USB 2 system behaves as if it's USB 1.1

[Dumb_Question]

Not sure if this should be in software...

Why is my computer giving me the message, for a long time now, on inserting any USB device, 'This device can perform faster....' ?

Detail
I get this message with every different device I hang onto a USB port, irrespective of which socket (USB port) on my PC I use (there are 6)- "This device can perform faster. If you connect it to a Hi-speed USB 2.0 port. For a list of available ports, click here" 

Clicking here does nothing.  I read somewhere (probably the HP website before they removed the specifications info on my PC) the my computer had USB 2. The m/b specification says "Two front, four back (ver. 2.0)".
I'm also attaching a screenshot of System Properties>Hardware>Device Manager, which I don't really understand, but at the bottom it definitely says "USB 2.0"...(PCI related entries).(H:\ScrSh_HWM_USB.jpg)

I have looked in BIOS setup, and I did not find anything about USB except legacy, set to auto/enable/disable.  I tried all of these without any change.  I have tried updating the drivers: I got the message from the update wizard that it couldn't find a better match than the drivers I already had.  I have tried re-installing the drivers for the three Universal Host Controllers and the three Root Hubs, this resulted in a change in the driver from 5.1.2600.0 to 5.1.2600.5512, but the message doesn't go away and there is no change to Hi-speed USB 2.0 operation.
I have the intel 845G/GE chipset and the intel driver finder tells me that I do not need to update that (driver is 'valid').  The host controller is an ICH4 intel 82801DB/DBM, which supports USB 2.0

One thing I have noticed is that under the first USB Controller, where the USB external ports seem to live, there is not (never, under no circumstances) an Enhanced controller or a USB 2.0 controller listed.  On my Pavilion a405.uk which has an extremely similar spec (m/b MS-6577 v3.1), which displays no message when connecting a USB device to it, the Universal Host Controllers and root hubs are listed, but there is an additional entry under USB Controllers for an Enhanced USB 2.0 Host Controller.  This is absent on my Presario.  Perhaps if I could make that entry appear it would give me USB 2.0 operation ?

Could anyone help me get my USB 2.0 system back up to speed ?

Dumb_Question
27.February.2014
Compaq Presario S5160UK DT261A under XP/SP3
Processor - Celeron 2.7 GHz
Motherboard - MSI MS-6577 v2.1

[recovering disk space, attachment deleted by admin]

[DaveLembke]

I have seen this message before with an older 2002 era motherboard also with the Intel 845GL chipset of my eMachine with Trigem Imperial 2002_1111 motherboard, and I now ignore it. I performed a USB speed test by copying a 100MB file to my 80GB external harddrive and used a stop watch. I then crunched out the math with time it took to transfer the 100MB file, divided by 100MB and the math showed that it was transferring at around 17MB/s, and USB 1.1 maxes out at 1.5MB/s, so its a bogus message from Windows XP. I also get this message when using a Belkin USB 2.0 HUB with a Laptop that also is USB 2.0 and I haven't done a speed test on that yet, but data seems to transfer the same speed with it connected or removed.

So it took 5.9 seconds to transfer 100MB and it comes to 17MB/s

If it were USB 1.1 it would have taken about 66.6 seconds to transfer 100MB

[Geek-9pm]

This is your motherboard - Right?

Do you have a free PCI slot? A simple upgrade is buy a USB card the fits into a free PCI slot. That would give you 4 USB 2.2 ports.
One example.
http://www.usb-pci.com/
You can also find similar items on Amazon and Newegg or even eBay.
USB 2.0 is High Speed 480 Mbps
Expect to pay about $20 plus shipping.

[Dumb_Question]

->Geek-9pm

Thank you for your response.  That certainly looks like my m/b, though the v2.1 is visibly indistinguishable from the v2.x and the v3.1.  If you did not find this on the HP website, I'd be grateful for a reference to your source of information on this m/b.

