Slashdot Mirror

← Back to Stories (view on slashdot.org)

USB 3 in 2008, 10 Times as Fast

Posted by CmdrTaco on from the i'll-believe-it-when-it's-on-my-laptop dept.

psychicsword writes "Intel and others plan to release a new version of the ubiquitous Universal Serial Bus technology in the first half of 2008, a revamp the chipmaker said will make data transfer rates more than 10 times as fast by adding fiber-optic links alongside the traditional copper wires." "The current USB 2.0 version has a top data-transfer rate of 480 megabits per second, so a tenfold increase would be 4.8 gigabits per second." This should make USB hard drives easier and faster to use."

24 of 381 comments (clear)

  1. I'm more concerned with latency. by jimstapleton · · Score: 4, Informative

    It seems current hard drives test to 40-80M/s (dunno if it's bit or byte, we'll assume byte since it is worst case for my example)., averaging between 50 and 60M/s

    480Mbit per second = 60MByte per second. That can handle the average case for a modern hard drive.

    4.8GBit/second - 600MByte/s? To utilize that with a drive, you'd need a RAID external enclosure!

    --
    34486853790
    Connection too slow for X forwarding? Try "ssh -CX user@host"
    1. Re:I'm more concerned with latency. by Laebshade · · Score: 5, Interesting

      I can't find the exact article, but you should read this one about the effective USB 2.0 speed. It states that the effective maximum speed is only about 40MB/sec, and that 60MB/sec can't be achieved due to overhead/software limitations; not sure if this is true now.

    2. Re:I'm more concerned with latency. by Televiper2000 · · Score: 5, Informative

      My experience with USB is that it has lousy thru-put do to the way the drivers manage data on the host side. We did some heuristic timing tests on USB to serial devices and found that most of them actually degraded the thru-put (time send and receive a packet) of an RS232 connection. We found that devices with Silicon labs ICs could get the desired thru-put, but it seemed they would take up an entire USB channel to do so.

      --
      New! Device Legs: These legs will help your poor OEM installed product escape any hamfistedness it may encounter. Ava
    3. Re:I'm more concerned with latency. by schwinn8 · · Score: 5, Interesting

      Exactly... Firewire400 works well for video streaming from a DV cam because it has very little overhead. Even though USB2 supposedly does 480Mbps, it can't do DV because there's too much overhead. Bottom line is, unless USB3 gets rid of the CPU dependency and overhead issues, I won't like it. Sure, with a "ten times" the performance, this won't hinder DV, but that doesn't make it good. I hope they make it systemically-efficient, instead of just ramping clock rates to reach these speeds.

    4. Re:I'm more concerned with latency. by Barny · · Score: 5, Interesting

      Mod parent up, hammer meets nail in their post :)

      Bonus points if you hook a 108mb/s wireless lan adapter via USB and throw some large data files over it, watch your data speeds closely, and monitor system performance even closer.

      Firewire (1394, ilink, DV port, whatever) really was the shit, not only fast, low overheads AND its a peer-2-peer setup, in a pinch you could daisy chain PCs with it for an impromptu 400mb/s lan.

      Why didn't they just hang USB out to dry and get power into the eSATA spec and use that? At least then no extra chips would be needed on a mobo, external HDD would hookup with no loss in performance and we might finally see thumb drives that work natively with ANY os as... drives.

      --
      ...
      /me sighs
  2. Re:Cable? by YrWrstNtmr · · Score: 5, Informative

    The last line in the article:
    "It will be backward compatible, so current USB 2.0 devices will be able to plug into USB 3.0 ports."

  3. Bottleneck? by mrjb · · Score: 4, Insightful

    Currently I'm getting transfer rates of about 16 megabyte per second on hard drives connected via USB. That's roughly 160 megabit per second, whereas USB 2.0 can transfer up to 480 megabit per second. While I'm all for faster and better, the bottleneck seems to be elsewhere in this case.

    --
    Visit http://ringbreak.dnd.utwente.nl/~mrjb/growingbettersoftware to download your free copy of the book
    1. Re:Bottleneck? by hcdejong · · Score: 4, Insightful

      In the controller, likely. I'm getting 30% higher transfer rates with FW400 than with USB 2.0 on the same external disk.

