Talk:USB/Archive 5

Another non-standard connector
Yet another non-standard but compatible plug is found on Sony's MicroVault series of USB solid-state memory devices. The pins are USB-compatible, but it lacks the metal shroud around the plastic part that is part of the standard plug. The purpose of that is to make the drive thinner, which it does, but it also makes it possible to insert the plug upside-down, or into the part that the upper part of the shroud goes into. If it is inserted wrong, nothing happens because the pins do not connect.

The device itself (and the packaging it came in, IIRC) does not have the USB(tm) logo because it is not conforming with the specs, but it does work with USB. Pic here for reference: http://www.66mobile.com/news/products/4GB-Sony-Micro-Vault.html I'll make a free pic of my own to add later. Identity0 (talk) 23:13, 16 October 2008 (UTC)
 * I have seen that basic style from a couple of other manufacturers previously. Plugwash (talk) 23:44, 16 October 2008 (UTC)

Green plug
Should this item be included in the power section of this article, or should a new article be written for it? Jim.henderson (talk) 16:42, 10 November 2008 (UTC)

Power Section
I have made a few minor changes to the Power section. The terms 'unit load' and 'load unit' were mixed and the wording was confusing. I have not added or changed any information. 150.101.206.3 (talk) 04:10, 8 December 2008 (UTC)

Confusing sentence
"Did not anticipate or pass-through monitors." To parse this, I have to conceive of a "monitor" as some kind of data event on the wire, and the standard as lacking in ESP mode. What does it really mean? -- Dispersion (talk) 19:49, 17 December 2008 (UTC)

Maximum Cable Length
Apparently... according to this page the maximum length of a cable is 5m due to some roundrip delay blah blah bullshit, but I am sitting here right now using a $3 cable that is well over 10 metres long which works fine. I think wikipedia is on the lies!!! —Preceding unsigned comment added by 59.167.100.191 (talk) 08:52, 6 January 2009 (UTC)
 * You see your computer must be more modern, maybe Vista or the newest mac OS. Some older computers require the response time of a USB device to be lower than a predefined threshold. And that particular threshold is achieved around 5.5m - 6m. All of us here are very proud that you know how long your USB cord is. From what i can see the article has been changed back to '5m'. I did not do this, but i think this is more correct, please let me know if you want to change back to 10m and we can debate about it here. Thanks! Mod.torrentrealm (talk) 11:37, 8 January 2009 (UTC)


 * One other thing I'd like to point out here. I'm not a USB expert by any means. But I do know how cables in general work, and how signal is passed through from one end to the other. In any specification, it will list what I'd refer to as the worst scenario. Different types of USB devices have different levels of demand when it comes to signaling. For example, something low-powered like a USB thumb drive would probably work with that 10m cable just fine. But use the same cable with something high-powered like a high-definition camcorder or an 80GB hard disk drive, and you will probably either see a dip in speed, or maybe even data loss. When consulting someone on what to purchase, I'd trust the 5m much more than the 10m. Groink (talk) 23:47, 8 January 2009 (UTC)

Image copyright problem with File:Otg usb.png
The image File:Otg usb.png is used in this article under a claim of fair use, but it does not have an adequate explanation for why it meets the requirements for such images when used here. In particular, for each page the image is used on, it must have an explanation linking to that page which explains why it needs to be used on that page. Please check


 * That there is a non-free use rationale on the image's description page for the use in this article.
 * That this article is linked to from the image description page.

This is an automated notice by FairuseBot. For assistance on the image use policy, see Media copyright questions. --23:13, 6 January 2009 (UTC)

USB 3 / SuperSpeed Connectors
I've put photos of the SuperSpeed connectors shown at CES on the Commons, they do need cropping but could then be added to this article. See and. --BenFranske (talk) 02:13, 15 January 2009 (UTC)

Topology
The article states:

USB networks use a tiered-star topology, while FireWire networks use a tree topology

while linking to Network topology (as a redirect for Tiered-star). The network topology article doesn't mention "tiered-star", and I don't understand how a "tiered-star" topology differs from a "tree" topology. -- 74.137.108.115 (talk) 01:10, 22 January 2009 (UTC)

