can't install UAD-1 PCIe (Windows boot freezes)

[cold jazz]

Hi everyone!

A few days ago I received my second UAD-1 card, the UAD-1 PCIe card.

This is what happened when I tried to install it.

The installation seemed to go through as smoothly as it could, with no problems at all. That was until I opened the \"meter program\" for the UAD-1. That program stated that I had no UAD-1 PCIe card installed and that the \"authorization chip\" was busted!

I have two UAD-1 cards (a normal PCI card and the PCIe card which I just received), but non of them worked when I installed the PCIe card! The meter program stated I had NO cards installed at all. But as soon as I removed the PCIe card, the PCI version of the UAD-1 worked again.

When I removed the PCI version and only tried to install the PCIe UAD-1 Windows XP didn’t start at all. Windows got stuck as it was loading up, and frooze at the “progress bar” in Windows.

Does anyone have a clue what the problem might be?
I’ve upgraded the BIOS, checked BIOS settings, moved the PCIe card to another slot...etc etc. But nothing helps.

Thanks in advance!

/CJ

 

[starfighter]

Check my post as well; about the PCIe causing blue screen.

Can you get someone from UA to check this out?

morten

 

[cold jazz]

I did check your post, and I did notice something...
we share the same motherboard, the Intel 925XECV2!

I wonder if that might be the problem.

the weird thing is that I've been out of luck with the MOTU 424 PCIe card as well! The same thing happened with that card as with the UAD-1.

It shouldn't be a faulty PCIe bus/slots though, since I already are using a PCIe graphics card (X16).

 

[starfighter]

I did check your post, and I did notice something...
we share the same motherboard, the Intel 925XECV2!

I wonder if that might be the problem.

the weird thing is that I've been out of luck with the MOTU 424 PCIe card as well! The same thing happened with that card as with the UAD-1.

It shouldn't be a faulty PCIe bus/slots though, since I already are using a PCIe graphics card (X16).

Have you noticed any errors with the SMBus controller? Service pack 1 or 2? I'll try running the driver tonight. Have even updated the BIOS to the latest version... I'll PM you if I find out what's causing the probs..

morten

 

[cold jazz]

I did check your post, and I did notice something...
we share the same motherboard, the Intel 925XECV2!

I wonder if that might be the problem.

the weird thing is that I've been out of luck with the MOTU 424 PCIe card as well! The same thing happened with that card as with the UAD-1.

It shouldn't be a faulty PCIe bus/slots though, since I already are using a PCIe graphics card (X16).

Have you noticed any errors with the SMBus controller? Service pack 1 or 2? I'll try running the driver tonight. Have even updated the BIOS to the latest version... I'll PM you if I find out what's causing the probs..

morten

The SMBus controller?

I'm running Service Pack 2. Latest BIOS.
And also the latest version of the UAD-1 software.

Please keep me updated if you come up with anything - Ill do the same.

 

[NuSkoolTone]

Man, I hope you guys get your problem resolved, and hope it's not the chipset. My motherboard has the 925xe chipset too, and I just ordered a PCI-E that comes tomorrow..


Fingers crossed.....

 

[starfighter]

Man, I hope you guys get your problem resolved, and hope it's not the chipset. My motherboard has the 925xe chipset too, and I just ordered a PCI-E that comes tomorrow..


Fingers crossed.....

No luck last night either. Well actually I didn't get a blue screen, Windows just wouldn't start... Black screen instead; although I'm not sure if that's any better )

NuSkoolTone, please let me know how things work out.

Does anyone know how to get UA on this? hmmm Guess I'll send them another mail

morten

 

[cold jazz]

No luck last night either. Well actually I didn't get a blue screen, Windows just wouldn't start... Black screen instead; although I'm not sure if that's any better )
morten

That's exactly what happened to me!
I kept getting a \"black screen\" right after the Windows \"progress bar\" was showing. And then nothing.

I've tried to remove every PCI card except the PCI RAID card and the graphics card (for obvious reasons), and tried to install the UAD-1 PCIe card in this stripped down system. But that didn't help either.

The only thing that UA has told me is that I might have a faulty card. But that doesn't seem very likely - since the same problem happened with the MOTU 424 PCIe card!

But who know's, with my luck I might have received two faulty PCIe cards! (that would make me lucky though - rather faulty PCIe that faulty MB).