      Which doesn't give me high hopes for USB3. High-speed links are all good and well, but if they keep including cheap-ass controllers, what's the point?

    2. Re:Bottleneck? by Jah-Wren+Ryel · · Score: 4, Insightful

      In the controller, likely. I'm getting 30% higher transfer rates with FW400 than with USB 2.0 on the same external disk. It's as least as likely the problem is in the protocol. USB is synchronous - data every packet received must be acknowledged in a return packet before the next data packet can be transmitted. That back and forth for each data packet means a lot of wasted time where the channel is essentially idle. Sometimes using a shorter cable can make a noticeable improvement.

      Firewire has both synchronous and asynchronous modes. In async mode, a bunch of packets can be transmitted before any acknowledgment back is required. That's bad if the cables is flakey, since it will result in a lot of retransmits, but bad firewire cables are the exception, not the rule. So async is almost always way more efficient than synch. I'm pretty certain that you are using the async mode for talking to your disk.
      --
      When information is power, privacy is freedom.
  4. Plug Shape by Crock23A · · Score: 5, Insightful

    Could be slightly off topic but I had to sound off on this one...

    While I appreciate USB's capability for backwards computability, I would much rather have a plug shaped in such a way that I didn't have to flip it over every time I try to plug it in. I don't know about you guys but this is one of the most annoying aspects of using my computer, and I run Windows!

    This would also be a great time to make a universal "other side" of the cable, rather than having a different plug for every single USB device. I have a mini plug for my camera, a big square one for my printer, a 2.5 mm jack to charge my MP3 player, etc. All these cables make a mess. If all my devices could share one cable, I'd be much happier.

  5. Re:Yeah, but.. by LiNKz · · Score: 4, Insightful

    Also, if you integrate an element of fiber optics into a cable that routinely wrapped up, stepped on, or just basically abused, wouldn't it fail far easier than a standard cable?

    --
    Proceed with Format (Y/N)? Y
  6. Re:Great. by morgan_greywolf · · Score: 5, Informative

    1.5 seconds if you all of your components were fast enough. The drive won't be. Exactly. There's still the slowdown associated with the mechanical aspects of the hard drive -- spin rate (RPM), average seek time (ms), etc. On top of that, most hard drive controllers are limited by the technology they use. For instance, a SATA hard drive, even plugged into a USB 2 or 3 port, is limited to 150 MB/s -- but, that's burst speed, not sustained transfer rate.
    --
    My blog
  7. Re:Great. by Anonymous+Brave+Guy · · Score: 4, Informative

    On top of that, most hard drive controllers are limited by the technology they use. For instance, a SATA hard drive, even plugged into a USB 2 or 3 port, is limited to 150 MB/s -- but, that's burst speed, not sustained transfer rate.

    Indeed. And realistically, it's going to be a pretty short burst: most hard drives today only have something like 8–16MB of cache that might be filled by a smart lookahead algorithm, so your best case with current hard drive technology is that you'll get perhaps 1/10 of a second of high-speed data transfer before hitting the physical barriers.

    I'm not sure this is directly applicable to this discussion, though, because AFAIK all current USB drives use different storage technology anyway. It's going to be the limits of that technology that tell us whether USB3's theoretical speeds will actually be useful with storage hardware available in the same time frame.

    --
    If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
  8. Screw bandwidth... by Balinares · · Score: 4, Insightful

    Give us a standard that actually delivers enough power that you don't need an additional power cord for just about every other device already... :/

    --

    -- B.
    This sig does in fact not have the property it claims not to have.
  9. Re:Yeah, but.. by ggeens · · Score: 4, Interesting

    When I was working on my Master's thesis, I had to splice optical fiber a few times. Believe me, it's not easy.

    Glass fiber is very flexible. You can bend it in any way you want, it won't break. You can cut it, but that takes considerable force. If you break the fiber, you'll break the copper wires as well.

    Personally, I think the weakest point in such a cable will be the connectors. Getting the light from one fiber to another requires careful alignment. Any deviation might causes loss of signal. Getting dirt into the connector is probably fatal.

    --
    WWTTD?
  10. Re:Great. by BosstonesOwn · · Score: 5, Insightful

    http://en.wikipedia.org/wiki/Flash_memory

    No Flash is slow to write , very fast to read. Hence Windows use of it for "ReadyBoost" caching. There is extremely low latency just not enough bandwidth to sustain high levels of I/O.