Mini Micro Section
I added and edited some references in the Mini Micro section. CaptainVideoJW (talk) 06:15, 14 January 2009 (UTC) I have a Bluetooth adapter that uses this micro USB 8 pin connector (also called "USB ON THE GO", "Agox connector", "mystery micro plug", "Mini USB 8P", "Mini B 8 Pin", "Mini 8 Pin", "8 pin flat", "I-USB7", etc). I need to fabricate a custom cable and I already figured out the pinout, but I cannot find the connector. Where can I buy this connector? Or the 8 wires cable with this connector at the end? —Preceding unsigned comment added by 99.253.208.187 (talk) 17:10, 8 February 2009 (UTC)

I'm a little confused by the claim that the micros are smaller than the minis when the measurements on the minis seem to be smaller. What's correct?--Tskandier (talk) 20:29, 11 February 2009 (UTC)


 * Yeah, the dimensions in the text are definitely wrong right now. Good eye.  What we really need is a diagram comparing them... ǝɹʎℲxoɯ ( contrib ) 21:26, 11 February 2009 (UTC)

Correlation of USB standards and Host Controller Interfaces
While editing Host Controller Interface (HCI) -- my simply adding a link to the USB page, there -- I noticed: It appears that there is a correlation indicated, between one standard for host controller interfaces (namely, EHCI) and one specific USB standard level -- namely, USB 2.0.

Though I am not familiar with the details of the USB standards, myself, I wonder if someone who would be already familiar with the said standards and their applications -- namely, someone with some reasonable experience about it -- that someone, as such, may be able to indicate this correlation, indicating it at least on the USB page. I propose that it would serve to clarify that there is such a correlation, and that it would serve to explain the relation, thereby further explaining the subject of the Universal Serial Bus specifications, overall.

As a second pip on the wish-list, maybe someone could fix up the HCI page, commensurate with such clarification on the USB page?

Whereas these two subjects -- so I understand -- these subject are closely and (at some semantic levels) are exclusively related, namely USB and HCI, perhaps it would be simply reasonable if the two subjects' respective pages would be made to be integrated -- at least, in terms of the textual content of each respective page, if not also in terms of structure and containment of the content, within the wiki.

Gimmal (talk) 15:45, 24 January 2009 (UTC)

date in history section wrong
The history section says USB 1.0 was introduced in 1906. I don't know much about USB, but I'm sure that's too early. Anyone?--Zachbe (talk) 14:25, 17 February 2009 (UTC)


 * No, it's not correct; that date was graffiti. I've reverted the article to the previous version. —Sasuke Sarutobi (talk) 15:29, 17 February 2009 (UTC)

Full speed?
Near the end of the USB 3 section:

"However, electronicdesign.com estimated cables will be limited to 3 m at full speed."

This might be confusing due to the use of the term 'full speed'. —Preceding unsigned comment added by 74.202.89.125 (talk) 17:19, 17 February 2009 (UTC)

Dimensions
What are the dimensions of the various plugs? That's mainly USB vs. mini and micro (I feel that micro is not smaller than mini), that's type A vs. B vs. AB and that's the "male" part vs. the full plug size. Apart from that, a merge of various info might be interesting: plug cycles, power (e.g. up to 1.8 A on micro USB, pins 1+5), pin count. --Traut (talk) 19:42, 4 March 2009 (UTC)

How do you distinguish between USB 1.1 and 2.0 Cables?
I have about 40 or 50 cables and most are not marked. Some have a 2.0 labeled tag, a few have high speed 2.0 printed on the cable. But most have no label or marking at all. Or Does it matter? 66.32.83.218 (talk) 18:11, 5 March 2009 (UTC)

What type of cables are these? USB to Mini USB? I would guess it doesn't matter, but I'm not sure. The worst that can happen is that the device will work slower, or (possibly) not work at all. Photographerguy (talk) 03:36, 6 March 2009 (UTC)

From my understanding, and several sources, the cables are identical and should not matter. It is the hardware and controllers which are rated, not the cables (similar to Ethernet/IEEE 802·3). Clear information in the article on this subject would disambiguate the issue. Lee Carré (talk) 11:45, 15 March 2009 (UTC)