 

[starfighter]

[/quote]But who know's, with my luck I might have received two faulty PCIe cards! (that would make me lucky though - rather faulty PCIe that faulty MB).[/quote]

Hate to break it to you, but I think the MB is the problem. Or rather some software issue with the MB. Experience tells me 99.5% of the problems you have with a computer are software related...

I'm gonna email UA and Intel and see what they tell me. I'll keep posting on this thread

morten

 

[NuSkoolTone]

Damn it. Install fails on mine too!

Installed the software then the card and rebooted. I had to MANUALLY update the other (PCI) cards back and they worked. However as soon as I try to upgrade the driver (4.3) on the PCI-e Card I get a BSOD:


stop:0x0000007E(0xC0000005,0x9CDCBF6F,0xF78DA948,0xF78DA648)

hypkern.sys -Address 9CDCB56F base at 9CDBB000, DateStamp 447cd74c

hypkern.sys is the UA driver.

My Motherbaord is an ASUS p5ad2-e premium, with a 925XE chipset.

It appears there is a problem with the PCI-E cards and the 925XE chipset.


One last thing I notice, and this may be the crux of the problem:
Under system devices, one of the \"PCI Standard PCI-to-PCI bridge\" entries has a yellow exclamation mark next to it, meaning it's not installed correctly. It say \"cannot find enough free resources it can use\" So let me see if I can free some IRQ's and hopefully that's the problem.

Cold Jazz, my SMbus controller wasn't working either until I installed the \"intel chipset inf update program\" I have no idea what this does though, or what the SM bus even is.

Time to write my first letter to UA support.....

 

[cold jazz]

One last thing I notice, and this may be the crux of the problem:
Under system devices, one of the "PCI Standard PCI-to-PCI bridge" entries has a yellow exclamation mark next to it, meaning it's not installed correctly. It say "cannot find enough free resources it can use" So let me see if I can free some IRQ's and hopefully that's the problem.

I'm off to the studio in a minute or two, Ill check if that's the case on my system as well. But I honestly don't think so, should have noticed that earlier.

The reason I stripped down my system from PCI-cards was to clear IRQ's, but that didn't help either. But who knows.

Ill try and see if I got any friends with PCIe-slots in their computers, then at least I can register my UAD-1!

 

[JamesR]

Yo guys,

email UA support - address on the web site - and they WILL get right back to you on this.

I have had the auth. chip fail in my mac and the reset it for me.

just call them in the US or email them, they're very helpful and they know what they're talking about...

J.

 

[NuSkoolTone]

Did some more tweaking, and no success but this is what I've seen happen.

-I brougth the computer back down to stock clock, no change.

-I took the other UAD-1 cards out to free, resources. The PCI-to-PCI bridge thing still says not enough resources error 12.

-Changed between \"Plug n play\" OS yes/no in Bios, no change.

-I DO notice when I put the PCI-e in my machine reported 300 mb less ram on POST, and when I took the two UAD-1 PCI's out, but kept the PCI-E in, the Mobo showed an additional 100mb of ram abouts in POST.

I have 3GB of ram currently. Stock clocked, SPD timed. Corsair all the way, same series even.

Ever since I did the \"Intel inf update\" on the machine, the PCI-E UAD doesn't show up anymore, just a problem with the PCI-to-PCI bridge issue.

I tried System restore, to where now it recognizes an unknown card, and now it's BSOD again with the same error from before.

Running out of options here.....

Going to have to see if UA knows a workaround...

 

[NuSkoolTone]

Talked to UAD-1 support, they're going to look at this thread and send it to the upper engineers and see what's up. Hopefully I'll hear from them tomorrow with some ideas.

James, you were right. Great support!

Let's hope there's a solution.

 

[cold jazz]

Yo guys,email UA support - address on the web site - and they WILL get right back to you on this.I have had the auth. chip fail in my mac and the reset it for me.just call them in the US or email them, they're very helpful and they know what they're talking about...J.

I know, they told me that they would reset my chip.
But that's not the issue here...the issue is that we can't install the UAD-1 PCIe card at all. To reset a chip wont help us (even if I got the message telling me that the auth chip was corrupted).

I have the same problem with the MOTU 424 PCIe card. Couldn't install that either.

 

[NuSkoolTone]

Cold Jazz, that is interesting that you couldn't use another PCIe card either. Was there a conclusion to the problem? It would be hard to imagine the 925X/XE chipset would have useless 1x slots. I mean it was a HUGE chipset release. Consequently my surprise for the UAD-1 incompatibility so far.