In fact, I have a PCI card such as that you refer to, with 4 external USB ports, 2 IEEE 1394 ports, 2 internal USB ports, 1 IEEE 1394 port, made by Sumvision (pictured).  It is described as having USB 2.0 ports. I bought it to save me having to forever be unplugging/plugging in various USB devices that I use (or not as the case may be !) but I've never used it.  A constructive suggestion, and it reminded me of the existence of the card.  I feel that it offers a way round the problem, but does not solve it.  Thank you.

Dumb_Question
27.February.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[recovering disk space, attachment deleted by admin]

[BC_Programmer]

This message can appear if if you plugin in a USB 1.1 device, (in which case it can't actually perform faster); if the device requires more power than it can draw from the port, is damaged to cause it to try to draw extra power or is otherwise defective; if the cable or port is damaged, or if the BIOS settings have been changed to 1.1.

[Dumb_Question]

->BC_Programmer
Thank you for your input.

None of the devices I connected are USB 1.1; it is very unlikely that all of them could be.  For example, I have just doubled checked the spec on my Canon Ixus60 camera - http://www.canon.co.uk/for_home/product_finder/cameras/digital_camera/ixus/digital_ixus_60/index.aspx?specs=1 - "USB 2.0 Hi-Speed".

I do not believe I am overloading the USB system.  I checked the power tabs on the root hubs and were well within the 500mA limit. 
 
It is again unlikely that all six external USB ports are damaged, though it is a remote possibility that something common inside the case on the m/b might be damaged, affecting all the ports; but how could I check that ?

My USB flash drives don't have cables.

About the BIOS settings, I tried to explain what was available in the BIOS set up menu in my initial post; there is no option to choose USB 2.0 as there is in some BIOS.  If I could change the setting from USB 1.1 to USB 2.0 I would.  Can you suggest a way of doing this that is possible on my system ?

Dumb_Question
28.February.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Dumb_Question]

-> DaveLembe

I had been ignoring this issue for a very long time.  I was able to leave stuff transferring for many hours and it did not matter.  It would take about 10 hours to transfer ~10GB from my internal HD to an external one.  That works out at about 2.2 Mb/s, which compares to nominal maximum of 12 Mb/s for USB 1.1 and 480 Mb/s for USB 2.0  I am sorry for not having more precise data (maybe do more careful measurement tomorrow).

I have been doing a bit of HD disk checking lately, with HD Tune amongst others, and when doing an error scan, my Presario scans at a reported rate of 0.9MB/s; this compares with my Pavilion a405.uk, which does not show this fault , which scans at 20MB/s.  When checking a 500GB disk, you can understand that a factor of 20 in such a scan time make a big difference.  HD Tune's error scan on my Presario seems to conk out after a few hours and after this it scans much faster and detects (falsely) ONLY bad blocks: it looks like all my disks have 90% bad sectors.  This does not happen on the Pavilion.  These are two more reasons why I would like to fix this problem  (the other being something not working properly for unknown reasons irritates me)

Dumb_Question
28.February.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Geek-9pm]

Dumb_Question, you should change your user name.

Back to topic. The photo was from HP
I think if you use your  PCI USB card the performance will close to USB 2.0 standard. If so, that would indicate the issue is not controlled or limited by the BIOS.

The built-in USB ports are part of the motherboard chip set.  They are often called 'north-bridge' and 'south-bridge'  chips. That are rather large chips that are soldered into the motherboard. If they are damaged, there is not feasible repair.
From HP for v3
http://h10025.www1.hp.com/ewfrf/wc/document?docname=c00100539&cc=us&dlc=en&lc=en

On eBay V2.x  motherboard goes for about $40 including shipping. Used, of course. Here in the western world you will not find a technician who will repair the board for f $40 or less.  {Maybe in South East Asia. }

Your may wish to consider a newer motherboard to improve your  investment.
Best I can offer. 
Thee is a very small possibility that a electrolytic capacitor on the mother bard has failed and it could be replace. But his is so dubious that it is not wroth the effort or the risk  of replacement. This is not base on my experience, rather the comments of more experienced  technicians elsewhere. Pragmatically, as one way to pout it, you are not able to repair the  motherboard

[BC_Programmer]

Apparently there are Some motherboards which flake out and their on-board USB 2.0 stops working. It can also happen if you are using certain incompatible peripherals which can force the entire bus to operate as USB 1.1 (A mouse, keyboard ,etc).

