The 6212 chip is FLAWED, face it...

[steven32272]

I bought a 6212 based card spanking new off the shelf.
I tried it in a intel 440bx chipset motherboard, under XP sp1, then SP2, then win98 FE.
the card does not behave right, and i tried every card slot in every configuration.

Then I tried it in my main system, a via kt400 mainboard. under XP sp1, sp2,
and win98 FE. throughput spits and sputters on every OS, usually with filesystem errors.

SO I said to heck with it, and ordered an ALI chipset off from pricewatch.
the vendor was out of them and was kind enough to send me a board based on...
The freakin' VIA 6212! I tried it anyway though, with hopes that maybe the first card
was flawed. same deal. simpy put, on any MOBO I have here under all available
OS's, and countless driver attempts, this 6212 chipset, on two different boards,
simply does not work.

And there are countless posts here complaining of 6212 problems.

I think VIA needs to step up to the plate
and admit the 6212 hardware is simply flawed.


Steven

[Stultify]

The 6212 chip is FLAWED

What is "FLAWED" is folks not reading the documentation about the stuff they are purchasing and jamming cards in their computers that aren't going to work.

These cards need a PCI 2.2 spec slot to go into, they also need a motherboard that is FULLY APIC/APCI compliant, and an operating system that allows the assignment of more than 16 IRQs.

The "bottom line" is: that these cards will only work well in a system that already has USB 2.0 ports.

Of course it doesn't help much that every "batch manufacturer" in China is buying VT6212 chips from VIA and spewing millions of these cards out daily to be sold to folks that don't have a clue, to drag home, stuff in their computers, only to find that they simply don't work as expected.

VIA is in the business of selling chips, I feel it is a shame that they don't have a qualification program for the folks they sell them to, but how VIA does business is not up to me, and "really" who am I to say?


At the end of the day, the "wise" choice in a USB 2.0 card for an "older" computer, is a card having ONLY 2 ports and a NEC controller chip.

[crc_73]

I'm not a software/hardware expert, but my experience of this chip tells me that it's not the chip that's flawed, but the drivers available, and problems/issues with Windows 2000 and XP. I don't know about other OS's.

I made a post already about the problems I've had with the vt6212l. My motherboard meets PCI2.2, ACPI, and apparently Windows 2000 is able to handle IRQ sharing through IRQ steering, but I still have problems with the USB2.0 PCI card I have. It works fine if I use it throttled with drivers to USB1.1, but when I update the drivers to give full speed USB2.0, that's when I get the problems. This leads me to think the problem is with the drivers available and the OS, but MS says that you should have USB2.0 support if you have SP4 installed, which I have, or SP2 with XP. I also looked for the vt6212l in the HCL, but it's not listed.

One strange behaviour I noticed with the same PCI card in an XP machine (with SP2 installed) was that it worked fine to begin with, but when I changed the "Policies" to "optimize for performance", the card then behaved in exactly the same way as when it was plugged into the machine with Windows 2000 (Windows 2000 doesn't have this option). I already reported this in my thread, but no one has commented on this observation. I think this is key to the problems Windows 2000 users are having with this chip, but then again I might be wrong, I'm a biochemist, not a computer engineer. Who's problem this is to sort, I don't know. My impression was that MS and VIA engineers got together to add chips to the HCL - am I wrong?

[Ghetto_Child]

Here's your first answer people. Yes I know the chip says VT6202 or VT6212L; I have 2 different PCI card with both these chips and they both detect in Lavalys Everest V2.20 as the models specified in this microsoft link.

http://support.microsoft.com/kb/q233163

So that's the first problem and Win2K users like me are pretty much out of luck. Hey I've hated the VIA company and their products from day one and never spent any money on them. This is why, get this my brother installed Win2K on his Asus P4VP-MX and wouldn't you know it uses these chipsets for the usb built on the motherboard too lmao. So the result for him is he has no working front usb ports in windows.

I'm not giving up yet I came up with the idea to install the separate VIA USB2.0 Drivers V270p download. It's a 12MB zip file with the usb 2.0 drivers. Don't use the 4 in 1 stuff as that did absolutely nothing for me when I tried installing it. BTW I'm using an Intel D850GB motherboard, when I put in these cards without installing drives and used them as detected they were unstable. My Verbatim Store n Go Pro 1GB drive can't unload properly and reconnects to windows immediately so fast that I don't even have time to pull out the flash drive. Sometimes it takes 3 unload attempts before it stays off for removal.

Oh look at that there's even a VT6214 USB 2.0 host controller on the website too.
On this page near ther bottom http://www.viaarena.com/defaul...atID=1260&SubCatID=122 there's a VIA USB 2.0 driver package for all windows. I'm going to try installing that to see what happens unless I get a better suggestion on here.

