Frequently freezes of mouse pointer and music playback

[Louk]

Hello,

As stated in the post title, my system freezes very frequently for short times (1 tot 5 seconds). When you move the mouse, the cursor suddenly stops for some seconds and then catches up. When playing music, the music stops for a about half a second and then resumes. I have the feeling that it happens more when the system is more busy (using more CPU time).

Some details about my setup:
- CPU Z530 @1,6MHz
- Stabilized power supply @ 12 V (Minibox M4)
- Andrea PureAudio USB Sound Adapter
- König WIFI 11g WLAN adapter
- Broadcom BlueTouth adapter

Relative Software installed:
- Latest BIOS (090721)
- Windows XP pro SP2
- All available FIT-PC2 drivers for Windows XP
- Centrafuse
- Blue Solleil

I have a serious suspicion against the USB hardware or drivers.

Note: I used the same setup before on a VIA EPIA PX 10000G system (which was to slow to handle sound processing) without this problem (no hick-ups).

Edit: changed audio output to on-board audio adapter and the music hick-ups where gone.


Louk.

[Louk]

Hi,

Any news about this item (hick-ups)?

Note: The problem is not solved; I found a work-around by connecting the sound adapter to a local USB port in stead off a port on the USB hub. I also need to be very carefully in using other (USB) devices.

At this moment music is stalling again, not as often as before, but it is getting really irritating.

Is there any change this will be solved in the next firmware release?

Louk.

[irads]

Try disabling C-States in BIOS.

[Louk]

Hi Irad,

I'm sorry about this late response, but I can't get into the BIOS with my monitor in the CAR (A 7" Touchscreen In-dash monitor from Yauto (CH) which only supports 800*600 resolution. So I will get the system out of the car this weekend and connect it to a multi sync monitor to adjust the setting.

By the way, disabling the C-states, does this mean the system is going to be emitting more heat as it already does?

Louk.

[irads]

Yes, with C-states disabled the system will use more power when idling.

[Louk]

No luck, With C-states disabled, I have still audio interruptions as soon as there is a lot of USB communication, like there is with an external audio adapter of an external USB WLAN device (Linksys WUSB300N in my case).

I still believe this is a firmware or an USB driver problem: something is blocking the bus or even the CPU.

Is there a way to get more information somehow? Like an USB sniffer or monitor tool?

B.t.w.: Sorry it took me so long to test this. My touchscreen monitor in the car does only support 800 x 600 mode, so my screen goes to blank when the system enters the BIOS settings menu in 640 x 480 mode. So I need a normal multi sync monitor and I have to park the car wright before the house to get power (240 V ~). But last week the road was closed for construction, so I could not get the car close enough to the house to get power.

So for me, an BIOS option to select the initial screen resolution would be a very nice option.

Louk.

[irads]

Freezes of 1-5 seconds are not a known behavior. I noticed that you specified installation of XP SP2, it is worth trying SP3.
Also, disable selectively the peripherals you are using, maybe one of them is blocking.

[Louk]

The interruptions are much smaller more .1 to .5 seconds (sounding like clicks to small gaps).

I've ran W7 RTM for a while and I thought the problems occurred also with this version, although I'm not sure about that anymore (can't remember). I have plans to update to the official release of W7 in the very near future.

When I disconnect the Linksys WUSB300N, the sound is OK. Maybe the USB hub is the problem. In the past where I had my Andrea USB SA connected through the (not powered), the problems where much worse.

I will try to exchange two devices and see if it still happens when the Linksys WUSB300N is directly connected to one of the 4 native USB ports and try to stress the other USB devices on the HUB. With a little bit of luck, I will get some tests done this weekend, maybe even with W7.

EDIT: I've moved the WUSB300N from the USB hub to one of the native USB ports of the FIT-PC2. The first results are very promising: Did not hear any glitches any more until now (about 15 minutes of testing). still need to do a stress test.

This all leads to one common component: the USB hub. In the beginning, I already tried different brand USB V2.0 HUBs, but it did not make any difference. I might also be a timing problem.

Best Regards,

Louk.