I have found the specs for the cable and according to them the two cables are interchangeable, if they were manufactured to the rev. 1.1 specs. [user: paul mychaluk —Preceding unsigned comment added by 141.151.16.147 (talk) 18:09, 16 March 2009 (UTC)

Licencing / Patents / Royalties
Information on these topics would seem to be rather important, in the same way that restrictive licencing on HDMI interfaces is an important factor. Lee Carré (talk) 11:53, 15 March 2009 (UTC)

USB protocol analyzers — examples
For USB protocol analyzers, surely some examples are needed? Both of software packages (Wireshark comes to mind), and methods for actually having software capture data transversing the interface. Lee Carré (talk) 12:05, 15 March 2009 (UTC)

3.0 Specs
According to, USB 3.0 transfer speeds are "...up to ten times faster than Hi-Speed USB (USB 2.0)..." This should be noted as 4.8 Gbps (600 MB/s) vs. 5.0 Gbps (625 MB/s) as indicated in the article. —Preceding unsigned comment added by 136.234.15.189 (talk) 17:14, 27 April 2009 (UTC)

Good work people
Only came here to find out max cable length as I'm trying to get a good signal from 3g usb broadband dongle. Found it so cheers, cracking article as a whole all involved without picking at the accuracy of the details. All aspects of this connector seem to be covered and I'm surprised this isn't a featured article. Thanks again all who work on this! Mark —Preceding unsigned comment added by 149.254.219.40 (talk) 09:18, 5 May 2009 (UTC)

What Micro USB will be the new standard?
What Micro USB will be the new standard for cell phones? Micro-AB or Micro-B? (note that both type of plugs fit into micro-AB sockets, so to use micro-AB sockets would be logical) Pizzaman79 (talk) 12:17, 14 May 2009 (UTC)

Inventor credit
As an encyclopedia, shouldn't inventor credit be important? How about naming the actual engineers who developed the technology, such as Ajay Bhatt? --68.45.218.70 (talk) 16:49, 25 May 2009 (UTC)
 * although giving credit to inventors is very important for an encyclopedia, i think nowadays teams of scores engineers and scientists like ajay bhatt must have worked on a new technology like usb. is it practical to name ALL of them on wikipedia? —Preceding unsigned comment added by Perryizgr8 (talk • contribs) 17:07, 25 May 2009 (UTC)

Real world USB 2.0 speed
On the bottom of section 4 USB signaling there is mention of real world speeds of USB 2.0 being around 2/3 of 53MB/s (about 35MB/s). I think this is accurate as I have never seen anything perform significantly faster than this... but not much attention is drawn to this. Most literature says 480Mb/s or 60MB/s which is just not true in practice. I don't know why it is so slow in real life, but if someone knows, that would be useful information to include. 131.203.80.9 (talk) 01:55, 17 June 2009 (UTC)

USB Plus + ?
On some newer laptops I noticed that instead of the standard USB logo they have one with an extra plus sign at the side of the central arrow. Anyone know what it means ? --Xerces8 (talk) 12:30, 22 June 2009 (UTC)


 * The port provides power even when the laptop is off. SchmuckyTheCat (talk)

Kodak U-8
Is the U-8 socket on some Kodak cameras a proprietary format or just a confuing name? Either way, if someone has a good explanation it seems to belong on this page. 85.210.67.232 (talk) 22:07, 24 May 2009 (UTC)


 * Maybe U-8 is "8 pin AGOX":
 * 8 pin AGOX pin-assign
 * 1 | GND
 * 2 | (Out) Composite Video
 * 3 | (Out) Audio
 * 4 | (In) AV Cable Detect
 * 5 | (In/Out) USB 2.0 Data+
 * 6 | GND
 * 7 | (In/Out) USB 2.0 Data-
 * 8 | (In) USB Cable & Host Detect
 * ¥---+ Cable Shield
 * http://www.mi-fo.de/forum/index.php?showtopic=22787
 * --220.210.186.56 (talk) 05:53, 2 July 2009 (UTC)
 * --220.210.186.56 (talk) 05:53, 2 July 2009 (UTC)

It's similar to 8 pin AGOX. It looks like they switched some of the pins around. 199.0.170.34 (talk) 21:46, 10 August 2009 (UTC)
 * http://www.allpinouts.org/index.php/8-pin_AGOX_Connector

USB signaling
The article reads:

The full speed rate of 12 Mbit/s is the basic USB data rate defined by USB 1.1. All USB hubs support full speed.