    On the other hand , by introducing fiber into the link doesn't that take away the greatest part of usb ? being able to just fold up the cable and stuff it in your pocket along with a small hard drive ? I know I use it for restoring machines after catastrophic failures (yeah windows) and some times I don't go right back to my desk with the cable and drive and have to toss it in my pocket. I can't do that with fiber, it would fracture.

    --
    This package Does Not Contain a Winner
  11. Re:Great. by antime · · Score: 5, Insightful

    The quoted speed of USB 3 is probably the bus speed, i.e. it's shared by all devices connected to the same host. So one disk won't saturate the bus, but if you plug in a bunch of them the bandwidth won't seem so incredibly massive anymore. Then you have to consider the bandwidth reserved by isochronous devices etc.

  12. Honest Question by martyb · · Score: 5, Interesting

    I've got a question that has been nagging at me for quite a while and was hoping someone here could phrase an answer in terms a mere mortal could understand.

    Why are there so many serial specifications?

    We've got, off the top of my head, SCSI, USB, Ethernet, FireWire, and SATA to name a few. I do understand there are different protocols (all the way up from the physical to the application layers). Different applications of these technologies permit some optimizations that might not be applicable in other situations. But, at some point, the underlying technology is fast enough

    Still, I can't help but think there should be some common denominator that ALL these communications standards can agree on, and through economies of scale, become universal standard(s). It just seems like people keep re-inventing the wheel with an eye toward THEIR favorite.

    I thought we were getting close when they released gigabit Ethernet over UTP (unshielded twisted pair).

    • can handle distances up to 100 meters
    • fast data rate (1000 Mbps)
    • supports lower data rates (100/10 Mbps)
    • development is underway for 10Gbps, too.

    So, for the sake of argument, why not have all of our serial devices just support gigabit Ethernet? Sure, you'd need a hub or switch in your PC to talk to all of the devices, but you already need something similar for the other protocols (USB hub, SCSI controller, etc.). It's a well-known technology with many implementations and is widely available. I'd willingly pay a few more bucks for each device if I could ditch all of these incompatible formats and just standardize on one SET of ports and cables for hooking things to (and within) my PC. And in those cases where a different connector is desired (e.g. for small form-factor devices like a digital camera), let me just get an adapter cable/plug that I can plug into my Ethernet port.

    Is there any good, technical reason that is keeping us from having truly UNIVERSAL serial communications?

    1. Re:Honest Question by mlBrianR · · Score: 4, Interesting

      I think the main problem is repeated insertion/removal vs. semi-permanent installation.

      USB connectors are designed to be inserted and removed over and over. They're held in by pressure against the connector, so they can be removed without having to push a tab or twist the connector to remove it.

      UTP cables are designed to be plugged in, and then generally left alone. The UTP cable in my computer bag is in terrible shape.. the RJ45 connector is coming loose, the plastic retaining tab is broken off (so the cable often pops out of the jack on its own), etc.

      I have USB devices which I've removed and inserted hundreds of times, and the connectors still work reliably.

    2. Re:Honest Question by smellsofbikes · · Score: 4, Informative

      A bunch of other people have chimed in with parts of the answer already, and I thought I'd add some more.
      Obviously each manufacturer wants you to use their standard and buy their hardware.
      Different implementations came through at different times, and have different amounts of software/hardware overheads, and try to do different things.
      RS232 has almost no necessary software overhead -- you do any and all the work with code you write. USB has *quite* a bit of software overhead to deal with device identification, and Ethernet has *enormous* amounts of overhead. In most small systems you have to buy an Ethernet software stack separately from the OS you're using.
      USB tries to provide power. People are trying to glue power into Ethernet although it hasn't yet taken off.
      People keep going off in odd wireless directions.
      The fundamental problem, I think, is that there are several different connectivity needs and manufacturers are trying to get you to buy their solution to what they think are the most important needs. What you're asking for is something good for the industry in the long term, and that's not really in the direct interest of any particular company, so nobody's building anything for it.
      The military embedded market seems to be moving towards gigabit or 10-gig fiber ethernet for all their interboard communications, but fiber has its own problems, and I'm not sure it's the right thing for a USB key you're carrying in your pocket all the time.
      I believe that the SCSI module in linux handles firewire and USB, so from that standpoint it looks like it's a start towards universal communications, except for Ethernet. (Even though old SCSI is nothing like serial: it's the the ultimate expression of parallel communications, with some similarities to the old HP/GPIB parallel communication standard that's still used in for test communication but used to be a hard drive standard.) I have no idea what Windows does.

