Motherboard Forums

Thread Tools Display Modes

Re: ASUS and its strange USB behaviour

Posts: n/a
      08-02-2008, 03:55 PM
chaovsky wrote:
> Hi, folks. I'm having some weird problems with ASUS mobos and USB
> behaviour, I hope you have heard of similar cases and can give me a hand
> here:
> First I had a desktop PC with a P5KPL-VM and Windows XP SP2. The USB
> ports would behave randomly, recognizing some devices always (memory
> stick, Creative Zen), some of them only depending on which port they
> were connected to (wrong port and I'd get a "this device has
> malfunctioned and may not work properly", like my CanonScan Lide50), and
> some of them never at all (Nokia phone, UPS).
> Sometimes I'd plug a device in a port and some other USB device (like
> the wireless receiver) would stop working.
> Furthermore, I have 8 physical ports (four 2-port hubs), but the device
> manager shows 5 hubs, 4 of which have 2 ports each, but the last one has
> 8! This would give a total of 16 ports! This number never varied, even
> when I physically disconnected all the hubs but one; the device manager
> would show the 5 hubs and the 16 ports no matter how many were really
> connected: 1, 2, 3 or 4, I tried them all.
> I've tried uninstalling and installing the drivers, the devices,
> changing BIOS settings... Nothing. USB behaviour is still totally
> erratic.
> Now I have a laptop, another ASUS (X53L series, comes with a model F3L
> mb), and it's giving me the same issues. Straight out of the box,
> Windows Vista SP1.
> It seems ASUS mobos have an awful way of handling USB, as I have read
> in other parts of this forum and other websites. But, has anyone come up
> with an answer to these problems? I would really appreciate any new
> ideas on this issue, as I think I may have tried all the "conventional"
> ones.
> Thanks to all for your time and hope you can help me out!

OK, here is a custom diagram for you.

| EHCI (Enhanced, 2.0) |
| | | | | | | |
| | | | | | | |
| USB1.1 | | USB1.1 | | USB1.1 | | USB1.1 |
| UHCI | | UHCI | | UHCI | | UHCI |
| | | | | | | | | | | | | | | |
x x x x x x x x x x x x x x x x
/ / / / / / / /
/ / / / / / / /
+ + + + + + + +
| | | | | | | |
Port0 Port1 Port2 Port3 Port4 Port5 Port6 Port7

What Device Manager shows, is "logic blocks" present in hardware. The
entries in Device Manager are not "ports". In the above diagram, there
are five logic blocks and eight ports.

In the figure, there are eight physical ports, arranged in "stacks of two".
The switches above each port, can swing to one of two positions. In one
position, a Port can connect to a USB1.1 UHCI logic block. In the other
position, a Port can connect to a USB2.0 EHCI logic block. The binding
is dynamic, at discovery time. The switch is flipped, as a function of
what the drivers determine has been connected to the port.

So in that diagram, there are eight Ports, but those don't have a
presence in Device Manager. There are five logic blocks. The four USB1.1
logic blocks, control (and share 12mbit/sec bandwidth over) two ports.
The single USB2.0 block happens to control (and share 480mb/sec bandwidth
over) eight ports.

With the switch settings shown above, half of the ports are running at
USB1.1 and half are running at USB2.0. (For that to happen, I might have
plugged in four USB1.1 devices and four USB2.0 devices.)

The next issue, is power consumption. Some USB devices are high powered,
and draw more current than other devices. As near as I can tell from
available information, the LIDE50 is bus powered, and doesn't have
a wall wart to power the scanner. It is rated to draw 5V @ 500mA when
active, which is the limit for USB. Sometimes, the current consumed can
cause the device to be disconnected. Especially if another high power
device is connected to the same stack_of_two connectors.

Ideally, USB devices should have serial numbers. Not that many of them
do. If a serial number is present, the device can be moved from USB port
to USB port, without needing to reinstall drivers. If there is no serial
number, then the device ends up being reinstalled over and over again.

For help with specific USB issues, there are a few resources.

Resetting the USB stack, is described here (for Safe Mode).

If you want additional hardware to play with, my might purchase a self-powered
USB2 hub, one that has a wall wart that plugs into the wall. Some of those
have a 5V @ 2A power supply. If the LIDE50 was plugged into one of those
self-powered USB2 hubs, and the hub connected to whatever computer you want,
there may be fewer issues (but no guarantees). If you have a few high powered
devices, then it helps to have a few extra "toys" in your kit, like the
self powered hub. Purely for experiments, and as a workaround for
any problems found.

Remember that the hardware manufacturers don't handle USB. The Microsoft
default driver is doing that for you.

For looking at USB devices, when they're plugged in, you can get a
copy of a USB viewer here. This tool used to be on the Microsoft web
page, but they removed it. This is the best I can do, in terms of
digging up a copy.

File size is 167,232 bytes.
MD5sum is 93244d84d79314898e62d21cecc4ca5e

The screen of that tool, looks like this. The ports get listed twice
here (and are not all visible at present), due to this interface
showing all the spigots on the logic blocks. There are several tools
that go by similar names, and again, this is the closest I can get
to a picture of the information content of UVCView.

Reply With Quote

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off

Similar Threads
Thread Thread Starter Forum Replies Last Post
Strange behaviour from 'About this Mac', System Profier and SoftwareUpdate Andy Apple 10 10-21-2007 11:37 AM
A8N32-SLI Deluxe strange fan behaviour JASE Asus 6 10-10-2007 09:46 PM
Strange behaviour with a mix of S-ATA and IDE drives - Epox 4PLAI Marcel Overweel Epox 0 04-20-2004 02:15 PM
Its easy and its legal jon hayes Intel 0 11-01-2003 04:12 AM
[8K9A2] Strange behaviour after BIOS upgrading Mauro Epox 3 10-28-2003 05:38 AM

All times are GMT. The time now is 11:18 PM.

Welcome to Motherboard Point