The link is for a person having an issue with a laptop that uses the chipset, but I'd imagine any pre-fab that used that chipset could have this problem.

You could try removing absolutely every single (uninstall) USB related item in Device Manager and rebooting, as well. But aside from that it looks like you ought to simply use the the PCI card.

[Dumb_Question]

->BC_Programmer

That's a very interesting post.

Maybe my USB kb/mouse transceiver is USB 1.1 and is forcing the rest of the USB ports to act as USB 1.1's.  I'll have to try with my PS/2 stuff, and check the specs for kb/mouse if I can find any.  The transceiver is the only item that could possibly not be USB 2.0 though those components are replacements for the original PS/2 mouse/kb

I did not delete the USB items which were labeled as USB 2.0 but related to PCI  in device manager (two of them, see screenshot), may be I should try deleting and reinstalling them as well.

It might just be my board.  Do you mean flake out due to passage of time, they simply give up  for no apparent reason ?  Aren't all m/bs whether in laptops or desktops 'pre-fab' - no-one builds their own systems from scratch (components, chips, resistors, capacitors etc) these days do they ?  So I regard it as immaterial whether the pc is a laptop or a desktop in that sense.  Laptops are subject to more wear and tear and stress maybe.

->DaveLembke
I measured the transfer rate from my HD to an external one for 490498944 bytes; it took 8 mins 16 secs, a transfer rate of ~7.9Mb/s, good for USB 1.1, pretty poor for USB 2.0

Dumb_Question
1.March.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Geek-9pm]

Referenced for anybody want to pursue this:

Specs for the HP motherboard of the OP.

Chipset "North Bridge" is a  SiS 651
Chipset "South Bridge" is a  SiS 962

Spes for SiS ship sets

Something is wrong. I give up

http://www.tomshardware.com/forum/151511-30-enabling-usb2-645dx-chipset-98se     1998 ? Time travel   

[BC_Programmer]

Do you mean flake out due to passage of time, they simply give up  for no apparent reason ?

Based on search results the issue is not uncommon and seems to be a problem with the chipset. Most of them say "over time" or mention it working at first, and then not working.

Aren't all m/bs whether in laptops or desktops 'pre-fab' - no-one builds their own systems from scratch (components, chips, resistors, capacitors etc) these days do they ?  So I regard it as immaterial whether the pc is a laptop or a desktop in that sense.  Laptops are subject to more wear and tear and stress maybe.

I was referring to prefabricated systems as in pre-built. Manufacturers like Dell, HP, etc. They do not manufacture their own components however they do get heavily lobotomized components, especially motherboards and add-on cards, with many components stripped out to reduce costs. I've had systems that have a chipset which supports USB that do not have any USB Ports at all. Even though the chipset used supports something doesn't mean the surrounding components to add that feature aren't cheap or trashy. In comparison motherboards sold on their own with a chipset will typically support all the features of the chipset

If you haven't done so, I would recommend uninstalling all the USB related devices (Root Hubs, Host controllers, etc) from device manager and then rebooting. It should redetect them all and hopefully the installation will fix any driver configuration error that could cause a problem like this.

Actually, I Just read back over and saw you specify the specific motherboard, whereas I've been searching about for the chipset, which is of course used on different motherboards.

MSI MS 6577 v2.1

It looks like it definitely has USB 2.0 capabilities.

If so you might get somewhere by uninstalling all USB-related devices in device manager, then rebooting. On startup Windows will redetect and reinstall the appropriate drivers. The reason this could help is that in the process of reinstallation it could fix a driver misconfiguration.

[Dumb_Question]

-> Geek-9pm
Thank you for your efforts, Geek.  I intended to answer the previous post of yours after I had done some investigation of the matter, and I still do

However this post concerns your post below:

Referenced for anybody want to pursue this:

Specs for the HP motherboard of the OP.

