Hardware Reference
In-Depth Information
But even with the overwhelming popularity of USB, a market for 1394 still exists, espe-
cially in DV (Digital Video) applications. One of the main differences between 1394 and
USB is that USB normally requires a PC as the host, providing the control and manage-
ment of the bus. This means you cannot connect two USB devices directly and transfer
data between them; instead, they must both be connected to a PC, which acts as the host
controller.WithFireWire,ontheotherhand,youcanconnecttwodevicesdirectlybecause
thebusisdesignedsothatanydevicecancontroltransfers.Forexample,1394candirectly
connect a DV camcorder to a DV-VCR for dubbing tapes or editing. This can play a big
part in performance because the transfer rate between devices over USB depends greatly
onthehorsepowerofthePCcontrollingthebus,whereasthetransferrateoverFireWireis
mostly independent of the PC and far more dependent on the devices doing the transfers.
FireWire support is built-in to most DV camcorders and video editing applications. For
example, when capturing video from tape you can connect a FireWire-enabled DV cam-
corder to a FireWire port in your PC, and then use Windows Movie Maker (included
with XP) or Windows Live Movie Maker (a free download from Microsoft for Windows
7/Vista)tocaptureordigitizethedatafromtapetothePC.OnceonthePCitcanbeedited,
copied, and or uploaded to video sharing sites. Although many systems include FireWire
built-in, if your system doesn't include a FireWire connection, you can easily add one via
an adapter card.
USB On-The-Go, introduced in December 2001, enables the same device-to-device con-
nectionsaswerecapablein1394(FireWire).Unfortunatelythishashadlittleeffectonthe
USB marketplace because few devices actually support USB On-The-Go.
USB/FireWire Performance Myths and Realities
A student once wrote me to inquire about a question that appeared on a test in a computer
class he and his classmates had taken. The question was, “Which is faster, USB 2.0 or
FireWire 400 (1394a)?” They had answered that USB 2.0 was faster, because USB 2.0
was rated 480Mbps (60MBps), whereas FireWire 400 was rated 400Mbps (50MBps).
Well, after the test was graded the students were shocked to find their answer had been
marked wrong! Upon discussing the question in class, the teacher confirmed the technical
specifications but then went on to say that due to less internal “overhead,” FireWire was
faster than USB in the real world, even though the raw transfer speed might be less on
paper. The student was basically asking me what the correct answer to the question really
was.
I had to sigh. I
despise
questions like that because, from a technical standpoint, the ques-
tion is so poorly and imprecisely written that
both
answers are arguably correct. For ex-
ample, it obviously wasn't clear whether the question meant “faster” as in the raw bus
speedorasinreal-worldthroughput.Andifitdidmeanreal-worldthroughput,underwhat