      --
      Nostalgia's not what it used to be.
    3. Re:Honest Question by gillbates · · Score: 4, Interesting

      Is there any good, technical reason that is keeping us from having truly UNIVERSAL serial communications?

      Yes.

      Let me explain:

      • USB, per standard (host side), must be able to source at least 500 mA. Even though there is a PoE (Power over Ethernet) standard, most choose not to implement it by default. Hence, ethernet can't power devices like flash drives and hard drives.
      • USB uses four wires, ethernet twice as many. USB is a synchronous bus, meaning that there are (theoretically, at least) no collisions. A properly operating USB device will not stomp on someone else's data packet. Thus, for the given bitrate, a higher portion of the bandwidth is available to applications. Unlike ethernet, adding devices to the same physical connection will not degrade the overall bandwidth of the network. In practice, I've found 100 Mbit ethernet devices operating with a maximum throughput of about 35-40 Mbits/second because of collisions. And this was with *two* devices! To get a better throughput requires using routers (which minimize or eliminate collisions).
      • RS232 is a pretty universal serial communication standard. However, it is also slow.
      • There are tradeoffs between maximum cable length and the speed of the bus.
      • There are tradeoffs between the number of signal lines and the cost of the device.
      • There are tradeoffs between bitrate and the device cost.

      So, the reason why we don't have a universal serial standard is because the different interfaces were designed with different goals in mind.

      --
      The society for a thought-free internet welcomes you.
  13. Re:Anyone else here see a problem?? by aadvancedGIR · · Score: 4, Funny

    Then I recommend you not to buy the USB-powered dildo.

  14. 0.9 TB / 4.8 Gb/s = 1500 seconds by Anonymous Coward · · Score: 5, Informative

    WTF, a chain of "Exactly" and "Indeed", and no one realizing that 1.5 seconds is wrong by a factor of 1000?

  15. Re:Cable? by ddstreet · · Score: 5, Informative

    > like if you plug a usb 1.X device onto a usb 2.0 bus, then everything slows to usb 1.X. IINM...

    This is wrong, if you plug a USB 1.1 device into a USB 2.0 "bus" then it does NOT slow everything down. Specifically there are 2 cases:

    1. You plug a USB 1.1 device directly into your computer (i.e. directly into the "host controller"). In this case, the USB 2.0 host controller (technically a EHCI chip) does NOT talk to your device. Instead, the EHCI chip has one or more USB 1.1 host controller chips (technically either a OHCI or UHCI chip, and called a "companion" chip when inside a EHCI chip) and your USB 1.1 device is connected electrically to that controller. You device is not on the USB 2.0 (EHCI) bus.

    2. You plug a USB 1.1 device into a USB 2.0 hub. In this case, the USB 2.0 hub creates a complete USB 1.1 environment specifically for your device. On the host-facing side of the USB 2.0 hub, all communication continues to take place at USB 2.0 (i.e. 480Mbps) speeds. When the host wants to talk to your USB 1.1 device, it uses what is called "split transactions" to talk to it. Basically (I'm simplifying), this involves sending a "start" packet to the USB 2.0 hub. Then, the USB 2.0 (EHCI) controller goes on to do other things, while the USB 2.0 hub initiates the transfer to your device at USB 1.1 speeds. And data transferred from the USB 1.1 device is stored temporarily in the USB 2.0 hub. Eventually the USB 2.0 (EHCI) host sends a "finish" packet to the USB 2.0 hub. If the USB 1.1 transation finished, the USB 2.0 hub responds successfully (either with the incoming data or a "ack" that the outgoing data was sent) which completes the transation.

    (There is also a combination case of those, where the EHCI chip does not contain a "companion" USB 1.1 chip, but instead contains an internal USB 2.0 partial hub - the "transaction translator" part - that handles talking to USB 1.1 devices. For bus usage purposes, this is effectively the same as using an external USB 2.0 hub, since the USB 1.1 devices do not appear on the USB 2.0 bus.)