Chipset "North Bridge" is a  SiS 651
Chipset "South Bridge" is a  SiS 962

Spes for SiS ship sets

Something is wrong. I give up

http://www.tomshardware.com/forum/151511-30-enabling-usb2-645dx-chipset-98se     1998 ? Time travel   

I don't think my chipset is SiS, and my m/b certainly is not the one you referred to I'm afraid.  There must have been a typo somewhere.  To state again, the m/b with the problem is an MSI MS-6577 v2.1 (Xenon 4), made for HP/Compaq and used in several of their models. This has an Intel chipset.  But I don't know what SiS means I'm not familiar with 'SiS' and it may be that they are the same.  Looking at the spec in the link though, that tells me max memory is 3GB, whereas my m/b has a max of 2GB.

There is one mistake that I made in my original post, that I would take this opportunity to correct, and that is my Pavilion a405.uk which does not show this problem (and it is beyond coincidence that in Dev Mngr an EHCI is shown), has the MSI MS-6577 v2.1 motherboard, not, as I incorrectly said, the v3.1.  This is not the m/b that came with case: I purchased that computer as a second hand refurbished unit, and assumed what it said on the tin was what was inside.  I have just this moment looked at the m/b and it has printed on MS-6477 v2.1

The drivers for ICH4 on the Xenon4 date from 2001. It may be that the chipset dates from as far as 1998, but I seem to remember reading that the USB 2.0 standard was introduced in 2001

I am aware that you are much more knowledgeable than I am, Geek-9pm,and I thank you for your post and the work put in to researching it.

Dumb_Question
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Dumb_Question]

Dumb_Question, you should change your user name.

What is wrong with the username Dumb_Question ?
 

Back to topic. The photo was from HP

Thank you, I thought maybe you had an alternative source which might provide additional information.

I think if you use your  PCI USB card the performance will close to USB 2.0 standard. If so, that would indicate the issue is not controlled or limited by the BIOS.

As both the entries associated with 'PCI' mention the phrases 'USB 2.0' or 'Enhanced', I would think this likely.  I believe that the problem is shown up by the absence of an Enhanced Host Controller entry in the first block of USB Controllers in Dev Mngr; is this caused by the O/S failing to recognize the 82801DB/DBM chip properly ?  On boot the computer should find the Enhanced USB Controller automatically 

The built-in USB ports are part of the motherboard chip set.  They are often called 'north-bridge' and 'south-bridge'  chips. That are rather large chips that are soldered into the motherboard. If they are damaged, there is not feasible repair.
From HP for v3
http://h10025.www1.hp.com/ewfrf/wc/document?docname=c00100539&cc=us&dlc=en&lc=en

On eBay V2.x  motherboard goes for about $40 including shipping. Used, of course. Here in the western world you will not find a technician who will repair the board for f $40 or less.  {Maybe in South East Asia. }

I wouldn't even dream of attempting any kind of m/b repair myself, I have never even got the hang of soldering simple things.  I found out how much labour costs were for one such technician and it was about $US 60 per hour (converted straight to US$).  On YouTube there are a few videos showing capacitor replacement - I like this one http://www.youtube.com/watch?v=wXCQ_EHG7kU - it shows how easy it is !
I have identified the chipset (well, partly) and know where the Northbridge and Southbridge are (I think) [aside: I have read that system temperature as determined by many temperature monitoring programs is read, on many systems, from a sensor 'in the Northbridge' - is this the case for the MS-6577 ? how can I find out ? Speccy, MBM5 and others quote a system temperature].
Cost of such a m/b is spot on.

Your may wish to consider a newer motherboard to improve your  investment.
Best I can offer. 

I've heard that a lot, and it sounds like good advice.  It seems like you need more advanced hardware these days to carry out the same tasks 

Thee is a very small possibility that a electrolytic capacitor on the mother bard has failed and it could be replace. But his is so dubious that it is not wroth the effort or the risk  of replacement. This is not base on my experience, rather the comments of more experienced  technicians elsewhere. Pragmatically, as one way to pout it, you are not able to repair the  motherboard

I agree. It's not cost effective in my circumstances.  A more practically skilled person might give it a try though.