¿Why not USB 1.0? 62.82.184.129 (talk) 21:36, 13 July 2009 (UTC)


 * USB 1.0 and USB 1.1 are essentially equivalent. USB 1.1 is the one that actually got used, so most specifications and references are to that version. SchmuckyTheCat (talk)

Upgrade Ability
Does anyone know if the 3.0 spec will require new hardware, or will some of the current equipment be firmware upgraded? --SDSpivey (talk) 03:11, 13 June 2009 (UTC)
 * USB devices are built to support a specific speed. They won't get magically faster and they won't stop working. If you want a 3.0 device, you'll need a 3.0 controller. SchmuckyTheCat (talk)
 * USB 3.0 devices will connect to and work with prior standard devices and cabling, but will not support USB 3.00 speeds in doing so. So backwards compatible pretty much (non-standard devices and cables are all over the lot and may cause either trouble (blown circuits), non operation, or work just the standard stuff mostly). But STC is correct that existing devices won't become 3.0 complicant just by connecting to a 3.0 device or cable. At least the standard stuff you've got will keep working as before. ww (talk) 03:45, 17 August 2009 (UTC)

dangerous loop?
"This two-connector-type scheme (A/B) prevents a user from accidentally creating a potentially dangerous electrical loop." The article does not explain HOW the distinction between A and B plugs would prevent the user from creating a loop. I'd find it a good idea to add an additional sentence to explain this, if anybody can?

Furthermore, "This two-connector-type scheme (A/B) prevents a user from accidentally creating a potentially dangerous electrical loop." contradicts this: "A USB connection is always between a host or hub at the "A" connector end, and a device or hub's "upstream" port at the other end. Originally, this was a "B' connector, preventing erroneous loop connections, but additional upstream connectors were specified, and some cable vendors designed and sold cables which permitted erroneous connections (and potential damage to the circuitry). USB interconnections are not as fool-proof or as simple as originally intended." --boarders paradise (talk) 13:20, 16 August 2009 (UTC)

USB types
I have been looking at "thumb drives" for my computer. I would like to find out how to tell if my computer has a 2.0 USB's or 1.0 USB's? I have checked the "SYSTEM" tab in my control pannel but the information listed does not tell me if I have 2.0 or 1.0 USB ports. —Preceding unsigned comment added by 71.200.175.28 (talk) 23:54, 30 August 2009 (UTC)

Limit on number of devices
I've just tagged as dubious the following statement:


 * Although the device address is a 7 bit value, giving a theoretical 127 devices on the bus, the host's address and the bus reset address (a binary zero) must be subtracted, making the maximum permitted number of devices connected to a single host interface a total of 125.

I'll admit that I haven't checked the USB spec but from memory this does not ring true. For a start a seven bit value gives 128 possible combinations: that is, 1-127 plus zero - you do not need to reserve one of the 127 non-zero combinations to be a zero. In addition, USB is strictly a host-controlled protocol: the host initiates all transfers (even "interrupts") and all transfers go either to or from the host - there are no direct device-device transfers. As such the host does not need an address assigning to it since this is always implicit from the comamnds issued controlling the bus: that is, a "read" will always have the host as the destiantion and a "write" the host as the source. Sicne this is always the case it does not need to be stated in the protocol. It is also (on the go notwithstanding) why you can't connect two USB hosts together with a simple patch cable: both play the role of host and confuse each other since their is no way to differentiate between the two. CrispMuncher (talk) 21:54, 12 September 2009 (UTC)


 * In the absence of any objections I am removing those comments now. CrispMuncher (talk) 16:52, 15 September 2009 (UTC)

