FireWire 400 versus USB 2.0

[feedback]

I just ran some tests on my laptop with an external Hard Drive to confirm that Firewire is indeed faster than USB 2.0 in real world application, regardless of raw data numbers.

Using firewire and USB 2.0, I basically copied a DVD source of 3036MB from my external 7200rpm Hard drive to my laptop (it also has a 7200rpm hard drive) using Nero's Recode, DVD Shrink and CloneDVD2.

Recode: With Firewire, my DVD source material transfer completed in 2:08 (2 minutes 8 seconds).
With USB 2.0 my DVD source material transfer completed in 2:41 (2min. 41 seconds).

DVD Shrink: With Firewire, my DVD source material transfer completed in 3:32
With USB 2.0 my DVD source material transfer completed in 4:26

CloneDVD2: With Firewire, my DVD source material transfer completed in 2:52
With USB 2.0, my DVD source material transfer completed in 4:14

Foxyshadis gave me a pretty good link to more firewire info. located here. http://www.1394ta.org/Technology/About/Firewire_FAQ_PC.pdf

There is no doubt, IMHO, that firewire is faster than USB 2.0.

Regards,

[mod]

Don't assume anything...Verify!

Nice! :D

[Sharktooth]

It all depends on the USB controller you have installed on your motherboard.
Firewire usually have lower latencies and less CPU use but some USB controllers are just on par if not faster.
The bandwidth difference (480mbps for USB2.0 vs 400mbps for firewire) is usually negligible.

[Andrew Kevin]

Try just having the slim CD drive attached and then enter the bios to see whether it is picked up as Master or Slave.

Then, you should be able to jumper your 2.5" HDD to be the opposite and then they should be able to be used together.

If this doesn't work, you could try looking for a firmware update for the CD drive which can change it's Master/Slave settings.

_________________________________________________

Portable Hard Drives (http://go.iomega.com/en/products/external-hard-drive-portable/?partner=4750) | Desktop Hard Drives (http://go.iomega.com/en/products/external-hard-drive-desktop/?partner=4750) | Network attached storage (http://iomega.com/nas/network-attached-storage.html)

[mariush]

Firewire will almost always be faster than USB 2.0, because the difference in architecture (firewire uses hardware for the communication and has some neat - but dangerous - tricks like direct memory access) while usb uses the processor a lot when transferring data.

See http://en.wikipedia.org/wiki/IEEE_1394_interface#Comparison_with_USB :


Although current high-speed USB 2.0 (introduced in 2001) is quoted as running at a higher signaling rate (480Mbps) than legacy FireWire 400 (400 Mbps, available since 1995), data transfers over S400 FireWire interfaces generally outperform similar transfers over USB 2.0 interfaces. Few if any USB 2.0 device implementations are able to saturate the entire 480 Mbps, but this can be achieved with multiple devices on the same bus. Typical USB PC hosts rarely can sustain transfers exceeding 280 Mbit/s, with 240 Mbit/s being more typical. This is likely due to USB's reliance on the host processor to manage low-level USB protocol, whereas FireWire delegates the same tasks to the interface hardware (requiring less or no CPU usage). For example, the FireWire host interface supports memory-mapped devices, allowing high-level protocols to run without loading the host CPU with interrupts and buffer-copy operations.[5]

Besides throughput, other differences are that FireWire uses simpler bus networking, provides more power over the chain and more reliable data transfer, and is less taxing of a CPU.[28]


and http://en.wikipedia.org/wiki/IEEE_1394_interface#Security_issues :


Security issues

Devices on a FireWire bus can communicate by direct memory access (DMA), where a device can use hardware to map internal memory to FireWire's "Physical Memory Space". The SBP-2 (Serial Bus Protocol 2) used by FireWire disk drives uses this capability to minimize interrupts and buffer copies. In SBP-2, the initiator (controlling device) sends a request by remotely writing a command into a specified area of the target's FireWire address space. This command usually includes buffer addresses in the initiator's FireWire "Physical Address Space", which the target is supposed to use for moving I/O data to and from the initiator.[43]

On many implementations, particularly those like PCs and Macs using the popular OHCI, the mapping between the FireWire "Physical Memory Space" and device physical memory is done in hardware, without operating system intervention. While this enables high-speed and low-latency communication between data sources and sinks without unnecessary copying (such as between a video camera and a software video recording application, or between a disk drive and the application buffers), this can also be a security risk if untrustworthy devices are attached to the bus. For this reason, high-security installations will typically either purchase newer machines which map a virtual memory space to the FireWire "Physical Memory Space" (such as a Power Mac G5, or any Sun workstation), disable the OHCI hardware mapping between FireWire and device memory, physically disable the entire FireWire interface, or do not have FireWire.

This feature can be used to debug a machine whose operating system has crashed, and in some systems for remote-console operations. On FreeBSD, the dcons driver provides both, using gdb as debugger. Under Linux, firescope[44] and fireproxy[45] exists.

[stax76]

I had about 20 MB/s with a USB 2.0 hard drive and 100 MB/s replacing it with a esata one, it was a Gigabyte Board with ATI chip set.

[Blue_MiSfit]

Firewire 800 is a lot faster than either USB 2.0 or FireWire 400, and USB 3.0 is a lot faster than any of them :devil:

I prefer full on eSATA in all cases though. I've sustained ~ 120MB per second using a 2TB seagate copying a large file to an internal RAID :D

To get back on topic, with Windows, the distinction isn't as big. On mac, USB is usually totally awful... FireWire is much faster.

Derek

[Ghitulescu]

Try just having the slim CD drive attached and then enter the bios to see whether it is picked up as Master or Slave.

Then, you should be able to jumper your 2.5" HDD to be the opposite and then they should be able to be used together.

If this doesn't work, you could try looking for a firmware update for the CD drive which can change it's Master/Slave settings.

_________________________________________________


If this is not advertising for Iomega products then you should know that there are no master and slave for USB/IEEE1394/SATA attached drives.
Besides, supposingly for the sake of argument there was a M/S problem, then neither drive would work, which is not the case.

[Przemek_Sperling]

Well, "comparable" FW will always have an advantage because it is full-duplex in contrast to USB 1.0/11.1/2.0 (I do not know if 3.0) which is only half-duplex.
Morever, USB controllers differ much from one to another. I take photos a lot (RAW mode written on Compact Flash UDMA cards) and I use a card reader connected to USB controller. Neither Intel P35 mobo controller nor nForce 720D mobo controller compare to my PCI card based on NEC chipset. Transfers are around 20% faster with the PCI card (they reach USB bandwidth saturation) and CPU utilisation is far, far lower.