So what do we need usb 3 for exactly?

i.e. can USB ports at last act for themselves with direct memory access or whatever, or are they still likely to be very limited in real-world speeds when compared to Firewire (even just mid-90s FW400 in my experience) and eSATA?

He didn't just say USB 3.0 v1.0!? I sincerely hope the USB tech group didn't come up with that double versioning thing and it was just the author.

Let us please not have the same old misnomer that afflicted the famous MP3 format! MP3, or MPEG3 is actually MPEG V1 layer 3.

Do you remember with the advent of DVDs and MPEG2, that MP3 didn't make much sense seeing as the audio format clearly pre-dated DVDs, and possibly the MPEG2 spec itself. The whole reason we now have MPEG4 because it was deemed too confusing to call it v3!

I understand fully that you are just emphasising the exiting of its beta status, but lets just call it USB 3.0 to avoid the catchphrasing? Then when theres a change, we can do the old 3.1 or 3.2 trick? Just like how we had USB 1.0 then USB 1.1 etc.

Just so long as we dont get USB27 prior to USB4 thats fine!

Thanks

So what I want to know is will USB 3.0 still be the same CPU-munching interface that USB 1 & 2 were. If it is, I'll stick to the undervalued and technically superior Firewire. It's been faster than USB in the previous versions in sustained transfers. And come to think of it, that's another question: This vaunted 10x the speed of USB 2.0... is that sustained or a burst speed like with USB 2.0. Given that the speed of an externally connected USB device is normally only a factor when transferring large amounts of data, e.g. to an external hard drive, the sustained speed is much more important. More technical details, please.

Firewire 3200 was announced nearly a year ago, and will probably be faster for continuous transfers than USB 3.

Yes please.

No, thanks. I'll stick with FireWire.

Nah, you "Super-Duper" guys are missing the obvious next implementation -- EXTREMEspeed (oh, sorry, XstreamSpeed). I'm surprised they didn't use the term for this implementation.

As for the question of whether it will have direct memory access... Didn't we already determine that giving an external device direct access to your memory, bypassing any security controls, is a bad thing?

Yeah, it's a bad idea to give external devices truly direct access to memory. So map it through virtual memory. Or you can disable the OHCI hardware mapping, albeit with an efficiency penalty. Of course if you're close enough to a machine to be plugging things into it, security is somewhat moot (unless someone cons you into hooking up a trojan device, but that's a bit too Mission Impossible to be plausible).

The real problem with USB vs 1394 is that USB is master/slave and 1394 is P2P. Unless that's going away, and I can't see how it would, real world 1394 performance is always going to exceed USB at a given nominal data rate. What does USB 2.0 really give you? Around 40-45% of nominal bandwidth in my experience. Firewire pledges to give you 97%. 45% of 4800 Mbits/s is less than 97% of 3200 Mbits/s.

exactly :)