When was the USB 3.0 specification released?
According to http://www.usb.org/developers/docs/, one of the article's two sources called from the last paragraph of section 1 "History", the date was November 12, 2008. Since there's no explanation for the sudden change to November 15 by 209.250.175.238, I'm reverting it. It looks like vandalism, especially given the recent history of the article. 212.166.222.244 (talk) 15:42, 28 September 2009 (UTC) also is very simple especialy to sipmlyfy work —Preceding unsigned comment added by 41.220.141.2 (talk) 08:51, 2 October 2009 (UTC)


 * Given the only other edit by that IP, a couple of minutes later, I think your assumption of vandalism is reasonable. David Biddulph (talk) 10:55, 2 October 2009 (UTC)

New Infobox
I think all the device Infoboxes should be updated with a creator/inventor's section whenever possible. These types of articles are incomplete without that basic information. --68.45.218.70 (talk) 18:15, 10 October 2009 (UTC)

The USB 1.0 specification was introduced in 1996. It was intended to replace the multitude of connectors at the back of PCs, as well as to simplify software configuration of communication devices. The original USB 1.0 specification had a data transfer rate of 12 Mbit/s. —Preceding unsigned comment added by 201.144.102.120 (talk) 15:27, 19 October 2009 (UTC)

ASUS Motherboard with USB 3.0 Review at theregister
See: http://www.reghardware.co.uk/2009/10/29/asus_superspeed_mobo/ —Preceding unsigned comment added by 80.68.93.104 (talk) 15:34, 29 October 2009 (UTC)

OSI Layer template / box
Why is this box present on the page though it neither contains any protocols of USB nor the article refers to it? - forky 62.225.105.244 (talk) 11:26, 30 October 2009 (UTC)


 * I tend to agree, I've experimented with removing it and it seems to, at least, improve the layout. Less 'white space'--220.101.28.25 (talk) 17:28, 31 October 2009 (UTC)


 * Yes, I think it can go. Presumably the argument for its inclusion is that it is a physical layer link but USB is not a host-host link, so I think that is pushing things a bit far.  More generally I have seen many other pages tagged as netowrking topics in one way or another (infoboxes being a common one) where the connection is tagential to say the least. CrispMuncher (talk) 18:47, 31 October 2009 (UTC)


 * USB doesn't even appear on the OSI Layer box. CrispMuncher, I take it you mean that the OSI layer model barely applies to USB? Other than as physical connection? But it has a data protocol of some sort doesn't it? I'm not an expert on networking, but I think I understand the concept of the OSI model. Maybe we need a mention of the applicable OSI layers, if any?


 * Quick side question. Is it standard Wikipedia practice to NOT have an 'Introduction' heading? ie I can't just click on [edit] (section edit).I have to click on the 'edit this page' tab and open the entire page to remove the OSI 'box'. This applies anytime you need to edit the intro. Seems to slow the editing process down a lot.


 * If there was an 'Intro' heading then the contents listing would also be at the very top of the page. Comment please?
 * --220.101.28.25 (talk) 03:10, 1 November 2009 (UTC)

Refimprove template
This article has a template, and also has 64 inline citations. If more inline citations are needed, the need must be more specific. Perhaps the details on USB 3.0 might be improved. --DThomsen8 (talk) 16:27, 8 November 2009 (UTC)

USB 3.0 PCI cards
As of September, 2009, USB3.com is shipping USB 3.0 PCI cards. This URL has more details. Of course, the cards and cables are only going to be useful when there is a variety of devices available to use with the higher speed made available. --DThomsen8 (talk) 16:36, 8 November 2009 (UTC)


 * [| http://www.usb3.com/]

Interested editors should follow up on this lead. --DThomsen8 (talk) 16:36, 8 November 2009 (UTC)

Power + Data Charging?
An increasing number of my devices (my Samsung mobile phone, my Sony eBook, and I think iPods/iPhones) require power on the data lines in addition to the USB power lines in order to charge. What is that standard (if indeed it is a standard) called? This would be extremely nice to know, in order to find a charger that is compatible with them. Several AC/Auto to USB adapters I have will not charge these devices. &mdash; Brianary (talk) 21:57, 24 November 2009 (UTC)
 * That isn't a standard and is probably a proprietary charger. SchmuckyTheCat (talk) 04:30, 25 November 2009 (UTC)

