Via chip to enable smaller, cheaper USB3 thumb drives

Via doesn’t get a whole lot of attention these days, but it’s still cranking out new products. Its latest creation is the VL750 USB 3.0 to NAND flash controller, which recently received certification from the USB Implementers Forum and has begun sampling. Via says this chip is the first of its kind to be USB-IF certified, and because it removes the need for an auxiliary bridge chip, it will enable “more compact, more power efficient and more cost effective Flash drives.”

Other details about the VL750 aren’t quite as exciting, except perhaps for the fact that it supports 2x-nm and 3x-nm flash memory and enables speeds as high as 100MB/s via USB 3.0 (or 35MB/s via USB 2.0). Now, some folks might point out that USB 3.0 offers top theoretical speeds of 625MB/s—and they’d be right—but 100MB/s is already pretty darned decent for a small, inexpensive thumb drive. Such products aren’t usually known for packing ultra-high-speed flash memory.

Here’s hoping the new Via silicon will help make cheap USB 3.0 thumb drives more ubiquitous. We already seem to be on the right track, with Super Talent having unleashed an 8GB offering priced at just $15 earlier this month. (That drive probably doesn’t use the Via controller, of course, since the VL750 is only at the sampling stage right now—but more like it may follow.) Too bad USB 3.0 support isn’t expected to appear in Intel chipsets until 2012.

Comments closed
    • phileasfogg
    • 9 years ago

    AFAIK, neither Vista nor WinXP will support USB3.0 – is this true? And Win7 support is only officially available with SP2 (ServicePack2). THis isn’t really an inconvenience per se; I suppose vendors will have to provide their own drivers to enable the SS (super-speed) capabilities of their chips; until the OS is automatically capable of detecting and enabling SS on a USB3 device+host.

    • phileasfogg
    • 9 years ago

    Cyril — You wrote: “Now, some folks might point out that USB 3.0 offers top theoretical speeds of 625MB/s—and they’d be right”

    well, that’s not true. Since the physical layer of USB3.0 performs 8b-10b encoding during transmission (and 10-8b decoding during receiving), the actual theoretical peak b/w is 625*0.8 = 500MB/s; the exact same ‘theoretical peak’ as a single-lane PCIe Gen2 link, which operates at the same physical layer speed of 5Gbits/second.

    • bdwilcox
    • 9 years ago

    VIA? Well, there goes USB3 compatibility…

      • phileasfogg
      • 9 years ago

      It does state clearly in this article that the VL750 chip is “USB-IF i.e. “implementer’s forum”, certified. If that isn’t the ‘gold standard’ for compliance, then perhaps you can come up with a better one?

        • bdwilcox
        • 9 years ago

        The gold standard is if it works the way it’s supposed to. And VIA’s always had a hard time with that one.

          • just brew it!
          • 9 years ago

          Well, in this case both VIA and the USB-IF would need to screw up to result in a borked product. Not saying it can’t happen, but at least there’s a (supposedly) independent third party trying to keep them honest.

    • Hattig
    • 9 years ago

    This will make those TV ‘tech crime’ shows more boring in those tense scenes when the spy is copying all of the incriminating data from the criminal’s supercomputer with a very slow, large animated ‘file copy’ progress bar on the screen, and the criminals just seconds away.

      • OneArmedScissor
      • 9 years ago

      Don’t worry. There are lots of Apple computers on TV.

        • StuG
        • 9 years ago

        This thread is full of lulz

      • ludi
      • 9 years ago

      This is why, when I take over the world with a crime syndicate, my supercomputer will store everything incriminating on magnetic tape at an artificially-low bit density. Then it doesn’t matter how fast the spy’s device is.

Pin It on Pinterest

Share This