Intel Releases USB 3.0 Controller Interface Spec
hardsky submitted thrilling news about everyone's favorite interconnect cable by saying "USB 3.0 is set to deliver data-transfer speeds of up to 5Gb/s, initially over tweaked connectors and wiring and, later, over optical links."
Still the same symmetrical plug design....stupid, stupid move. Would have been that hard to add a ridge on one side or something, so you don't have to stare at the end??
TODO: Something witty here...
Doesn't matter. It is a universal serial BUS.
This means that traffic to and from many slower storage devices can share the path so any speed increase is still a good thing, right?
Not without a connector for it first.
<sig> </sig>
And how are you supposed to work out which way is "up" with a socket that is on a tower case or PCI bracket?
Intel has provided chipset makers with a draft specification for a USB 3.0 eXtensible Host Controller Interface (XHCI), making good a promise it made a couple of months ago.
I thought we had a standards body that would release such a spec to developers. This development in my opinion, might have other chip makers release a "renegade USB 4.0" promising new features and the like.
Question is; is it up to manufacturers to think of ideas, name them and release these to the general public? What's up with IEEE Standards group, whose global standards include Biomedical and Healthcare, Nanotechnology, Information Technology and Information Assurance among others?
It's all well and good to quote the new speed but what will be get int the real world? USB2 never meets expectations due ot the huge (compared to fire wire) host CPU requirements.
Will Intel be integrating the Larabee core into it's USB 3 host chips?
It will be interesting to see if USB 3.0 relies on the processor as much as the USB 2.0. This has led to firewire (400mbps) outperforming USB 2.0 (480mbps) in real world tests. In todays multicore world this may be non-issue on most machines by the time it ships. In a way I hope USB 3.0 does perform well. I would be OK with firewire going away if Firewire 3200 is outperformed by USB 3.0 without hogging to many clock cycles.
Yup. I have seen the research, and I clearly needed to step away from the keyboard.
Although I think my point stands, when the entire discussion on this article is on my crappy spelling and grammar, rather then the oh-so-exciting USB 3.0
In theory, you could take two SATA 3GB drives and put them in a dedicated box that treated them as a software-driven RAID-0. That would give you peak theoretical data transfer of 6Gb/sec, but that's likely to happen only if you hit the drives' on-board caches. Connect that to your box using USB 3.0.
Of course, I'd probably prefer 1Gb/sec Ethernet, so I could see the data from my network not just one machine.
Seriously though, widespread use of the full bandwidth will probably not show up until 6-12 months after this hits the market. But it will come. It will be a competitor to eSATA.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
What they should have done, from the beginning of USB, was to have the connector truly symmetric, so that you could plug it in either way.
Connecting +5V to ground with a wire is inadvisable. The magic smoke is let out of the chip, which then ceases to work.
Seriously, how many connectors out there do you know of that let you plug it in any way you feel like? All connectors have to be oriented so that the signals and power goes to the right place.
Please do not come if I ask for someone to jump my car.
Remember, You are unique...just like everyone else.
Headphones, RCA Jacks, Car accessory power adapters, most laptop and wall-wart power adapters, F connectors, bnc connectors.
Do you want me to continue?
It really wouldn't have been _that_ hard to make it symmetrical or have a tactile key.
What they should have done, from the beginning of USB, was to have the connector truly symmetric, so that you could plug it in either way.
Connecting +5V to ground with a wire is inadvisable. The magic smoke is let out of the chip, which then ceases to work.
Seriously, how many connectors out there do you know of that let you plug it in any way you feel like? All connectors have to be oriented so that the signals and power goes to the right place.
Please do not come if I ask for someone to jump my car.
Trivial, my dear Watson - just have two sets of contacts on the male connector
that are centrally symmetric. This way regardless of orientation you have proper polarity.
If you are worried about EM properties of the connector make the host have two sets as well - this way you will not have dangling ends.
Very simple. Use an odd number of contacts plus shield, +5V on the center contact, GND on the shield, each differential data connection symmetrical around the center contact. Plugging that in the wrong way will only reverse the polarity on the differential signals, which isn't a problem.
Picture this, 8 pins in a row.
1 VCC
2 GND
3 D-
4 D+
5 D+
6 D-
7 GND
8 VCC
No matter which way you plug it in, the pinout would've worked fine. The additional cost to manufacture an 8-pin USB cabling system versus the current 4-pin scheme would've been negligible.
"you're reading for meaning" ... "where most people are pretty sloppy"
That's fine if people are trying to express "standard" or "normal" aka boring stuff.
If you are trying to communicate unusual meanings to somebody else, it doesn't work so well if you are sloppy.
On Slashdot I'm expecting the discourse to be on a higher level than "Me hungry. Want food", and that at least some people here will post stuff that is out of the ordinary and hopefully interesting.
In such a case, in addition to figuring out whether the writer made a mistake in spelling, grammer, you also have to figure out whether the writer made a mistake in reasoning, or is trying to be funny, or is saying something really _different_, or is saying multiple things at the same time, or is just plain crazy, or whatever.
It's all very easy to parse if you only have to expect people to say boring stuff.