Maximum useful distance
I removed a section describing "special extender cables" that clearly referenced UNPOWERED 1 port hubs used as repeaters. This section was restored saying "please look at the spec section 7.1.19.1" First, that link to the spec is obsolete as it is in the USB 2 spec. The USB spec is at version 3. Second, the section is about signal propagation, not power. The illustration shows a 5 hub layout. Those hubs would be required to be POWERED hubs. The rule of thumb is "You can't connect a bus powered hub to a bus powered hub". A bus powered hub plugged into a bus powered hub would work, but only the hub function would work, no downstream ports would be powered. A bus powered hub can only provide one unit load of power to a downstream port, if you connected five bus powered hubs in a row, you would run out of power on the second hub.

The reality is that these extender cables falsely report as self-powered hubs (which violates the spec). A self-powered hub can provide 5 unit loads to any port. This allows these extender cables, 5 of them in a row, to keep being powered even though all power is coming from the bus. These devices violate the spec, and so does the layout described in this section of the article. SchmuckyTheCat (talk)


 * Oh dear.


 * USB2.0 is still a current standard, it has not been obsoleted. USB3.0 explicitly includes the USB2.0 spec at section 1.3 (scope of the document).  In any case, how is this relevant to the substantive issue?
 * Where does section 7.1.19.1 of the USB2.0 spec make any mention of how the hubs are powered?
 * Where does the text you have removed from the article state that the hubs are unpowered? On the contrary, it explicitly refers to the bus power.  Bus powered != unpowered but "no external power" still holds.
 * Where does anything state that an bus-powered hub is unable to provide only one unit bus power? Quite the oppsite.  The is a large section specifying how devices and the hubs they are connected to are able to negotiate the power usage while remaining within the spec.  The common unpowered 4 port hubs are unable to provide more that one power unit to each of their ports becasue 100mA has to be available to each port.  However this does not apply to a single port hub.  Bus power is intelligently negotiated, not a simple 1 unit/5 units all-or-nothing. If the repeater can draw 500mA in total it may take the full amount, take 100mA for itself and pass 400mA downstream.  The second device is permitted to draw 4 units, of which it may take another 100mA for itself and pass 300mA downstream and so on. Provided each device in the chain negotiates for and is granted the power it receives this is entirely within the spec.  Indeed, purely from a power perspective, since the power is actually negotiated in 2mA units there is theoretically scope for more than 4 repeaters in the chain: if each hub knows that it does not use more than e.g. 20mA the last in the chain may negotiate for 120mA and so on, but there are architectural limits to how deep hubs may be nested.


 * I'll be very interested to hear your responses. I suspect you have an overly-simplistic view of USB power management. CrispMuncher (talk) 21:03, 1 December 2009 (UTC)


 * "External ports in a bus-powered hub can supply only one unit load per port regardless of the current draw on the other ports of that hub." Which is part of the definition of bus-powered hub in section 7.2.1 of the USB Spec V2. A bus powered port should never support a high powered device plugged into it (though many do ganged power, so it works in the unsupported configuration).  You should also note in the power definition of a high-power bus-powered device that the device must enumerate, but does not have to function, at one unit load specifically to handle bus-powered hubs.  You should note again the USB-IF compliance testing for hubs never requires testing of greater than 100ma to an external port from a bus-powered hub.  The compliance testing requires trees of hubs, but no tree testing layout has daisy-chained bus-powered hubs.  These are a critical absence, if either daisy-chained bus-powered hubs, or supporting greater than 100ma output on a bus-powered external port, was a supported configuration there would be compliance testing for that. Another link for you is the USB-IF developer FAQ about cables.  Question 3 is this exact question.  Daisy-chaining bus-powered hubs isn't a solution, and they would have listed it as a solution to this question if that was a supported configuration. SchmuckyTheCat (talk) 00:43, 2 December 2009 (UTC)