Thread: computer 1 usb 1<====>usb 2 computer 2

  1. #1
    Registered User kryptkat's Avatar
    Join Date
    Dec 2002
    Posts
    638

    computer 1 usb 1<====>usb 2 computer 2

    computer 1 usb <====> computer 2 usb

    i bought a usb cable with two male ends. then plugged that in to an xp box and a vista box.

    nothing.

    so how come the computers do not identify them selves to each other ? if you buy a usb memory device it will identify itself to the computer. they will see each other over wifi but not usb ? think there is something not designed in to the computers that should be there. future suggestion. include usb to usb id.

    thought it would make a good quick null modem to transfer files from one computer to another.

  2. #2
    [](){}(); manasij7479's Avatar
    Join Date
    Feb 2011
    Location
    *nullptr
    Posts
    2,657
    I think those kinds of devices are available.(can't remember the name).
    All I remember is that it had a little cube somewhere in the middle.
    Some extra driver is needed though, and I could not get it working for a linux distro.

  3. #3
    Banned
    Join Date
    Aug 2010
    Location
    Ontario Canada
    Posts
    9,547
    You need a special cable for this... Computer-to-Computer USB Networking Cable | Overstock.com ... note the electronics pod in the middle. Something has to handshake with the usb ports before the computer will recognize them.

    Also be aware that with power in the connectors it is possible to damage both computers with the wrong cable.

  4. #4
    spurious conceit MK27's Avatar
    Join Date
    Jul 2008
    Location
    segmentation fault
    Posts
    8,300
    A modem is a serial/bit-streaming device. A usb port is more abstract, like an IDE or SATA connection. It represents a block device. So this is like saying, "Why can't I plug two hard drives directly together with an IDE cable to transfer data? Why does it have to go thru the computer?"

    Ie, you need some additional hardware and software to accomplish this (eg, the cables mentioned, which I am sure require software drivers to be useful).
    C programming resources:
    GNU C Function and Macro Index -- glibc reference manual
    The C Book -- nice online learner guide
    Current ISO draft standard
    CCAN -- new CPAN like open source library repository
    3 (different) GNU debugger tutorials: #1 -- #2 -- #3
    cpwiki -- our wiki on sourceforge

  5. #5
    Banned
    Join Date
    Aug 2010
    Location
    Ontario Canada
    Posts
    9,547
    Quote Originally Posted by MK27 View Post
    A modem is a serial/bit-streaming device. A usb port is more abstract, like an IDE or SATA connection. It represents a block device. So this is like saying, "Why can't I plug two hard drives directly together with an IDE cable to transfer data? Why does it have to go thru the computer?"

    Ie, you need some additional hardware and software to accomplish this (eg, the cables mentioned, which I am sure require software drivers to be useful).
    Yes they do.... But Windows since XP/SP2 has the driver in it's built in repository. If the OS has the data transfer module (used to move stuff from old computer to new computer) the computer to computer cable should work.

    Also, a little known USB trick is that a device can actually store it's own driver which is uploaded to the host as part of the Plug and Play handshaking sequence...

  6. #6
    Registered User kryptkat's Avatar
    Join Date
    Dec 2002
    Posts
    638
    Item#: 934629

    This cable provides the perfect Plug-and-Play solution for networking. It is USB 2.0 compliant for transfer speeds up to 480Mbps, but also works in USB 1.1 ports at USB 1.1 speeds!

    Sold out!
    This product is not in stock.

    We will notify you when this product is in stock. Please enter your email address below, then click 'Submit'.
    Nowhere on the product description, the packaging, the installation CD or on the cable itself is the manufacturer listed. This was a big problem when I found that the installation CD was blank, and could not load the drivers -rendering the product useless.
    serilously ? <bangs forehead on wall> why meow ?

    by the one and only review i would guess that the cable is older ie made before roxio changed the cd format which is why i think the cd is not blank just not readable with the newer format.

    [img]www.starmount.co.uk/usb_skta.gif[/img]
    this is only four lines. vc+ data+ data- and gnd so the data lines are only one way ? ok i see that the - needs to connect to the + data line ie one in and the other out line for data flow. i knewd that the xp had a driver for the usb but not sure if it would work connecting two computers.

    the 5v to 5v should not do any damage. it is when the v is diff is when probs happen.

    thank you for the replies and suggestions.