I have a VIA Firewire 1394A card. 3 or 4 ports and it works perfect didn't have to install any drivers on Win2K SP4 and I use that for high speed data transfer. I try to always purchase enclosures with both usb 2.0 and firewire for hard drives and optical drives. The model number on the chip of the VIA firewire card is VT6306 and the card's model number says VT6306-B01. According to Lavalys Everest v2.20 it's a VIA Fire II the rest I don't remember.

Here's another site I came across that might help troubleshoot your usb issues.
http://www.usbman.com/Guides/V...0and%20Tricks.htm#Bios

[steven32272]

Originally posted by: Stultify
The 6212 chip is FLAWED

What is "FLAWED" is folks not reading the documentation about the stuff they are purchasing and jamming cards in their computers that aren't going to work.

You assume a LOT sir. I did not "JAM" the card into a computer that is not
going to work. What ticks me off is people who read half a complaint message and spout
off garbage. If you had actually read my post, and knew what a KT400 chipset is, you
would know it meets all of your "Special" criteria. Jamming cards into computers indeed.

These cards need a PCI 2.2 spec slot to go into, they also need a motherboard that is FULLY APIC/APCI compliant, and an operating system that allows the assignment of more than 16 IRQs.

Again, you did not read my message, card was placed into a kt400 chipset under XP service pack2.
please read my messages in entirety before rudely spouting off.

The "bottom line" is: that these cards will only work well in a system that already has USB 2.0 ports.

The KT400 already has 2.0 ports. But you are clearly wrong here. What is the point
of buying a usb2.0 pci card for a system that ALREADY has those ports? That notion
is as ill thought out and as silly as your response.

Of course it doesn't help much that every "batch manufacturer" in China is buying VT6212 chips from VIA and spewing millions of these cards out daily to be sold to folks that don't have a clue, to drag home, stuff in their computers, only to find that they simply don't work as expected.

This has long been a history with VIA. Back in the day when they made questionable motherboards
for questionable companies like cyrix and AMD, I knew what they were. However, with the kt333 and
kt400 chipsets along with AMD finally moving into mainstream quality, I was sucked into
beliveing VIA had re-invented themselves. My mistake.

VIA is in the business of selling chips, I feel it is a shame that they don't have a qualification program for the folks they sell them to, but how VIA does business is not up to me, and "really" who am I to say?

I agree. Really...Since you didnt read my message in it's entirety, and clearly know nothing
about the hardware I posted about. Then, who ARE you to say?

At the end of the day, the "wise" choice in a USB 2.0 card for an "older" computer, is a card having ONLY 2 ports and a NEC controller chip.

I agree. ANY chip EXCEPT VIA.

Steven

[steven32272]

Originally posted by: crc_73
I'm not a software/hardware expert, but my experience of this chip tells me that it's not the chip that's flawed, but the drivers available, and problems/issues with Windows 2000 and XP. I don't know about other OS's.

You do NOT know about other OS's. I have also tried this chip under linux with the
2.6 series kernels. the same problems I experience under XP and even 98 still exist. Also,
VIA does not even provide drivers for this chipset. you need to rely on microsoft.

I made a post already about the problems I've had with the vt6212l. My motherboard meets PCI2.2, ACPI, and apparently Windows 2000 is able to handle IRQ sharing through IRQ steering, but I still have problems with the USB2.0 PCI card I have. It works fine if I use it throttled with drivers to USB1.1, but when I update the drivers to give full speed USB2.0, that's when I get the problems. This leads me to think the problem is with the drivers available and the OS, but MS says that you should have USB2.0 support if you have SP4 installed, which I have, or SP2 with XP. I also looked for the vt6212l in the HCL, but it's not listed.

I also meet the critera our resident "Computer genius" attacked me about, and I also
meet your criteria. Yes, both of my cards work fine when the EHCI controller
is disabled. I have also tried 3 different versions of usbehci.sys for winXP, all
dated differently, all different sizes. I guess thats what we get for "Dragging"
chips home and "Jamming" them into our computers.
[/quote]

One strange behaviour I noticed with the same PCI card in an XP machine (with SP2 installed) was that it worked fine to begin with, but when I changed the "Policies" to "optimize for performance", the card then behaved in exactly the same way as when it was plugged into the machine with Windows 2000 (Windows 2000 doesn't have this option). I already reported this in my thread, but no one has commented on this observation. I think this is key to the problems Windows 2000 users are having with this chip, but then again I might be wrong, I'm a biochemist, not a computer engineer. Who's problem this is to sort, I don't know. My impression was that MS and VIA engineers got together to add chips to the HCL - am I wrong?[/quote]

Who knows? VIA allows themselves to be put into legal agreements
where they cannot even make DRIVERS for thier own chipsets?

Steven

[steven32272]

My original post was that I believe the VIA 6212 chips is "Flawed".

Here is what microsoft has to say about the problem:

These chip sets are unsupported because of problems in the input/output (I/O) port aliasing that can only be resolved in the chips themselves.