Once again, I am indebted to you and the other members of Computer Hope for your comments.

Dumb_Question
1.March.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Geek-9pm]

After having a night's rest,  it would seem the suggestion by BC_programmer is the best idea. For some reason the the device manager 'forgot' that the chip-set supports USB 2.0 speeds.  So it is not an issue of the chip set drivers per se, rather Windows is  supposed to know how to make it run a USB 2.0 speed using just software.

So you need to find some way to get Windows to reload the  USB drivers again.
The Motherboard was a standard design years ago an d Windows at that time did run it at USB 2.0 using only the chip set plus stuff windows has in its driver archive.

Did you say you have some spare hardiness? If so, put one in and try a new install of an older version of Windows of the spare drive. It just might use bette3r drivers. But if  it does work --- what next??

[Dumb_Question]

Due to a suggestion by BC_Programmer I appear to have discovered the cause of the problem, and the simple, if inconvenient, solution follows directly from this.  (I didn't have do anything remotely difficult or that was likely to be irreversible, thankfully)

Forgetting DaveLembke's advice, having connected my PS/2mouse and keyboard, I started uninstalling the USB drivers, this time including the seemingly PCI ones.  When I uninstalled the one that corresponded to my wireless kb/mouse, I couldn't do anything more apart from power down because I found that PS/2 kb/mouse were not working  .

OK, so finally when I rebooted again, the computer went through the process of reinstalling the USB hardware and drivers.  I now connected a flash drive to the USB, it had to reinstall it, but there was no "This device can perform faster....etc" message  I ejected and tried it again with the same result !

Now the Device Manager looked like figure ScrSh_HWM_USB_S5160UK_PS2_010314.JPG (attached).  As a reminder, this is very similar to Dev Mngr of my a405.uk (attached), which had the same PS/2 mouse/kb (a405uk_DevMgr_USBcontrollers_270214.JPG), and importantly has the device 82801DB/DBM USB 2.0 Enhanced Host Controller 24CD displayed, whereas with the USB wireless mouse/kb in the S5160UK this Enhanced controller is not found (ScrSh_HWM_USB.JPG) (attached)

I am sure I don't understand it; my USB kb/mouse is a PCLINE PCL-K350,  and a quick look around did not say anything its USB 2.0 or USB1.1 capability or compatibility.

This a bit inconvenient because I have one set of PS/2 stuff, ad the wireless/USB mouse/kb was much easier to use.

Dumb_Question
1.March.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1
 

[recovering disk space, attachment deleted by admin]

[Dumb_Question]

Thanks Geek, to reinstall the OS using another drive was my last resort, I had already started that process. 
But there were some other things I had to try first....

As my wireless USB kb/mouse were bought considerably after the computer, I'd be surprised if that was a USB1.1 sub-system, but stranger things have happened. Also, I don't know if it is coincidental that the PCI connected USB Controllers in Dev Mngr have gone away.  Would my PCI modem still work for example ? (not that I see myself using it)

Dumb_Question
1.March.2014
Compaq Presario S5160UK DT261A under XP/SP3
Motherboard - MSI MS-6577 v2.1

[Geek-9pm]

Some devices are called 'PCI ' even when they are not in a PCI slot.

(1) Short for Peripheral Component Interconnect, a local bus standard developed by Intel Corporation. Most modern PCs include a PCI bus in addition to a more general ISA expansion bus. PCI is also used on some versions of the Macintosh computer. PCI is a 64-bit bus, though it is usually implemented as a 32-bit bus. It can run at clock speeds of 33 or 66 MHz. At 32 bits and 33 MHz, it yields a throughput rate of 133 MBps. Although it was developed by Intel, PCI is not tied to any particular family of microprocessors. ..

http://www.webopedia.com/TERM/P/PCI.html
From the above, it is a set of standards and does not mean the device is actually plugged into a slot.