  7. #7
    Registered User
    Join Date
    Aug 2003
    Posts
    1,218
    What you try can not be done without any in the middle electronics. What you are trying to do is hook up 2 USB Host devices together, but the Host is expecting a Client device to be plugged in, so you have 2 Hosts waiting for a client to be plugged in. This has nothing to do with the wiring and everything to do with the specification of the USB protocol.

  8. #8
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    this is only four lines. vc+ data+ data- and gnd so the data lines are only one way ? ok i see that the - needs to connect to the + data line ie one in and the other out line for data flow. i knewd that the xp had a driver for the usb but not sure if it would work connecting two computers.
    NO! USB is NOT serial port (where you have to exchange Rx and Tx).

    D+ and D- are used in differential signaling. Meaning, they only carry one signal. Theoretically, USB can be made a 3 pin connector, with just 1 data line. They used 2 to improve high speed signal integrity. Differential signaling - Wikipedia, the free encyclopedia

    I don't know what will happen if you swap the 2 lines. In the very best case, it just won't work. Worst case is a lot worse.

    The USB protocol is host/client based. A device has to either be host, or client (there is a new class of USB devices that can work in what's called On-The-Go mode, as either device or client, but they can still only do 1 at a time).

    5V to 5V will most certainly do damage. In the real world, you'll never get 5V +-0%. In fact, I believe USB specifies something like 4.3V to 5.3V. One side will always be higher, and will be feeding current backwards into the lower side, and it probably won't like that. In the best case, the USB controller will just lock up and shut off the port. In the worst case...

    The USB standard says the host must supply power to the device, and the device must not feed current back into the host (draw 0-500mA depending on power negotiation, negative current is NOT allowed).

    As you can see, it's a REALLY bad idea to plug host into host.

    The USB people thought about that, so to make sure that never happens, the standard says all hosts should have Type-A ports (the USB port on your computer), and all devices should have Type-B ports (including mini and micro variants, which are electrically identical). And cables will all to A to B, and no one can plug anything wrong. If it fits, it's right. It's physically impossible to make wrong connections.

    And then of course some bastards decided to make A to A cables... like the one you have. They are designed for non-standard compliant devices, and should never be used to connect 2 standard-compliant devices (like PCs) together. For their one time convenience, they broke the robust protection built into USB standard, and now every unsuspecting person can ruin their devices with those cables.

    At very high level, how USB works is you have a host that awaits connection from up to 127 devices, optionally through a tree of hubs.

    Every time a device is connected, the host gives it a unique address. Each device identifies itself with 2 16-bit IDs - VID (Vendor ID, indicating the, well, vendor), and PID (Product ID, allocated by the vendor, to distinguish between their own devices). The OS then uses those 2 IDs to attempt to find a driver to talk to the device. If it's successful, the OS hands control to the driver that says it can handle the VID/PID pair.

    When the driver, through the host controller, wants to talk to a device, it sends the message along with the address to ALL devices. All except the addressed device ignores it. If the message is a read request, the addressed slave then starts transmitting back to the host.

    All transactions are initiated by the host. There is absolutely no way a device can initiate a transaction (it could lead to very bad things if multiple devices try to transmit at the same time). They only speak when spoken to.

    As you can see, there is no way 2 hosts can talk to each other. You should just throw out the cable (it has absolutely no use for you unless you have a non-standard device that happens to need it) and send unpleasant things to the manufacturer.

    The cables with an electronics box in the middle work because they have a microcontroller inside that talks to both computers as slaves (everything else is driver-level magic on the computer side).

  9. #9
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    Also, a little known USB trick is that a device can actually store it's own driver which is uploaded to the host as part of the Plug and Play handshaking sequence...
    Where did you hear that from? I have never heard of anything like that. Last time I checked, USB is completely platform-independent.

  10. #10
    Banned
    Join Date
    Aug 2010
    Location
    Ontario Canada
    Posts
    9,547
    Quote Originally Posted by kryptkat View Post
    the 5v to 5v should not do any damage. it is when the v is diff is when probs happen.
    You wanna try that again?

    If you have a simple wire cable --no interface chip-- you can have some major problems with the power connections...
    You're talking about a cable that most likely has an end to end resistance of less than 2 ohms... So, lets do just a little math, based on normal variations in supply voltage... Computer A's +5vdc supply is actually 5.25v (within tolerance)... Computer B's +5vdc supply is running 4.75v (also well within tolerance)...

