Angel - Wednesday, May 16, 2012 - link
This sentence is a bit misleading:"a specific RPM at any particular time is dependent on both access requirements and sensed operating temperature." - This would suggest that this HDD runs with varying RPMs. This is not true. One HDD runs with one RPM. However, between the various HDDs among one charge there can be different RPM settings. So one Caviar Green "IntelliPower" runs with 5400 while the next runs with 5900. You will not see one 5400 unit suddenly run at 5900 because the temperatures are particularly low.
bdipert - Saturday, May 19, 2012 - link
Dear Death666Angel, I don't think you're correct. WD has told me on many occasions that the IntelliPower algorithm varies a drive's rotational speed over time as a function of performance demand and operating temperature. And past AnandTech writeups make similar statements:http://www.anandtech.com/show/2385/2
for example
wolfgang123usa - Thursday, July 12, 2012 - link
The real advantage of Thunderbolt should be that it carries power, but special this feature is very limited, as the power pin on the connector is very small and shall carry 10Watts, Sata connectors, USB and other have bigger contact surfaces and/or multiple pins to carry power. The display port adapter was designed for a lot less power then 10watts, that's the current issue with Thunderbolt today. Some details can be found here:http://technology.coolodman.com
When a self powered thunderbolt device is used it works fine and uses only about 22% overhead to encapsulate the PCIe bus, so the speed with fast SSD's like Samsung 830 series is really not bad.
marmot_animal - Friday, July 13, 2012 - link
Two months after this article was written, I find Thunderbolt options at the same phase, quite limited. So I bought the Seagate GoFlex Thunderbolt adapter for $$$ from B&H, plus the cable. Although expensive, it seems for the time being the least expensive approach, albeit a DIY approach, to use raw hard drives, of which I have an expanding collection for my video and audio projects.If my research is correct, there are no official generic Tbolt docks, unlike the common USB3 (and earlier interfaces) docks on Tiger Direct. Although LaCie 2big bays appear to physically allow users to swap raw hard drives, LaCie voids the warranty in doing so. Obviously LaCie discourages a DIY approach and of course favors consumers to procure LaCie's aftermarket harddrive-installed bays. This is less of a diehard obstacle as opposed to WD's Caviar requirement. Conversely, I'm not into opening up enclosures to swap drives. It requires tools, is time-consuming, and is semi-permanent. My preference is to swap raw drives on-the-fly.
At least the GoFlex DIY hacker method works, and I secure the drive on the dock via rubber band technology. I report 'so far so good' with raw drives. I don't know precise speed details as long as I'm able to put the Tbolt bus to use, and it appears faster than any other bus I've used with FCPX. Too many companies try to hijack Tbolt by instilling proprietary constraints. Nevertheless, I find the GoFlex design with its generic SATA interface to be an extremely niche approach because most consumers won't want to seat a raw hard drive somewhat precariously in a dock not designed for such use.
spanading - Saturday, October 6, 2012 - link
Honestly peps,Why are people compairing the two. Yes USB is great if you are conneting things like HHD, Mice, Tablets, and the like, but remember you only have 5Gb in a single channel, one way maximum to play with, while thunderbolt has 20Gb when using both channels. True neither USB 3 nor Thunderbolt would be maxed out by a single HDD, but when connecting several data hungry devices at once, (which with thunderbolt can include monitors, and possible legacy port hubs, if they ever arrive) then USB will show the strain a lot quicker than thunderbolt.
Hence why you need a thunderbolt port? Well if all you are doing is plugging in one HHD over it it is probably overkill, BUT, if you like to have one of those new shiny Ultra books to use on the run but then want a fully fledged PC when you get home, Thunderbolt gives you this potential, but only having to connect 1 cable, rather than 3 or 4. Plus with it being compatible with PCi then you can have expansion cards to further enhance performance with out the need for a completely separate PC. Now before you complain ultra books have low spec CPUs, go and play with one that uses a SSD and see how much speed a traditional HDD sucks up while the system waits for data. As an example, my 2011 Mac book air has a 1.7 GHz dual core i5 with integrated graphic and only 4GB of memory, and on day to day tasks, is faster than my 2011 mac min server with a 2.0Ghz Quad core i7 with the same integrated graphics but 16 GB of RAM. In fact my macbook only struggles when the RAM is all used up, which is not very often .
I guess this is exactly what Dan Neely meant. Good on you Dan, I guess the USB people have not quite grasped the fundamental difference yet.
biswa60 - Saturday, October 19, 2019 - link
Each Thunderbolt port handles 40 Gbps of aggregate bandwidth, consisting of two pairs' worth of distinct 10 Gbps transmit and receive laneshttps://theprintcalendar.com/
https://vicschoolholidays.com/