I've been doing some looking around, and this all appears to point to the pci-to-pci bridge thing. Since I have seen differences in how windows reacts to the hardware when different drivers are installed. There may be hope it is just a software issue.

The following is a Windox XP hotfix, that might be a possibility. In essence, it looks like it replaces PCI.sys.

http://support.microsoft.com/kb/890463 has the descritpion.

The only problem? Apparantly this is a \"Special\" hotfix that you must CALL micro$uck to get. Meaning it's going to cost you money, just to try it out.

I've tried looking on the net for it, but haven't found a working link. Supposedly it was part of an SP3 for XP that was never released?

Anyway, has anyone tried or have access to hotfix 890463?

It should have the following versions of pci.sys:

Date Time Version Size File name
--------------------------------------------------------------
16-Dec-2004 02:44 5.1.2600.1615 62,976 Pci.sys
16-Dec-2004 18:34 5.1.2600.2585 68,224 Pci.sys

Anyone even have this version of the file to drop in and try?

Any other ideas?

 

[cold jazz]

Cold Jazz, that is interesting that you couldn't use another PCIe card either. Was there a conclusion to the problem?

No conclusions yet. I returned my PCIe 424-card, and a new PCIe card is on its way. I'm not if they are sending a new card back, or if they are sending the same one back...because they tested the card for errors, and I'm still not sure if they found any (but I've mailed them about it).

But I have a hard time beleiving there's something wrong with the card. There's at least three of us on here, sharing the same chip, who has problems installing PCIe cards (not only the UAD-1 in my case).

But I agree...it's very VERY weird if Intel released a chip with dodgy PCIe suport.

 

[NuSkoolTone]

I found this explanation why my memory shows as different sizes at post when a different amount of cards are installed:


\"# Why do I not see all 4GB of memory during POST?

The problem that you are seeing is based on an older architecture design for memory addressing. All the systems architecture up to this point were based on a maximum of 4GB of total memory. Nobody really thought, when this standard was designed, that this amount of memory would actually be in use. The problem that has happened is that you have PCI devices that require memory address ranges so that they can properly execute their commands. These address ranges were mapped in the upper sections of this maximum amount. Since nobody thought you would be using up to 4GB these address ranges started around the last 500MB of the memory ranges. This range is called the T.O.M. or Top of Memory range. This is the point in the bios where it places on hold the amount of memory that is required by the various PCI devices that are found on the motherboard. Thus when you have PCI cards or AGP cards installed on your motherboard these devices hold on to memory for their own use and take away from the maximum amount of memory that is available for other tasks. This amount of memory can vary from a little as 200MB all the way to 1GB of memory (or even more in select cases). It just depends on the PCI devices you have and the amount of PCI (including AGP) that you have installed all at once.

There is really no way to get around this basic design limitation. The only way to get around these type of issues is to use certain new designs that have brand new architectures (i.e 64-bit designs) that allow memory to be mapped in area's above 4GB. The brand new Intel Xeon designs and the AMD Opteron designs are built around 64-bit technology. This is only ½ of the equation that you would need to find success. You would also need to use an OS that is actually PAE or PAE aware so that it is able to address memory above the 4GB level. To find out about PAE you can search Microsofts website for PAE (Physical Address Extensions) and it will explain this concept and what OS's actually are capable of providing this benefit. Windows 2000 and Windows 2003 would fit both of these criteria. Windows XP on the hand would not allow this type of ability.

Microsoft has addressed this type of issue in the following Microsoft Article
(291988)http://support.microsoft.com/default.aspx?scid=kb;en-us;291988\"


Incidentally, messing with the 3GB switch in boot.ini did not change how much RAM Windows saw, so this is being allocated in Hardware I suppose.

I have 3GB installed, and POST/Windows is only recognizing 2560mb of it.
Not the most of my worries right now, but strange.

 

[starfighter]

Hi! Emailed both UA and Intel last night (about 7PM CET) and got a quick reply from Intel with a bunch of stuff. Had to \"escalate\" to another level with Intel support since their suggested solutions didn't work out.

Hopefully they or UA will get back to me in a few days

morten

 

[NuSkoolTone]

Hi! Emailed both UA and Intel last night (about 7PM CET) and got a quick reply from Intel with a bunch of stuff. Had to "escalate" to another level with Intel support since their suggested solutions didn't work out.

Hopefully they or UA will get back to me in a few days

morten

What did Intel suggest?