[BC_Programmer]

Here is my theory on why the keyboard/mouse may be affecting it.

"USB Legacy" settings. These settings, when enabled, will let you use the (USB) keyboard during boot-up to manipulate the BIOS. The setting essentially serves as an "adapter" that will make the USB keyboard cause the same interrupts as a standard keyboard, whereas normally the interrupts triggered would be the USB device interrupts that won't be handled by anything. (essentially the BIOS sets up a handler for those interrupts and translates keyboard-class device input to keyboard interrupts).

Now the problem (again, just a theory) could be that in doing so it locks the root hub to operate only in USB 1.1 mode.

To better explain this we really have three "modes" of USB. There is Low-speed (1.5MBit/s, Full Speed(12Mbit/s), and high-speed (480 Mbit/s) 1.1 had Low-speed and Full Speed, and High-speed was added by 2.0.

Thus a keyboard or mouse, which will undoubtedly use the 1.5MBit/s Low speed bus, will operate the same on any USB hub. What I'm thinking is that if the USB legacy option is enabled it may, at least for that machine, somehow prevent the USB 2.0 driver from initializing properly, since the BUS is already being used for the keyboard; so the 480 Mbit transfer speed is unavailable.

That's just a spit-balled theory though so I don't know how true it is. I can say for certain that keyboards/mice almost always use the 1.5Mbit low speed option, however.

Also, I don't know if it is coincidental that the PCI connected USB Controllers in Dev Mngr have gone away.

They haven't gone away, they have been properly identified. the "PCI Class USB 2.0 Enhanced Host Controller" is basically Window's Device identification being able to determine that it is attached via PCI and appears to be a USB 2.0 host controller, but for whatever reason not being able to find out much else about it. In the "fixed" screenshot, that same device is now properly identified as a Intel USB 2.0 Enhanced host controller.

My suggestion at this point would be to see if there is a "Legacy USB" option, and see if disabling it will allow you to use the system with the USB keyboard/mouse without forcing everything to work at Full Speed (USB 1.1).

[Dumb_Question]

->BC_Programmer

Your hypothesis sounds very plausible.  There is one part of it that does not fit though.  When I was looking in the BIOS set up for an option for 'set USB mode to 2.0', which apparently is present in some BIOS set-ups, I did see a 'USB Legacy' option.  This was one of the first things I tried.  To remind people of a section in my long and unmemorable first post in this thread:

...I have looked in BIOS setup, and I did not find anything about USB except legacy, set to auto/enable/disable.  I tried all of these without any change.....

Originally, before I tried all the settings, it was set to 'auto'.

It is my opinion that the system should permit simultaneous USB 1.1 and 2.0 operation, but maybe this just not possible in Hardware.

I have a wireless keyboard and a wireless mouse which connect over an RF radio link (2.4GHz ?) to the transceiver which in turn is connected to the PC by a wired USB link (i.e. both keyboard and mouse share the same transceiver). I can't remember if I installed drivers for this manually from a CD, or whether I simply connected it and it worked.  That sub-system is a PCLine PCL-K350: I couldn't find any spec for it on the internet except that it was 'USB connected' - I was unable to determine its compatibility wrt 1.1 or 2.0.  Also checked in manual (written in four languages).  If anyone reading this can determine that spec, it might help.

Dumb_Question
2.March.2014
Compaq Presario S5160UK DT261A under XP/SP3 
Motherboard - MSI MS-6577 v2.1

[BC_Programmer]

Actually, I just read back on something I Missed as well:

I measured the transfer rate from my HD to an external one for 490498944 bytes; it took 8 mins 16 secs, a transfer rate of ~7.9Mb/s, good for USB 1.1, pretty poor for USB 2.0

USB 1.1, Full Speed, tops out at 12 megabits per second. a transfer rate of 7.9 Megabytes per second is around 80 Megabits per second, which is far in excess of anything USB 1.1 is capable of, which means it is in fact operating via USB 2.0.