For additional information, or to inquire about the availability of a software update that corrects this issue, please contact Via.

After dragging this chip home and jamming it into my computer with
no regard for documentation, in the end I am glad microsoft
is on my side.

VIA needs to fix the problem with this hardware.

VIA needs to be in a position where they can write drivers
that compensate for flaws in hardware.

VIA needs to publicly admit this is a problem.


Steven

[Stultify]

Originally posted by: steven32272

If you had actually read my post, and knew what a KT400 chipset is, you would know it meets all of your "Special" criteria. Jamming cards into computers indeed.

My "special criteria"
1) These cards need a PCI 2.2 spec slot to go into (do you have one on your as yet unidentified mainboard?)
2) they also need a motherboard that is FULLY APIC/APCI compliant (again try identifying your hardware, a KT400 is not and never will be fully compliant, a KT400a is "kinda" compliant, or as compliant as VIA bothered to make it)
3) an operating system that allows the assignment of more than 16 IRQs. ( USB 2.0 "roots" are "primary" devices, if you bought a 5 port card you have 3 "roots", requiring 3 IRQs and interrupts, not sharing with ANY other device)

Your unidentified Intel chipset antique meets none of the above, a KT400 board also does not quite make it, a KT400a "kinda" makes it, maybe before "spouting off" it would help if you understand a bit about the hardware you choose to "jam" together.

card was placed into a kt400 chipset under XP service pack2.

This means what exactly? You can run XP on your Intel antique, but this card is NEVER going to work with it.

What is the point of buying a usb2.0 pci card for a system that ALREADY has those ports?

I really don't know, I didn't make it, offer it for sale to idiots, or buy it, and VIA did not make it either.

However, with the kt333 and kt400 chipsets along with AMD finally moving into mainstream quality, I was sucked into beliveing VIA had re-invented themselves. My mistake.

Neither was/is any "special treat", or they would still be making them, carry on making mistakes you seem to keep yourself amused full time with them.

Since you didnt read my message in it's entirety

I managed to read it, and believe it, or not, FULLY understand "what is wrong with this picture".

You have a "nice" day now!

[steven32272]

Originally posted by: Stultify
Originally posted by: steven32272

If you had actually read my post, and knew what a KT400 chipset is, you would know it meets all of your "Special" criteria. Jamming cards into computers indeed.

My "special criteria"
1) These cards need a PCI 2.2 spec slot to go into (do you have one on your as yet unidentified mainboard?)

Not unidentified, it is a kt400 as you tend to see sometimes, other times you "forget"
what I am running and you call it "Unidentified"

2) they also need a motherboard that is FULLY APIC/APCI compliant (again try identifying your hardware, a KT400 is not and never will be fully compliant, a KT400a is "kinda" compliant, or as compliant as VIA bothered to make it)

That is not in the documentation that I did not read (So you claim).

3) an operating system that allows the assignment of more than 16 IRQs. ( USB 2.0 "roots" are "primary" devices, if you bought a 5 port card you have 3 "roots", requiring 3 IRQs and interrupts, not sharing with ANY other device)

I clearly stated I was running service pack 2 on XP on at least 4 instances. Sometimes you can read,
othertimes you cannot. Why should I trust your information...I mean seriously, if you
cannot understand my messages what makes you think you can understand documentation.

Your unidentified Intel chipset antique meets none of the above, a KT400 board also does not quite make it, a KT400a "kinda" makes it, maybe before "spouting off" it would help if you understand a bit about the hardware you choose to "jam" together.

A VIA pci card that will not work in a 3 year old VIA mother board meeets MY criteria of "Flawed".

I understand much about the hardware I "Jammed" together, I have been fitting hardware together since
the 386 was all the rage. Your problem is you start first with the assumption that people
are total idiots, then you go from there. Your arrogance amuses me.

card was placed into a kt400 chipset under XP service pack2.

This means what exactly? You can run XP on your Intel antique, but this card is NEVER going to work with it.

I said nothing about antiques. This proves my point that sometimes you can read, and
other times you cannot. the quote clearly states the card was placed into a kt400
chipset under service pack2, yet you respond with something about intel? C.A.N Y.O.U. R.E.A.D.?

What is the point of buying a usb2.0 pci card for a system that ALREADY has those ports?

I really don't know, I didn't make it, offer it for sale to idiots, or buy it, and VIA did not make it either.

Oh VIA made it, and I consider IDIOTS to be people who simply CANNOT read.

[steven32272]

I LOVE this statement, from the VIA docs. (Which I don't read)

VIA has of course worked with Microsoft and other manufacturers to develop driver support for VIA's USB 2.0 controllers but due to licensing issues, the support for the drivers is the responsibility of device manufacturers.


And then:

That means, do not assume that the BIOS or drivers that were supplied with your mainboard, PC or USB device is the latest version.




Rely on your device manufacturer. But then again, don't.

Kills me.

Steven