    Ohm's law tells us that current = voltage / resitance ... so we can create a formula here...

    current = (5.25 - 4.75) / 2 = .5 /2 = .250 AMPS ... and that's only from normal, in tolerance, supply variations...

    You don't just simply plug stuff in, like that...

  11. #11
    Banned
    Join Date
    Aug 2010
    Location
    Ontario Canada
    Posts
    9,547
    Quote Originally Posted by cyberfish View Post
    Where did you hear that from? I have never heard of anything like that. Last time I checked, USB is completely platform-independent.
    Oh gees, it's been a while... I don't have the link anymore. But there was at least one chip design I saw with memory space for imbedded drivers.

  12. #12
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    Quote Originally Posted by CommonTater View Post
    Oh gees, it's been a while... I don't have the link anymore. But there was at least one chip design I saw with memory space for imbedded drivers.
    Ah, it probably enumerates as a composite device with 2 "virtual" interfaces. One is mass storage device that just contains the driver, and the second one is the real interface. This way driver can be delivered with the device, and computer will (initially) see it as a USB drive and an unknown device.

    I'm pretty sure that's not part of the USB standard, though. They worked very hard to achieve platform-independence and forward-compatibility. Allowing this in the standard would kill both things by tying a device to one (or a few) version of one (or a few) OS.

  13. #13
    Banned
    Join Date
    Aug 2010
    Location
    Ontario Canada
    Posts
    9,547
    Quote Originally Posted by cyberfish View Post
    Ah, it probably enumerates as a composite device with 2 "virtual" interfaces. One is mass storage device that just contains the driver, and the second one is the real interface. This way driver can be delivered with the device, and computer will (initially) see it as a USB drive and an unknown device.

    I'm pretty sure that's not part of the USB standard, though. They worked very hard to achieve platform-independence and forward-compatibility. Allowing this in the standard would kill both things by tying a device to one (or a few) version of one (or a few) OS.
    I've got to visit a few hardware sites over the next few days (audio parts) and while I'm surfing I'll try to locate that chip again... No promises.

    What you say makes complete sense to me but I'm not exactly a USB protocal expert...

  14. #14
    Unregistered User Yarin's Avatar
    Join Date
    Jul 2007
    Posts
    2,158
    Quote Originally Posted by CommonTater View Post
    You wanna try that again?

    If you have a simple wire cable --no interface chip-- you can have some major problems with the power connections...
    You're talking about a cable that most likely has an end to end resistance of less than 2 ohms... So, lets do just a little math, based on normal variations in supply voltage... Computer A's +5vdc supply is actually 5.25v (within tolerance)... Computer B's +5vdc supply is running 4.75v (also well within tolerance)...

    Ohm's law tells us that current = voltage / resitance ... so we can create a formula here...

    current = (5.25 - 4.75) / 2 = .5 /2 = .250 AMPS ... and that's only from normal, in tolerance, supply variations...

    You don't just simply plug stuff in, like that...
    While this may be true (and definitely a safe practice to follow)...
    I'm pretty sure that so long as both computers USB bus' are running off of the same reference voltage, it's perfectly safe to run a high logic voltage from each computer directly the other, head on. (In the case of high to low logic, it should also be fine given adequate short circuiting protection)

  15. #15
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    You mean both computers running on the same PSU...?

    It's also not logic voltage. Logic voltage for USB is much lower. 3.3V for low and full speed, something lower for high speed.

Popular pages Recent additions subscribe to a feed

Similar Threads

  1. Computer Science vs Computer Engineering degree
    By PCG33K in forum A Brief History of Cprogramming.com
    Replies: 4
    Last Post: 10-03-2007, 07:13 AM
  2. PHP on my Computer!
    By xxxrugby in forum Tech Board
    Replies: 4
    Last Post: 03-15-2005, 09:34 AM
  3. Omg, My Computer!!!
    By jverkoey in forum Tech Board
    Replies: 20
    Last Post: 03-04-2003, 09:48 PM
  4. Replies: 23
    Last Post: 01-31-2003, 03:13 AM
  5. computer science or computer engineering??
    By pkananen in forum A Brief History of Cprogramming.com
    Replies: 8
    Last Post: 02-26-2002, 12:10 PM