I don't know why you would be getting the message. However I would suspect in this case that one of the two devices (keyboard/mouse) may have a error in their implementation that is causing devices to be recognized incorrectly, but still be used properly. I was looking through the USB spec and it looks like identification commands are sent to all devices via a broadcast, and basically it says "Device 4, identify yourself!" when you plug it in.

What could be happening is that the device you had (keyboard/mouse) was responding to everything you plugged in and giving the USB driver false information to go on, that made it think the port was 1.1.

That's all I can really think of, But it makes sense, assuming I Interpreted the specifications document correctly. However so far it is clear that the devices you were using were in fact using the USB 2.0 High Speed interface since otherwise they would have topped out at less than around 1.5 megabytes a second.

[Dumb_Question]

-> BC_Programmer

I meant 1b = 1 bit, 1B = 1Byte
so by 7.9Mb/s I meant 7.9 x 1E6 bits per second (~(490498944 x 8 )/{(8 x 60 + 16) x 1E6)}
which is less than 12 megabits per second.
I realise that this is only approximate as does not include error bits, stop&start bits, etc, the time is approximate to a couple of seconds.

I apologise for not having made myself clear on this point.

Dumb_Question
2.March.2014
Compaq Presario S5160UK DT261A under XP/SP3 
Motherboard - MSI MS-6577 v2.1

[Dumb_Question]

So, a little more experimentation:

starting with the configuration shown in ScrSh_HWM_USB_S5160UK_PS2_010314.JPG,  (but without the Mass Storage Device), with only my PS/2 mouse and kb plugged in:

I connected the USB mouse/kb receiever to a USB port.  The bubbles appeared which said 'New Hardware Found' and 'Your new hardware is installed and ready to use' etc. The Device Manager now showed an additional composite device.  The wireless mouse/kb functioned properly it seemed.

I removed the PS/2 mouse and kb, there was no obvious change to DeviceManager.  I now connected a WD MyPassport external HD to a USB port.  There was no 'This device can perform faster...' message, the Device Manager changed to add 'Mass Storage Device' under the (only) USB Controllers category. 

I inspected the power tabs on the Root Hubs.  One said "one port 500 mA required, 5 ports available", another two said "Two ports available", the third said "one port 100mA required, 1 port available"   [my digests of the content of the power tabs].  From what I had seen previously, the Root Hubs with 2 ports each belong to USB 1.1, and this is consistent with the USB wireless kb/mouse being a USB 1.1 sub-system.

Now for the acid test: reboot without PS/2 mouse/kb, only the USB sub-system.  I ejected the MyPassport and restarted....the PC at this point decided it would fail to restart (another issue).  I left it switched off to try again later. Later, I turned on the PC.  After waiting for it to start up properly, I opened Device Manager, and it looked like I described it earlier, one USB 2.0 Enhanced Host Controller 24CD entry, three Universal Host Controller entries (24C2, 24C4, 24C7), the composite device entry, 3 Root Hub entries each capable of supporting 2 devices, and finally another Root Hub entry, capable of supporting 6 devices.  The second category for USB Controllers (where the PCI related entries had been) was absent. 

As a last test I connected the external HD to a USB port, and did not get the message  'This device can perform faster...'.  I have yet to measure the transfer speed, or restart n times, but it seems the problem has gone away, somehow the PC was reminded of what it should do by connecting the PS/2mouse and kb.

Dumb_Question
3.March.2014
Compaq Presario S5160UK DT261A under XP/SP3 
Motherboard - MSI MS-6577 v2.1

[Geek-9pm]

Thank you for the report. This information can be helpful later.
That fact is that you did an extensive effort to resolve an issue and found that the answers did not follow a consistent pattern.

I still think the problem goes away when you  either:
A) Use and older version of Windows.
OR
B ) Install the  PCI USB card, which has silicon from another maker.

Here is yet another resource:
http://www.usbman.com/WinME%20USB%20Guide.htm
Windows  XP   -USB Guide

A USB device that is connected to a USB 2.0 hub is not detected in Windows XP Service Pack 2, Windows Server 2003 Service Pack 1, or Windows Server 2003 x64 versions