Better stability by ditchin ONBOARD USB!!

[Lord Evermore]

I've never seen a 645 CMOS setup screen directly (I keep talking myself out of building a new system), but I've read a lot of reviews and none mentioned this particular setting (and I did check the Asus manual for the P4S533, so didn't see it, and I looked and Gigabyte's manual for the GA-8SRX and 8SDX don't have any mention of this). And I've never seen it for any other modern SiS chipsets so it is a new thing to me, and I assume it is a new function with the 645DX, not available on the original 645.

 

[WetWilly]

Gigabyte's manual for the GA-8SRX and 8SDX don't have any mention of this

The Gigabyte manual for the 8SRX (rev 1.003) only covers BIOS revision F1; they're now up to F9. I don't have the F1 or F2 BIOSes, and since I don't have the F9 BIOS handy, the earliest BIOS I dug up quickly with the "USB0/USB1/IDE Interface Access" selections was the F3 BIOS. However, in the F3 BIOS the Interface Access options are there but set not to display. Here's where it gets interesting. While the "IDE Interface Access" selection is in essence forced to "Embedded Bus," the "USB0/USB1/Audio Interface Access" selections are all forced to "PCI Bus" unless you modify the BIOS. Not exactly what you'd expect. Epox' 4SDA BIOS is configured exactly the same; only difference is the Interface Access selections are visible to the user.

I assume it is a new function with the 645DX, not available on the original 645.

Nope. The Gigabyte 8SRX, Epox 4SDA and Abit SD7-533 are all "old" 645 boards. IIRC, the options are even available on some SiS 735/745 boards.

 

[Iron Woode]

All of this could have been avoided if Sudheer Anne would have not included the whining.

It should have went like this:

I was overclocking my Asus P4S333 and hit a stability/functionality limitation with the onboard USB controller. It was limiting my ability to overclock and retain USB stability. I bought a USB2.0 PCI card with 5 ports. It works great at overclocked speeds because the PCI bus is kept within spec and not overclocking the USB like before. I am very happy now.

This would not only have been informative but would avoid ranting like a spoiled kid. Calling all onboard USB controllers garbage is just stupid.

 

[OneOfTheseDays]

obviously you guys have not read any of my replies. this thread needs to freakin die, you guys take everything and analyze every word to the nth degree. All i'm trying to say is that the onboard USB controllers on many motherboards aren't very stable when it comes to overclocking. For those lookin for a little more Ocing and stability should ditch the onboard and get a nice PCI controller (cuz u can run the pci in spec if your mobo supports it). ok, time for thread to die. no more replies.

 

[Barrei]

Hey Anne , tough crowd on this thread .

 

[OneOfTheseDays]

tough crowd.....no? anal retentive crowd.......yes

 

[Lord Evermore]

All i'm trying to say is that the onboard USB controllers on many motherboards aren't very stable when it comes to overclocking.

Then, why didn't you say that instead of what you did say?

There's far too many people that generalize based on their one experience with a product or situation, giving their opinions as if they were written in stone by a deity, and leading others to believe the same thing is always true. Your initial post didn't say that you've found there are problems with on-board USB under certain conditions, it said they're all garbage. If one were to listen to you, they'd think there's never a situation where you'd want to use the onchip USB controller.

 

[OneOfTheseDays]

well it's pretty obvious that I was specifically talking about overclocking STABILITY and not just plain WORKING. jeezus i would figure that you would use a little common sense when readin.

 

[Lord Evermore]

I do use common sense. Not everyone does. You're the one that made the blanket statement that onchip USB is garbage; you didn't state that you meant ONLY during overclocking. In fact in your second post you said specifically that it sucks on many motherboards, without any mention of overclocking, and that people should just get a PCI adapter.

I'm not stupid, nor lacking in common sense. I'm just literal.

 

[Iron Woode]

Originally posted by: Sudheer Anne
well it's pretty obvious that I was specifically talking about overclocking STABILITY and not just plain WORKING. jeezus i would figure that you would use a little common sense when readin.

I wish people would use a little common sense when they post.

Your original post was vague and full of blanket statements. This is almost as bad as x86 (formally docsmarts) post about amd cpus causing fires which have to be put out by pouring water on the motherboard while it is powered up.

 

[mschell]

I've got a P4S 533 1.6A running at 150MHz and have zero problems with my USB devices or stability in general. They include 2 different digital cameras which are nortoriously finicky about USB bus conditions. The key here is is it's my setup which my be different than your setup or many people's setup for that mater. My success does not mean everyone will have perfectly a functioning USB bus at 150MHz and you failure doesen't mean that the SiS controller is "utter garbage" or everyone will have your problem.

Instead of blaming the motherboard, I would check your BIOS settings and confirm your memory's stability as it has an effect on the USB bus functioning.

 

[WetWilly]

Thought I'd resurrect this thread from the dead ...

functional onboard USB @ 175fsb? lol - yea right!

FYI, I've been using the onboard SiS 7001 USB controller on a MSI 745 Ultra at a 183MHz FSB for a few days now and it's working perfectly.

 

[WangButter]

Oops, dead thread.